The British Journal of Social Work

Golightley, Malcom;Holloway, Margaret

doi: 10.1093/bjsw/bcy013pmid: N/A

Of the so-called specialisms in social work which actually cut across all client groups, alcohol and substance misuse is one of particular impact and a problem shared internationally. A recent news item in the UK focused on the high and increasing rate of drug-related deaths and the human stories behind them; parents questioning where they went wrong with their child’s upbringing, helpless in the face of serious addiction, unable to help but unable to turn their back; men and women, living on the streets, vulnerable, wanting a better life each morning while knowing that nothing else will matter later in the day but getting their next fix; scared and lonely young runaways, seeking to obliterate what they have been subjected to in the past and what they have to do in the present to survive. The routes by which each addict encounters social and criminal justice services are various but relentlessly circular and bring with them further human costs. Drug addiction features as a significant factor in mental health breakdown and self-harm, in the repeated removal of children from their birth mother, in theft, sometimes involving assault against older people in their own homes, in the break-up of family units and destroying of close relationships, not to mention the complicated grief caused to those who experience the double blow of loss of the person to drugs in life and again in death (Guy and Holloway, 2007). Social work is so often involved in dealing with the problems caused as a result of addiction and there probably isn’t a social worker around who has not encountered substance misuse somewhere on their caseload. Yet social work research into addiction and specialist practice interventions has a relatively low profile. One approach – motivational interviewing – seeks to tackle the core of the problem of addiction, but as Watson (2011) has pointed out, it is a method with which social work has barely engaged, despite the profession’s commitment to facilitating change. Yet, as one man commented (Channel 4 News 19 February 2018) finding some other reason to get up in the morning, was what finally enabled him to become drug-free after decades of addiction. The first article in this issue, from Australia, analyses the experiences of single fathers experiencing homelessness. Roche, Barker and McArthur explain how the social construction of homelessness leans towards individual pathology and inadequacy (including, though the authors do not specifically highlight this, substance misuse) and they show how the fathers in their study experienced ‘performance failure’ as a result of the interaction with constructions of fatherhood. The authors argue that social work must reconceptualise fathering when homeless to counter the distancing responses of such fathers from their children. Premised on relating to people as service users rather than clients, the next article, also from Australia, uses an ethnographic study of service provision with the Gain model as both theory and practice. The authors (Parsell, Stambe and Baxter) postulate that rather than offer affordable housing to end homelessness the Gain model can be used to create the conditions for people to improve their lives to the extent that they can access housing and other resources without the need for long term social service involvement. They suggest that this approach challenges social workers to think differently about service users and to re-evaluate with them their aspirations. Addressing a related social problem but in a very different context and with women and mothers, Saar-Heiman, Krumer-Nevo and Lavie-Ajayi (from Israel) nevertheless make a very similar core point. In a study of women participating in a poverty intervention programme, they argue that learning from the everyday realities of service users’ lives and conceptualising their practice from the service users’ lived experience is an essential ‘political act’ for social workers. Also from Israel, Savaya, Melamed and Altschuler report on the findings of a study that compared service providers’ and service users’ perceptions of service providers’ burnout. The findings have several implications, not least of which is the importance for social workers to check their service users’ perceptions of the quality and effectiveness of the services they receive, as well as of their treatment by the service provider. Finally in this first group of articles, de Jong, Schout, and Abma report their multiple case study of Family Group Conferences in mental health services in the Netherlands. Their key message resonates with those of the aforegoing articles: social workers should resist taking over as professionals and learn to work effectively with the service user’s existing social network, however broken and inadequate that may seem to be. The next two articles tackle the difficulty of ensuring high standards of care for older people. Funded by the Swedish research council FORTE, the study reported by Hanberger, Nygren and Andersson provides useful insight into the effectiveness of Swedish state supervision of eldercare that had both intended and unintended effects. The authors conclude that although supervision was found to do little to improve the quality of eldercare, it did enable providers to be held accountable for the care that they provide. Providing quality healthcare within the context of dwindling budgets and increasing demand from an aging population provides a challenge for Inter-Agency collaboration for the provision of community based services for older adults. Writing from a Canadian perspective, Tong, Franke, Larcombe and Gould highlight that successful collaborations are those that include relationship-building, sharing of resources and vision and they offer a method which they suggest assists the collaborative process. More broadly posited within care ethics, Julia Shaw challenges the individualist, liberal, ‘spiritual-but-not-religious’ approaches of many earlier writers from the caring professions on contemporary spirituality, arguing that in (what she terms) a ‘post-secular’ context, faith based social action should be seen as an alternative communitarian model of care to the bureaucratic, consumerist systems pursued under neo-liberalism. This article is our Editors’ Choice. The next three articles all examine ways in which social work practice can be improved. The first article is from Sweden (Tafvelin, Isaksson and Westerberg) and examined first year managers and the factors that hinder transformational leadership. The impact of financial restraints, continual change and administrative loads all provide evidence for organisations and management to consider. The next article, also from Sweden, (Pia Tham) was based on a repeat survey of Swedish social workers that enabled the author to examine how working conditions had developed in the eleven years between the surveys. Two general patterns were evident: the first was the deterioration of working conditions with higher work demands, role conflicts and the like and the second was the nature of the job itself. Third, Milena Heinsch, through an empirical study conducted in Australia, looks at factors which may facilitate social workers’ use of research to improve their practice, suggesting that the mode of interaction between researchers and the field is crucial but complex. Whilst arguing for a more engaged, relational approach from researchers, Heinsch’s findings suggest that translational research should not be abandoned and integrated approaches may be called for to address the range of organisational, practical and attitudinal barriers to research-aware practice. Two articles look at developments in the vanguard of social work education. In the UK a contentious government-driven scheme (Frontline) introduced a fast track training scheme for children’s social workers. The aim was to attract ‘outstanding’ graduates to the social work profession based on the premise that they will impact positively on the service and remain committed to it. The authors (Maxwell, Scourfield, de Villiers, Pithouse, and Le Zhang) provide a discussion of findings drawn from an independent evaluation of the scheme’s pilot stage, focusing on the pre-training characteristics of the first cohort. From China, where the social work profession is newly established but growing rapidly, Lei and Huang offer an interesting take on social work competencies. Of note, and worthy of further comment, is that the authors describe a de-politicised activity which they characterize as ‘pragmatic professionalism’. The last two articles are ‘short replies’ to previously published articles and make interesting and informative contributions. Ian Paylor returns us to the opening topic of this Editorial in a response about harm reduction approaches in drug services and Terri Combs-Orme takes up the debate on epigenetics. References Guy P., Holloway M. ( 2007) ‘ Drug-related Deaths and the Special Deaths of Late Modernity’, Sociology , 41( 1), 83– 96. Google Scholar CrossRef Search ADS   Watson J. ( 2011) ‘ Resistance is futile? Exploring the potential of motivational interviewing’, Journal of Social Work Practice , 25( 4), 465– 79. Google Scholar CrossRef Search ADS   © The Author(s) 2018. Published by Oxford University Press on behalf of The British Association of Social Workers. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Roche, Steven;Barker, Justin;McArthur, Morag

doi: 10.1093/bjsw/bcx050pmid: N/A

Abstract This article links sociological perspectives of fatherhood and homelessness to qualitative research with fathers experiencing homelessness. Reporting on the findings from forty in-depth interviews and two focus groups, we explore participants’ conceptions of fathering and detail the material and symbolic barriers to fathering while experiencing homelessness. Fathers describe a ‘performance failure’ that results in ‘distancing’ themselves from their children, acting to conceal their inabilities to fulfil the prescriptive social norms and dominant representations of fatherhood. In response, we apply a critical social work approach to practice that concentrates on highlighting the social processes and dominant meanings of fathering and homelessness, as well as reconceptualising performances of fathering within these contexts. This is a new contribution to the literature on homelessness that expands understandings of homelessness beyond individualistic and causal explanations, as well as supports critical social work approaches to practice. Homelessness, fathering, qualitative methods, social work and sociology, children and families Introduction Welfare services, policy makers and researchers frequently overlook the parental status of people experiencing homelessness, despite its impact on children, families and communities. In response, this article explores the perspectives of fathers experiencing homelessness and situates them within dominant cultural meanings of fatherhood. Drawing on data from forty individual in-depth interviews and two focus groups with homeless fathers in Melbourne and Canberra, Australia, we illustrate how performances of fathering and experiences of homelessness intertwine as a social process, and can provide insights into how fathers negotiate and manage homelessness, and how social workers can respond. To do so, we draw on Erving Goffman’s (1959) dramaturgical account of human interaction to provide a framework to interpret the experiences of fathers experiencing homelessness and their interactions with their children. We argue that viewing fathering as a socially and culturally constructed ‘performance’ can provide additional insights into the complex social relations of homelessness and social work practice. We explore the implications for critical social work practice with this group (Fook, 2016; Pease and Nipperess, 2016) by linking an awareness of the social processes of homelessness with everyday experiences of homeless fathers, and supporting efforts to reconceptualise performances of fathering and homelessness within contexts of practice. This is a new contribution to the social work literature that expands explanations of homelessness and provides insights into this population. We begin by describing the characteristics of the homeless father population in Australia and outline an understanding of fathering as a fluid socio-cultural construct. Next, we explain the methods of this study and present the findings, beginning with a summary of participants’ understandings of fathering and the impact of homelessness on ‘performing’ fathering. We conclude with a discussion centred on the ramifications of these findings for social work practice. Core concepts Homelessness and fatherhood Homelessness is an enduring concept that categorises some of society’s most disadvantaged people via defining living scenarios that are socially and culturally constructed as materially inadequate, centred on the absence of a home. Experienced in various ways and in multiple contexts, it is a pervasive social problem that extends across both the Global North (Toro et al., 2007) and south (Toro, 2007). This study draws on the Australian Bureau of Statistics’ most recent definition of homelessness used in official statistics and governmental publications, as well as in research, service delivery and policy (ABS, 2012, p. 8). It states that experiences of homelessness occur: When a person does not have suitable accommodation alternatives they are considered homeless if their current living arrangement: is in a dwelling that is inadequate; or has no tenure, or if their initial tenure is short and not extendable; or does not allow them to have control of, and access to space for social relations (ABS, 2012, p. 8). The most recently published Australian census estimates that 105,237 people were homeless (ABS, 2011). Amid this group are a large number of parents and an undetermined number of fathers. Adult family members and their children accounted for 83,211 presentations to Australian homelessness services in 2012–13 and, of this group, 61 per cent were single parents with children (AIHW, 2013). The homelessness literature suggests that the number of single fathers who have experienced homelessness is consistently underestimated (Bui and Graham, 2006; Chamberlain and McKenzie, 2008; McArthur et al., 2006). Parental status is frequently disregarded by homelessness service data collection resulting in its removal from dominant understandings of homelessness in service, policy and community responses to homelessness (Barker et al., 2011; Barrow and Laborde, 2008). For example, in the Specialist Homelessness Service data collection conducted by the Australian Institute of Health and Welfare, the number of single fathers or mothers accompanied by children is not recorded (AIHW, 2013). These data do not describe the single or co-parenting status of people utilising homelessness services with children, nor provide details concerning single men who access homeless services who are fathers without accompanying children. In this respect, parental status is frequently unacknowledged despite research clearly articulating the importance of fathers for children’s development and well-being. Children’s attachments to fathers are highly important for social relationships later in life (Mattanah et al., 2011), while Adamsons and Johnson (2013) suggest that disruptions to attachments between children and fathers are minimised when non-resident fathers are able to maintain positive relationships with their children and model positive relationships that work to reduce the likelihood of poor social outcomes. Previous research has found that homeless fathers struggle with the experience (Barker et al., 2011; Schindler and Coley, 2007). Homelessness restricts the ability of fathers to perform nurturing and providing tasks and experiences of uncertainty, insecurity and distress are common (Barker et al., 2011; McArthur et al., 2006). Additionally, the experience of homelessness is oppressive, stigmatising and burdensome (Roschelle and Kaufman, 2004; Zufferey and Kerr, 2004). Qualitative research with people experiencing homelessness reveals a ‘symbolic burden’ in which structural inequality is frequently reduced to the characteristics of an individual (Farrugia, 2011), resulting in an experience interpreted to comprise problematic pathologies, irresponsibility and deviance (Parsell and Parsell, 2012). This affects the homeless individual’s processes of negotiating their social world and managing their self-concept within dominant and discrediting interpretations of homelessness (Roche, 2015). The socially constructed meanings around homelessness, often perceived as passivity and irresponsibility (Farrugia, 2011), and perceptions of economic irrationality (Roche, 2015) have the potential to marginalise men experiencing homelessness, particularly in contemporary neo-liberal contexts. In response, this article aims to provide additional insights into homelessness through an exploration of homeless fathers’ experiences and perspectives, and to reveal how culturally situated performances of fathering intersect with experiences of homelessness. Culturally situated performances of fathering The multiple characteristics of fathering are fluid social constructions influenced by socio-historical context, culture and social background (Lupton and Barclay, 1997; Marsiglio, 1993; White, 1994). Fathering can be understood as performances that negotiate the prescriptions of the socio-cultural and familial contexts that they are set within (Featherstone, 2001, 2009). Erving Goffman’s (1959) dramaturgical account of social life provides a framework to understand fathering as a ‘performance’. Widely held understandings of fathering (including children’s) can be understood by all social actors as socio-cultural scripts—a guide as to how fathers should successfully ‘act’ or ‘perform’ fathering. Goffman (1959) contends that people are social actors and that the world is a stage on which to perform roles, controlling how we are interpreted by others, and trying to set ourselves in the best light possible. The intention is to fulfil socially desirable characteristics and perform a ‘self’ that is interpreted as successful, competent and coherent. To achieve a successful performance, individuals must demonstrate a ‘front’—a collection of convincing props, attitudes and expressions, combining into a set of ‘stereotyped expectations’ (Goffman, 1959, p. 37). Goffman (1959) highlights the importance of these tools to manage impressions and influence the definition of the situation at hand, and project a positive impression that prepare audiences for an individual’s performance of a specific role, such as that of a father (Goffman, 1959, p. 3). Goffman’s theorising has been extended by Beck and Beck-Gernsheim (2002), who posit that social actors are tasked with the responsibility of performing tasks that relate to their identity, as well as the responsibility for failing to perform those associated tasks and, in the case of this research, fathering. Other, more recent scholarship confirms Goffman’s views of social interaction as central to understanding human behaviour (Brickell, 2005; Manning, 2013). Brickell (2005) explains that performances are central to the course of everyday life, highlighting that the ‘self’ is assembled through interactions and reflections on and with the social world (Brickell, 2005). Subsequently, performances and presentations of ‘self’ involve managing a performance in social interactions (Brickell, 2005). Manning (2013) confirms this, arguing that performances are a set of idealised actions, designed to be fully compatible with a culture’s norms and values, and that performances are both realised and idealised to meet and manage the social demands of any situation. Dominant conceptions of fathering The multiple roles of fatherhood are valued by society because they have great influence over the development of children (Lamb, 2010; Marsiglio et al., 2000). They are also understood as a project within one’s life and an opportunity to demonstrate rationality, autonomy and responsibility (Lupton and Barclay, 1997). Accordingly, parenting and relationships with children are integral to a strong sense of ‘self’ and indicate social competence and success. Scholars have identified the multiple and evolving characteristics of fathering and organised them into the frameworks of ‘traditional’ and ‘new’. ‘Traditional’ conceptions of fathering centre on behaviours of ‘breadwinning’ and ‘providing’ which combine with elements of power, authority and status in a gender differentiated home (Brandth and Kvande, 1998; Finn and Henwood, 2009). Achievement in the public sphere, being a good provider, paternal authority in the home and virility have been associated with masculine honour within a family unit (Brandth and Kvande, 1998; Gregory and Milner, 2011). Throughout the twentieth century, economic provision was viewed as the central role of fathers in families (Featherstone, 2009). However, the traditional breadwinning model of fathering no longer fulfils contemporary expectations of fathers in many societies (Brandth and Kvande, 1998; Hamilton and De Jonge, 2010). Featherstone (2009) details that the contemporary role of men as economic providers has evolved, along with wider economic changes to employment such as mothers entering the paid labour force. Research has documented the evolution from traditional behaviours to more involved care-giving and emotionally present fathering (Humberd et al., 2015). Changes in labour force demographics, ideologies and dominant cultural constructions of fatherhood have seen the emergence of a ‘new fatherhood’ characterised by changes in the roles of fathers across dimensions of attitudes, parenting roles, economic contributions, legalities, and rights of custody and care (Flood, 2010; Gregory and Milner, 2011; Hamilton and De Jonge, 2010; Ranson, 2001). This diversity of roles has led to increased engagement between contemporary fathers and their children (Flood, 2010). ‘New fatherhood’ finds both parents responsible for care and finances, with roles shifting from perceptions of automatic rights to having both rights and duties, with more egalitarian sharing and caring responsibilities (Gregory and Milner, 2011; Hearn, 2004). This has translated into fathers spending more time with children, having greater responsibility for children’s welfare and an increase in emotional engagement in children’s lives (Finn and Henwood, 2009; Ranson, 2001). The contemporary hybridity of fathering Researchers have found that socio-cultural interpretations of fathering in Western contexts are more dynamic than the categories of ‘new’ and ‘traditional’. Fatherhood has never been a fixed institution, experience or taken on a unitary character and reflects a range of meanings (Brandth and Kvande, 1998; Featherstone, 2009; Humberd et al., 2015; Thompson et al., 2013). As Doucet (2006) argues, fathering occurs in larger sets of social relations within which fathering practices are perceived and judged. For successive generations, there have been increases in child-care activities of fathers (Pleck and Mascidrelli, 2004). In their study, Brandth and Kvande (1998) found that ‘providing’ was not emphasised by fathers, identifying that the most important aspect of a good father was one who was caring (Brandth and Kvande, 1998). Doucet (2006) describes how fathers highlight their play with children, as well as their active approach to care-giving. Similarly, an interplay between ‘new fathering’ and ‘traditional’ fathering roles has been found to exist among expectant fathers who imagine themselves as ‘rational, responsible, strong and morally-wise fathers’ (Finn and Henwood, 2009, p. 559). Various research on fathering reveals a multiplicity of images of fathering. A recent study found that young men view employment as dominant in their imaginations of their future lives, particularly as a perceived prerequisite to family-related goals (Thompson et al., 2013). In the same study, participants emphasise the importance of active involvement with family, yet view a successful father as one who supports their family financially, in a traditional, ‘breadwinning’ sense. Dermott (2008) argues that, while fathers remain highly engaged in the labour force, this does not necessarily mean that fathers do not provide emotional and nurturing roles. Further, research by Barclay and Lupton (1999) found that fathers frequently accept nurturing and caring roles, which include fulfilling a purpose in a child’s life or being a teacher, and combine these with providing. Fatherhood is a privileged position of power and influence across time, space and context, and a culturally celebrated social centre of power and status (Lupton and Barclay, 1997; Featherstone, 2009). Doucet’s (2006) research suggests that professional status and class influence fathering behaviours. For example, in the UK, fathers are highly economically active and fathers are more likely than non-fathers to be working long hours (Featherstone, 2009). Enduring features of fathering, such as providing for family, are positioned within traditional conceptions of men’s family roles as material providers. Elements of fathering, including those of a ‘new father’, are set within these socially privileged activities of fathering. These studies show that socio-cultural scripts of fatherhood are made up of a ‘hybrid’ collection of fathering activities that include ‘providing’, which directs status and power outside of the family, as well as ‘purpose’, such as teaching, care and being available. Fathering is made up of myriad roles, supported by dominant, socially produced correlates of meanings, performed and adapted to multiple contexts. Our study suggests that fathers who are experiencing homelessness contend with the dramaturgical task of convincing others that they are fathers as prescribed by the dominant socio-cultural of fatherhood that pervade contemporary society. This article applies these concepts to the participants in a study that explored the experiences of fathers who are homeless, but first we present the details of the study. The study This research was originally completed for a project undertaken in partnership between the Institute of Child Protection Studies, Hanover Welfare Services and Melbourne City Mission (Barker et al., 2011). It was funded by the Australian government through the National Homelessness research agenda and was carried out during 2011–12. Although the research project had a broader range of interests including how services respond to homeless fathers, this paper focuses on the experiences of single fathers who are or have been homeless. To best explore the meaning fathers attach to their experiences of homelessness, this qualitative study is informed by the interpretive tradition of sociology that focuses on the meanings constructed through interactions in social life and explores the subjective states of individuals (Crotty, 1998). In-depth interviews and focus groups were conducted to explore experiences of fathering and homelessness. Two focus groups (four participants in Melbourne and seven in Canberra) were completed with homeless single fathers recruited through homelessness agencies. The focus groups identified the issues experienced by single fathers who were homeless and discussions assisted in the development of the interview schedule. The in-depth interviews explored fathers’ experiences with fathering, contact and relationships with their children, as well as the role of services. Forty interviews were conducted with fathers who were homeless (thirty in Melbourne, ten in Canberra). The interviews took between sixty and ninety minutes and, with participants’ permission, were recorded and fully transcribed. Ethics The project was designed to safeguard the rights of all who were involved and was conducted with the approval of the Australian Catholic University’s Human Research Ethics Committee (HREC). Pseudonyms are used in place of participants’ names. Data analysis The data were analysed using the qualitative data analysis program NVivo. Impression sheets written by researchers directly after interviews captured the researchers’ first impressions and other notable dimensions of the interview that may not necessarily be in the transcript. Categorical and emergent themes from the data were identified by researchers who then developed codes (Bryman, 2008). Transcripts were double-coded to ensure the meaning of codes aligned between researchers. Some of the codes used for the analysis of the data included ‘pathways into homelessness’, ‘contact with children’, ‘father identity’ and ‘homelessness and parenting’. There was a reanalysis of the data for this article that focused more specifically on dimensions of parenting and participants’ interpretations of fathering. Sample and recruitment The research used purposive sampling techniques to recruit participants through homelessness services in Canberra and Melbourne. The criteria for participant selection were the following: a father who had been homeless in the past twelve months and aged eighteen years or over, as well as fathers with or without active involvement with their children. To compensate for their time, participants were offered a $25 gift voucher. Profile of participants The age of participants ranged from twenty to forty-six years, with an average age of thirty-five years. Of the forty interview participants, four identified as Aboriginal and a further two said that their children identified as Aboriginal. Three additional participants identified as culturally and linguistically diverse. Most participants reported lengthy periods of homelessness that were either a continuous experience or scenarios where they cycled in and out of homeless. The length of time participants were homeless varied from one month to thirty-eight years. The contact participants had with their children varied. Around half of the fathers had regular face-to-face contact, while a small number of fathers were primary carers for their children (N = 5). Over one-third of participants had little or no contact with their children. Almost three-quarters of the sample were fathers to more than one child. Participants lived in various living circumstances at the time of interview. Eighteen (45 per cent) participants were living in crisis accommodation, (e.g. refuges, hostels, emergency accommodation) or were ‘couch surfing’. Eleven (27.5 per cent) participants were living in transitional accommodation or in a boarding house, while three participants were sleeping rough. At the time of interview, eight (20 per cent) participants had recently moved from being homeless to public housing and one participant had moved to private rental accommodation. Findings Homeless fathers and hybrid fathering Participants held detailed ideas of fatherhood and articulated comprehensive perspectives on their own roles as fathers. The presence of ‘providing’ and ‘guiding’ in participants’ accounts of fathering reflect a strong engagement with contemporary hybrid views of fathering found in previous research (Brandth and Kvande, 1998; Greaves et al., 2010; Lupton and Barclay, 1997). Most associated themselves with a ‘hybrid’ understanding of fathering, viewing the role of fathers as active carers and contributors to family life, defining fathering as the strong responsibilities that fathers have to provide for their children. Exemplifying this was Jack, a forty-six-year-old father of three children who had experienced homelessness intermittently since a teenager, who reflected on the importance of incorporating both ‘traditional’ and ‘new’ fathering roles: Children have different needs, they have those emotional needs, and they have the physical needs, so physically I need to provide for them, ideally decent shelter, clothing, education, finance, and emotionally I need to be able to show them how to love unconditionally, and that’s pretty well it (Jack). Many fathers spoke about the importance of having a purposeful role in their children’s lives despite, for a range of reasons, frequent absence. Fathers considered socialising and modelling behaviours a crucial component of performing fathering roles. For instance, Rohan, a twenty-six-year-old with three sons, described an acute awareness of this role with his children: … you don’t even know you’re teaching them, they watch you. Things you do, they learn from you, without you even trying to teach them, they watch you because you’re their role model, to a degree. You’re not a superstar or anything on TV, you are still dad, but you’re their dad (Rohan). Another participant, a couch-surfing twenty-six-year-old father of two, who had sporadic contact with his children, recognised the importance of having a purposeful relationship with his children. He viewed his role as providing a relationship that is consistent, guiding and ‘sculpting’: I mean it’s a huge responsibility. You’ve got to mould an individual into being a productive member in society … . It’s like a sculptor with a sculpture, you’ve got to create a masterpiece, you can’t just go bang bang and it’s done. It’s a slow and steady process (Dominic). Participants demonstrated thoughtful engagement with parenting activities that reflect dominant meanings and frameworks of contemporary parenting. This article now explores the interaction between participants’ accounts of fathering performances and their experiences of homelessness. Fathering while homeless Fathering while experiencing homelessness was difficult for most participants. Their accounts provided details of a range of barriers to performing ‘hybrid’ fathering roles. They discussed both ‘material’ and ‘symbolic’ obstacles to performing fathering. ‘Material’ barriers involve the conditions of homelessness that make it problematic to provide material support to children; ‘symbolic’ barriers involved the undesirable symbolic arrangements of homelessness that hold negative implications for self-concept and can produce a conscious withdrawal from fathering. Participants described how the material conditions of homelessness disrupt their ability to provide their children with a home, financial security and food. Reflecting on a period of homelessness with his four children, forty-three-year-old Angus felt anguish at not being able to provide material comforts for his children, made more acute when combined with caring responsibilities: I felt so bad because I couldn’t give the kids what they wanted. I couldn’t give them their home as much as I tried and tried, there was just nothing I could do, there was nowhere to go to. There was nothing we could do. Now they expect me … to look after them and give them somewhere to live and food and that and I couldn’t do that as much as I tried (Angus). Many participants despaired at their inability to perform ‘providing’ and ‘guiding’ roles, and articulated concerns about how their homelessness was interpreted by others, including by their children. They believed that their children were aware of the stigmatising meanings of homelessness, and were concerned that they were unable to provide a convincing performance of fathering to their children. Participants absorb the dominant social constructs and negative meanings of homelessness, which impact on their self-concept and their relationship with their children. Fathers in the study believed their children’s interpretations of their homelessness equalled the dominant negative meanings of homelessness, and were aware of their inability to perform key father roles of providing, being present and guiding. A number of participants reflected on how they thought their children interpret their homelessness. For illustration, Jesse, a father of five and self-described drifter, wanted his children to view him as a provider—someone who is ‘constructive’ and a participant in the labour market. He believed his children felt ‘uncomfortable’ because they were not adequately provided for: Q: How would you describe your role as father? A: Not to where I want to be at the moment. Q: Where would you like to be? A: Just to see, if they’re having a look at their dad doing things, being constructive, doing this, that and they know I provide so they can go everywhere they want to go. Just fit in so they don’t have to feel uncomfortable because they can’t go because I ain’t got nothin’ if you know what I mean, just so that they’re comfortable in life. Q: So if they ask for something, you’re able to give it to them? A: Yeah that’s right, provide for them. No that’s the way I look at it, that’s what I want it to be (Jesse). Performance failure Emerging from the interview data was a discrepancy in the dominant socio-cultural scripts of fathering and the fathering performances of participants. Navigating the discrepancies between idealised notions of fathering and the difficulties of conditions of homelessness reveal the insufficient assemblage of characteristics to demonstrate a ‘performance’ that is typical of fathering. We suggest this discrepancy between their lives and the idealised and culturally produced understandings of fatherhood is a performance failure. Participants assessed their performance as fathers in relation to their position in the labour market, because providing and having a strong labour force connection demonstrates a well-recognised performance of fathering. According to participants, connections to the labour market is an ideal way to fulfil fathering roles. The experience of homelessness involved limited access to the labour market for participants, with many of the belief that future connections to the labour market, in particular through business or enterprise, would assist to fulfil fathering roles and improve their status in the eyes of their children. A twenty-six-year-old father with infrequent contact with his three sons believed that salvaging his performance as a father could occur through ownership of a business: ‘I want to show them what they [children] can do, what they can aspire [to]; their dad’s not fucking hopeless, they can be proud of him … once I’ve got my own business and everything’ (Rohan). Similarly, Jay, a twenty-year-old father of a one-year-old son, in and out of homelessness since his mid-teens, viewed owning a business as a way his son would be proud of him in the future, illustrating how economic success intersects homelessness and fathering: ‘I’ll get my business going and my kid will be able to be proud of me’ (Jay). Sustaining relations to a range of fathering performances was difficult for participants and the social ramifications of this were acute. Participants associated emotionally and economically absent fathering with their social status. For example, Jeff, a thirty-seven-year-old father of four, found that the social interpretations of his status as a homeless father meant that he felt inferior to other parents: I felt like I was down here, everyone else was up there basically—and I couldn’t—you’d speak to other parents or whatever, or whatever, and where do you live? You don’t want to tell them and stuff like that (Jeff). Many participants renegotiated their roles as fathers as they came to terms with homelessness and confronted difficulties providing and being present for their children. Chris, homeless for the last four years, felt excluded from his son’s life and unable to perform a fathering role due to his ex-partner’s repartnering: I just don’t feel part of his [son’s] life, I feel like I’m outside the dome … I don’t know how you could put it, outside the dome or something, knocking trying to get in sometimes. My ex-partner, she re-partnered last year, and that was even worse for me. I found that was probably the worst four or five months of my life because there was another man looking after my kid; that was horrible, that was probably the most depressed I’ve ever been (Chris). For Chris, not ‘feeling’ a part of his child’s life had ramifications for his own happiness and sense of purpose. Similarly, for James, a father of two who had been homeless for nearly nine years, his marginalisation from performing fathering were made acute by the financial success of his previous wife’s new partner: ‘Christmas day I met him and he pulled up in his brand new ute [small truck]. It made me feel small’ (James). James’s marginalised position as a carer and provider for his two children is confirmed by the man now fulfilling these roles, symbolised by his new car that demonstrates material wealth and an economic competence crucial to performing the ‘providing’ component of fathering. A distancing effect Participants were aware of their failure to perform prevailing socio-cultural scripts of fathering and acted to distance themselves from their children. Homelessness created uncertainty around performances as a parent and their capacity to be a part of their children’s lives, viewing homelessness as a symbol of a subordinate position in relation to the valued roles of fathering. Some participants articulated an unwillingness to allow their children access to their lives where their inability to perform fatherly roles was observable. Accordingly, these fathers withdrew from fulfilling the aspects of fathering throughout their experience of homelessness. Concerned about his children’s adverse interpretations of his living arrangements, a participant residing at a homelessness service remarked that: ‘I’m not going to let my child, my son, come here and see me in a place like this’ (Lynden). Similarly, Neville, living in a car and on the streets for several years with limited contact with his five children, centred his reluctance to be involved with his children on the belief that they would be upset about the material conditions of his life and hesitant to see him in the future: I mean I can’t have my kids anyway, I am living on the street you know and they are not used to my type of lifestyle. They would be probably shocked and want to have nothing to do with me again (Neville). For Sam, the ramifications of his identity as a homeless father with little to offer materially meant that contacting his twelve-year-old son was not possible until these circumstances changed: Q: What does it mean to you to be a dad when you have no home? A: Useless, it makes you feel useless … it’s the worst feeling of all mate because you’ve got nothing to represent you. That’s why I’m not contacting him [my son] until I have something to offer him (Sam). The experience of homelessness operates to distance fathers from their children in a symbolic manner. When participants believed their children were aware of the negative imputations of homelessness, fathers withdrew from their children’s scrutiny, despite a strong desire to perform fathering roles: ‘Yeah. And especially if there’s a child involved, you don’t want to look bad in their eyes, you don’t want to look bad at all. And I think that it stops things from happening sometimes’ (Max). The dominant meaning around homelessness also penetrated into other areas such as ethnicity and class. A reflective father of four was intensely aware of the perceptions, feelings and attitudes of others towards him and expressed his torment as a homeless father removed from various culturally celebrated positions of social power: I felt like a piece of shit to be honest having to drag my kids into this place and that place looking for accommodation. You feel like sixth class citizens—not second-class citizens and especially being black, you’re just basically a fucking loser—that’s excuses but that’s what it just feels like. I’m letting the kids down; they’re looking at me (Bernie). Bernie was conscious of his children’s interpretation of him occupying a socially marginalised position, viewing these as also embedded in the thoughts, ideas and impressions of his children. The ramification of performance failure and distancing behaviours are now discussed in relation to social work practice. Discussion and implications for practice Our findings outline the complex social processes for fathers during experiences of homelessness, frequently encountered by social workers in multiple contexts of practice. We suggest that critical approaches to social work provide a valuable lens that can support practice through recognising the social processes of fathering and homelessness, while assisting to reconceptualise both socio-cultural and practical roles of fathering within constraints of homelessness. Utilising a critical social work approach Critical social work aims to reveal the structural and social processes that affect individuals (Fook, 2016; Pease and Nipperess, 2016), emphasising structural inequities, over individual problems and deficiencies (Mendes, 2009) and allowing a more holistic and contextualised approach to practice (Briskman et al., 2009; Fook, 2016). Importantly, critical social work practice is concerned with how power, powerlessness and the impact of structural domination is expressed and experienced in the ‘richness of everyday relations’ (Healy, 2000, p. 45). This approach questions the oppressive aspects of systems and structures, and explores how policies, practices, conformist social values and frameworks of meanings, such as those around homelessness or fathering, are influential in shaping discourses, attitudes and behaviours, looking beyond remedial solutions to symptoms, and focusing on broader structural issues and their implications for clients (Fook, 2016). Approaches to practice Fook (2016) summarises core approaches to critical social work practice, focusing on the construction of social realities, both externally and internally, and on the connections between structural domination and personal self-limitation, while viewing social context as fundamental to informing practice. Resistance and change for clients involve challenging power relations and structures through practice that consists of dialogue, interaction, self-reflection and analysis, with knowledge obtained empirically (such as the research presented here), and an exploration of how power relations are expressed and used in particular situations (Fook, 2016). Practices of self-reflection can challenge dominant structures and social relations in everyday life (Fook, 2016), while individual consciousness raising can connect private troubles with structural sources (Mendes, 2009) and, in doing so, resist and question dominant constructions of meaning that impact negatively (Fook, 2016). Implications for practice with fathers experiencing homelessness Incorporating an appreciation of the complex social relations and processes of homelessness can assist social workers in their work with clients. Critical social work practice suggests inquiry into the socio-cultural construction of clients’ situation and their lives (Pease and Nipperess, 2016). The interviews and focus groups reveal that homeless fathers situate their understandings of fathering in the dominant views of contemporary fatherhood. Thus, recognising fathering as a collection of dominant socio-cultural meanings provides insight into the absence of homeless fathers from overarching frameworks of fathering roles. This has ramifications for self-perception among this cohort, who internalise their fathering performance failures and act to suspend their parenting roles. We suggest that social work interventions could assist fathers to renegotiate their understandings of parenting roles and relationships with their children, and explore the ways in which homelessness has influenced these perspectives by incorporating understandings of the structural inequalities that bring about personal difficulties (Pease and Nipperess, 2016). Dialogue, reflective practice and shared analysis can reveal the multiple expressions of power in social relations that sit at the intersection of homelessness and fathering embedded in daily life and offer insights into ‘distancing’ practices and inabilities to fulfil dominant socio-cultural norms of fathering (Featherstone, 2009). Critical social work practice would also involve a critical reconceptualisation of homelessness and fatherhood by assisting homeless fathers to deconstruct their social performances and roles (Fook, 2016) and reconceptualise performances of fathering as separate from experiences of homelessness. Through a process of reconceptualising dominant constructions of homelessness and fathering, and reflecting on how performance failure is created and sustained in everyday contexts, clients can, as encouraged by Fook (2016) and Barak (2016), engage and reflect on presenting issues, such as performance failure, and reconceptualise acts of power within fathering practices and contexts of homelessness. By reconceptualising and expanding ideas of fathering as broader than ‘providing’ or ‘caring’, symbolic barriers to performing fathering can be replaced with other positive roles that can be played in the lives of their children that can reduce the ‘distancing effect’ between fathers and their children, and improve father–child relationships. In addition, fathering might be a way to re-engage with an identity and a positive view of ‘self’ which can provide motivation for change. Highlighting the possibilities for fathering within the constraints of homelessness can contribute to fathers maintaining or restoring their relationships and connections with their children. Parenting roles, such as providing care or support, or contributing to the development of children, may maintain relationships throughout experiences of homelessness and connect fathers with important fathering activities. Conclusion This article presents analysis of homelessness that moves beyond individualistic and causal accounts, instead considering how prevailing socio-cultural meanings of fathering intersect with homelessness and impact on performances of parenting roles. A strong sense of hybrid fathering emerged from participants’ narratives that incorporated an array of contemporary interpretations of fathering that include elements of ‘providing’ and ‘guiding’. These interpretations fit within dominant discourses of both ‘traditional’ and ‘new’ fathering. However, many of the fathers in this study recognise that their performances of fathering do not match the demands of the socially produced performances of fathering. Participants articulated an unwillingness to allow their children access to their lives because of their inability to fulfil the prescriptive social norms and symbolic representations of fathering. They find themselves distant from the dominant socio-cultural understandings of fathering and, as a result, navigate symbolic and material barriers while experiencing ‘performance failure’ and working to distance themselves from parenting roles. We suggest that a critical social work practice response would focus on broadening the dominant socio-cultural understandings of fathering, emphasise reducing the barriers to performing fathering roles within the constraints of homelessness, while incorporating understandings of the social processes of homelessness for fathers in social work practice. Acknowledgements We are grateful to the participants who shared their experiences. Thank you also to the reviewers whose recommendations significantly improved this article. References ABS ( 2011) Census of Population and Housing: Estimating Homelessness , Canberra, Australian Bureau of Statistics. ABS ( 2012) Information Paper: A Statistical Definition of Homelessness , Canberra, Australian Bureau of Statistics. Adamsons K., Johnson S. K. ( 2013) ‘ An updated and expanded meta-analysis of nonresident fathering and child well-being’, Journal of Family Psychology , 27( 4), p. 589. Google Scholar CrossRef Search ADS PubMed  AIHW ( 2013) Specialist Homeless Services: 2012–2013 , Canberra, Australian Institute of Health and Welfare. Barclay L., Lupton D. ( 1999) ‘ The experiences of new fatherhood: A socio‐cultural analysis’, Journal of Advanced Nursing , 29( 4), pp. 1013– 20. Google Scholar CrossRef Search ADS PubMed  Barak A. ( 2016) ‘ Critical consciousness in critical social work: Learning from the theatre of the oppressed’, British Journal of Social Work , 46( 6), 1776– 92. Google Scholar CrossRef Search ADS   Barker J., Violet K., Mallett S., McArthur M., Saunders V. ( 2011) More Than Just Me: Supporting Fathers Who Are Homeless , Canberra, Institute of Child Protection Studies, Australian Catholic University. Barrow S. M., Laborde N. D. ( 2008) ‘ Invisible mothers: Parenting by homeless women separated from their children’, Gender Issues , 25( 3), pp. 157– 72. Google Scholar CrossRef Search ADS   Beck U., Beck-Gernsheim E. ( 2002) Individualization , London, Sage. Brandth B., Kvande E. ( 1998) ‘ Masculinity and child care: The reconstruction of fathering’, The Sociological Review , 46( 2), pp. 293– 313. Google Scholar CrossRef Search ADS   Brickell C. ( 2005) ‘ Masculinities, performativity, and subversion: A sociological reappraisal’, Men and Masculinities , 8( 1), pp. 24– 43. Google Scholar CrossRef Search ADS   Briskman L., Pease B., Allan J. ( 2009) ‘1—Introducing critical theories for social work in a neo-liberal context’, in Briskman L, Pease B., Allan J. (eds), Critical Social Work: Theories and Practices for a Socially Just World , Crows Nest, NSW, Allen and Unwin. Bryman A. ( 2008) Social research methods , Oxford, Oxford University Press. Bui B. H., Graham A. ( 2006) Support Issues for Homeless Single Fathers and Their Children , Victoria, Victoria University. Chamberlain C., McKenzie D. ( 2008) Australian Census Analytic Program: Counting the Homeless Australia , Canberra, Australian Bureau of Statistics. Crotty M. ( 1998) The Foundations of Social Research: Meaning and Perspective in the Research Process , St Leonards, NSW, Allen & Unwin. Dermott E. ( 2008) Intimate Fatherhood , London, Routledge. Doucet A. ( 2006) ‘ “Estrogen‐filled worlds”: Fathers as primary caregivers and embodiment’, The Sociological Review , 54( 4), pp. 696– 716. Google Scholar CrossRef Search ADS   Farrugia D. ( 2011) ‘ The symbolic burden of homelessness: Towards a theory of youth homelessness as embodied subjectivity’, Journal of Sociology , 47( 1), pp. 71– 87. Google Scholar CrossRef Search ADS   Featherstone B. ( 2001) ‘ Putting fathers on the child welfare agenda’, Child and Family Social Work , 6( 2), p. 179. Google Scholar CrossRef Search ADS   Featherstone B. ( 2009) Contemporary Fathering: Theory, Policy and Practice , Bristol, The Policy Press. Finn M., Henwood K. ( 2009) ‘ Exploring masculinities within men’s identificatory imaginings of first‐time fatherhood’, British Journal of Social Psychology , 48( 3), pp. 547– 62. Google Scholar CrossRef Search ADS PubMed  Flood M. ( 2010) ‘ “Fathers’ rights” and the defense of paternal authority in Australia’, Violence against Women , 16( 3), pp. 328– 47. Google Scholar CrossRef Search ADS PubMed  Fook J. ( 2016) Social Work: A Critical Approach to Practice , London, Sage. Goffman E. ( 1959) The Presentation of Self in Everyday Life , London, Penguin. Greaves L., Oliffe J. L., Ponic P., Kelly M. T., Bottorff J. L. ( 2010) ‘ Unclean fathers, responsible men: Smoking, stigma and fatherhood’, Health Sociology Review , 19( 4), pp. 522– 33. Google Scholar CrossRef Search ADS   Gregory A., Milner S. ( 2011) ‘ What is “new” about fatherhood? The social construction of fatherhood in France and the UK’, Men and Masculinities , 14( 5), pp. 588– 606. Google Scholar CrossRef Search ADS   Hamilton A., De Jonge D. ( 2010) ‘ The impact of becoming a father on other roles: An ethnographic study’, Journal of Occupational Science , 17( 1), pp. 40– 6. Google Scholar CrossRef Search ADS   Healy K. ( 2000) Social Work Practices , London, Sage. Hearn J. ( 2004) ‘ From hegemonic masculinity to the hegemony of men’, Feminist Theory , 5( 1), pp. 49– 72. Google Scholar CrossRef Search ADS   Humberd B., Ladge J. J., Harrington B. ( 2015) ‘ The “new” dad: Navigating fathering identity within organizational contexts’, Journal of Business and Psychology , 30( 2), pp. 249– 66. Google Scholar CrossRef Search ADS   Lamb M. E. ( 2010) ‘Chapter 1—How do fathers influence children’s development? Let me count the ways’, in Lamb M. E. (ed.), The Role of the Father in Child Development , 5th edn, Hoboken, NJ, John, Wiley and Sons inc. Lupton D., Barclay L. ( 1997) Constructing Fatherhood: Discourses and Experiences , London, Sage. Manning P. ( 2013) Erving Goffman and Modern Sociology , Cambridge, Wiley. Marsiglio W. ( 1993) ‘ Contemporary scholarship on fatherhood culture, identity, and conduct’, Journal of Family Issues , 14( 4), pp. 484– 509. Google Scholar CrossRef Search ADS   Marsiglio W., Amato P., Day R. D., Lamb M. E. ( 2000) ‘ Scholarship on fatherhood in the 1990s and beyond’, Journal of Marriage and Family , 62( 4), pp. 1173– 91. Google Scholar CrossRef Search ADS   Mattanah J. F., Lopez F. G., Govern J. M. ( 2011) ‘ The contributions of parental attachment bonds to college student development and adjustment: A meta-analytic review’, Journal of Counseling Psychology , 58( 4), p. 565. Google Scholar CrossRef Search ADS PubMed  McArthur M., Zubrzycki J., Rochester A., Thomson L. ( 2006) ‘ “Dad, where are we going to live now?”: Exploring fathers’ experiences of homelessness’, Australian Social Work , 59( 3), pp. 288– 300. Google Scholar CrossRef Search ADS   Mendes P. ( 2009) ‘2—Tracing the origins of critical social work practice’, in Allan J., Briskman L., Pease B. (eds), Critical Social Work: Theories and Practices for a Socially Just World , Crows Nest, Allen and Unwin. Parsell C., Parsell M. ( 2012) ‘ Homelessness as a choice’, Housing, Theory and Society , 29( 4), pp. 420– 34. Google Scholar CrossRef Search ADS   Pease B., Nipperess S. ( 2016) ‘1—Doing critical social work in the neoliberal context: Working on the contradictions’, in Pease B., Goldingay S., Hosken N., Nipperess S. (eds), Doing Critical Social Work: Transformative Practices for Social Justice , Crows Nest, Allen and Unwin. Pleck, J. H. and Mascidrelli, B. P. (2004) ‘Paternal involvement by U.S. residential fathers: Levels, Sources, and Consequences’, in Lamb M. E. (ed.), The role of the father in child development, 4th edn, pp. 222–71, Hoboken, NJ, US, John Wiley & Sons. Ranson G. ( 2001) ‘ Men at work change—or no change?—in the era of the “new father”’, Men and Masculinities , 4( 1), pp. 3– 26. Google Scholar CrossRef Search ADS   Roche S. ( 2015) ‘ The salvaging of identities among homeless men: Reflections for social work’, Australian Social Work , 68( 2), pp. 228– 43. Google Scholar CrossRef Search ADS   Roschelle A. R., Kaufman P. ( 2004) ‘ Fitting in and fighting back: Stigma management strategies among homeless kids’, Symbolic Interaction , 27( 1), pp. 23– 46. Google Scholar CrossRef Search ADS   Schindler H. S., Coley R. L. ( 2007) ‘ A qualitative study of homeless fathers: Exploring parenting and gender role transitions’, Family Relations , 56( 1), pp. 40– 51. Google Scholar CrossRef Search ADS   Thompson R., Lee C., Adams J. ( 2013) ‘ Imagining fatherhood: Young Australian men’s perspectives on fathering’, International Journal of Men’s Health , 12( 2), pp. 150– 65. Google Scholar CrossRef Search ADS   Toro P. A. ( 2007) ‘ Toward an international understanding of homelessness’, Journal of Social Issues , 63( 3), pp. 461– 81. Google Scholar CrossRef Search ADS   Toro P. A., Tompsett C. J., Lombardo S., Philippot P., Nachtergael H., Galand B., Schlienz N., Stammel N., Yabar Y., Blume M., MacKay L., Harvey K. ( 2007) ‘ Homelessness in Europe and the United States: A comparison of prevalence and public opinion’, Journal of Social Issues , 63, pp. 1540– 60. White N. R. ( 1994) ‘ About fathers: Masculinity and the social construction of fatherhood’, Journal of Sociology , 30( 2), pp. 119– 31. Google Scholar CrossRef Search ADS   Zufferey C., Kerr L. ( 2004) ‘ Identity and everyday experiences of homelessness: Some implications for social work’, Australian Social Work , 57( 4), pp. 343– 53. Google Scholar CrossRef Search ADS   © The Author 2017. Published by Oxford University Press on behalf of The British Association of Social Workers. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Parsell, Cameron;Stambe, Rose;Baxter, Janeen

doi: 10.1093/bjsw/bcx045pmid: N/A

Abstract Social work rarely considers how the delivery of social services shape the behaviours of, and potentially creates dependencies for, people receiving services. Drawing on an ethnographic examination, this article presents the Gain model as a theory and a practice. Delivered in temporary homeless accommodation, the Gain model rejects wraparound service provision. Premised on service users as people rather than clients, it challenges the idea that more social services are required to assist people experiencing material deprivation. The article outlines two contributions to social work. First, it challenges social work to develop a clear narrative about people with the desire and potential to live optimistically aspirational lives. Second, the article puts forward a view of an autonomous good life that is at odds with the conditions and even resources associated with accessing social services. Rather than practising to disrupt oppression, the Gain model pushes social work to think about how providing services to excluded groups is premised on limited life expectations. The model’s value lies in the significance of a reflective practice premised on people using social work services as no different to the social workers providing the services. Autonomy, dependence, homelessness, optimistic lives, social services Introduction In the 1930s, Orwell asked the questions that this article grapples with today. After living in poverty in Paris and London, he pondered: The problem is how to turn the tramp from a bored, half-alive vagrant into a self-respecting human being. A mere increase in comfort cannot do this. Even if the casual wards became positively luxurious, a tramp’s life would still be wasted. He would still be a pauper, cut off from marriage and home life, and a dead loss to the community. What is needed is to depauperise him (Orwell, 2013, p. 208). If we can get past the derogatory language, Orwell’s analysis of people living excluded and impoverished lives not only captures the conditions of some citizens of contemporary advanced society, but his question about how to improve their lives is as apt today as it was then. The existence of rough sleeping and multiple exclusion homelessness (Fitzpatrick et al., 2013) shares similarities with Orwell’s depiction. Despite massive improvements in material standards of living and the development of Welfare States, in contemporary society, people experiencing homelessness live in ways people did in the 1930s. This article examines theories and practices of a model that aims to not simply respond to people experiencing homelessness, but a model that aims to demonstrably change how they live. As opposed to Housing First that provides immediate access to housing and support to solve a person’s homelessness (Padgett et al., 2016), the model examined aims to improve the lives of people who are homeless—to change how they behave—without providing housing. The model is referred to as the Gain model. Rather than offering affordable housing to end homelessness (i.e. Housing First), the Gain model aims to create conditions for people to improve their lives so that they access housing and other resources without the future need of social services. Whereas Housing First involves wraparound multidisciplinary support (Padgett et al., 2016), the Gain model rejects wraparound support. The Gain model is directed towards freedom from social services and intends to enable individual change so that social services are superfluous. In Australia, responding to social problems with wraparound support is the norm, advocated by the Australian Government (2015), large not-for-profit organisations (Mission Australia, 2014; Partners in Recovery, 2016), social investors (Social Ventures Australia, 2016) and the Australian Association of Social Workers (2011). The wraparound approach to support has shaped service delivery in myriad contexts, including: mental health, education, child welfare and juvenile justice (Burchard et al., 2002). Indeed, wraparound multidisciplinary supports are fundamental to the rigorous evidence base for the effectiveness of the Housing First approach (Tsemberis, 2010). Padgett, Henwood and Tsemberis (2016) demonstrate the profound success Housing First has achieved internationally in transforming systems: they refer to Housing First as a paradigm shift. On the other hand, and consistently with neo-liberal analysis (Gray et al., 2015), the Gain model sees well-being and individuals’ interests best realised when people are free from wraparound services. Differing from neo-liberal tenets, however, the Gain model assumes that well-resourced social services are required to disrupt deep exclusion and dependence. Moreover, the model assumes, contrary to neo-conservative scholars (see below), that socially and materially deprived people should not be negatively judged for and characterised by their deprivation. Management established the Gain model because of dissatisfaction with the deficient assumptions inherent in the former case management model. The Gain model positions those experiencing exclusion as people first and foremost. Drawing on ethnographic research in homeless accommodation where the Gain model is delivered, this article addresses two research questions. First, what constitutes, as a theory and a practice, the Gain model? Second, does the Gain model, or elements of it, have anything to offer social work? The remainder of the article proceeds over four sections. First, we review seminal theories of the consequences of social services, before describing the study site and the ethnographic research driving the article. The third section outlines empirical material to demonstrate key elements of the Gain model. Finally, we analyse the findings and ask how the Gain model contributes to social work. Social services: theories and debates Ever since individuals, charitable organisations or later states have delivered social services, scholars, politicians and service providers have theorised and debated what social services should aim to achieve—what function they have in society—and what form they should take (Field, 1998; Katz, 1986; Mead, 1986; Murray, 1984; Titmuss, 1958). Underpinning debate are unresolved questions of whether social services are a collective form of justice. Are they a means of redistribution to disrupt inequities? Conversely, do social services undermine autonomy, subvert independence and exacerbate exclusion from the market? The literature is vast; here we locate our research by reviewing debates about the behavioural consequences of social services. Ignoring behaviour We define social services to encompass not only cash transfers from the state through welfare, but also the array of services delivered and funded through state and community providers. Social services are not something distinct from society, but rather integral to industrialisation (Titmuss, 1958). Deacon’s (2002) analyses of the function and history of social services in the UK are instructive. He identified redistribution as an expression of altruism and collective good, on the one hand, and social services as a mechanism for recipients to improve their lives, including the transition to employment, on the other. Social services as a collective good drove the British post-war Welfare State. Titmuss (1958) provided a seminal rational for universalism and collectivism to promote equity and social justice. Universalism intended to address deficits in the English Poor Laws that focused on, and excluded people because of, behaviours (Deacon, 2002). Non-judgementalism avoids behaviour as an exclusion from entitlements and transcends identification of deserving groups (Katz, 1986). Universalism provides social protection as a right of citizenship (Spicker, 2013). Characterised as the quasi-Titmuss paradigm, Deacon (2002, p. 14) argued that British social policy scholars displayed disinterest in human agency and hostility ‘to the idea that one of the purposes of welfare is to shape the behaviours and aspirations of those who receive it’. Titmuss and his followers created a norm in social policy of ignoring how social services may induce dependencies and constitute barriers to accessing employment. Debate continues about whether social work research considers human agency (Parsell et al., 2017), but the normative position of ignoring behaviours resonates with Garrett’s (2016) argument that social work needs to be more sociologically informed and reject pressure to individualise social problems. Similarly, Spicker (2013, p. 133) says that social services should not consider behaviours because ‘it does not follow because problems are experienced by individuals that the responses should be individual too’. Focusing on behaviour Notwithstanding moral arguments for ignoring behaviours, in the USA, Wilson (1987) observed that overlooking agency has created a vacuum allowing conservatives to dominate debate about social services and how recipients behave. Indeed, for millennia, concern has existed that social services or any type of aid eroded incentives to work and created dependencies (Katz, 1986; Spicker, 2002). Mead (2011) influentially argued that social services encourage passivity through providing resources and requiring individuals to do nothing in return. To ensure recipients engage with employment, the state needs to help, but it also needs to enforce people into work (Mead, 2011). Overlooking evidence of the working poor (Newman, 1999), Mead (2011) believed that the key to enforce people into work is that the penalty be immediate and reliable, not that it be severe. Murray (1984) also provided influential theories about social services shaping behaviours. He argued that people accessed social services based on rational calculations that they represented more effective means to achieve their self-interests. For Murray, social service policy must be predicated on citizens as reflective and active decision makers. Field’s (1998) work restructuring the British Welfare State assumed that social service delivery directly influences human behaviour. Likewise, Dunn (2014) showed that people’s values and assessments of desirable employment and welfare entitlement availability determines whether they will seek employment or not. Even Dean (2015), who ideologically questions Dunn’s analysis, recognises that motives and actions of social service recipients need to be debated. Deacon (2002) argues that interest in the behaviours of social service recipients, particularly ideas about what behaviours are appropriate, constitute central elements of British welfare reform under the Third Way. Interest in recipients’ behaviours was tied to the responsibility of the state to create opportunities for social justice. Although policy has moved beyond the Third Way, British policy is, and has long been, interested in shaping recipients’ behaviours (Fletcher, 2015). Research design Research project Accepting that behaviours of social service recipients and service providers are a meaningful foci of analysis, this article presents the first ten months of an ethnographic study examining the service provision—service receipt interface in homelessness accommodation in an Australian city. In this article, we focus on data from service providers. The experiences of service users, and especially the meaning they ascribe to working with the Gain model, constitute a significant contribution to social work. Data from service users will be reported elsewhere, as will the results from longitudinal research examining the impacts of the Gain model. Research site The homeless accommodation and model examined are referred to using the pseudonym Gain. Gain has eighty-nine beds, including seventeen for women and seventy-two for men. Gain provides crisis accommodation for up to three months and transitional accommodation for up to fifteen months. Breakfast and an evening meal are served in a communal dining room. Gain has 32.6 staff, including: one manager, one chaplain, two team leaders, three senior case workers, 8.6 case workers, twelve case workers in training, two maintenance staff, two cleaners and one chef/catering trainer. We do not have precise information on the qualifications and professional background of all staff, but, reflecting Gain’s formal position of recruiting new university graduates, more than a dozen staff were either qualified social workers or social work students. Fieldwork Ethnographic fieldwork included in-depth interviews and participant observations. We conducted eight in-depth interviews, including: two team leaders, a senior case worker, four case workers and a chaplain. We selected in-depth interview participants using maximum variation sampling (Kitto et al., 2008). We sought service providers in different organisational roles with different service provision responsibilities. As demonstrated below, we sampled service providers to reflect diversity in how the Gain model was conceptualised and practised. In-depth interviews enabled service providers to outline the Gain model. Drawing on an interview schedule, interview participants were asked: to articulate the model objectives, what practices underpinned the model, how the model was delivered in practice, what facilitated and undermined success and how the model sat within the social service context. In-depth interviews were conducted several months after participant observations commenced. Thus, interview questions were informed by learnings from participant observations. Conducting interviews after we had developed relationships and where we had observed, listened to and learned about practice meant we conducted interviews from an informed position. Complementing ethnographic research where researcher and interview participants have more than a one-off relationship, in-depth interviews promoted ontological power that helped understand people in their practice environment (Floersch et al., 2014). Participant observations involved observing practice, including encounters between service providers and services users (Carr, 2011), team meetings, as well as two training workshops. Over ten months and approximately 210 hours of observation, we observed the eight interview participants and all other Gain staff. Consistently with the literature (Brewer, 2000), participant observations involved more than detached observing. Observations involved conversations and informal interviews with service providers. We conducted participant observations to examine practice in action, including how practitioners reflected on and made sense of practice (Forrester et al., 2013). From observations of first-hand practice, we asked practitioners how what we observed sat with the model they described. We follow Ferguson (2016), who acknowledges that ethnography does not represent a straightforward notion of truth. Rather, ethnography enabled us to examine the espoused practice model as an ideal theory, and then to observe, interpret and analyse practice. Noting the gap between theory and practice, Floersch et al. (2014) observe that social workers use perceptual capacities to gather, think and feel information about service users in their environment to inform practice. Although we present differences in how the model is explained and how we observed it delivered—accounts of action and action itself (Desmond, 2016)—we do not see the practised model as more true than the espoused ideal model. Realities of service provision to people materially excluded means that practitioners respond to their assessment of immediate need. Theoretical models that structure practice serve as ideal aspirations that set out intentions and normative positions, but these ideals are challenged when delivering or withholding services to people requesting assistance. Analytical strategy In-depth interviews were digitally recorded and transcribed. We recorded participant observations in a fieldwork journal. We quickly recorded observations in note form in a private office at Gain or, infrequently and when in line with the social context, we recorded notes at the time of observation. Following each day’s observation, we extended notes and wrote reflections on fieldwork (Emerson et al., 2001). We analysed interview transcripts and the fieldwork journal thematically. Similarly to Desmond (2016), we did not use software in the analytical process. Instead, driven by the research questions, the first and second author who conducted the fieldwork read and re-read the data over and over until we became intimate with the material. Next we coded and analysed data separately, and met periodically until we agreed on themes. We then scrutinised the analysis in a near complete stage, checking for accuracy, the audit trail and identifying negative case analyses. With prolonged engagement, these strategies promoted rigour (Padgett, 2008). An organisational director granted access to the research site. We then met with Gain’s leadership team. We agree with Ferguson (2016)—it is courageous for managers and practitioners to allow researchers to observe, critique and write about their work. We conducted research overtly; all staff knew our researcher status and that our presence was for research. On the basis of being overt, and providing ongoing opportunities to consent, the research received ethical approval from our university ethics review committee. Empirical findings Outcomes are not about results. The outcome for us is changing the way we do things, changing the way we look at people, changing the way we respect people, changing the way we see people and start calling them people, not residents or clients because that tends to have that power disparity automatically. One of the things we talk a lot about here is language because, if you have this perception that ‘I'm no different to anyone else, I'm going to treat you normally even though you're experiencing homelessness’ (Service Provider 01). The Gain model is predicated on the above. Practices of staff, together with the intended outcomes of practices, assume normality of service users and their status as people. In the months prior to the study, Gain implemented physical changes informed by the belief that it was people, not clients, using the service. The building, a purpose-built shelter with dormitories, was renovated so the dining room resembled a café, with new seating and bench tables added. In the dining room, likewise, the bar was removed where people previously queued to receive meals. All but one of the dormitories were transformed into single rooms: the capacity of Gain went from 128 to eighty-nine beds. In the reception, a security glass barricade separating workers and service users was removed. Removing physical barriers not only reflects the intention to minimise service provider–service user distinctions; it also intended to facilitate service providers freely moving throughout the building to engage service users. Staff argued that sitting behind desks and managing service users was passive, condescending and ineffective. We observed team leaders directing staff to move from their computer screens and ‘get out onto the floor’. In a staff meeting, minimal hours of service user contact in a given month was cited to argue that service providers needed to make themselves more available outside of the office. Management were also introducing a book which comprised one-page biographies, without photos, of caseworkers so that people presenting could nominate whom they wanted to work with. These changes were implemented so the physical setting reflected the philosophical position of people beyond their client status. A café environment, single rooms rather than dormitories, staff working away from desks and service users choosing whom they worked with aimed to disrupt power structures of clients as subordinate. Service providers believed the changes brought Gain closer in line with mainstream societal expectations. From the premise of service users as people, the Gain model key principles include: service provision as disabling; service users as passive and dependent; and enacting autonomous change. Service provision as disabling The fundamental value position of service users as people fed into the reasoning that social services had disabling effects. Service providers evoked the normality of the people they worked with to argue that it was unhelpful for service users, as normal people, to have social services in their lives. Instead, it was community and informal influences that people using Gain required: [We’re] trying to get away from that service first approach. We’re trying to have more links to natural supports within the community … we’re trying to get away from wrapping around. Some of the people we work with will have up to 15 services working with them that just wraparound, and we just find, from reports back from the people we work with, it’s making them more unwell (Service Provider 02). It is noteworthy that it is service providers themselves who argued that social services in people’s lives was problematic. The negative appraisal was not only about the number of services, but that services differentiated people as clients. They believed that service provision created disabling effects. Describing the context of the Gain model, Service Provider 03 explained: ‘… we always say if it doesn’t happen in the community, why does it happen here?’ What occurred in the community was presented as normal; this notion of normality was used to renounce social services. The proposition that the Gain model should replicate normality was debated during a training workshop where staff discussed ovens in homeless accommodation. Participants reflected on the response to a child who was burnt in one of the provider’s self-contained homeless accommodation units (not Gain) after the parent had opened the oven door to heat the unit. When the burning occurred, all ovens were removed to mitigate risk. Service providers, some of whom removed the ovens, ridiculed the removal because it treated people as irresponsible, failed to prepare them for the community and because it was inconsistent with community expectations. They described services as disabling by placing counterproductive restrictions on service users. Conditions undermined service users’ capacity for autonomous decision making. Citing the oven removal example, Service Provider 04 said that, prior to the Gain model, whenever a problem arose, management would impose new regulations on service users’ behaviours to prevent future problems. The service provider argued that short-term solutions perpetuated service users’ reliance on social services. By enacting new policies to respond to service users’ mistakes, the dominant service system created ‘a parent–child dynamic’. Service providers argued that traditional social services assumed people were deficient; people’s deficiencies had to be managed, thus service providers required rules and procedures to protect deficient clients. This idea of paternalist social services vis-à-vis deficient clients differentiated the Gain model. Underpinned by the valuing of service users as people, the Gain model was presented as antithetical to the dominant service system; because many service providers described themselves as participants in the dominant service system both outside of Gain and at Gain before the new model, they evidenced the model’s superiority with first-hand experiences. The Gain model assumed that service users were no different from service providers. Treating service users normally meant not restricting their capacities for autonomous decision making, and not practising to protect service users from the negative consequences of their decisions. It was not simply the imposition of restrictions that was presented as disabling. Instead, service providers debated how providing any services, even if not restrictive, constitutes a counterproductive intrusion. During another training workshop, Service Provider 04 suggested that, in the transitional accommodation where service users conduct the majority of cleaning themselves, it would be reasonable to have commercial cleaners ‘conduct a big spring-clean every now and again’. Others challenged the idea: cleaners were discredited as inappropriate because ‘it sets up a distinction with what happens in the community’. Further, cleaners subverted independence and undermined learnings for the development of normative behaviour required post Gain. Service Provider 03 explained that the Gain model required understanding that ‘you’re not here to save people’. When describing the Gain model, Service Provider 05 said that engagement with service users had to recognise that whatever they did was ‘ultimately their decision: consequences are theirs if they don’t do what they’ve said they’ll do’. The assertion that the dominant service system sought to save people, and thereby stifled capacities for decision making and self-responsibility, was articulated as ‘Giving someone something on a silver platter is doing them no favours’ (Service Provider 02). From a positioning of people as normal and an objective to replicate a notion of normality in mainstream society, the Gain model rejected social services that were likened to either a ‘silver platter’ or paternalistically saving people through a ‘child–parent dynamic’. The Gain model was distinct from wraparound social services because the latter had disabling consequences by controlling people. Although there was debate about what practices have disabling affects, such as cleaners, service providers recognised the Gain model did not always live up to the ideal. They acknowledged the curfew for women staying at Gain was akin to a child–parent dynamic. They identified the curfew, which they enforced, as inconsistent with the espoused model to promote autonomous decision making and opportunities for making mistakes. From observations of practice and team meetings, a tension emerged between an approach that assumes people’s normality and need for autonomy, on the one hand, and restrictive practice that monitors behaviour, on the other. Self-determination was difficult to achieve in practice. When senior staff learned of a service users’ room infested with bedbugs, service providers were reprimanded for failing to identify the warning signs during daily room inspections. Service providers were accused of ‘abandoning the men’ (in the men’s section), as bedbug infestation was the consequence of not thoroughly monitoring. Rather than identifying room checks as contradictory, senior staff argued that service providers allowing bedbugs was the antithesis to the Gain model. Bedbugs signalled service providers ‘setting the bar too low for service users’. Bedbugs were described as less than what people deserve. Passivity and dependence Following the argument that wraparound services were disabling to normal people, the Gain model assumed that people using homeless accommodation were passive and dependent. Participants believed that service users, because they were enculturated in a system characterised by wraparound supports, were unaccustomed to making decisions in their best interest. Service Provider 01 describes the Gain model and concludes: One of the barriers is the actual client themselves. They don’t even realise it and they don’t mean to be. They’ve had so much exposure to all these other services. They come here and it’s a total different way of doing things. It’s, ‘no, you’ve got to do this for me. No, over there their support workers do those. How come you’re not doing them?’ For a while there we had regular complaints about your staff aren’t doing their job. It’s not what their job is. Even though people using Gain were positioned as normal, service providers assessed that they acted in abnormal and unrealistic ways. Service providers believed that it was the intrusive presence of services in people’s lives that undermined autonomy and prevented people taking control of their lives. Service Provider 01 elaborated: So what we say, ‘OK, its $22 [per night] but if you’re on [unemployment benefit] you’re on $19 and if you’re on youth allowance its $13. Just bring your income statement,’ and they forget to do that on intake so that’s OK, we’ll give them a couple of days grace. Now, we have a [government benefits] office in [location removed], which is less than a ten minute walk and you should hear the carry on sometimes when we ask them to take some responsibility and go, ‘look, you’re going to save yourself sometimes $8 and $9 a night by going down and getting an income statement.’ [Service users will say that the benefits workers] ‘will be here in five days’ time. I’ll get it then.’ ‘Well that’s OK, but until that time you’re going to be charged $22 a night and we’re not going to backdate it.’ So it’s about providing them with an opportunity to have some responsibility and to meet some sort of obligation. Even though we’re a homeless service and it’s, ‘we should be doing this for them and everything for them,’ which is what other services do. The example illustrates how service users act against their own interests—by paying more money than they are required—and it shows how they expect and are dissatisfied when service providers do not address their problems. Service providers will not mitigate the consequences of service users’ actions, such as reducing rent without individuals ‘showing responsibility’. The remark also demonstrates how the problem is interpreted. Service users’ passivity is not framed as individual deficit. Service providers assume that the prevailing service system subverts people’s capacities and desire to take responsibility for their own actions. Passive behaviour is a structural problem. System problems are identified to advocate the Gain model which aims to disrupt existing social service systems. Responding to questions about the Gain model, Service Provider 03 observed: We try to get them to think because at the end of the day if they do find themselves in the situation again they can draw upon this experience, which means that they can call upon any natural resources, whether it was family they accessed or whatever the case may be, whereas some organisations will automatically say, ‘right, we will contact [service provider] for you and we’ll get brokerage for you,’ and they’ve learned nothing except that there are organisations out there who will support them if they spend their money on something which they shouldn’t have. The model sought to overcome individual problems so that people would change and not need a service in the future. At the centre was a long-ranging vision: services were delivered, not motivated by meeting immediate needs or protecting people from failed decision making, but to enact change so that people would address problems themselves instead of presenting to services expecting solutions. The focus of change was individual expectation rather than individual capacity. Because the model assumed people’s normality, it intended to alter their expectations about doing things for themselves so that they could realise their innate normality. Enacting autonomous change Operationalising the Gain model was difficult. The extract below is from observations between a service users and Service Provider 05: As I was sitting with [service user], he called out to [Service Provider 05]; he asked [Service Provider 05] whether he could get a lift to court tomorrow. The service provider said that ‘we do not do transport anymore.’ The service user said that Gain previously did transport; [Service Provider 05] responded that they no longer did. She added that the train station was just down the road (Fieldwork journal, 10 November 2015). In a later interview, Service Provider 05 described concern about people in her caseload not achieving the outcomes she wanted: ‘I had been stressed a couple of weeks ago because I was trying to make sure everyone was on top of their rent, everyone was doing their housing stuff.’ She described a dilemma with complying with the Gain model by feeling unable to meaningfully contribute to people’s lives. But, she reflected, in the way she was comfortable with advising people to use public transport, the Gain model was consistent with self-determination principles as a social worker: Since supervision I’ve taken a step back because [supervisor] was like, ‘I’ve noticed that you’ve become more directive instead of instructive or less suggestive and more directive in terms of saying, “you have to do this at this time.”’ I noticed I was doing it, but I didn’t know why I was doing it. It was just the pressure of the work. But it was good. Now I’ve taken a step back and I’ve said, ‘these are your options. We can do it if you want,’ and it’s helped. I feel less stressed (Service Provider 05). The service provider shows how practising the Gain model can be at odds with the expectations of service users. Moreover, although, after supervision, she felt that the Gain model was consistent with her professional values, she recognised the approach meant that some service users were failing to achieve outcomes. The challenges assisting people achieve outcomes also meant that some service providers practised contrary to the model. We observed Service Provider 06 make a telephone call for a service user to organise his appointment. When asked how making a call sat with the Gain model, Service Provider 06 stated the individual required accommodation and making the appointment, even though contrary to letting people do things for themselves, was justified because ‘we are supposed to be helping people find accommodation’. The service provider justified making the appointment because she did not want the service user to again miss an opportunity to secure housing. Nevertheless, she acknowledged that, if her supervisor was aware, she would be ‘pulled aside to justify my practice’. The above examples highlight a crucial aspect of the Gain model challenging the way service providers think, what they say and how they practise. Alongside supervision, they articulated the importance of reflection upon the model’s fundamental pillars. Reflectivity is core to realising the intended difference. Service providers understood the disabling service system not just theoretically, but because they identified themselves as agents of the system prior to the Gain model. Without ongoing self-monitoring, they risked regressing to wraparound support: ‘There will be times when I go into the fairy godmother mode where I do everything for them and then I quickly realise and I step back’ (Service Provider 07). Service providers both identified the need to monitor practice to ensure consistency with the model and ridiculed alternative approaches. The alternative approach was characterised as directing clients; the Gain model, where reflection was critical, involved informal conversations with service users and led by service users. Discussion and conclusion The Gain model aims to facilitate an environment for people experiencing homelessness to improve their lives so they no longer require social services. Material deprivation in advanced Welfare States, and social services to address it, have existed for decades. How is the Gain model novel? Social work theories, for example, have long identified practitioners making themselves redundant (Dominelli, 2004). Reflective practice, empowerment and working alongside people, rather than doing tasks for them, are social work tropes. In the remainder of the article, we theorise the Gain model and argue that, through positioning service users as people, the Gain model offers two contributions: advancing a vision for ambitious life trajectories; and presenting a normative good life and ideas for realising it. On the basis of personhood, the model proceeds assuming that people desire and have capacities to live positively. Driving the model is an optimistic assumption of human potential. We agree with Webber and Joubert (2015, p. i2): notwithstanding a commitment to self-determination and well-being, ‘social workers frequently have low expectations for the people they work with’. Structural exclusion and the way biography and life chances explain people’s poverty are well understood (Padgett et al., 2016). These analyses represent sophisticated understanding of inequities and life outcomes. Principles driving the Global Definition of Social Work are couched as overcoming inequities, injustices, violence and violation of rights (International Federation of Social Workers, 2014). These are unambiguously positive ideals, but day-to-day service provision, policy work and advocacy saturated with disadvantage and injustices represent barriers to social work articulating positive narratives for people’s potential. Identities matter, especially through stigmatising and characterising people by what they lack and what services they require (Stevenson et al., 2014). The Gain model forces reflection on identities that are imposed upon people by virtue of receiving social services, and how this identity closes off ambitious life opportunities. Expecting an ambitiously optimistic life logically follows positioning service users as people. The Gain model contains a normative position on the good life. A good life is autonomous. Autonomy is pivotal to social work values (Leece and Peace, 2010); the Gain model is innovative by conceptualising autonomy as freedom from social services. Social work advocates resources to promote autonomy through positive self-determination (Banks, 2006). Social problems are explained as an absence of services; solutions are presented through the introduction of services (Australian Association of Social Work, 2011). The Gain model rejects that people’s problems, even continuation of problems for people securely housed, are addressed with more services (Padgett et al., 2016). Social services are presented as antithetical to autonomy. Although social services are rejected for preventing autonomy, autonomy is not narrowly constructed as independence and self-reliance, as neo-liberal analysis would suggest (Gray et al., 2015). The Gain model recognises all of our lives as interdependent (Young, 2011); however, dependences should be informal, as opposed to dependence on social services. The model assumes social services—intervention of service providers—impedes recipients from exercising the characteristics of autonomy, including: agency, control and freedom (Collopy, 1988). Drawing on ideas close to Murray (1984), social services are rejected, not because people are irresponsible, but because social services are believed to undermine autonomy. Ideas about autonomy are consistent with the role service providers should play in contributing towards the good life. Differing from the norm in homelessness services where abstinence and sobriety are enforced as morally superior (Carr, 2011), the Gain model does not impose a moral position on people’s actions. Although alcohol and illicit substances are formally prohibited on site at Gain, service providers view substance misuse and addiction as a mechanism to have conversations with service users. The Gain model is not relativist about what people should do to realise a good life. Rather than sociological rationalisations of poor decision making, such as spending benefits on lavish meals (Desmond, 2016), the Gain model assumes that people should act in rational ways. The Gain model has a normative position on what constitutes the good life and what actions will achieve it, but these are not enforced. The model of autonomy means that people need to deal with the consequences of their choices as necessary to realise the good life. Implicit in the model is an assumption that telling people how to act is counterproductive, as it would subvert autonomy. Even though service providers view alcohol and substance misuse negatively, they believe that enforcing abstinence and sobriety would be contrary to the fundamental principles of autonomy. Telling people how to behave would perpetuate their reliance on service providers. Debate about social services creating dependencies has largely been the domain of neo-conservatives and neo-liberals who express ideological opposition to social services. The Gain model opens these questions to social work. It does so from a premise of social service users as people who aspire to positive life trajectories, but that social services hamper these trajectories. Is social work silent about dependencies because of awareness that some practices do indeed create inadvertent negative outcomes that render people dependent upon more services? Our data demonstrated how practitioners’ efforts to prevent dependencies can be challenging and disheartening. Creating autonomy for service users can disempower service providers (Leece and Peace, 2010). We showed how practitioners are conscious of creating possible dependencies, but this potential is rationalised because service provision nevertheless has greater, more immediate positive impacts. Through operationalising the Gain model, normative ideas and practice merge. Practice expects that human behaviour is socially bound. Service delivery constitutes a social product that shapes service users’ behaviours. Service providers did not articulate simplistic views of people’s problems (Agllias et al., 2016). They argued that passive behaviour was symptomatic of a system people had reasonably adapted. Individuals were not blamed for their passivity. The Gain model offers two contributions to social work. First, it is consistent with and builds on arguments for anti-oppressive social work. Endorsing the need for social work to develop non-hierarchical architecture where power differentials between service user and service provider are acknowledged and mitigated, the model rejects what is presented as traditional service provision that is experienced by service users as perpetuating their oppression (Strier and Binyamin, 2010). The Gain model not only aims to re-envisage service delivery to counter oppression and realise empowerment, but the model fundamentally rejects what its proponents see as the limited aspirations that are implicitly driving social services. Rather than practising to disrupt oppressions, the Gain model pushes social work to think about how providing services to excluded groups is premised on limited life expectations. Second, it is easy to write about autonomy and social work that enables it but it is another matter to achieve it in practice. Rather than uncritically presenting the Gain model as an exemplar, our ethnographic data present evidence, and a reminder for some, of how challenging it is to achieve autonomy and freedom from social services. In addition to people’s problems having an aetiology that is far more complex than a passive social service system (Fitzpatrick et al., 2013), we have shown how, despite intentions to move beyond traditional casework, the realities of meeting the needs of people requiring shelter, for example, challenge theoretical models. Despite identifying with an innovative model, some service providers continued practising what is traditional casework. We presented examples where service users expect—and use—services consistent with a casework model. The Gain model should not be romanticised as a panacea to people’s complex problems. Instead, its value lies in the significance of a reflective practice premised on people using social work services as no different to the social workers providing the services. The Gain model is directed towards changing service providers rather than service users. As the opening observation from Service Provider 01 suggested, outcomes are not ultimately about service users. Outcomes are about ‘changing the way we do things, changing the way we look at people, changing the way we respect people, changing the way we see people and start calling them people, not residents or clients’. Acknowledgements We are grateful to the Salvation Army for their significant support accessing their homeless accommodation and staff to enable the research. The research was funded by an Australian Research Council DECRA Fellowship and by a grant from the Australian Research Council Centre of Excellence for Children and Families over the Life Course. References Agllias K., Howard A., Schubert L., Gray M. ( 2016) ‘ Australian workers’ narratives about emergency relief and employment service clients: Complex issues, simple solutions’, Australian Social Work , 69( 3), pp. 297– 310. Google Scholar CrossRef Search ADS   Australian Association of Social Workers ( 2011) Income Management Position Paper, available online at http://www.aasw.asn.au/document/item/2213 (accessed 11 May 2017). Australian Government ( 2015) Domestic Violence in Australia, Finance and Public Administration References Committee, available online at http://www.aph.gov.au/Parliamentary_Business/Committees/Senate/Finance_and_Public_Administration/Domestic_Violence/Report (accessed 11 May 2017). Banks S. ( 2006) Ethics and Values in Social Work , Basingstoke, Palgrave. Brewer D. ( 2000) Ethnography , Buckingham, Open University Press. Burchard J., Bruns E., Burchard S. ( 2002) ‘The wraparound approach’, in Burns B., Hoagwood K. (eds), Community Treatment for Youth: Evidence-Based Interventions for Severe Emotional and Behavioral Disorders , New York, Oxford University Press, pp. 69– 90. Google Scholar CrossRef Search ADS   Carr E. ( 2011) Scripting Addiction: The Politics of Therapeutic Talk and American Sobriety , Princeton, Princeton University Press. Collopy B. ( 1988) ‘ Autonomy in long-term care: Some crucial distinctions’, The Gerontologist , 28, pp. 10– 17. Google Scholar CrossRef Search ADS PubMed  Deacon A. ( 2002) Perspectives on Welfare: Ideas, Ideologies and Policy Debates , Buckingham, Open University Press. Dean H. ( 2015) ‘ Rethinking unemployment and the work ethic: Beyond the “quasi-Titmuss” paradigm’, Journal of Social Policy , 44( 3), pp. 641– 3. Google Scholar CrossRef Search ADS   Desmond M. ( 2016) Evicted: Poverty and Profit in the American City , New York, Crown. Dominelli L. ( 2004) Social Work: Theory and Practice for a Changing Profession , Oxford, Polity Press. Dunn A. ( 2014) Rethinking Unemployment and the Work Ethic: Beyond the ‘Quasi-Titmuss’ Paradigm , Basingstoke, Palgrave Macmillan. Emerson R., Fretz R., Shaw L. ( 2001) ‘Participant observation and fieldnotes’, in Atkinson P., Coffey A., Delamont S., Lofland J., Lofland L. (eds), Handbook of Ethnography , London, Sage. Ferguson H. ( 2016) ‘ Researching social work practice close up: Using ethnographic and mobile methods to understand encounters between social workers, children and families’, British Journal of Social Work , 46( 1), pp. 153– 68. Google Scholar CrossRef Search ADS   Field F. ( 1998) Reflections on Welfare Reform , London, Social Market Foundation. Fitzpatrick S., Bramley G., Johnsen S. ( 2013) ‘ Pathways into multiple exclusion homelessness in seven UK cities’, Urban Studies , 50( 1), pp. 148– 68. Google Scholar CrossRef Search ADS   Fletcher D. ( 2015) ‘ Workfare—a blast from the past? Contemporary work conditionality for the unemployed in historical perspective’, Social Policy and Society , 14( 3), pp. 329– 39. Google Scholar CrossRef Search ADS   Floersch J., Longhofer J., Suskewicz J. ( 2014) ‘ The use of ethnography in social work research’, Qualitative Social Work , 13( 1), pp. 3– 7. Google Scholar CrossRef Search ADS   Forrester D., Westlake D., McCann M., Thurnham A., Shefer G., Glynn G., Killian K. ( 2013) Reclaiming Social Work? An Evaluation of Systemic Units as an Approach to Delivering Children’s Services , Bedford, University of Bedford. Garrett P. ( 2016) ‘ “Introducing Michael Gove to Loic Wacquant”: Why social work needs critical sociology’, British Journal of Social Work , 46( 4), pp. 873– 89. Google Scholar CrossRef Search ADS PubMed  Gray M., Dean M., Agllias K., Howard A., Schubert L. ( 2015) ‘ Perspectives on neoliberalism for human service professionals’, Social Service Review , 89( 2), pp. 368– 92. Google Scholar CrossRef Search ADS   International Federation of Social Workers ( 2014) Global Definition of Social Work, available online at http://ifsw.org/get-involved/global-definition-of-social-work/ (accessed 11 May 2017). Katz M. ( 1986) In the Shadow of the Poorhouse: A Social History of Welfare in America , New York, Basic Books. Kitto S., Chesters J., Grbich C. ( 2008) ‘ Quality in qualitative research’, Medical Journal of Australia , 188( 4), pp. 243– 6. Google Scholar PubMed  Leece J., Peace S. ( 2010) ‘ Developing new understandings of independence and autonomy in the personalised relationship’, British Journal of Social Work , 40( 6), pp. 1847– 65. Google Scholar CrossRef Search ADS   Mead L. ( 1986) Beyond entitlement: The social obligations of citizenship , New York, Free Press. Mead L. ( 2011) Expanding Work Programs for Poor Men , Washington, American Enterprise Institute Press. Mission Australia ( 2014) From Homelessness to Sustained Housing 2010–2013, available online at https://www.missionaustralia.com.au/what-we-do/research-evaluation/misha (accessed 11 May 2017). Murray C. ( 1984) Losing Ground: American Social Policy, 1950–1980 , New York, Basic Books. Newman K. ( 1999) No Shame in My Game: The Working Poor in the Inner City , New York, Russell Sage Foundation. Orwell G. ( 2013 [1933]) Down and Out in Paris and London , London, Penguin Books. Padgett D. ( 2008) Qualitative Methods in Social Work Research , Los Angeles, Sage. Padgett D., Henwood B., Tsemberis S. ( 2016) Housing First: Ending Homelessness, Transforming Systems, and Changing Lives , New York, Oxford University Press. Parsell C., Eggins E., Marston G. ( 2017) ‘ Human agency and social work research: A systematic search and synthesis of social work literature’, British Journal of Social Work , 47( 1), pp. 238– 55. Partners in Recovery ( 2016) A White Paper on Partners in Recovery in Victoria, available online at file:///C:\Users\uqcparse\Desktop\Victorian%20PIR%20White%20Paper%20full.pdf (accessed 20 May 2017). Social Ventures Australia ( 2016) Housing, available online at http://www.socialventures.com.au/assets/Housing-Perspective-web.pdf (accessed 11 May 2017). Spicker P. ( 2002) Poverty and the Welfare State: Dispelling the Myths , London, Catalyst. Spicker P. ( 2013) Reclaiming Individualism: Perspectives on Public Policy , Bristol, The Policy Press. Google Scholar CrossRef Search ADS   Stevenson C., McNamara N., Muldoon O. ( 2014) ‘ Stigmatised identity and service usage in disadvantaged communities: Residents’, community workers’ and service providers’ perspectives’, Journal of Community and Applied Psychology , 24( 6), pp. 453– 66. Google Scholar CrossRef Search ADS   Strier R., Binyamin S. ( 2010) ‘ Developing anti-oppressive services for the poor: A theoretical and organisational rationale’, British Journal of Social Work , 40( 6), pp. 1908– 26. Google Scholar CrossRef Search ADS   Titmuss R. ( 1958) Essays on The Welfare State , London, Unwin University Books. Tsemberis S. ( 2010) ‘Housing First: Ending homelessness, promoting recovery, and reducing costs’, in Ellen I., O’Flaherty B. (eds), How to House the Homeless , New York, Russell Sage Foundation, pp. 37– 56. Webber M., Joubert L. ( 2015) ‘ Editorial: Social work and recovery’, British Journal of Social Work , 1, pp. i1– 8. Google Scholar CrossRef Search ADS   Wilson W. ( 1987) The Truly Disadvantaged: The Inner City, The Underclass, and Public Policy , Chicago, Chicago University Press. Young I. ( 2011) Responsibility for Justice , Oxford, Oxford University Press. Google Scholar CrossRef Search ADS   © The Author 2017. Published by Oxford University Press on behalf of The British Association of Social Workers. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Saar-Heiman, Yuval;Krumer-Nevo, Michal;Lavie-Ajayi, Maya

doi: 10.1093/bjsw/bcx054pmid: N/A

Abstract The spatial aspects of therapeutic intervention receive marginal attention in the practice and discourse of therapeutic professions. Currently, influenced by the classic Freudian tradition, the central therapeutic space of social workers and psychologists is the agency and the office. Despite the fact that models of home therapy have been developed over the years, its practice has remained marginal in the professional discourse. This article introduces the concept of intervention in a real-life context as a theoretical and practical framework for work with people in poverty that takes place outside the agency. The article is based on a qualitative study that examined the ongoing experiences of nine women who participated in an intervention programme based on the poverty-aware social work paradigm. The findings revealed that intervention in a real-life context takes place in four major realms: the home space, in which meetings were held regularly; the utilisation-of-rights space; the emergency space; and the professional-friendly space. The discussion sheds light on intervention in a real-life context as a means to practise the psychodynamic concept of holding in working with people who live in poverty. Poverty-aware social work, critical practice, therapeutic space, in-home intervention, holding Introduction The spatial aspects of therapeutic intervention receive marginal attention in the professional practice and discourse of therapeutic professions (Ferguson, 2016). Generally, influenced by the classic Freudian tradition, the central therapeutic space of social workers and psychologists is found in the agency, and specifically in the office. The current study aims to explore the spatial aspects of the developing poverty-aware social work paradigm (PAP) (Krumer-Nevo, 2016; Saar-Heiman et al. , 2016), which offers theoretical and ethical premises for practice with people in poverty. The article examines the spatial aspects of the paradigm from service users’ perspectives, conceptualising what we call intervention in a real-life context and discussing its therapeutic potential. Literature review Poverty is a social problem that influences and shapes the lives of individuals and families. Social workers have been committed to supporting families who live in poverty since the establishment of the profession. However, through the years, scholars have argued that social work had failed to develop effective modes of direct practice, relying on a critical analysis of poverty as a social problem (Davis and Wainwright, 2005; Krumer-Nevo, 2016). One of the critics regarding dominant social work practice with people in poverty points to its focus on a pre-determined spatial setting, namely the office. Although it is possible to identify prominent examples of therapeutic interventions that took place outside of the agency, in the life spaces of service users (e.g. Fraiberg et al., 1980; Redl, 1966), with time, this idea became marginal (Allen and Tracy, 2004; Woodford, 1999). Instead, the analytical therapeutic setting designed by Freud has permeated the professional activity, and therapeutic interventions have become identified with a fixed setting that has a clear structure and rules. Those characteristics were perceived as necessary in order to form the conditions for the development of transference and expression of neuroses (Freud, 1958). Though, over the years, many scholars have extended and developed Freudian psychoanalytic theory to make it more relevant for the understanding of social problems (e.g. Deleuze and Guattari, 1972; Cooper and Lousada 2005; Altman, 1995), these developments did not converse into a systematic change in the way therapeutic setting is perceived. The current neo-liberal ideology and the new public management, with their emphasis on formalisation, documentation and inspection, led to a further distancing between professionals and service users, and continue to reinforce therapy’s confinement to the office (Murphy et al., 2013). Nonetheless, situations in which service users meet social workers outside of the agency do exist, and can be divided into three categories. The first of these consists of in-home visits in the framework of welfare services. In-home visits are mainly one-time encounters that serve narrow, focused goals, mainly concerning families’ evaluation of capability or risk. Thus, they are not continuous interventions and are not perceived as playing an important role in establishing relationships of help (Sousa and Rodrigues, 2012; Winter and Cree, 2015; Ferguson, 2014). The second category is the framework of reaching out to clients labelled as ‘hard to reach’ (Lavie-Ajayi and Krumer-Nevo, 2013). In this framework, contacts with populations such as socially excluded youth (Connolly and Joly, 2012), women in prostitution (Kurtz et al., 2005) or homeless people (Ferguson, 2007) might take place in public spaces such as the street, public buildings, clubs or parks. Although, in this category, conducting meetings outside of the safe space of the office gains significance and importance, it still serves only as a strategy for bringing clients to the agency, where ‘real’ therapeutic intervention will take place. Arising from family therapy approaches, intervention in the third category takes place in special programmes that consist of continuous in-home meetings (Cortes, 2004). This type of intervention is applied mainly with poor families that have not cooperated with mainstream interventions in the past, and often occurs in mental health and child welfare services (McWey, 2007; Waisbrod et al., 2012). The specific kinds of intervention used in this approach vary from family therapy (McWey, 2007; Cortes, 2004) and psychotherapy (Rogers, 2014; Fraiberg et al., 1980) to crisis-management (Fuller, 2004) and community-oriented practice (Wong, 2014; Scarborough et al., 2013). Although this kind of intervention was found suitable for the needs of service users (Sousa and Rodrigues, 2012; McWey, 2007; Cooley, 2011), successful and effective (Al et al., 2012; Ammerman et al., 2011), it remains marginal and of limited scale in mainstream professional literature and practice (Waisbrod et al., 2012). Research context Similarly to other OECD countries, poverty is measured in Israel in comparison to the median disposable income per standardised person. According to current available data, the incidence of family poverty in Israel was the highest amongst the OECD countries (OECD, 2017). Poverty-aware social work is a new paradigm developed in Israel that aims to capture the complex relationships between the individual, family and the social worker, between the material and the emotional, and between therapy and social activism. By challenging the separations between all these spheres, the paradigm corresponds with different scholars that place therapeutic practice inside a social and political context (e.g. Altman, 1995; Cushman, 2015). Combining a critical social analysis of poverty with a therapeutic stance, the paradigm involves three theoretical and ethical principles. The first is the conceptualisation of poverty as a violation of human rights. This leads to the recognition of the continuous daily resistance of families to their poverty. The second principle holds that, since knowledge is legitimised or de-legitimised in the context of power relations, professional knowledge about families in poverty should be based on careful attention to their perspectives through close relationships of social workers and service users. The third principle calls for social workers to stand by their clients in their struggle against poverty, to represent their knowledge and advocate their interests in order to change societal organisations (Krumer-Nevo, 2016). These premises have been translated into practice in the framework of a special B.A. fieldwork programme, Casework for Social Change, which has been running since 2010 at Ben-Gurion University of the Negev. (For a detailed account of the educational framework of the programme, see Krumer-Nevo et al., 2009.) The programme aims to: (i) improve the lives of poor families, (ii) educate students in the poverty-aware paradigm and (iii) serve as a laboratory for the development of the paradigm which aims to combine a critical social analysis of poverty with a therapeutic stance to practice. The students meet the families regularly every week for two academic years, at a place convenient to the families, mainly their homes. This setting sets the stage for working with the whole family as a unit of intervention, although, in all the cases, it is the woman who is identified as the main client. The intervention plan is determined by the family and the student together, but it always combines dealing with material issues such as debts or housing problems and working through clients’ past and present emotional experiences and relationships. In order to do so, the students receive theoretical courses, individual and group instruction, and a practicum. This article focuses on one principle of the practice enacted in the programme (Krumer-Nevo, 2016)—intervention outside the agency, through weekly meetings at the families’ homes and the accompaniment of family members to a variety of institutions in accordance with their needs. The systematic use of work outside the agency makes the programme an appropriate research arena for an in-depth examination of practices carried out outside the agency. The choice to focus on service users’ perspectives stemmed from the notion that learning about social services cannot be complete if it does not include service users’ perspectives (Teater, 2011). Methodology This article is based on a secondary analysis of interviews with women who participated in the programme Casework for Social Change (Saar-Heiman et al., 2016). The initial research questions were: (i) What is the programme’s practice in the eyes of its users? and (ii) What kind of experiences does this practice elicit? A detailed account of the experiences of the programme’s users is described elsewhere (Saar-Heiman et al., 2016). The current article focuses specifically on the spatial characteristics of the intervention. The study comprised twenty-four interviews, three interviews with nine women who participated in the programme over two years from 2012 to 2014. These nine women were all the participants who started the programme in the first month of the academic year of 2012. The first set of interviews took place at the beginning of the relationship with the students, the second about nine months later and the last after the end of the relationship with the students. In the second round of interviews, eight women were interviewed, since one student discontinued her studies. In the third round, seven women were interviewed because one woman was ill at the time of the interviews. Initial contact was made with the women through a telephone conversation in which they were provided with a brief explanation of the research and a time and place for an interview were set. All of the women requested to be interviewed in their homes. The study was approved by the ethics committee of the university and the women signed an informed consent form. The participants The average age of the women was 44.5 (range thirty to fifty-two) and the average number of children was 3.5 (range one to five). Five women were divorced, three were married and one was living with a partner. All of the women were Jewish. Apart from one woman who was born in Ethiopia, all were born in Israel. Six of the women lived in rented public housing, one lived in an apartment rented on the private market and two owned their apartments. Five women did not work and four others worked part time in temporary jobs. Seven of the women struggled with health problems. The interviews The women were interviewed by the first author using in-depth semi-structured interviews (Polkinghorn, 2005), which began with an open question: ‘Tell me about your encounters with the student.’ Next, interviewees were asked about their expectations for changes, the practices used by the students, the way in which they experienced and understood these practices, and changes that had taken place in their lives. The interviews lasted an average of one hour each. Data analysis Data was analysed using the interpretive interactionism approach (Denzin, 2001). The interpretive interactionism approach, with its focus on the reciprocal relations between people and their social environments (Denzin, 2001), is especially suitable for this research, because it makes it possible to place service users’ experiences within the context of poverty, oppression and social exclusion. After the initial analysis that had focused on the experiences of the women (Saar-Heiman et al., 2016), a further analysis was carried out focusing on the characteristics of the practice. At this stage, we identified four spaces in which the students’ practice took place and the women’s meanings to those spaces. Findings We refer to intervention in various life spaces in accordance with service users’ needs as intervention in a real-life context. The interviews revealed that work outside the agency, in a real-life context, makes it possible to turn diverse spaces not usually considered therapeutic into therapeutic spaces. Intervention in a real-life context has two characteristics: first, it does not occur in an artificial space, such as the office, but rather in the everyday life spaces of the families; second, it is not limited to one fixed space, but is dynamic and moves between different spaces relevant in service users’ lives. As the current study demonstrates, the intervention was spread over four different types of public and private spaces: the home space; the utilisation-of-rights space; the emergency space; and the professional-friendly space. The home space Holding meetings at the families’ homes is a prominent and central practice in the experience of the interviewees, because the home was the central place in which the therapeutic relationship took place throughout its duration. The women described the home meetings positively, saw them as tailored to their needs and expanded upon the advantages of the home setting as compared to the agency. At the concrete level, they saw meeting the students at their home as adapted to their needs because it spared them the trouble of coping with the mobility limitations associated with poverty, such as a lack of private vehicles and the inability to pay for babysitting services. The literature identifies these everyday problems as barriers that might prevent families who live in poverty from consuming social and health services (Gelberg et al., 2004): First of all, it makes it easier for me, because I don’t own a car, so I do everything on foot. And it’s much easier for me that she [the student] comes to me than it would be for me to take two busses each way, which comes at the expense of children (Rakefet). The home was described as suitable for the women’s needs also because it allowed them to talk calmly and privately, in contrast to meetings at the agency, where they felt their privacy was invaded. The experience of privacy was maintained at home even when meetings took place against the background of household noise or the unexpected arrival of family members or neighbours. Women experienced the home meetings as allowing an emotional availability that was impossible in the agency: ‘It was also extremely helpful that I didn’t have to go over there … and hear all the stories. I have my story. What do I need more stories for? We sit here, it’s personal, ours, we have privacy’ (Gali). Home was described by the women as a space that allows, and perhaps even requires, the students to see for themselves their needs and the difficulties they face. Importantly, being in the home allowed the students to see two separate components of the women’s lives—first, the actual material lacks and deprivation, such as having no food in the refrigerator, and, second, the women’s continuous struggle with their problems, namely their ongoing resistance to poverty: I tell her [the student] ‘open it, look at the fridge.’ She sees. She knows everything (Rikki). I … give up my life, everything, because of the kids. And Gal knows it and sees it (Shani). The fact that the students have the opportunity to see the reality of the women has therapeutic value because it allows them to witness what Ziv (2012) refers to as insidious trauma, namely trauma generated by living in conditions of social oppression and violence. As in the treatment of extreme trauma, the visibility and naming of the traumatic events are of great therapeutic value. In our case, this visibility might be achieved by pointing to the terrible difficulties associated with material lacks and the constant struggle to cope (Saar-Heiman et al., 2016). These two components of the women’s visibility stand in contradiction to other experiences in which they felt that they faced disbelief regarding their difficulties and/or a lack of recognition of their struggle. Another feature of the home setting is that it has enabled women to reposition themselves vis-à-vis the students. The home space creates a change in the familiar positions of the helper–helped relationships and facilitates an experience in which the inter-personal elements are dominant: ‘The student comes to you, so you’re relaxed. A person has come to you, so you must be relaxed … you cannot attack him [laughs] or bang on the table. It’s a different experience, it’s not like going there’ (Dalit). The different position made possible in the therapeutic encounter at home stems from, among other things, the fact that the meetings do not take place only in a conversational space, but also contain intimate situations that cannot exist in the office. Thus, for example, students took part in family activities such as meals, watching television, doing homework and so on. This created close, intimate and mutual relationships. One participant noted: ‘She was no longer a social worker, she was already one of the family who was used to the home, who ate with us, sat with us’ (Shoshi). Although the general attitude of the women to holding sessions in the home was very positive, and they mentioned various benefits of this practice, they also noted having had hesitations and concerns over this practice at the beginning of the relationship. Primarily, they were concerned about the exposure of their vulnerability and being subject to supervision. These concerns highlight the risk of romanticising the visibility achieved in the home space and remind us the tendency to use in home visits as a tool for surveillance (Winter and Cree, 2015). Nonetheless, all the women affirmed that, as the helping relationship developed, they became convinced that the students saw the home setting as a space for creating connections and not for evaluation and supervision, and their suspicions waned accordingly: ‘I was a bit hesitant at first. I didn’t know what it was. Suddenly a person comes to you, starts to investigate you … but then I saw that it was for me, that it was helpful’ (Rina). The utilisation-of-rights space Paradoxically, vulnerable populations are the very ones that tend to under-utilise their rights (Torczyner, 2001). The main reasons for this phenomenon concern lack of information or misinformation about rights, feelings of shame and fear or lack of confidence in the system: a perception that the price of implementing eligibility is higher than the profit involved in it or lack of access to the entity that provides the rights (Currie, 2004). Administrative authorities that deal with housing, education, health, employment, allowances and the handling of debts raise various barriers through bureaucracy disregard the difficulties connected to poverty and use language unfamiliar to people who live in poverty (Currie, 2004). Our findings indicate that ‘doing errands’—namely coping actively with various administrative authorities—was a major concern for the families and constituted a decisive part of their everyday lives. In spite of their efforts, their success in utilising their rights was meagre. While the conventional social work practice of advocacy in such cases includes mediation or letter writing, the students in the programme practised what we call active advocacy—that is, they took an active role in doing everything necessary for their clients to be able to utilise their rights, beginning by accompanying them to the various administrative authorities in question when they asked for it (Saar-Heiman et al., 2016). Thus, students met family members in institutions such as the social security office, the public housing office, the bailiff’s office, at municipal and court hearings, as well as in hospitals and at children’s schools. Shoshi differentiated active advocacy from the kind of advocacy she was used to receiving from social workers in the past: I had Dina and Einat [social workers], who wrote letters for me. ‘Take this letter, see you later, bye’. They didn’t come with me. It’s not like Naama who says, ‘Come, I’ll go with you to all those offices, I’ll help you’. Intervention in the utilisation-of-rights space included four characteristics. The first concerned assistance in filling out forms and dealing with the labyrinths of the bureaucracy. This aspect was found to be particularly important in situations that involved language barriers. It is important to note that this assistance was not merely technical, since difficulty in reading and writing also involves feelings of extreme shame and inadequacy. Thus, the situation requires exposing these emotions and coping with them: ‘Sometimes you’re too shy to talk, you don’t know what to say, you don’t know what to write, but he [the student] will be with me during this time’ (Dalit). The second point mentioned by the women was that the presence of the students in the various coping situations involved in the ‘errands’ strengthened the relationship between them. The women reported a sense of enjoyment and relief that stemmed from the fact that they were not alone in difficult situations created by the interaction with administration officials. Significantly, they also reported this in cases where students were ultimately not successful in fulfilling the goal they had come for: ‘It’s a matter for the court and there are things here that you cannot move, but when she came with me it made me feel good, it reassured me’ (Shani). The third point involves what the women called ‘professional backing’. In addition to its effect on their feelings, the presence of the students also affected the way in which the women were perceived by the various authorities. The mere involvement of students in the direct interactions with the authorities gave the women ‘backing’—as Gali defined it: led to them being treated more seriously and receiving a higher level of service. Accordingly, their chances of utilising their rights also improved: ‘When I come with him [the student], he says to the clerk, “Listen, we’re from Welfare as well, we’re trying to help her. Come and help her, too”’ (Dalit). Finally, similarly to the witnessing function of the students within the home space, the students’ presence in the various spaces of active advocacy allowed them to witness instances of micro-aggression (Pierce, 1970) connected with poverty. Micro-aggression is everyday minor instances of aggression that may be manifested through contempt, distancing, humiliation, stigma and otherness. Unlike witnessing in the home space, which is essentially witnessing of material lacks, in the spaces of active advocacy, the students witnessed micro-aggression, which Lister (2004) calls lack of symbolic capital and is characteristic of poverty. The students’ presence allowed them to see with their own eyes these various forms of oppression and enabled them to recognise and validate the irrationality of the women’s reality and thus made it easier for them to cope: I think that when Shimrit [the student] leaves here she is miserable. I swear, my heart breaks for her. In every office I have to cry and beg and this is very difficult for her, really hard, I believe. But never mind, she gives me strength (Gali). The emergency space The third space of intervention in the real-life context concerns emergency situations in which the women urgently need immediate and close assistance. Living in poverty invites a multiplicity of emergencies (Sousa and Rodrigues, 2012). Ribner and Knei-Paz (2002) found that the mobilisation of social workers in emergency and crisis situations was perceived by their clients as extremely valuable. Indeed, the women talked about emergencies such as imprisonment, medical crises, power cuts and severe shortages of food, in which they contacted the students or were assisted by them. The nature of the relationship and the fact that the women had the students’ personal cell phone numbers allowed them to turn to the students in these situations, in order both to inform them and to seek assistance. The first characteristic of intervention in crisis situations is that it is a kind of ‘first aid’—that is, an immediate and concrete solution provided at a time and in a place where it is needed: ‘When there is a critical situation, it’s hard, and there’s no food, we tell her, it’s like first aid. She tries here, there, runs around’ (Ricki). This concrete aid goes hand in hand with a sense of support and welfare. The students’ availability and the women’s knowledge that they are not alone when they encounter a crisis constituted an important anchor for the women, even in cases where the student could not get to them. In fact, the knowledge that the student would like to come, even if he or she could not, was a source of support: The social worker tells you that he’s there until one o’clock, if I have a serious problem I have to wait until the next day. Here I can call, and talk to the social worker, tell him ‘listen, I have a problem, the bailiff’s office is coming to arrest me’ (Rina). In another case, in which the response did not involve anything material, the fact that the student came to the emergency space, which evoked feelings of great loneliness and anxiety, was of great importance, and helped the woman to feel safer and calmer: I told him I didn’t feel well, I was at the HMO. He said, ‘Dalit, I will come.’ The fact that he is sitting next to you already says that he is available and he’s helping you with his presence. It doesn’t have to be concrete things. His presence is enough (Dalit). The professional-friendly space The last space of intervention in the real-life context consists of situations in which the students and the women, as part of their professional relationship, were in places usually associated with informal, friendly encounters, such as cafes, shopping malls or cinemas. Being involved in friendly activities is controversial in the professional literature, mainly because it challenges the boundaries of the professional relationship, constitutes a deviation from its constraints, and could jeopardise both the therapeutic relationship and the client (Pope and Keith-Spiegel, 2008). Conversely, various studies that addressed the friendly context of professional relationships found that social workers and their clients attributed positive meaning to closeness with friendly features. In addition, social workers’ willingness to modify the limits of traditional therapy within the context of a professional helping relationship was perceived as beneficial (Beresford et al., 2008; Doel, 2010; Reimer, 2014). The women described two main benefits of being in friendly spaces with the students. First, intervention in these spaces allowed them to engage in activities they would not have engaged in otherwise, such as going to the cinema: I don’t go out in the evening. It’s enough that she took me out, gave me some air, a different atmosphere. You’re alive, beginning to live. It is good because I won’t go out alone. Where would I go to the movies, with whom? (Gali) The second point involves the fact that being with the women in spaces associated with friendship enabled the women to feel that they could give something to the students and allowed the relationship to be more reciprocal: I invited her to coffee and a croissant when I received the allowance because she went with me to all the ‘errands.’ I wanted to do something for her, because she deserves everything good, and I wanted us to be happy together (Ricky). In accordance with Sousa and Rodrigues (2012), we distinguish between a symmetrical friendly relationship, in which the needs of both sides of the relationship receive equal treatment, and friendly actions, which contain elements of social relationships, but occur within the professional context (Saar-Heiman et al., 2016). Discussion Based on service users’ perspectives, this article seeks to introduce the concept of intervention in a real-life context as a practical principle in poverty-aware social work. Intervention in a real-life context is not limited to one pre-determined physical space, but rather takes place in the community and in everyday life spaces, which are determined and vary in accordance with families’ needs. The findings indicate four main spaces in which intervention took place—the home space, the utilisation-of-rights space, the emergency space and the professional-friendly space—and detail the characteristics of intervention in each space. There is a growing body of scholarly calls to connect in therapy the personal and social aspects of families’ lives and to connect their internal and external worlds (Prilleltensky, 2001; Lavie-Ajayi and Nakash, 2015). This is especially necessary when dealing with poverty, which shapes and influences both the real alternatives and circumstances of families’ lives as well as their emotional experiences. Often, the complex situations created by poverty make professionals think that poor families are not suitable for therapy. The difficulties of families to commit themselves to regular therapy sessions in the agency are seen in this context as evidence of their unwillingness to explore their emotional world, either because they are too occupied with their daily struggle or because they are not interested in or capable of therapeutic work. Using Winnicott’s concepts ‘holding’ and ‘holding environment’, we will conceptualise how the characteristics of intervention in the real-life context, which are seen as a violation of the therapeutic rules, can become therapeutic components in work with families in poverty. The concept of holding carries many meanings. At the most basic level, it conceptualises the mother’s physical holding of her baby in the early stages of life. Beyond physical holding, the concept reflects all the physical and mental activity, the state of primary maternal preoccupation (Winnicot, 1953; Winnicott, 1956a), that constitutes the holding environment, namely that guards the baby from impingements and creates for him the conditions necessary for the development and advancement of the ability to ‘go on being’ (Ogden, 2004). The holding function can serve as a metaphor for the therapeutic relationship (Slochower, 1991; Applegate, 1997), pointing to the importance of creating an environment characterised by the empathetic, guiding and reliable presence of the therapist. Based on our findings, we wish to extend the metaphor of the holding environment to the understanding of the role of the societal environment, especially in shaping poor people’s experiences. To borrow from Winnicott, a society is ‘good enough’ if it provides its citizens with the basic experience of protection, investment and response to needs. This kind of experience does not exist for people who live in poverty, who experience society as abandoning them and neglecting them, on the one hand, and, on the other, as intrusive and invasive (Nice, 2011; Piven and Cloward, 1971). Intervention in the real-life context provides the kind of holding needed after years of societal abuse and neglect. As the findings demonstrate, it allowed the students to respond to various emotional and social needs of the women that had been unmet due to their social position. A look at the four different spaces of intervention in real-life contexts reveals how being and operating in each space and moving between spaces make it possible to fulfil various holding functions. First, the home meetings and active advocacy produce visibility. Visibility, a significant component in any therapy, becomes even more significant in light of the considerable social transparency suffered by families in poverty: transparency in the public discourse and the media (Chauhan and Foster, 2013); the transparency of the oppression that comes with poverty (Frost and Hoggett, 2008; Ziv, 2012); and the transparency of people’s daily struggle against poverty (Lister, 2004). One could say that intervention in the real-life context is a practice that opposes social transparency and fulfils the basic need for social recognition. Second, accompanying the women to the various administrative authorities makes it possible to promote the utilisation of fundamental, necessary social rights. Sucharov (2013), who conceptualises the act of accompanying patients to assist them in dealing with officials from various authorities as ‘therapeutic advocacy’, claims that, in addition to fulfilling the actual social needs of clients, which is essential in its own right, the act of accompanying them is also a function of creating space for clients’ unique voices, which are frequently silenced by social power structures. Third, mobilisation during emergencies prevents isolation. Ostensibly, moments of crisis are individual situations that are met first of all by family members. However, as the findings of this study show, many of the extreme situations faced by the women are the direct result of their poverty—food shortages, disconnection from essential services and coping with extreme situations when confronted with various enforcement agencies. In addition, because formal and informal support networks are limited for people in poverty (Sousa and Rodrigues, 2012), the fact that the students were available for them comprised a critical function of social and personal support. Finally, being in friendly spaces produces reciprocity in the power relations with a representative of the establishment and can heal wounds created by the lack of symbolic capital. The relationships created with the students allowed women to experience a beneficent connection with those who are often seen by them as agents of power. The professional-friendly space makes it possible to create a rich and mutual expression of relationships in general and of power relations vis-à-vis the establishment specifically. This function is significant for those whose relationship history with society is marked by failures and disappointments. The various examples presented above illustrate how intervention in the real-life context produces a holding therapeutic space at the inter-personal level, which is necessary following a long-standing deficit of holding at the societal level. In other words, the various functions of the intervention and of any space in and of itself are compatible with the women’s dynamic needs and ongoing deprivation. It is important to note that, in other contexts, or other populations, intervention in a real-life context may include other spaces and different specificities. Moreover, not all interventions must include all four spaces, since the guiding principle is not intervention in specific spaces, but rather intervention in life spaces relevant to the needs of service users. In this context, it is possible to describe a relevant intervention in which the treatment room or office space in itself constitutes the most relevant space for specific service users. Understanding the practice of intervention in a real-life context as fulfilling holding functions enables us to understand its potential therapeutic benefits. However, it also raises various challenges (Adams and Maynard, 2000; Rogers, 2014): whether and how it is possible to establish and maintain therapeutic boundaries in such interventions; how to use the closeness such interventions create to benefit clients; how social workers should act to maintain separation between the personal and the professional. Indeed, further study is needed to examine the perspectives of professionals regarding the difficulties and ethical and professional issues created by intervention in the real-life context. In addition, the current study has several limitations. The research arena is relatively small and addresses a specific intervention of students in a unique programme. The ability of other professionals to implement such practices and the responses of clients to these practices should also be examined. Finally, in this study, data were collected only from the point of view of service users, so they do not reflect the points of view of the students and programme staff. Moreover, it is not within the scope of this study to shed light on the students’ motives, perceptions and experiences regarding the intervention and on the different ways in which they translated the theoretical premises of poverty-aware social work into practice. Learning social work practice and conceptualising it based on service users’ descriptions and meanings of it is in itself a political act. However, there is a need for further research on students’ and social workers’ perspectives that will allow us to expand our understanding and assessment of the therapeutic and political aspects of intervention in a real-life context and the possibilities and challenges it offers. References Adams J. F., Maynard P. E. ( 2000) ‘ Evaluating training needs for home-based family therapy: A focus group approach’, American Journal of Family Therapy , 28( 1), pp. 41– 52. Google Scholar CrossRef Search ADS   Al C. M., Stams G. J. J., Bek M. S., Damen E. M., Asscher J. J., van der Laan P. H. ( 2012) ‘ A meta-analysis of intensive family preservation programs: Placement prevention and improvement of family functioning’, Children and Youth Services Review , 34( 8), pp. 1472– 9. Google Scholar CrossRef Search ADS   Allen S. F., Tracy E. M. ( 2004) ‘ Revitalizing the role of home visiting by school social workers’, Children & Schools , 26( 4), pp. 197– 208. Google Scholar CrossRef Search ADS   Altman N. ( 1995) The Analyst in the Inner City: Race, Class, and Culture through a Psychoanalytic Lens , Hillsdale, The Analytic Press. Ammerman R. T., Putnam F. W., Stevens J., Bosse N. R., Short J. A., Bodley A. L., Van Ginkel J. B. ( 2011) ‘ An open trial of in-home CBT for depressed mothers in home visitation’, Maternal and Child Health Journal , 15( 8), pp. 1333– 41. Google Scholar CrossRef Search ADS   Applegate J. S. ( 1997) ‘ The holding environment: An organizing metaphor for social work theory and practice’, Smith College Studies in Social Work , 68( 1), pp. 7– 29. Google Scholar CrossRef Search ADS   Beresford P., Croft S., Adshead L. ( 2008) ‘ “We don’t see her as a social worker”: A service user case study of the importance of the social worker’s relationship and humanity’, British Journal of Social Work , 38( 7), pp. 1388– 407. Google Scholar CrossRef Search ADS   Chauhan A., Foster J. ( 2013) ‘ Representations of poverty in British newspapers: A case of “othering” the threat?’, Journal of Community & Applied Social Psychology , 24( 5), pp. 390– 405. Google Scholar CrossRef Search ADS   Connolly J. A., Joly L. E. ( 2012) ‘ Outreach with street-involved youth: A quantitative and qualitative review of the literature’, Clinical Psychology Review , 32( 6), pp. 524– 34. Google Scholar CrossRef Search ADS   Cooley M. ( 2011) ‘ Short-term home visiting for parents’, Journal of Family Social Work , 14( 2), pp. 179– 88. Google Scholar CrossRef Search ADS   Cooper A., Lousada J. ( 2005) Borderline Welfare: Feeling and Fear of Feeling in Modern Welfare , London, Karnac Books. Cortes L. ( 2004) ‘ Home-based family therapy: A misunderstanding of the role and a new challenge for therapists’, The Family Journal , 12( 2), pp. 184– 8. Google Scholar CrossRef Search ADS   Currie J. ( 2004) The Take Up of Social Benefits , No. w10488, Cambridge, MA, The National Bureau of Economic Research. Cushman P. ( 2015) ‘ Relational psychoanalysis as political resistance’, Contemporary Psychoanalysis , 51( 3), pp. 423– 59. Google Scholar CrossRef Search ADS   Davis A., Wainwright S. ( 2005) ‘ Combating poverty and social exclusion: Implications for social work education’, Social Work Education , 24( 3), pp. 259– 73. Google Scholar CrossRef Search ADS   Deleuze G., Guattari F. (1972/ 2004) Anti-Oedipus , London, Continuum. Denzin N. K. ( 2001) Interpretive Interactionism , Thousand Oaks, CA, SAGE. Google Scholar CrossRef Search ADS   Doel M. ( 2010) ‘Service users perspectives on relationships’, in G. Ruch, D. Turney, A. Ward (eds), Relationship-based social work: Getting to the heart of practice . (pp. 199–214). London, Jessica Kingsley Publishers. Ferguson H. ( 2014) ‘ What social workers do in performing child protection work: Evidence from research into face-to-face practice’, Child and Family Social Work , 21( 3), pp. 283– 94. Google Scholar CrossRef Search ADS   Ferguson H. ( 2016) ‘ Making home visits: Creativity and the embodied practices of home visiting in social work and child protection’, Qualitative Social Work  Advance Access published July 6, 2016, 10.1177/1473325016656751. Ferguson K. M. ( 2007) ‘ Implementing a social enterprise intervention with homeless, street-living youths in Los Angeles’, Social Work , 52( 2), pp. 103– 12. Google Scholar CrossRef Search ADS   Fraiberg S., Shapiro V., Spitz Cherniss D. ( 1980) ‘Treatment modalities’, in Fraiberg S. (ed.), Clinical Studies in Infant Mental Health: The First Year of Life , New York, Basic Books. Freud S. ( 1958) ‘Remembering, repeating and working through’, in J. Strachey (ed. and trans.), The standard edition of the complete psychological works of Sigmund Freud  (Vol. 12, pp. 145–163). London, UK, Hogarth Press and the Institute of Psycho-Analysis. (Originaly published in 1914). Frost L., Hoggett P. ( 2008) ‘ Human agency and social suffering’, Critical Social Policy , 28( 4), pp. 438– 60. Google Scholar CrossRef Search ADS   Fuller A. ( 2004) ‘ Crisis: Home‐based family therapy’, Australian and New Zealand Journal of Family Therapy , 25( 4), pp. 177– 82. Google Scholar CrossRef Search ADS   Gelberg L., Browner C., Lejano E., Arangua L. ( 2004) ‘ Access to women’s health care: A qualitative study of barriers perceived by homeless women’, Women & Health , 40( 2), pp. 87– 100. Google Scholar CrossRef Search ADS   Krumer-Nevo M. ( 2016) ‘ Poverty-aware social work: A paradigm for social work practice with people in poverty’, British Journal of Social Work , 46( 6), pp. 1793– 808. Google Scholar CrossRef Search ADS   Krumer-Nevo M., Weiss-Gal I., Monnickendam M. ( 2009) ‘ Poverty-aware social work practice: A conceptual framework for social work education’, Journal of Social Work Education , 45( 2), pp. 225– 43. Google Scholar CrossRef Search ADS   Kurtz S. P., Surratt H. L., Kiley M. C., Inciardi J. A. ( 2005) ‘ Barriers to health and social services for street-based sex workers’, Journal of Health Care for the Poor and Underserved , 16( 2), pp. 345– 61. Google Scholar CrossRef Search ADS   Lavie-Ajayi M., Krumer-Nevo M. ( 2013) ‘ In a different mindset: Critical youth work with marginalized youth’, Children and Youth Services Review , 35( 10), pp. 1698– 704. Google Scholar CrossRef Search ADS   Lavie-Ajayi M., Nakash O. ( 2015) ‘ “If she had helped me to solve the problem at my workplace, she would have cured me”: A critical discourse analysis of a mental health intake’, Qualitative Social Work , 16( 1), pp. 60– 77. Google Scholar CrossRef Search ADS   Lister R. ( 2004) Poverty , Cambridge, Polity. McWey L. M. ( 2007) ‘ In-home family therapy as a prevention of foster care placement: Clients’ opinions about therapeutic services’, American Journal of Family Therapy , 36( 1), pp. 48– 59. Google Scholar CrossRef Search ADS   Murphy D., Duggan M., Joseph S. ( 2013) ‘ Relationship-based social work and its compatibility with the person-centered approach: Principled versus instrumental perspectives’, British Journal of Social Work , 43( 4), pp. 703– 19. Google Scholar CrossRef Search ADS   Nice J. A. ( 2011) ‘Poverty as an everyday state of exception’, in Feldman S., Geisler C. C., Menon G. A. (eds), Accumulating Insecurity: Violence and Dispossession in the Making of Everyday Life , Vol. 9, Athens, GA, University of Georgia Press, pp. 67– 110. OECD ( 2017) Poverty Rate (Indicator), available online at https://data.oecd.org/inequality/poverty-rate.htm, 10.1787/0fe1315d-en (accessed 24 April 2017). Ogden T. H. ( 2004) ‘ On holding and containing, being and dreaming’, International Journal of Psychoanalysis , 85( 6), pp. 1349– 64. Google Scholar CrossRef Search ADS   Pierce C. ( 1970) ‘Offensive mechanisms’, in Barbour F. B. (ed.), The Black Seventies , Boston, Porter Sargent Publisher, pp. 265– 82. Piven F. F, Cloward R. ( 1971) Regulating the Poor: The Functions of Public Welfare , New York, Vintage Books. Polkinghorn D. E. ( 2005) ‘ Language and meaning: Data collection in qualitative research’, Journal of Counseling Psychology , 52, pp. 137– 45. Google Scholar CrossRef Search ADS   Pope K. S., Keith-Spiegel P. ( 2008) ‘ A practical approach to boundaries in psychotherapy: Making decisions, bypassing blunders and mending fences’, Journal of Clinical Psychology , 64( 5), pp. 638– 52. Google Scholar CrossRef Search ADS   Prilleltensky I. ( 2001) ‘ Value-based praxis in community psychology: Moving toward social justice and social action’, American Journal of Community Psychology , 29( 5), pp. 747– 78. Google Scholar CrossRef Search ADS   Redl F. ( 1966) When We Deal with Children: Selected Writings , New York, The Free Press. Reimer E. C. ( 2014) ‘ Using friendship to build professional family work relationships where child neglect is an issue: Worker perceptions’, Australian Social Work , 67( 3), pp. 315– 31. Google Scholar CrossRef Search ADS   Ribner D. S., Knei-Paz C. ( 2002) ‘ Client’s view of a successful helping relationship’, Social Work , 47( 4), pp. 379– 87. Google Scholar CrossRef Search ADS   Rogers S. ( 2014) ‘ The moving psychoanalytic frame: Ethical challenges for community practitioners’, International Journal of Applied Psychoanalytic Studies , 11( 2), pp. 151– 62. Google Scholar CrossRef Search ADS   Saar-Heiman Y., Lavie-Ajayi M., Krumer-Nevo M. ( 2016) ‘ Poverty‐aware social work practice: Service users’ perspectives’, Child & Family Social Work , 22( 2), pp. 1054– 63. Google Scholar CrossRef Search ADS   Scarborough N., Taylor B., Tuttle A. ( 2013) ‘ Collaborative home-based therapy (CHBT): A culturally responsive model for treating children and adolescents involved in child protective service systems’, Contemporary Family Therapy , 35( 3), pp. 465– 77. Google Scholar CrossRef Search ADS   Slochower J. ( 1991) ‘ Variations in the analytic holding environment’, International Journal of Psychoanalysis , 72( 4), pp. 709– 18. Sousa L., Rodrigues S. ( 2012) ‘ The collaborative professional: Towards empowering vulnerable families’, Journal of Social Work Practice , 26( 4), pp. 411– 25. Google Scholar CrossRef Search ADS   Sucharov M. ( 2013) ‘ Politics, race, and class in the analytic space: The healing power of therapeutic advocacy’, International Journal of Psychoanalytic Self Psychology , 8( 1), pp. 29– 45. Google Scholar CrossRef Search ADS   Teater B. A. ( 2011) ‘ A qualitative evaluation of the Section 8 Housing Choice Voucher program: The recipients' perspectives’, Qualitative Social Work , 10( 4), 503– 19. Google Scholar CrossRef Search ADS   Torczyner J. ( 2001) ‘ The application of human rights advocacy theory to organizational innovation in Israel: The Community Advocacy/Genesis Israel experience’, International Journal of Social Welfare , 10( 2), pp. 85– 96. Google Scholar CrossRef Search ADS   Waisbrod N., Buchbinder E., Possick C. ( 2012) ‘ In-home intervention with families in distress: Changing places to promote change’, Social Work , 57( 2), pp. 121– 32. Google Scholar CrossRef Search ADS   Winnicott D. W. ( 1953) ‘ Transitional objects and transitional phenomena: A study of the first not-me possession’, International Journal of Psycho-Analysis , 34, pp. 89– 97. Winnicott D. W. ( 1956a) ‘Primary maternal preoccupation’, in D. W. Winnicott (ed.), The Maturational Processes and the Facilitating Environment , New York, International Universities Press, pp. 300– 5. Winter K., Cree V. E. ( 2015) ‘ Social work home visits to children and families in the UK: A Foucauldian perspective’, British Journal of Social Work , 46( 5), pp. 1175– 90. Google Scholar CrossRef Search ADS   Wong V. ( 2014) ‘ Social work with youth in social withdrawal: In-home and beyond-home intervention modalities’, China Journal of Social Work , 7( 2), pp. 161– 74. Google Scholar CrossRef Search ADS   Woodford M. S. ( 1999) ‘ Home-based family therapy: Theory and process from “friendly visitors” to multisystemic therapy’, Family Journal , 7( 3), pp. 265– 9. Ziv E. ( 2012) ‘ Insidious trauma’, Mafteah , 5, pp. 55– 74 (in Hebrew). © The Author 2017. Published by Oxford University Press on behalf of The British Association of Social Workers. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Savaya, Riki;Melamed, Sharon;Altschuler, Dorit

doi: 10.1093/bjsw/bcx051pmid: N/A

Abstract This paper reports the findings of a study comparing service providers’ and service users’ perceptions of providers’ burnout. It addresses two issues: the similarities and differences in their perceptions; and the associations between any gaps in their perceptions and the service users’ satisfaction and perceptions of change. The study was conducted on 270 matched pairs of service users and service providers in a human service agency in a major city in Israel, using the Maslach Burnout Inventory. The findings show that the service users viewed their providers as less emotionally exhausted, as having a lower level of accomplishment and as more depersonalising than the providers viewed themselves. They also show that user-versus-provider discrepancies in perceptions of the providers’ accomplishments and depersonalisation contributed significantly to the users’ satisfaction with the agency and their provider, as well as to their perception of changes in their presenting problem. Burnout, gaps in perceptions of burnout, perceptions of change, satisfaction with services, service providers, service users Introduction Burnout is known to be a serious problem in human service professions. As defined by Maslach and Jackson (1981), burnout is characterised by three key dimensions: emotional exhaustion, depersonalisation and reduced personal accomplishment. Emotional exhaustion is manifested in service providers feeling that they are no longer able to give of themselves emotionally. Depersonalisation means that providers respond to persistent stress by developing negative, cynical attitudes and feelings about their clients. Reduced personal accomplishment is manifested in providers’ negative perceptions of their work and feelings of dissatisfaction with their accomplishments (Maslach et al., 1996). A great deal has been written about the deleterious consequences of burnout for both service providers and their clients. Consequences for service providers include impaired emotional and physical health, and diminished sense of well-being (Acker, 2010; Kim and Kao, 2011; Lizano, 2015; Peterson et al., 2008). With respect to clients, service providers suffering from burnout are less able to be empathic, collaborative and attentive, and are less able to bond with clients and to form a co-operative working alliance with them (Corrigan, 1990). They have high turnover rates, which disrupt the continuity of care (Boyer and Bond, 1999); and generally show poorer job performance (Taris, 2006) and provide poorer quality services than those who do not suffer from burnout (Carney et al., 1993; Parrish and Rubin, 2012). Almost all the research on burnout to date has queried service providers. To the best of our knowledge, only two studies have asked service users to report on the burnout of their service providers. The first, by McCarthy and Frieze (1999), examined associations between therapists’ behaviours and clients’ perceptions of the quality of the therapy. The findings showed that perceived therapist burnout, one of the behaviours investigated, was positively associated therapists’ use of personal coercive influence strategies, direct expert influence strategies and indirect expert influence strategies. The second, by the authors of the present paper, found that service users’ perceptions of their providers’ burnout mediated the association between the working alliance they formed with their provided and their perceptions of the effectiveness of the intervention (Savaya et al., 2016). No study, however, has investigated possible differences in service providers’ and service users’ perceptions of the service providers’ burnout. The present paper examines the latter issue: namely the congruence between service providers’ and service users’ perceptions of providers’ burnout. It addresses two questions. The first is: how similar or different are their perceptions of service providers’ burnout? Although the literature assumes that burnout has concrete manifestations in service providers’ behaviours, it cannot be certain that service users necessarily perceive those manifestations or sense them to be as serious as the service providers do. The second question concerns the predictive power of differences in their perceptions: are discrepancies in their perceptions associated with service users’ satisfaction with their service providers and the agency, and with their perceived changes in their presenting problems? The rational for the examination is anchored in theories emphasising the importance of consensus or congruence of perceptions between service users and service providers to achieving positive outcomes. Among other things, such congruence is an indicator of the quality of the relationship between the service user and service provider. In particular, importance has been attributed to congruence, or agreement, with respect to the aims of treatment and the ways of attaining them. For example, a meta-analysis by Horvath, Del Re, Flückiger and Symonds (2011) shows that the quality of the working alliance, defined as the positive emotional bond between therapist and client and their agreement on the treatment goals and the means attaining them, is one of the strongest predictors of the success of therapy. A meta-analysis by Tryon and Winograd (2011) found that service user–service provider goal consensus and collaboration are associated with service users’ well-being and positive outcomes in psychotherapy. In a similar vein, several researchers found that client–therapist agreement on what is happening in the therapy are associated with better outcomes (Cummings et al., 1992; Kivlighan and Arthur, 2000; Reis and Brown, 1999). Conversely, discrepancies between client and therapist ratings of the relationship and outcome measures are associated with negative outcomes. Corning et al. (2007) found that greater discrepancies between clients’ and therapists’ perceptions of the severity of the presenting problem lowers the odds of mutual treatment termination. From a somewhat different perspective, Taber and colleagues (2011) found that client–therapist personality congruence is associated with the quality of the therapeutic bond, which in turn is associated with the therapeutic outcome. Context of the study The study was conducted in the Social Welfare Department (SWD) of the Tel Aviv-Yafo Municipality. This is among Israel’s largest municipalities. Its SWD provides a large variety of services to close to 39,032 persons from some 27,470 households. In the early 2000s, the senior management of the SWD introduced a major change in the department’s approach, from heuristic practice to planned, systematic, outcome-oriented practice (Savaya et al., 2013). As part of the endeavour, management and workers joined in developing guidelines for assessment and intervention. Among the key principles of the guidelines were the importance of co-operation between service providers and service users in the assessment process and their agreement on the aims of the intervention and ways of attaining them. The endeavour, which is still in process, has been accompanied by ongoing evaluation and research. In light of the literature reviewed above and within the context of the evaluation and research that accompanied the changes, the present paper presents findings on two questions: How congruent are the service users’ and service providers’ perceptions of the providers’ burnout? Are discrepancies in their perceptions associated with service users’ satisfaction and perception of change? Method Sample In 2012, we conducted two studies in the Department of Social Welfare of the Tel Aviv-Yafo municipality. One study queried service users; the other queried service providers. Among the variables examined in both samples was service providers’ burnout. Service providers answered questions on their own levels of burnout; service users were asked to rate their service providers’ levels of burnout. The questionnaires were administered to all the service providers in the Department of Social Welfare. (For details on the service user sampling procedure, see Savaya et al., 2016.) Since participants from both groups were asked to identify themselves on the questionnaire, we were able to create a matched service provider × service user sample. Both studies were approved by the Ethics Committee of Tel Aviv University. All the participants signed an informed consent form prior to filling out the questionnaire. Sample characteristics The sample consisted of 270 matched pairs of service provider × service users. About two-thirds of the service user sample were women. Their mean age was 43.95 years (SD = 10.38). About a third of them were married; almost half were separated or divorced. About a third had no more than a junior high-school education; under half had completed high school. A little over half were employed, the others not. Most of the service providers were female (85.3 per cent). More than half (56.1 per cent) had a BSW, the entry-level degree to practise social work in Israel; 42.4 per cent had an MSW and 1 per cent a Ph.D. Their mean number of years in the profession was 15.7 (SD=12.4) and their mean number of years on the job was 11.3 (SD=14.6). Measures Background information Service providers were asked their gender, education and number of years in the profession and on the job. Service users were asked their gender, age, family status, education and employment status. Perceived social worker burnout This was was assessed on the Maslach Burnout Inventory (MBI; Maslach and Jackson, 1986). This is a twenty-one-item questionnaire which asks respondents to indicate how often they experience feelings pointing to burnout on a seven-point scale (0 = never to 6 = every day). The items cover three factors: emotional exhaustion, (e.g. I feel emotionally drained from my work), depersonalisation (e.g. I feel I treat some recipients as if they were impersonal ‘objects’) and personal accomplishment (e.g. I deal very effectively with the problems of my recipients). The MBI is a well-known and validated self-report measure of professionals’ burnout. Reliabilities for the service providers in the present sample were: α = 0.92 for emotional exhaustion, α = 0.75 for depersonalisation and α = 0.88 for personal accomplishment. Service users completed the same questionnaire, in which they were asked to report their perceptions of their service providers’ burnout. The questionnaire covered the same three factors as that completed by the service providers: emotional exhaustion (e.g. ‘My social worker seems emotionally drained’), depersonalisation (e.g. ‘My social worker treats me as if I were impersonal object’) and personal accomplishment (e.g. ‘My social worker deals effectively with my problems’). The construct validity of the version of the MBI that was filled out by the service users was examined using confirmatory factor analysis. The findings showed a good fit to the model: /df = 2.16, CFI = 0.963, TLI = 0.957 and RMSEA = 0.058, thereby confirming the construct validity of the measure. In other words, these findings show that the theoretical three-part structure of the original questionnaire remained almost entirely intact. Only one item, which had a loading of less than 0.4, had to be removed from the scale. Reliabilities were very good: α = 0.91 for emotional exhaustion, α = 0.83 for depersonalisation and α = 0.93 for personal accomplishment. For both samples, scores were calculated as the mean of the responses. The higher the score, the greater the perception of service providers’ emotional exhaustion, depersonalisation and personal accomplishment. Two aspects of service users’ satisfaction were measured: satisfaction with their social worker, and satisfaction with the agency and the service it provided. Satisfaction with the social worker This was assessed by four items, querying the respondents’ satisfaction with their social worker, with the frequency of their contact, with the help they received from the social worker and with their social worker’s attitude. Ratings were made on a five-point scale (1 = not at all satisfied, 5 = very satisfied). Factor analysis with Varimax rotation yielded one factor, which explained 84.36 per cent of the variance. Reliability was α =  0.94. Scores were calculated as the mean of the responses. The higher the score, the greater the satisfaction with the social worker. Quality of services received from the agency This was assessed by the twenty-one-item Client Satisfaction Inventory (CSI; McMurtry and Hudson, 2000). Service users were asked to rate the services they received from the agency (e.g. ‘The services I get here are a big help to me’; ‘People here want to do things their way, instead of helping me find my way’; ‘I thought no one could help me until I came here’) on a seven-point scale (1 = none of the time, 7 = all of the time). Although factor analysis with Varimax rotation yielded the same two factors that were reported by the developers of the measure, a single CSI score was calculated for the purpose of the present analyses. This explained 69.4 per cent of the variance; its reliability was 0.95. The CSI was calculated as the mean of the responses. The higher the mean, the higher the service users’ assessment of the quality of the services they received. Perceived change To ascertain their perceived change, service users were asked to name the main problem for which they had sought help from the welfare department and to indicate its severity at the time of their initial help seeking. Ratings were made on a ten-point scale, with 1 = not at all serious, 10 = extremely serious. Then, the service users were asked to indicate the amount of change they perceived in the problem on a ten-point scale (1 = a lot worse, 5 = no change, 10 = much better). Perceived change was determined by the ratings on the second question. All the measures were administered in Hebrew, the language of the study’s respondents. Findings Table 1 presents the means and standard deviations of the service users’ satisfaction with their service providers and the agency and its services, as well as the changes they perceived in their presenting problem. Table 1 Means and SD of service users’ satisfaction and perceived changes   Range  Mean  Std. deviation  Satisfaction with service providers  1–5  3.59  1.28  Satisfaction with agency and services  1–7  4.42  1.75  Perceived change in presenting problem  1–10  6.26  2.04    Range  Mean  Std. deviation  Satisfaction with service providers  1–5  3.59  1.28  Satisfaction with agency and services  1–7  4.42  1.75  Perceived change in presenting problem  1–10  6.26  2.04  As can be seen, the respondents’ satisfaction with the service providers and the services was in the high-to-moderate range, while their perceptions of the changes in their presenting problem were moderate. Table 2 presents the means, standard deviations and t-values of service users’ × service providers’ perceptions of service providers’ burnout. Table 2 Means, SD and t-values of service providers’ and service users’ perceptions of the service providers’ burnout   Service providers   Service users     Burnout  Mean  SD  Mean  SD  T  Emotional exhaustion  3.30  1.14  2.37  1.26  –8.74***  Personal accomplishment  5.34  0.80  4.01  1.44  –13.14***  Depersonalisation  1.99  0.85  2.40  1.33  4.15***    Service providers   Service users     Burnout  Mean  SD  Mean  SD  T  Emotional exhaustion  3.30  1.14  2.37  1.26  –8.74***  Personal accomplishment  5.34  0.80  4.01  1.44  –13.14***  Depersonalisation  1.99  0.85  2.40  1.33  4.15***  *** p<0.001. As can be seen, both the providers and the users perceived the providers’ burnout to be low to moderate. With this, there were some differences in their perceptions. The service providers reported a fairly high sense of personal accomplishment, rather low depersonalisation and moderate emotional exhaustion. The service users, in contrast, viewed their service providers as being less emotionally exhausted, as more depersonalising (i.e. as objectifying them more) and accomplishing less than the service providers rated themselves. All the differences are statistically significant. In addition, matched pair correlations were calculated between the perceptions of each of the service providers and his or her service users. None of the correlations was significant. To measure the discrepancies between the service users’ and service providers’ perceptions of the service providers’ burnout, we calculated the differences in the burnout scores they gave. Table 3 presents the ranges, means and standard deviations on the three burnout subscales. Table 3 Discrepancies in perceptions of service provider burnout: ranges, means and standard deviations   Minimum  Maximum  Means  Std. deviations  Emotional exhaustion  –5.13  3.78  –0.92  1.74  Personal accomplishment  –5.13  2.38  –1.32  1.66  Depersonalisation  –4.00  4.20  0.41  1.63    Minimum  Maximum  Means  Std. deviations  Emotional exhaustion  –5.13  3.78  –0.92  1.74  Personal accomplishment  –5.13  2.38  –1.32  1.66  Depersonalisation  –4.00  4.20  0.41  1.63  As can be seen, the discrepancies were about eight points on each subscale. The minuses preceding the emotional exhaustion and personal accomplishment scores indicate that the service users rated their service providers as less emotionally exhausted and as having accomplished less than the service providers rated themselves. The minus preceding the depersonalisation score indicates that the service users viewed their providers as more depersonalised than the providers viewed themselves. Three hierarchical regressions were carried out: the first to determine whether the discrepancies predicted the service users’ satisfaction with the agency; the second to determine whether they predicted the service users’ satisfaction with their service provider; and the third to ascertain whether they predicted the service users’ perceived changes. In the first step of each regression, the service users’ employment, education, gender, age and marital status were entered as control variables. The findings are presented in Table 4. Table 4 Hierarchical regression analysis predicting CSI, satisfaction with service provider and perceived changes Service users’ variables  CSI   Satisfaction with service provider   Perceived changes     B  SE  β  B  SE  β  B  SE  β  Model 1                    Employment  –0.29  0.23  –0.08  –0.12  0.18  –0.04  0.47  0.32  0.09  Education  –0.07  0.08  –0.06  –0.00  0.06  –0.00  –0.01  0.10  –0.00  Gender  0.24  0.24  0.06  0.22  0.19  0.07  –0.29  0.32  –0.05  Age  –0.01  0.01  –0.06  –0.01  0.01  –0.10  –0.03  0.01  –0.13*  Marital status  –0.07  0.24  –0.02  0.06  0.19  0.02  0.04  0.32  0.01  R2  0.02  0.01  0.04*  Model 2                    Work  –0.11  0.14  –0.03  0.03  0.11  0.01  0.57  0.30  0.12  Education  –0.09  0.05  –0.07  –0.02  0.04  –0.02  –0.02  0.10  –0.01  Gender  0.12  0.15  0.03  0.09  0.11  0.032  –0.35  0.31  –0.07  Age  –0.00  0.00  –0.01  –0.00  0.00  –0.05  –0.03  0.01  –0.11  Marital status  –0.12  0.14  –0.03  0.02  0.11  0.01  0.00  0.31  0.00  Discrepancy in perceived providers’ emotional exhaustion  –0.02  0.04  –0.02  0.01  0.03  0.02  –0.05  0.09  –0.04  Discrepancy in perceived personal accomplishments  0.66  0.04  0.63***  0.43  0.03  0.52***  0.35  0.09  0.24***  Discrepancy in perceived depersonalisation  –0.24  0.05  –0.23***  –0.33  0.04  –0.39***  –0.15  0.10  –0.10  R2  0.63***  0.63***  0.15***  Service users’ variables  CSI   Satisfaction with service provider   Perceived changes     B  SE  β  B  SE  β  B  SE  β  Model 1                    Employment  –0.29  0.23  –0.08  –0.12  0.18  –0.04  0.47  0.32  0.09  Education  –0.07  0.08  –0.06  –0.00  0.06  –0.00  –0.01  0.10  –0.00  Gender  0.24  0.24  0.06  0.22  0.19  0.07  –0.29  0.32  –0.05  Age  –0.01  0.01  –0.06  –0.01  0.01  –0.10  –0.03  0.01  –0.13*  Marital status  –0.07  0.24  –0.02  0.06  0.19  0.02  0.04  0.32  0.01  R2  0.02  0.01  0.04*  Model 2                    Work  –0.11  0.14  –0.03  0.03  0.11  0.01  0.57  0.30  0.12  Education  –0.09  0.05  –0.07  –0.02  0.04  –0.02  –0.02  0.10  –0.01  Gender  0.12  0.15  0.03  0.09  0.11  0.032  –0.35  0.31  –0.07  Age  –0.00  0.00  –0.01  –0.00  0.00  –0.05  –0.03  0.01  –0.11  Marital status  –0.12  0.14  –0.03  0.02  0.11  0.01  0.00  0.31  0.00  Discrepancy in perceived providers’ emotional exhaustion  –0.02  0.04  –0.02  0.01  0.03  0.02  –0.05  0.09  –0.04  Discrepancy in perceived personal accomplishments  0.66  0.04  0.63***  0.43  0.03  0.52***  0.35  0.09  0.24***  Discrepancy in perceived depersonalisation  –0.24  0.05  –0.23***  –0.33  0.04  –0.39***  –0.15  0.10  –0.10  R2  0.63***  0.63***  0.15***  N of pairs – service users × service providers = 270. * p<0.05; ***p<0.001. As can be seen, these socio-demographic variables explain 2 per cent of the variance in satisfaction with the agency, 1 per cent of the variance in satisfaction with the service provider and 4 per cent of the variance in perceived changes. However, of all the socio-demographic variables, the only one that made a significant contribution was age and this only to the service users’ perceptions of change. Older service users perceived less change than younger ones. In the second step, the three discrepancies were entered into the regressions. These added 61 per cent to the explained variance in satisfaction with the agency, 62 per cent to the explained variance in satisfaction with the service provider and 11 per cent to the explained variance in perceived changes. Only two of the three discrepancies, however, made significant contributions to the users’ satisfaction: the service users’ perceptions of their providers’ accomplishments and their perceptions of their providers’ depersonalisation. The more the users viewed their service providers’ accomplishments as greater than the providers themselves did, the more satisfied the users were with their providers and agency. The more depersonalised the service users perceived their service providers to be in comparison to the providers’ own perceptions, the less satisfied the users were with their service providers and agency. With regard to perceived changes, the contribution of age disappeared and the only discrepancy that made a significant contribution to the variance was the discrepancy in their views of the service providers’ accomplishments. The larger the discrepancy, the greater the perceived change the service users reported in their presenting problem. In other words, the more the service users viewed their service providers’ accomplishments as greater than the service providers themselves did, the more change the users reported in their presenting problem. Discussion Overall, the service users and service providers drew a similar picture of the latter’s burnout. The service providers reported moderate emotional exhaustion, but rather low depersonalisation and a fairly high sense of personal accomplishment. The service users, too, viewed the providers as being moderately emotionally exhausted, as not depersonalising them and as attaining a fairly high level of personal accomplishment. With this, the findings show significant differences in their perceptions. The service users viewed their providers as less emotionally exhausted, as having a lower level of accomplishment and as more depersonalising than the providers viewed themselves. Put differently, the findings suggest that the service users’ view of their service providers is somewhat more critical than the view that the providers have of themselves and, moreover, that the service users do not realise the extent of the emotional burden that the service providers feel. Either or both of two explanations may account for their non-realisation. One is that the service users are focused not on the provider, but on their own needs. The other is that the service providers successfully hide their emotional exhaustion and sense of burden. The findings also indicate a certain incongruity within the service users’ own perceptions. That is, while they rated their satisfaction with the service providers and the service in the high moderate range, they rated the changes in their presenting problem only in the moderate range. In other words, both their satisfaction with their service provider and their satisfaction with the agency were somewhat greater than the changes they perceived in their presenting problems. Since persons presumably request services in order to solve their presenting problems, one may wonder what the service users were satisfied with. Two non-exclusive answers may be suggested. One is that they were satisfied with a partial resolution of their presenting problems, or with the resolution of some other problem that they brought up only in the course of their counselling. The other is that they derived satisfaction from the attention they received from their service providers. The latter possibility is consistent with findings in the literature showing that service users value being listened to and treated with respect (Cahalane, 1997; Ribner and Knei-Paz, 2002; Winefield and Barlow, 1995). Two of the three discrepancies in the burnout perceptions contributed significantly to the explained variance in the outcome variables. Namely, the gap between the providers’ and users’ perceptions of the providers’ personal accomplishment contributed significantly to the users’ satisfaction with the agency and provider, as well as to the users’ perceived changes in their presenting problem. The more the users viewed their providers as more accomplished than the providers viewed themselves, the more satisfied the users were and the more positive changes they reported. The discrepancy in perceived depersonalisation worked in the opposite direction. The more the users viewed their providers as depersonalising them compared to how the providers viewed themselves, the less satisfied they were with the service provider and agency. These findings provide further evidence of the importance emphasised in the literature for shared perceptions between client and therapist (Cooley and Lajoy, 1980; Corning et al., 2007; Kivlighan and Arthur, 2000; Reis and Brown, 1999; Tryon and Winograd, 2011). The discrepancy in the perceptions of the service providers’ emotional exhaustion was not associated with any of the outcomes. This is surprising in view of the deleterious effects of service providers’ emotional exhaustion on their ability to connect emotionally to their clients, as well as of the importance generally attributed to that connection. Nonetheless, the finding both reinforces and may be explained by the apparently greater importance to clients of service providers’ professional ability, sympathy and understanding (Horvath and Symonds, 1991; Howard et al., 2006; Ribner and Knei-Paz, 2002; Taber et al., 2011; Tryon et al., 2007). The study has two main limitations. One is its use of a non-representative sample drawn from a single agency in one city. This means that further study using a representative sample of social workers from a variety of agencies and locations is needed before we can generalise from the results. The other limitation is the study’s cross-sectional design, which precludes drawing conclusions about causality. The findings do not enable us to determine whether the discrepancies in perceptions led to the users’ dissatisfaction or their dissatisfaction led them to view their providers less favourably than the providers viewed themselves. To resolve this issue, longitudinal study is recommended. Its limitations notwithstanding, the findings have implications for theory, research and practice. In terms of theory, the findings reinforce the importance attributed to congruence of perceptions between service users and service providers to achieving positive outcomes. With this, however, they suggest that when congruence is incomplete, as was the case in this study, the incompleteness itself may matter less than the direction of the difference. That is, when service uses view their providers as being more accomplished and less depersonalising than the providers themselves, they may still be quite satisfied with the services they receive and perceive them as effective. In terms of research, to the knowledge of the authors, this study is the first examination of the discrepancies in service providers’ and service users’ perceptions of providers’ burnout and of the associations between the discrepancies and service users’ satisfaction and perceptions of change. Further study is needed, however, to better understand the implications that the direction of the incongruent perceptions and evaluations may have for service users’ satisfaction and perceptions of treatment effectiveness. In terms of practice, the findings have several implications. They underscore the importance of finding ways and implementing means to reduce the risk of service provider burnout, not only for their own sake, but also for the good of their service users. In addition, they suggest that it may be worthwhile for practitioners to check their service users’ perceptions of the quality and effectiveness of the services they receive, as well as of their treatment by the service provider. That is, do they view the provider as engaged and giving or as distant and objectifying them? To engage in such needed discourse, service providers should offer more collaborative and participatory service provision, in which the service users are partners in the intervention process. References Acker G. M. ( 2010) ‘ The challenges in providing services to clients with mental illness: Managed care, burnout and somatic symptoms among social workers’, Community Mental Health Journal , 46, pp. 591– 600. Google Scholar CrossRef Search ADS PubMed  Boyer S. L., Bond G. R. ( 1999) ‘Does assertive community treatment reduce burnout? A comparison with traditional case management’, Mental Health Services Research , 1, pp. 31– 45. Google Scholar CrossRef Search ADS   Cahalane H. G. ( 1997) ‘Relationship between family satisfaction and outcome for children and adolescents in inpatient psychiatric treatment’, Dissertation Abstracts International , 57( 12-A), US, University Microfilms International. Carney J., Donovan R., Yurdin M., Starr R. ( 1993) ‘ Incidence of burnout among New York City intensive case managers: Summary of findings’, Psychosocial Rehabilitation Journal , 16, pp. 25– 38. Google Scholar CrossRef Search ADS   Cooley E. J., Lajoy R. ( 1980) ‘ Therapeutic relationship and improvement as perceived by clients and therapists’, Journal of Clinical Psychology , 36, pp. 562– 70. Google Scholar CrossRef Search ADS PubMed  Corning A. F., Malofeeva E. V., Bucchianeri M. M. ( 2007) ‘ Predicting termination type from client-therapist agreement on the severity of the presenting problem’, Psychotherapy: Theory, Research, Practice, Training , 44, pp. 193– 204. Google Scholar CrossRef Search ADS   Corrigan P. W. ( 1990) ‘ Consumer satisfaction with institutional and community care’, Community Mental Health Journal , 26, pp. 151– 65. Google Scholar CrossRef Search ADS PubMed  Cummings A. L., Martin J., Hallberg E. T., Slemon A. ( 1992) ‘ Memory for therapeutic events, session effectiveness, and working alliance in short-term counseling’, Journal of Counseling Psychology , 39, pp. 306– 12. Google Scholar CrossRef Search ADS   Horvath A. O., Symonds B. D. ( 1991) ‘ Relation between working alliance and outcome in psychotherapy: A meta-analysis’, Journal of Counseling Psychology , 38, pp. 139– 49. Google Scholar CrossRef Search ADS   Horvath A. O., Del Re A. C., Flückiger C., Symonds D. ( 2011) ‘ Alliance in individual psychotherapy’, Psychotherapy , 48, pp. 9– 16. Google Scholar CrossRef Search ADS PubMed  Howard I., Turner R., Olkin R., Mohr D. C. ( 2006) ‘ Therapeutic alliance mediates the relationship between interpersonal problems and depression outcome in a cohort of multiple sclerosis patients’, Journal of Clinical Psychology , 62, pp. 1197– 204. Google Scholar CrossRef Search ADS PubMed  Kim H., Ji J., Kao D. ( 2011) ‘ Burnout and physical health among social workers: A three-year longitudinal study’, Social Work , 56, pp. 258– 68. Google Scholar CrossRef Search ADS PubMed  Kivlighan D. M., Arthur E. G. ( 2000) ‘ Convergence in client and counselor recall of important session events’, Journal of Counseling Psychology , 47, pp. 79– 84. Google Scholar CrossRef Search ADS   Lizano E. L. ( 2015) ‘ Examining the impact of job burnout on the health and well-being of human service workers: A systematic review and synthesis’, Human Service Organizations: Management, Leadership & Governance , 39, pp. 167– 81. Google Scholar CrossRef Search ADS   Maslach C., Jackson S. E. ( 1981) ‘ The measurement of experienced burnout’, Journal of Organizational Behavior , 2, pp. 99– 113. Google Scholar CrossRef Search ADS   Maslach C., Jackson S. E. ( 1986) Maslach Burnout Inventory , 2nd edn, Palo Alto, CA, Consulting Psychologists Press. Maslach C., Jackson S. E., Leiter M. P. ( 1996) MBI: The Maslach Burnout Inventory: Manual, Palo Alto, CA, Consulting Psychologists Press. McCarthy W. C., Frieze I. H. ( 1999) ‘ Negative aspects of psychotherapy: Client perceptions of psychotherapists’ social influence, burnout, and quality of care’, Journal of Social Issues , 55, pp. 33– 50. Google Scholar CrossRef Search ADS   McMurtry S. L., Hudson W. W. ( 2000) ‘The client satisfaction inventory: Results of an initial validation study’, Research on Social Work Practice , 10, pp. 644– 63. Google Scholar CrossRef Search ADS   Parrish D. E., Rubin A. ( 2012) ‘ Social workers’ orientations toward the evidence-based practice process: A comparison with psychologists and licensed marriage and family therapists’, Social Work , 57, pp. 201– 10. Google Scholar CrossRef Search ADS PubMed  Reis B. F., Brown L. G. ( 1999) ‘ Reducing psychotherapy dropouts: Maximizing perspective convergence in the psychotherapy dyad’, Psychotherapy , 36, pp. 123– 36. Google Scholar CrossRef Search ADS   Peterson V., Demerouti E., Bergström G., Samuelsson M., Asberg M., Nygren A. ( 2008) ‘ Burnout and physical and mental health among Swedish healthcare workers’, Journal of Advanced Nursing , 62, pp. 84– 95. Google Scholar CrossRef Search ADS PubMed  Ribner D. S., Knei-Paz C. ( 2002) ‘Client’s view of a successful helping relationship’, Social Work , 47, pp. 379– 87. Google Scholar CrossRef Search ADS PubMed  Savaya R., Altschuler D., Melamed S. ( 2013) ‘ Apprehensions and expectations of the adoption of systematically planned, outcome oriented practice’, Research on Social Work Practice , 23, pp. 437– 46. Google Scholar CrossRef Search ADS   Savaya R., Bartov Y., Melamed S., Altschuler D. ( 2016) ‘ Predictors of perceived changes by service users: Working alliance, hope, and burnout’, Social Work Research , 40, pp. 183– 91. Google Scholar CrossRef Search ADS   Taber B. J., Leibert T. W., Agaskar V. R. ( 2011) ‘Relationships among client-therapist personality congruence, working alliance, and therapeutic outcome’, Psychotherapy , 48, pp. 376– 80. Google Scholar CrossRef Search ADS PubMed  Taris T.W. ( 2006) ‘ Is there a relationship between burnout and objective performance? A critical review of 16 studies’, Work & Stress , 20, pp. 316– 34. Google Scholar CrossRef Search ADS   Tryon G. S., Winograd G. ( 2011) ‘ Goal consensus and collaboration’, Psychotherapy , 48, pp. 50– 7. Google Scholar CrossRef Search ADS PubMed  Tryon G. S., Blackwell S. C., Hammel E. F. ( 2007) ‘ A meta-analytic examination of client- therapist perspectives of the working alliance’, Psychotherapy Research , 17, pp. 629– 42. Google Scholar CrossRef Search ADS   Winefield H. R., Barlow J. A. ( 1995) ‘ Worker satisfaction in a child protection agency’, Child Abuse & Neglect , 19, pp. 897– 905. Google Scholar CrossRef Search ADS PubMed  © The Author 2017. Published by Oxford University Press on behalf of The British Association of Social Workers. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

de Jong, Gideon;Schout, Gert;Abma, Tineke

doi: 10.1093/bjsw/bcx037pmid: N/A

Abstract From 2011 until 2013, forty-one family group conferences were organised for clients in a public mental health care (PMHC) setting in the north of the Netherlands. In total, 312 semi-structured interviews were done out of a possible number of 473 Family Group Conferencing (FGC) participants. A multiple case study brought four dynamics to the surface on the process of FGC in PMHC: (i) overcoming resistance, breaking through isolation, sharing shameful feelings and grievances; (ii) you’ll change for your mother, not for professionals; (iii) the complex role of FGC coordinators; and (iv) professionals who can resist the temptation to take over. To understand these dynamics, the role of the social network as ‘shock absorbers’ and the necessity of a collaboration between FGC coordinators and professionals are further explored. Eventually, the impact of FGC on the client’s quality of life is influenced by four factors, namely if: (i) clients are willing to invite and extend their social network; (ii) clients and their network are willing to share shameful feelings and grievances; (iii) there is mutual trust between clients and FGC coordinators and (iv) professionals reinforce the self-direction of the group and prevent clients from falling back into individual care trajectories. Family Group Conferencing, multiple case study, public mental health care, responsive evaluation Introduction In recent years, Family Group Conferencing (FGC) has increasingly become popular in the Netherlands. The rise of FGC can be seen as the forerunner of various initiatives that entail changes in the relationship between the government and citizens. Also, in several other Western countries, a reconsideration is taking place of what can be expected from the welfare state and how resources from the civil society can be mobilised (Blond, 2010; Pennell, 2006). In the Netherlands, this is reflected in the introduction of the Social Support Act (‘Wet maatschappelijke ondersteuning’) and Social Welfare New Style (‘Welzijn Nieuwe Stijl’) and the emergence of social interventions such as ‘wraparound care’ and ‘community support’. In our study, we focused on what happens when people who have a history of disempowerment and state interference—namely the client group of public mental health care (PMHC)—are given the chance to establish a FGC plan with the help of their social network. Verstehen was the goal of our study (Schwandt, 1999): gaining an understanding from the perspectives’ of insiders of how FGC proceeds in PMHC—a field with limited experience in FGC—and what this decision-making model can offer to socially isolated people who have languished for a long time, and whether it helps solving liveability problems in neighbourhoods. Background Family Group Conferencing A family group conference is a meeting in which a client and his network are stimulated to find solutions to an identified problem and develop a joint plan on their own. In these meetings, the social network of the client are brought together with the formal system of care agencies. This involves mobilising the resilience and capabilities of clients and their social network combined with the expertise of professionals. There are a number of different types of people who may participate in a family group conference. Generally, the client and their family are present, but friends, neighbours and colleagues may also participate. Through democratic decision making, a plan is developed which each participant of the conference needs to agree on. The aim of FGC is to create an environment in which all participants are able to share their ideas for possible solutions to an identified problem, as well as being able to express any dissatisfaction or grievances. The outcome of the conference is a plan that describes the roles and responsibilities of the different people involved in the conference (De Jong and Schout, 2011). The conference involves three stages. First, information is shared between all participants, including professionals. The second stage is the private family time. Supplied with advice from professionals, the client and the family draw up a plan to address the problem. In the last stage, the plan is shared with professionals and the FGC coordinator. The FGC coordinator is a fellow citizen, who preferably has no background in social work or other care fields, who positions himself as organiser, listener, reliever and encourager (Natland and Malmberg-Heimonen, 2014). The coordinator ensures that the right people are invited, while finding a balance between supporting the aims of FGC and respecting the wishes of their clients. Unlike professionals, coordinators do not have an interest in the outcome of the conference. Once the conference is completed, their role comes to an end. Public mental health care Clients in PMHC often have to contend with multiple problems. Various social institutions use a variety of descriptors for this group: problem tenants, people who avoid care, trouble makers within neighbourhoods, addicted homeless, homeless youth and people who are deteriorating in silence. Often, clients are vulnerable and do not possess the capabilities to hold their own ground in society. Generally, they lack (the help of) a social network. They often do not use existing facilities and services. Some clients may recognise their need for help, but often do not know where to access services (Schout et al., 2011). In the Netherlands, the PMHC client group is defined according to five criteria: the presence of a psychiatric disorder (including addiction), or severe psycho-social problems; the presence of multiple problems in several areas at the same time; leading to an insufficient provision in their living conditions (housing, income, social contacts, self-care and so forth); a lack of ability to solve problems themselves; and an absence of a proper demand for help. The challenge for PMHC professionals is to provide help to those who apparently do not want to be helped. PMHC is not a traditional agency with an office and paid staff, but rather a safety net consisting of collaborating mental health professionals from different disciplines (usually social workers and community mental health nurses, headed by psychiatrists). Understanding the social environment PMHC clients are often deprived of social resources. They may experience a number of social problems such as domestic violence, addiction, social isolation and neighbourhood conflicts. Therefore, FGC is deployed in social environments which are characterised by a lack of resources, inequities in participation across class, ethnicity and education levels, and the difficulties of engaging in complex services (e.g. Heitzmann et al., 2009). There may also be a lack of willingness of clients to contribute to the realisation of collective goals, declining civic engagement, loosening connections and the loss of associations (e.g. Lorentzen and Hustinx, 2007). Given this social environment, we wanted to explore whether PMHC clients who receive limited help from an often reduced and broken social network could still benefit from the social resources which are mobilised with the help of FGC. Additionally, PMHC clients often lack motivation to seek help and therefore avoid mental health services. We were interested in exploring whether FGC could help in overcoming this lack of motivation and stimulate clients to accept social and professional support. Third, we wanted to examine whether FGC could help in breaking the cycle of deterioration and languishing experienced by PMHC clients, ultimately restoring contact with family, friends and neighbours or extending existing social networks with new contacts. The study Research design From 2011 until 2013, forty-one family group conferences were organised for clients in a PMHC setting in the north of the Netherlands. Each conference was examined using a case-study design. We considered each conference as a case which was unique and complex. We were interested in the social context in which the conference was held, and in patterns explaining its process and outcomes (e.g. Stake, 1995). We used a responsive evaluation approach to integrate various stakeholder perspectives (client, social network, FGC coordinator, professionals) and give voice to the ‘silenced voices’ (Abma, 2005). PMHC clients often have a history of disempowerment and their perceptions are barely taken into account by professionals or researchers. Population, recruitment and selection We carefully instructed potential referrers in which situations FGC could be deployed. These included the following: when the social network should be extended or its co-operation improved; when clients are not motivated for treatment and FGC can offer an alternative; when threatened housing evictions, child protection measures and involuntary admissions need to be averted; in troubled living conditions where the capabilities of the extended network should be mobilised; when liveability problems and conflicts with neighbourhoods should be solved. Each of the twenty-three PMHC networks in the province of Groningen was asked to refer cases to our project. A relatively high number of seemingly hopeless cases were included in this study. When clients were motivated for FGC, their case managers referred them to the FGC regional manager who, in addition, appointed a FGC coordinator to plan a first exploratory meeting with the client. In almost all of the forty-one cases, psycho-social problems were paramount, usually interwoven with psychiatric and addiction problems. In eleven cases, the problem situation revolved around neighbourhood conflicts. These cases revolved around several actors (versus individual clients) who were in conflict with each other. Research participants For each case, everyone who participated in the conference and/or could reflect on the situation was approached for an interview. Four stakeholder groups can be distinguished: (i) main actors (individual clients, couples, families, community members); (ii) participants from the social network (family members, friends, neighbours, concerned bystanders); (iii) professionals (everyone with a professional background, such as social workers, community mental health nurses, policy workers, counsellors of social housing associations and police officers) and (iv) FGC coordinators. When many participants took part in a conference (more than fifteen), then respondents were selectively approached (purposive sampling; see Silverman, 2013). Ensuring diversity of stakeholder perspectives was important. We therefore provided a sufficient spread of respondents from the four stakeholder groups. Data collection Semi-structured interviews were completed at sites designated by respondents (home, work or a neutral environment) and at times which were convenient to them. The interviews were recorded and transcribed verbatim. In the interviews, respondents were asked to reflect retrospectively on the process of the conference, including four topics: (i) a description of the referral; (ii) the preparation stage; (iii) the process of the conference itself (information sharing, private family time, decision making and establishing the plan) and (iv) the implementation of the plan (see Hayes and Houston, 2007, p. 995). The exploration of the conference included multiple subtopics such as a description of the problem situation, expectations prior to the conference, the decision-making process during the conference and the role of the FGC coordinator. The total number of semi-structured interviews in the forty-one case studies was 312 (an average of seven or eight respondents per case) out of a possible number of 473 FGC participants. On average, interviews lasted sixty to ninety minutes, ranging between fifteen minutes and four hours. Sometimes, respondents were interviewed a second time when new insights emerged during a later stage of the case study. Data analysis There were two parts to the analysis. First, each conference was analysed separately, resulting in an individual case-study report. The purpose of each case study was to assess whether the conference had a positive effect or not, and what the specific patterns were that helped in understanding its success or failure. Second, we carried out qualitative meta-analyses to reveal common patterns that could be observed in several cases. In this article, we only reflect on findings from the meta-analyses. During the meta-analyses, we used insights from the multiple case-study approach (see Stake, 2006) to reveal patterns in several cases, and explore similarities and differences between cases. Patterns were discussed between all authors and further refined, and they finally resulted in four dynamics that are highlighted in the findings section. As Baxter and Jack (2008) emphasise, it is not a strict rule within the multiple case-study approach that contexts in several cases should be completely similar to formulate conclusions that account for several cases at the same time. In that sense, it was possible for us to observe similarities between analysed family group conferences in which the main actors had different backgrounds and problems. Measures for quality and rigour Involving various stakeholders is an indication of credibility in case-study research (Small, 2009). Triangulation was used to enhance credibility and reliability. In addition to interviews, field notes were written to broaden the scope and check data. A logbook was used by the researchers to reflect on their assumptions. The conferences were analysed using inter-coder agreement (e.g. Silverman, 2013), which means that two researchers analysed the data and agreed on the coding that would be assigned to a certain interview section as well as the clustering of several codes under the umbrella of one theme (dynamic). The outcome codes and themes were also discussed with the research supervisor. In each case study, we carried out group member checks where we presented interim findings from the interviews to respondents (see Guba and Lincoln, 1989), both for validation and to broaden the scope. These meetings were organised at a time when the majority of respondents could be present, with the strict condition that at least the main actors (clients) were given the opportunity to reflect on interim findings. A total of 144 participants took part in the forty-one group member checks. Each member check lasted ninety minutes; on average, three or four participants took part in each one. Strengths and limitations of the study The various case studies have a strong ecological validity, namely that the research methods were aligned and connected with the field of research so that real-world phenomena could be articulated (e.g. Schmuckler, 2001)—in the case of our study: the process and outcomes of the forty-one family group conferences. Because we were aware of the necessity for gaining respondents’ trust, we created conditions in which the respondents felt at ease to share their perspectives. This, however, also had an impact on the reliability of the study results: mutual trust and strong bonds between respondents and researchers influence the extent to which researchers objectively gather and interpret data. This was captured in each case study as: (i) the interviews were organised using a standardised topic list; (ii) the interviews were recorded and later on transcribed verbatim; (iii) the interviews and analyses of interview findings were co-checked between several researchers; (iv) memos were tracked in a logbook on specific twists and turns during the process of data gathering and analysis and (v) interim findings were validated during a group member check. Ethical considerations PMHC is a network organisation which operates in an ethical-laden context. The contributing organisations have their own ethics committees. Executives of these organisations formed an advisory board to assess the legal and ethical implications of the study and approved the procedures. The following safeguards for privacy, anonymity and repeatability of information were taken (Silverman, 2013). All names of persons, streets, places and institutions have been replaced by a unique code. We were aware of the ethics of confidentiality, and therefore respondents in the same case study were not informed about the content of the other interviews. Each case-study report was shared with respondents, but not distributed to other parties. All data files are archived and will be kept private until 2018 and then subsequently deleted and erased. Finally, all researchers have signed a confidentiality agreement that meets the ethical standards of client-related research. Findings The multiple case-study analysis revealed four dynamics (or themes) that help to understand the process of FGC in PMHC. Overcoming resistance, breaking through isolation, sharing shameful feelings and grievances Clients were often reluctant to invite members from their social network, who potentially could have had a major contribution to the conference, or with whom it would have been fruitful to restore contact with, to the FGC: Later, accidently, I had contact with B [client] and then he said: ‘My brother was standing at the door.’ ‘Your brother?’, I asked. ‘Do you still have contact with your brother?’ ‘Yes, he was waiting in front of the door. He had heard that I was admitted to a clinic and he was waiting at the door unannounced.’ I asked him: ‘And now?’ He responded: ‘Yeah, well, a lot needs to be discussed.’ Nothing else came out of that. He said finally: ‘I certainly do not want my brother or sister to be involved’ (Ex-colleague client, Case 5). Clients sometimes deliberately excluded people from participating in the conference. Some cases indicate that it is unlikely that FGC will produce progress when clients are consistently refusing to invite those to a conference with whom restoring relationships is actually recommended, especially with clients who lost contact with their family and friends as a result of their addictive lifestyle: You can do it [organising a conference] with your family or you can do it with friends and neighbours. Thus I first thought positive about that. Who should then be invited? Family was not any longer involved. … The coordinator still asked about them. I said: ‘No! I also don’t want to talk about it [inviting his family for the conference] any longer. I am done with them!’ (Client, Case 7). The same man—who is an exemplar of other clients in our study—was overall very difficult to convince of the necessity to let people participate in the conference: I wanted the professional of the housing assistance services to participate. He is the person who can activate him and who can figure out what could be a suitable job for him. But of course he was hesitating. Because, just imagine: he should work again! Now, every week he can tell his neighbour that he still feels ill and that he is still not able to work again (Case manager addiction care, Case 7). An underlying pillar of FGC is to empower citizens to make their own decisions. FGC coordinators, however, often needed to act in an arena of tensions where they were continuously forced to decide whether to assist clients in fulfilling their wishes to include and exclude particular members from their social network from participating in the conference, or to encourage clients to critically reflect on the composition of participants: Then I received from the police the names and addresses of the main actors [young people committing troubles in the neighbourhood], both from the side of the perpetrator and the victim [a young man threatened a girl and her boyfriend]. Those people I visited at home. I spoke with the young people, but also with their parents. I continuously listened, and during these discussions at their homes you hear names of the other young people who are also involved. My task was to involve as many actors as possible. I spoke with all these people, and all their stories had a sense of truth. … I’ve visited all these families, talked with everyone and listened. … It were particularly the fathers who could reflect neutrally on the problems and who were open for a conference to solve the problems [both the young people and their mothers acted very emotional and angry towards the other side]. … Finally I brought the fathers of both sides together and let them come to a mutual agreement (FGC coordinator, Case 28). Many clients in PMHC are socially isolated and deprived of help from a social network. Usually, a small network is considered a contra-indication for FGC. However, a prominent result of our study is that social isolation is actually a reason for organising FGC in PMHC. Our study illustrates that FGC has the potential to restore damaged and diluted contacts between clients and their social network, so that downward spirals of deterioration can be broken. This is specifically important to PMHC, as a large part of the population finds itself in isolated and neglected conditions. Some cases reveal that, even after a conference, relationships with people who did not participate can be restored. For example, a man who was addicted to alcohol and lived in deplorable circumstances was able to remain sober after the FGC paid attention to his household and did not relapse into destructive behaviour (De Jong and Schout, 2013). After this first case-study report, we interviewed the man again. It appeared that contact with his daughter, whom he had not seen for more than seven years, was restored. He told us that his daughter stated that she always loved him as a father, but that she broke contact after he started drinking excessively to cope with his divorce, eventually becoming addicted and not being able to fulfil his role as a father any longer. The daughter did not want to face the pain and misery that accompanied his deterioration and decided to cut all ties with him. It was partly because the man felt so ashamed about his deplorable living situation that he did not dare to invite former loved ones to his house any longer. Not only were father and daughter reunited; one year after the conference, the man also met his grandchild for the first time in his life. He repeated many times that he does not want to fall back into languishing and the shame that comes along with it. The role of shame as an engine for the prevention of relapse emerged in both the conferences on marginalised living conditions of socially isolated clients as well as in conferences on liveability problems. FGC holds the potential to provide a platform on which victims feel at ease to share their displeasure and shameful feelings: ‘Everyone had the chance to cry out his heart. People were sincerely listening to each other’ (Resident, Case 16). It also creates opportunities for participants to express their inconvenience with troublemakers, while troublemakers have the chance to explain their side of the story: It was a sore point to discuss. It was such an issue that eventually both S [client] and C [client’s ex-girlfriend who was overburdened taking care of her] started to cry. The finger was truly put on the salient point. And it became clear for everyone that it was just too much for C. I consider that aspect was really good about the conference. … If the atmosphere would had been different, I perhaps would not had dared to be so open. Everyone was very open (Friend of client, Case 33). In this type of situation, clients can explain why they have languished for such a long time, creating understanding among their social network for how they ended up in such deplorable living conditions. We saw that the FGC process ultimately led to a recovery of relationships and paved the way for a new beginning of meaningful contact with the social network. The case studies reveal that FGC is a useful intervention to improve the lives and living conditions of clients in PMHC. The usefulness of FGC depends on clients’ willingness to invite enough social and diverse resources for the conference, coordinators encouraging clients to think critically about the best composition of possible participants and when conditions are created in which people feel safe to share shameful feelings and grievances. You’ll change for your mother, not for professionals Many clients in our study did not want to change for professionals, but were willing to change for their family, friends and even former colleagues. Their loved ones could use harsh words that professionals would (and legally could) never employ, but that are sometimes needed to encourage clients to change their behaviour for the better. Several of our cases demonstrated that, when the atmosphere at home was tense and threatening, the organisation of a meeting in a neutral environment with a larger group of participants created conditions for progress that could never be reached with only professional help. One case in particular exemplified the usefulness of FGC in allowing family members the space to tell clients about the impact of their behaviour on those around them. The case involved two brothers in their twenties in a meeting with extended family members. The aunt and stepsister—who had not been present before when issues were discussed in informal meetings between the brothers and their mother and stepfather—were the ones who were able to confront the brothers about their unacceptable behaviour (drug use, drinking, criminal acts, stealing from the household). They had the ability to escalate when they deemed this necessary (eventually to such an extent that the youngest brother burst into tears): I said to them: ‘When are you both finally gonna say sorry to my sister?! When are you finally gonna do so?! You have treated her like a beast! Why are you making the life of my sister such a misery?!’ (Sister of the mother, Case 38). The stepdaughter also started yelling: ‘Why are you not working?! Why is the police all the time getting involved?! Why are you troubling the life of my father and your mother?!’ That was really tough, but very necessary (Mother, Case 38). This was the moment that the sons finally realised the impact of their behaviour, as the youngest brother described: When my mother started crying, I felt pity for her. And finally I thought: something needs to be changed, because I have never heard my mother’s side of the story. At home you always leave the kitchen table angrily and you never hear of the other side. … I looked at my brother, and thought: shit, you need to convince it all to them. Then they got to know why we had so many debts. Well, that was because of cannabis use, buying liquor, partying every night. … Before I always kept this secret, I never told anyone about this. … And when I finally told them, I felt relieved (Youngest brother, Case 38). During a conference with extended family members, it is difficult for participants to escape from the consequences of their behaviour. This is because members from the social network can use words that could never be used by professionals, but that are sometimes needed to let clients realise the misery they have caused to others. Indeed, clients in our cases wanted to change for their family rather than for professionals. Family and friends can also absorb shocks when emotions are getting intense, and can prevent impasses. Our study revealed that confrontation with the misery that has been caused to others is difficult for clients, but that it also lays the foundation for recovery. The complex role of FGC coordinators FGC coordinators frequently struggled with their role as facilitator (who aim to give clients the power to determine their own life), especially when clients remained passive and the social network was kept away: It cannot be the case that the coordinator is the only one who ensures that everything will be fine. They [referring to the conference’s participants] should ensure by themselves it would be okay. But I am already involved with the people during the process, as well as with M [client] and her daughter. And I see what they want and what they can do, but I also see that they still cannot live together. And, well, I feel very powerless as coordinator: incomplete and insufficiently equipped to do something about it. But is it my role to do something? No, my role is to be involved for a little while. So it is a constant balancing (FGC coordinator, Case 24). In many cases, the impact and the quality of the plan were linked to the role of the coordinators. Their role as independent fellow citizens who were not making plans for families, but empowering them to come up with their own plan, was highly appreciated by clients, their network and professionals: For that matter, I do need to give that man a compliment, because I know how difficult such situations can be, and sometimes [as a coordinator] you have to be very tough: or you are gonna talk, or you need to act more directly. And that man really found the golden mean … . For that I have a lot of respect (Father, Case 28). She knows how to act diplomatically and how to start a conversation with the people involved tactically. She creates proximity and commitment. But she also knows when to keep distance, in the sense of: ‘My role ends here’ (Social network, Case 11). She also visited all families at their home. She had no judgment, and that was very pleasant. She did not react like: ‘But you did do this, and you did do that!’ She did perform as a care giver. And that whole different approach I consider very effective because the people involved therefore trusted her. Hence she was able to act as an efficient chairman, and consequently people really listened to what she had to say, in the sense that she was not choosing sides … . And therefore I considered she fulfilled her role really well (Social worker, Case 15). Coordinators were easily accessible, even during evenings and on weekends, and they had a non-judgemental attitude towards the situation. They often fulfilled the role clients wanted their care providers to have taken. Several of the cases we analysed reveal that clients who had a troubled history with representatives of the professional society had less trouble with representatives of the civil society. As coordinators were seen as independent and free of ties with agencies, they could also use words that a professional could not use: One time he wanted to quit [the client lost motivation for the conference]; he had a breakdown and was no optimistic anymore about the process. He called me and said: ‘I want to quit and I’d like you to come over right now because I want to clarify this to you.’ I told him that I’m not gonna spend my time on that: ‘If you stop, fine, but I won’t come to you right now. I’m not gonna drive one hour to hear what you already told me. I don’t care what your reason is, I’m not gonna spend my time on that. So I will hang up the phone, and if you tell me that it is over, then it is.’ ‘Wait, wait!’, he responded. That really gave him an enormous drive to continue with the preparation towards the conference, like: ‘Damn, I lose things, this is quicksand.’ So I think that worked really well for him. From that moment he was way more motivated (FGC coordinator, Case 7). Mutual trust between coordinators and clients appeared to be important in the cases we analysed. For example, in the case we described of the man in his sixties who had been addicted to alcohol, who was living in isolated circumstances and who had languished for a long time (Case 5; see also De Jong and Schout, 2013), the role of the coordinator was crucial. From the first moment, the man had trust in the coordinator, probably also as there was a minimum age difference between them. In other cases, it was striking that, if there was no mutual trust, or when coordinators did not collect feedback on their own performance, the likelihood of a successful conference and sustainable outcomes decreased. Professionals who can resist the temptation to take over Most of the professionals we interviewed in the forty-one cases were social workers and community mental health nurses. Some of them had contact with their clients over a long period during which they saw little to no progress. They expressed doubts about whether, under the messy circumstances of client systems in PMHC, FGC could create momentum for change: I had expected that it would be clear for everyone what their position and role would be and how we would proceed [towards the conference]. For me it was a bit like: the neighbours do their thing and we as professionals do our thing, but we do not thwart each other. … It did not go well. I actually found it [the conference] from the start, yet again, just a repetition. But it went on, it was already getting a bit personal. And that is not a family group conference, but an evaluation of his [the client’s] treatment (Community mental health nurse 2, Case 29). A colleague of this nurse also shared her doubts about whether the conference could result in a workable plan: They believed that A. [client] suddenly after three months would use his pills. … We have known A. for fifteen years and we never have succeeded in that period, so why would it now after three months suddenly succeed? (Community mental health nurse 1, Case 29) We observed that it was sometimes hard for professionals to resist the temptation to organise individual care trajectories when FGC plans were likely to fail. That professionals are tempted to take over care is understandable. Often they have chosen their profession because they want to help others. With the policy transformations in the Netherlands, they are forced to ensure that clients take care of themselves first before calling in professional help. Professionals need to share responsibilities and cede power to clients and their social network, and it is understandable that they struggle with this process. Discussion Our empirical findings reveal two central outcomes. We reflect on these in conjunction with the research literature and provide implications for practice. The social network as ‘shock absorbers’ Our findings demonstrate that FGC in PMHC creates conditions for recovery of contacts within networks, highlighting the problem situation from different angles, as all of those involved are invited to participate, including the client. FGC combines the knowledge of both family and communities with the expertise of professionals. In doing so, an integrated solution with a social base is reached rather than individual care trajectories led by professionals. Our research findings show that, in PMHC, the involvement of the social network produces progress, because they can confront clients about their unacceptable behaviour, but also mobilise socio-emotional support and provide a sense of belonging. The network functions as a ‘shock absorber’ that can prevent the conference from getting stalled when conflicts break out and can keep clients motivated to work towards a plan. During our years of investigation, we often heard the question: What is the added value of FGC over a professional who helps clients solve their problems through an informal meeting at the kitchen table? Critique was raised on why such a conference should be organised when it is apparently cheaper to discuss and solve problems under the watchful eye of a professional. In recent years, professionals in the Netherlands are encouraged to clarify clients’ demand for care, and in addition assess their self-reliance and possible support from their social network before it is decided which professional help should be applied for. In situations where problems have got out of hand, when the situation is unclear and no one has an overview, when interests collide or when too many actors are involved, it is difficult to address and strengthen the client’s capabilities by just an exploratory meeting between the client and professional (this was especially true in Case 38 with the two brothers). Such a meeting holds the risk that a dialogue between clients and professionals never occurs as professionals push through their own agenda (e.g. Burns and Früchtel, 2014; Garrett, 2009, 2010; Hayes and Houston, 2007). Professionals are by definition better informed and equipped than clients, so there is a power imbalance (Ney et al., 2013). In particular, with PMHC clients who have a history of disempowerment, there is the risk that they have difficulties in expressing their own wishes in a dialogue with professionals. When clients cannot express themselves, they are often labelled as passive and lacking initiative. Consequently, professionals quickly take over, resulting in clients, once again, being deprived of the space in which to make decisions for themselves. The complexity and multiplicity of problems in PMHC, however, require the involvement of a diverse group (De Jong et al., 2015). The necessity for a collaboration between FGC coordinators and professionals Our study reveals that FGC coordinators are crucial actors in carrying out conferences. This finding is in line with the literature on FGC which shows that the coordinators are the ones who create the conditions where the empowerment of clients and their network can flourish (Connolly, 2006a, 2006b; Macgowan and Pennell, 2002; Natland and Malmberg-Heimonen, 2014). The coordinators who were involved in our research project frequently struggled with their role as independent citizens. The coordinator in Case 24 as described above had difficulties in finding the right balance between remaining the neutral actor who does not judge clients and offering them help when they deem this necessary. It is advisable for FGC coordinators to build a certain degree of sensitivity. On the one hand, it is recommended that clients are motivated to invite certain people for the conference, while, on the other hand, it is better that coordinators share their doubts on the possible composition of the FGC participants. Coordinators need to decide when to ask—what do you need so your family will take part in the conference—and when it is better to leave such a question aside. To establish contact and gain the trust of a client group that is characterised by suspicion and mistrust, tenacity and compassionate interference are required (e.g. Metze et al., 2015). The role and attitude of professionals in the FGC process, and their influence, especially during the preparation of the conference and implementation of the action plan, should not be overlooked. FGC requires a different professional attitude and a task perception that changes into what Gerritsen (2013) describes as ‘egoless care’: professionals who enjoy making themselves redundant and who give clients and their social network the feeling that they solve the problems themselves. What is needed in this new social welfare era is a professional who creates conditions in which clients are not only postulated as the owners of their problems, but also as the owners of the potential solutions. On the other hand, discovering, promoting and encouraging the strengths and positive attributes of clients should not be accompanied by downplaying clients’ impairments and disabilities. The language of FGC is strongly focused on capabilities and opportunities, but it is also important not to ignore or overlook defects and disorders which may impact on the process. Professionals should resist the temptation to organise individual care trajectories when FGC plans (are likely to) fail. Involving support from social networks, and thinking in terms of strengths and capabilities, demand a different professional culture. It is therefore a fundamental challenge to describe the role of professionals in the different stages of FGC with adult clients, and how they can support clients and their network. When a collaboration between coordinators and professionals is initiated, then both are consequently able to benefit from each other’s qualities. Professionals often have a good relationship with clients, but have little insight about their network. FGC coordinators can learn from professionals how to establish contact and gain trust, and accordingly will be accepted and trusted by clients. Professionals can benefit from the coordinators’ qualities to think in terms of ‘widening the circle’ and creatively mobilising possible resources they would not have thought of themselves. FGC coordinators and professionals can complement each other, and subsequently prepare a conference with the potential to reach the best outcome. Conclusion The multiple case study reveals that the success or failure of the process of FGC in PMHC is influenced by four factors, namely when: clients are willing to invite enough social and diverse resources for the conference, and extend their network; clients and their network are willing to share shameful feelings and grievances; there is mutual trust between clients and FGC coordinators; and professionals reinforce the self-direction of the group and prevent clients from falling back into individual care trajectories. If just a few close family members and friends take part in the conference, then the social and emotional support that the client could rely on remains limited. If no outsiders and concerned bystanders participate, a conference lacks creative angles on possible solutions (De Jong et al., 2015). Without connections to the wider community, it is difficult to achieve progress. In the cases where the FGC process successfully resulted in a plan, we saw that coordinators encouraged the clients to critically reflect on who to invite for the conference, and that they were assisted by professionals in how to maintain the motivation of clients and their social networks. When coordinators and professionals work closely together, they ensure that clients do not lose their motivation and fall back into individual care trajectories. Acknowledgements We would like to thank all study participants and the students of the Hanze University of Applied Sciences in Groningen for their contribution to the case studies. The forty-one family group conferences were funded by the province and municipality of Groningen, the Netherlands. For our research we received grants from the Dutch Ministry of National Health, Welfare and Sport (Ministerie van Volksgezondheid Welzijn en Sport), Netherlands Foundation of Mental Health (Fonds Psychische Gezondheid) and NutsOhra. References Abma T. A. ( 2005) ‘ Responsive evaluation: Its meaning and special contribution to health promotion’, Evaluation and Program Planning , 28( 3), pp. 279– 89. Google Scholar CrossRef Search ADS   Baxter P., Jack S. ( 2008) ‘ Qualitative case study methodology: Study design and implementation for novice researchers’, The Qualitative Report , 13( 4), pp. 544– 59. Blond P. ( 2010) Red Tory: How the Left and Right Have Broken Britain and How We Can Fix It , London, Faber and Faber. Burns G., Früchtel F. ( 2014) ‘ Family Group Conference: A bridge between lifeworld and system’, British Journal of Social Work , 44( 5), pp. 1147– 61. Google Scholar CrossRef Search ADS   Connolly M. ( 2006a) ‘ Fifteen years of family group conferencing: Coordinators talk about their experiences in Aotearoa New Zealand’, British Journal of Social Work , 36( 3), pp. 523– 40. Connolly M. ( 2006b) ‘ Up front and personal: Confronting dynamics in the Family Group Conference’, Family Process , 45( 3), pp. 345– 57. Google Scholar CrossRef Search ADS   De Jong G., Schout G. ( 2011) ‘ Family group conferences in public mental health care: An exploration of opportunities’, International Journal of Mental Health Nursing , 20( 1), pp. 63– 74. Google Scholar CrossRef Search ADS PubMed  De Jong G., Schout G. ( 2013) ‘ Breaking through marginalisation in public mental health care with Family Group Conferencing: Shame as risk and protective factor’, British Journal of Social Work , 43( 7), pp. 1439– 54. Google Scholar CrossRef Search ADS   De Jong G., Schout G., Pennell J., Abma T. ( 2015) ‘ Family Group Conferencing in public mental health and social capital theory’, Journal of Social Work , 15( 3), pp. 277– 96. Google Scholar CrossRef Search ADS   Garrett P. M. ( 2009) ‘ Questioning Habermasian social work: A note on some alternative theoretical resources’, British Journal of Social Work , 39( 5), pp. 867– 83. Google Scholar CrossRef Search ADS   Garrett P. M. ( 2010) ‘ “Making social work more Habermasian?”: A rejoinder in the debate on Habermas’, British Journal of Social Work , 40( 6), pp. 1754– 8. Google Scholar CrossRef Search ADS   Gerritsen E. ( 2013) De Wonderen zijn de Wereld nog niet uit [Wonders Will Never Cease], available online at www.binnenlandsbestuur.nl/sociaal/opinie/columns/de-wonderen-zijn-de-wereld-nog-niet-uit.8974453.lynkx (accessed 29 April 2017). Guba E. G., Lincoln Y. S. ( 1989) Fourth Generation Evaluation , Newbury Park, CA, SAGE Publications. Hayes D., Houston S. ( 2007) ‘ “Lifeworld”, “system” and Family Group Conferences: Habermas’s contribution to discourse in child protection’, British Journal of Social Work , 37( 6), pp. 987– 1006. Google Scholar CrossRef Search ADS   Heitzmann K., Hofbauer J., Mackerle-Bixa S., Strunk G. ( 2009) ‘ Where there’s a will, there’s a way? Civic participation and social inequality’, Journal of Civil Society , 5( 3), pp. 283– 301. Google Scholar CrossRef Search ADS   Lorentzen H., Hustinx L. ( 2007) ‘ Civic involvement and modernization’, Journal of Civil Society , 3( 2), pp. 101– 8. Google Scholar CrossRef Search ADS   Macgowan M. J., Pennell J. ( 2002) ‘ Building social responsibility through family group conferencing’, Social Work with Groups , 24( 3–4), pp. 67– 87. Google Scholar CrossRef Search ADS   Metze R. N., Kwekkeboom R. H., Abma T. A. ( 2015) ‘ “You don’t show everyone your weakness”: Older adults’ views on using Family Group Conferencing to regain control and autonomy’, Journal of Aging Studies , 34, pp. 57– 67. Google Scholar CrossRef Search ADS PubMed  Natland S., Malmberg-Heimonen I. ( 2014) ‘ A study of coordinator positionings in family group conferences’, Nordic Social Work Research , 4( 2), pp. 158– 72. Google Scholar CrossRef Search ADS   Ney T., Stoltz J. A., Maloney M. ( 2013) ‘ Voice, power and discourse: Experiences of participants in family group conferences in the context of child protection’, Journal of Social Work , 13( 2), pp. 184– 202. Google Scholar CrossRef Search ADS   Pennell J. ( 2006) ‘ Restorative practices and child welfare: Toward an inclusive civil society’, Journal of Social Issues , 62( 2), pp. 259– 79. Google Scholar CrossRef Search ADS   Schmuckler M. A. ( 2001) ‘ What is ecological validity? A dimensional analysis’, Infancy , 2( 4), pp. 419– 36. Google Scholar CrossRef Search ADS   Schout G., De Jong G., Zeelen J. ( 2011) ‘ Beyond care avoidance and care paralysis: Theorizing public mental health care’, Sociology , 45( 4), pp. 665– 81. Google Scholar CrossRef Search ADS   Schwandt T. A. ( 1999) ‘ On understanding understanding’, Qualitative Inquiry , 5( 4), pp. 451– 64. Google Scholar CrossRef Search ADS   Silverman D. ( 2013) Doing Qualitative Research , 4th edn, London, SAGE Publications. Small M. L. ( 2009) ‘ “How many cases do I need?”: On science and the logic of case selection in field-based research’, Ethnography , 10( 1), pp. 5– 38. Google Scholar CrossRef Search ADS   Stake R. E. ( 1995) The Art of Case Study Research , London, SAGE Publications, Inc. Stake R. E. ( 2006) Multiple Case Study Analysis , New York, NY, Guilford. © The Author 2017. Published by Oxford University Press on behalf of The British Association of Social Workers. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Hanberger, Anders;Nygren, Lennart;Andersson, Katarina

doi: 10.1093/bjsw/bcx032pmid: N/A

Abstract This article examines the assumptions regarding how Swedish state supervision (SSV) of eldercare is to achieve its intended effects. It explores how SSV is intended to work to ensure and improve eldercare quality, and theoretically and empirically assesses the validity of its guiding assumptions with programme theory methodology. The theoretical assessment suggests that most intended effects are partly achieved, though the quality-enhancing assumption finds little support in caring research. The assumption that the supervised parties will improve their compliance with laws and regulations has some validity, but this compliance is temporary and confined to the aspects of eldercare being supervised. Twenty-four interviews with the chairs of Social Welfare Committees and care unit managers provide empirical support for all but two intended effects. SSV has not increased ‘awareness of national regulative demands in eldercare’ or contributed to ‘general quality improvement in eldercare’. Four unintended effects of SSV were also recognised in the interviews—for example, unsupervised caring activities were less prioritised. The authors conclude that, although SSV does little to improve eldercare quality, it is needed for transparency and accountability as well as to hold local governments and public and private service providers to account for compliance with national statutes. State supervision, Sweden, eldercare quality, intended effects, unintended effects Introduction In the last decade, policy makers in Sweden have increasingly regarded state supervision (SSV) as an effective way to detect and alleviate unhealthy conditions and improve public welfare services (Prop. 2008/09:160; Statskontoret, 2012). Recurrent monitoring of compliance with laws and regulations is expected to ensure the equal distribution of safe and high-quality eldercare (Socialstyrelsen, 2011a, 2011b) and it is assumed that well-functioning SSV can improve quality in both public and private eldercare services (Blomgren and Waks, 2011; Johansson, 2011; Ek, 2012). However, we know very little about how SSV works in practice. Although some research examines state supervision and inspection (Hood et al., 1999; Power, 1997; May and Wood, 2003; Johansson, 2006; Johansson, 2011; Nygren and Hanberger, 2011; Hämberg, 2013), such studies are rare in the field of social work and eldercare. One of the few relevant studies demonstrates that the external regulation of nursing homes in Australia, the UK and the USA has failed to contribute to continuous improvement because of various kinds of ‘regulatory ritualism’ (Braithwaite et al., 2007, p. 11). For example, nursing homes prioritise completing forms on outputs and other required information instead of resolving important problems, in what is called ‘documentation ritualism’ or ‘get[ting] the documents right but the care wrong’ (Braithwaite et al., 2007, p. 221). Similarly, Furness (2009) studied the influence of inspection on the quality of private care homes in one region of England, finding that the greatest change in care practices was increased documentation. The research and policy community disagrees on the need for SSV and its effect on welfare services and on whether external supervision does more harm than good (e.g. only promoting increased documentation). Furthermore, introducing new public-management (NPM) ideas and market mechanisms in eldercare is highly contested due to ideological and ethical disagreements about their value in eldercare (Szebehely and Trydegård, 2012). More knowledge of SSV can facilitate better-informed discourse on what can be expected from SSV. Exploring how the SSV of eldercare is intended to and actually works in Sweden, as is done here, is illustrative because considerable resources are devoted to eldercare in Sweden (Huber et al., 2009). SSV has strong political support and has been allocated extra resources (SOU 2007:82; Prop. 2008/09:160). If the Swedish SSV model can ensure and improve eldercare quality, it may serve as a model for other countries’ eldercare systems. This article examines the intended and unintended effects of the Swedish state supervision of eldercare. It explores how the SSV model is intended to work to ensure eldercare quality, and theoretically and empirically probes the validity of its guiding assumptions. Further aims are to contribute to research into eldercare evaluation and supervision and to promote better-informed discussions of what SSV can achieve. The research project is funded by the Swedish research council FORTE. Swedish eldercare Swedish eldercare has a long tradition of universality and has been taken to represent the Scandinavian welfare model, being publicly financed, comprehensive and serving all citizens according to need, not the ability to pay. In recent decades, there has been a strong marketisation trend in public eldercare, characterised by competition between service providers, stricter regulation and free choice of service provider for the service users (Meagher and Szebehely, 2013). Eldercare is governed at three levels in Sweden and monitored by SSV, local auditors and citizens. At the national level, the government uses legislation, policies and state subsidies to govern eldercare. At the regional level, twenty county/regional councils are responsible for the areas of hospital care, primary care, psychiatric care and dental care for all citizens and govern eldercare in these areas. At the local level, 290 local governments govern residential care homes and home-help services in terms of level of service provision and permitting or not permitting private service providers. Local auditors are commissioned to monitor county council and municipal implementation of eldercare policy. During elections, citizens hold the elected officials at all three levels to account for eldercare performance. An elderly person in need of care must first apply for care services, after which a care manager assesses his or her needs according to national eldercare standards. The decision made can be appealed to an administrative court if the care-recipient is dissatisfied. State supervision of eldercare The National Board of Health and Welfare (NBHW) was responsible for the state supervision of eldercare from 2010 to 2013, constituting the SSV system explored here. A new state agency, the Health and Social Care Inspectorate (IVO), has been commissioned with this supervision since July 2013. The supervision model analysed here is largely the same as that of the new agency. NBHW’s supervision included desktop supervision as well as the preannounced and unannounced inspection of residential care homes and home-help services. SSV was guided by a supervision policy, special government assignments for thematic supervision, and laws and regulations for eldercare. The SSV covered five assessment areas (planning and steering; staff competence; internal and external co-operation; management and documentation; care content) and NBHW developed forty criteria and 205 variables to guide the supervision (Socialstyrelsen, 2012b). A single supervision focused on one or several of the areas and applied a selection of the measures (Socialstyrelsen, 2012a). After each supervision, a decision was made concerning any noted deficiencies and needed improvements. The SSV system also included follow-up of the implementation of supervision decisions. A yearly summary report compiled for all supervisions consisted of conclusions regarding the state of eldercare and recommendations for action. In the years 2010–13, SSV’s supervisions focused on various themes such as: ‘coordination of care planning between caregivers’ and ‘documentation of provided care’, ‘staff competence’, ‘content of care’ and ‘follow-up on implementation of allotted care’ (Socialstyrelsen, 2012a, pp. 11–13). In the summer of 2010, NBWH conducted unannounced inspections of homes for elderly suffering from dementia and short-term accommodations, based on risk assessment of possible deficiencies in quality. This inspection round focused on five of these themes. In 2011, NBHW inspected 123 residential homes and forty-five short-term accommodations and, in 78 per cent of the inspected homes, deficiencies were noted (Socialstyrelsen, 2012a, p. 12). As indicated, the Swedish SSV model entails different types of supervision covering five assessment areas and many different themes. In the next section, the research used in the analysis of the Swedish SSV is summarised. Theory and framework Knowledge from supervision, evaluation and eldercare research is used here as these research areas concern important aspects of state supervision. Using supervision research is logical as SSV is a supervision system. We conceive SSV as an evaluation system that collects recurrent evaluation data, which justifies integrating knowledge from evaluation system research. As the focus is on SSV’s ability to contribute to the continuous improvement of eldercare, research into eldercare quality is also used. Supervision research To ensure eldercare quality, the SSV of eldercare should be in line with the general supervision policy for state-run social welfare services in Sweden and enact a strengthened standardised supervision model (Prop. 2008/209:160). Whether SSV is an effective way to expose unhealthy conditions and improve quality in eldercare, however, is debated in the research community (Cooper, 2006; Beddoe, 2012). SSV and other kinds of external supervision tend to challenge professional norms and competences. Critics of NPM warn that external supervision, although well intentioned, can negatively affect professions in that it creates mistrust and lowers the level of confidence (Cooper, 2006, p. 29; Beddoe, 2012). Following this line of reasoning, SSV is conceived as an intervention in organisations containing professionals with considerable competence, discretion and responsibility to deal with multifaceted social problems (Green, 2007; Beddoe, 2010). In supervised organisations, peer supervision is generally in place to support the professions and sustain reflective practices (Bradley and Höjer, 2009). In contrast to SSV, many studies of peer supervision in social work and nursing indicate that peer supervision can positively affect care quality if certain conditions are met (Fish and Twinn, 1997; Bishop, 1998; Bradley and Höjer, 2009; Beddoe, 2012; Hyrkäs and Paunonen-Ilmonen, 2001). For example, clinical supervision was found to have a positive impact on the supervisee’s work and well-being if a constructive alliance was built (Livni et al., 2012). The government and two state commissions, however, claimed that both public and private eldercare should also be subjected to national scrutiny and recurrent control (SOU 2004:100; 2007:82). The state and its supervisors convey high expectations of state control as a way to ensure quality; however, professionals tend to have lower expectations of such control, sometimes expressing resistance to it. This reflects a conflict between external control and democratic accountability for performance, on the one hand, and internal control and trust in professionals, on the other (Power, 1997; Bouckaert and Halligan, 2006, p. 444; Beddoe, 2012). Studies of regulatory enforcement, state supervision, inspection and evaluation in different policy sectors indicate that adjustment to, and compliance with, what is inspected/evaluated is reinforced to some extent and depends on many factors and conditions. The overall picture is that the effects of SSV and inspection are mixed, temporary and depend on whether and how control and advice are combined in supervision. For example, compliance with regulations depends on the inspectors’ enforcement styles mainly if they act as ‘cops’ or ‘consultants’ (Ayres and Braithwaite, 1992; Kagan, 1994; Sparrow, 2000) enforcing the clarity of expectations and consistency of procedures (May and Wood, 2003). If health and safety regulation goals are adaptable to various risks and working conditions, if practical guidance is available to help organisations and professionals and, if their competence is ensured, then compliance with health and safety laws is promoted (Löfstedt, 2011, p. 36). In general, if targets are clear and easily measurable, achievement of targets (i.e. national objectives and demands) can be expected to increase (Bevan and Hood, 2006; Hood, 2012). Several indirect and unintended effects of supervision/evaluation are also recognised. For example, numbers and targets used in evaluation tend to become ineffective over time as they promote strategic behaviour (Hood, 2012). According to Hood (2012), targets can create three negative behaviours in managers and organisations: threshold effects, which is the tendency not to perform above the threshold target; ratchet effects, which is the tendency to perform below an organisation’s maximum potential if targets are expected to be set higher later on; and output distortion, which is a focus on targeted values at the expense of other important values. Using targets in hierarchical settings, such as the setting in which SSV is used, can be expected to produce unintended effects (Hood, 2012), among service providers, for example. Inspections of nursing homes tend to create various types of ritualism. Besides the ‘documentation ritualism’ mentioned in the introduction, ‘rule ritualism’ (writing a rule instead of solving the problem) and ‘legal ritualism’ (following the letter but not the spirit of the law) are recognised as well (Braithwaite et al., 2007, p. 221). Evaluation system research Kusek and Rist’s (2004) handbook of monitoring and evaluation (M&E) systems provides insights and practical guidelines for when and how to set up traditional and results-based M&E systems. Traditional M&E systems, designed to monitor implementer compliance with requirements, have long been used in various sectors and countries. Results-based M&E systems, developed in many sectors in recent decades, are essentially a public-management tool that governments use to measure and evaluate the achievement of objectives and to provide feedback on the ongoing processes of governing and decision making. Kusek and Rist (2004) provide a ten-step practical guide to designing and sustaining results-based M&E systems. To work well, such systems must be designed to suit their intended purpose. Key actors must agree on the outcomes to be monitored and evaluated, while the system owners must collect baseline data on indicators and spend time sustaining the M&E system within the organisation (Kusek and Rist, 2004). Leeuw and Furubo (2008) have a more critical approach, assuming that evaluation systems produce routinised information for single-loop learning and that the systems are largely used as providers of (procedural) assurance. The systems produce information that confirms rather than questions policies. Evaluation systems can also have ‘wash-back effects’ or lead to ‘tunnel vision’—that is, the things being measured become goals in themselves, leading to narrow thinking (cf. output distortion). Empirical research into evaluation systems indicates that these systems tend to create ‘performance demonstration behaviour’, ‘uncritical reporting of effects’ and ‘post-rationalization of action taken to meet the performativity and accountability norm’ (Hanberger, 2011, p. 31). Research into eldercare quality The quality of care can be defined in relation to the structural attributes of eldercare settings, the processes of care and the outcomes of care (Donabedian, 1988/1997; Blom and Morén, 2012). Eldercare research emphasises that the quality of care concerns the relationship between care-giver and care-receiver, which contains several important qualities that make the relationship personal and genuine. The meeting between care-giver and recipient is important and needs to be based on mutuality and trust. Continuity is also integral to quality because it centres the care relationship (cf. Noddings, 1984; Waerness, 1984; Fisher and Tronto, 1990). Furthermore, sufficient and competent staff and low employee turnover are important aspects of structural quality necessary to maintain continuity. Other important conditions for ensuring quality are time for care, flexibility (i.e. care must be performed individually according to personal needs) and influence on care (i.e. participation/self-determination regarding how and when care is to be provided) (Szebehely, 2011). The quality of eldercare includes various aspects that are somewhat interrelated (e.g. structural quality can affect building and maintaining trust). Framework Against this background, we formulate a theoretical framework for exploring SSV. It conceives SSV as both an evaluation system (Leeuw and Furubo, 2008; Hanberger, 2011, 2013) and a regulatory enforcement tool (Hood et al., 1999; May and Wood, 2003) using regulation, economic means and information (Bemelmans-Videc et al., 1998) to enforce compliance with statutes and to ensure safety and good quality in eldercare services (Noddings, 1984; Fisher and Tronto, 1990; Szebehely, 2011). From the cited literature and with reference to the relevant Government Bill (Prop. 2008/09:160), we have identified several possible intended and unintended effects of SSV, which we have compiled and organised for use in assessing the effects of SSV on supervisees (see Table 1). Table 1 Possible intended and unintended effects of state supervision on supervisees Effects on supervisees  Examples of effect manifestations  Intended    Instrumental compliance with statutes  Care unit managers change the conditions or routines criticised by inspection  Action to enhance safety and quality in eldercare  Local decision makers increase the number of care-givers per care recipients  Awareness of national regulative demands in eldercare  Care unit managers pay attention to and discuss national guidelines with staff  Action to improve achievement of national objectives  Local decision makers increase the responsible committee’s monitoring of these  Check whether documentation is in good order  Care unit managers go through routines and file all documents  Foster an improvement culture  Decision makers and managers encourage collective learning for improvement  General quality improvement in eldercare  SSV is used as a resource in committees’ and service providers’ quality work  Unintended    Ratchet effects: tendency to provide services below an organisation’s maximum potential  A service provider thinks the service level is sufficient, referring to the SSV inspection  Threshold effects: disincentivise performance exceeding the minimum level stipulated in regulations  An eldercare committee reduces staffing exceeding the national standard  Output distortion: focus on what is ‘targetised’, overlooking other important values  Decision makers and managers focus on and allocate resources only to supervised areas  Ritualism  Managers get the documents and routines right but leave the problem unsolved  Deprofessionalisation  Managers and staff feel that their professional competence is not respected  Professional resistance and questioning how SSV defines quality in eldercare  Care-givers criticise SSV’s quality focus and what is overlooked  Distrust in SSV and eldercare governance  Local actors express distrust in SSV and national eldercare governance  Effects on supervisees  Examples of effect manifestations  Intended    Instrumental compliance with statutes  Care unit managers change the conditions or routines criticised by inspection  Action to enhance safety and quality in eldercare  Local decision makers increase the number of care-givers per care recipients  Awareness of national regulative demands in eldercare  Care unit managers pay attention to and discuss national guidelines with staff  Action to improve achievement of national objectives  Local decision makers increase the responsible committee’s monitoring of these  Check whether documentation is in good order  Care unit managers go through routines and file all documents  Foster an improvement culture  Decision makers and managers encourage collective learning for improvement  General quality improvement in eldercare  SSV is used as a resource in committees’ and service providers’ quality work  Unintended    Ratchet effects: tendency to provide services below an organisation’s maximum potential  A service provider thinks the service level is sufficient, referring to the SSV inspection  Threshold effects: disincentivise performance exceeding the minimum level stipulated in regulations  An eldercare committee reduces staffing exceeding the national standard  Output distortion: focus on what is ‘targetised’, overlooking other important values  Decision makers and managers focus on and allocate resources only to supervised areas  Ritualism  Managers get the documents and routines right but leave the problem unsolved  Deprofessionalisation  Managers and staff feel that their professional competence is not respected  Professional resistance and questioning how SSV defines quality in eldercare  Care-givers criticise SSV’s quality focus and what is overlooked  Distrust in SSV and eldercare governance  Local actors express distrust in SSV and national eldercare governance  Key: The effects can occur before and after SSV and be major or minor. The Government Bill, Coordinated and Transparent Supervision of Social Services (Prop. 2008/09:160), refers to seven intended effects of state supervision (on supervisees): instrumental compliance with the statutes, action to enhance safety and quality in eldercare, awareness of national regulations in eldercare, action to improve achievement of national objectives, checking whether documentation is in good order, fostering an improvement culture and achieving general quality improvement in Swedish eldercare. From the cited literature, we can identify at least seven unintended effects: ratchet effects, which is a tendency to provide services below an organisation’s maximum potential (Bevan and Hood, 2006; Hood, 2006); threshold effects, which is a tendency to disincentivise excellent performance (in this context, performance exceeding the minimum level stipulated in regulations) (Bevan and Hood, 2006; Hood, 2006); output distortion, which is a focus on what is ‘targetised’, overlooking other important values (Bevan and Hood, 2006; Hood, 2006); types of ritualism (Braithwaite et al., 2007); deprofessionalisation (Johansson, 2011; Meagher and Szebehely, 2013); professional resistance and questioning of how SSV defines quality in eldercare (Cooper, 2006; Beddoe, 2012); and distrust of SSV and eldercare governance (Cooper, 2006; Beddoe, 2012). Methodology The framework is applied in the methodology, mainly in how intended and unintended effects are assessed. The Swedish SSV system is examined via a programme theory (PT) analysis and the effects of SSV are assessed against the criteria listed in Table 1. PT refers to the assumptions as to how a programme can achieve its intended effects (Leeuw, 2003). In this article, the PT notion is applied to the SSV system. A PT analysis is chosen because it uncovers these assumptions and probes their consistency in a systematic and effective way. PT analysis A policy-scientific approach is used to reconstruct the SSV’s PT (Leeuw, 2003). This approach starts with the policy makers’ intentions and assumptions and uses knowledge from research to probe them. The methodology has been used, with some variations, in several PT analyses of evaluation systems (e.g. Leeuw, 2003; Hanberger, 2014). The PT analysis includes three main steps and seeks answers to the following questions: Step 1: Reconstructing the SSV’s PT: Q1. What assumptions underlie SSV? Q2. What problem(s) was SSV intended to resolve? Q3. What are the prerequisites for SSV and what activities are assumed to produce what intended effects? Q4. Are the problems that SSV is to address described and substantiated? Q5. Are the intended effects clearly described/specified? Step 2: Assessing the PT’s internal validity: Q6. Is SSV consistent? Step 3: Assessing the PT’s external validity: Q7. Does SSV have scientific support? Q8. Have intended effects evolved in practice? Step 1, reconstructing the SSV’s PT, focuses on how the state assumes that SSV should achieve the intended effects of supervision. Step 2 assesses the PT’s internal validity. If the intended effects appear unlikely to occur, this indicates that the PT’s internal validity is (somewhat) inconsistent. The PT’s internal validity was assessed by the three authors by means of an argument analysis (Leeuw, 2003, p. 7). Step 3 assesses the PT’s external validity, namely the extent to which SSV is supported by the cited research and whether it provides knowledge useful in resolving the problems it is intended to resolve. The SSV’s PT can be in accordance with or more or less supported by the cited research. The PT’s external validity was assessed by the three authors. The PT was reconstructed (Step 1) based on the official supervision policy intended to guide SSV (Socialstyrelsen, 2011b), one interview with a key person at the NBHW involved in developing the SSV system and national reports on SSV of eldercare (Socialstyrelsen, 2011a, 2012a, b, c; IVO, 2013). As said, to probe the PT’s consistency (Step 2), the authors conducted an argument analysis (Leeuw, 2003). In the analysis of the PT’s external validity (Step 3), we used the cited evaluation, supervision and caring research together with the results of interviews with two groups of local key actors in Swedish municipalities: chairs of Social Welfare Committees and care unit managers. The interviews were here used to probe the PT’s empirical support. The selection of municipalities and interviewees were based on six senior regional inspectors’ (one from each region) selection of severe supervision cases. They were first asked to select five case files each, which they recalled to be severe or complex supervisions conducted between 2010 and 2012. We assumed that these cases would be easier to recall and thus not be passed unnoticed. If these supervisions were not recalled or had no effect, less severe cases would too. During this time span, approximately 1,100 supervision cases were documented—the majority desktop supervisions. One hundred and seventy-five of these supervisions noted deficiencies and included demands on the supervised. Among these 175, the twenty-eight severe cases were selected from different municipalities. We read these twenty-eight closely in terms of arguments and responses to SSV. Next, among the twenty-eight cases, fifteen municipalities were selected to obtain sufficient geographical distribution and a variety of supervisions in terms of both depth and breadth. Fifteen chairs of municipal Social Welfare Committees and fifteen care unit managers were identified and contacted by e-mail to participate in an interview study about how SSV works in practice. Only two declined to participate, while four did not reply, resulting in twelve telephone interviews with chairs and twelve with managers in the fifteen municipalities. All the interviewees consented to be audio recorded. The interviews lasted between thirty minutes up to one hour, and were conducted according to ethical guidelines. The interviews focused on the actors’ experiences of and responses to SSV, their views on eldercare and care quality, and their routines in eldercare practice. For the purpose of this article, the interviewees’ experiences of SSV were analysed in terms of intended and unintended effects, as shown in Table 1, applying a directed content analysis (Hsieh and Shannon, 2005). The results from interviews were presented to maintain respondent confidentiality, and for not revealing municipalities and care institutions. Assessment of effects of SSV Table 1, developed from the referred literature and Government Bill (Prop. 2008/09:160), was used in assessing the intended and unintended effects of SSV based on twenty-four interviews and on NBHW’s follow-up of the implementation of measures decided on to alleviate unhealthy conditions revealed by inspections. Results The PT of SSV Three basic assumptions underpin SSV (Q1): first, the supervision system needs a clear division between the roles of inspection and of advice. Second, effective SSV is a prerequisite for management by objectives and national eldercare governance. Third, the multiplicity of public and private service providers and frequent changes in eldercare require recurrent quality control, monitoring and evaluation. What problem(s) is SSV intended to resolve (Q2)? The problem is constructed as the need for a strengthened SSV (see the introduction). Another challenge is to maintain confidence in SSV and to sustain recurrent quality control to ensure compliance with applicable laws and regulations. What are the prerequisites for SSV and what activities are assumed to produce what intended effects (Q3)? This question is answered by reconstructing SSV’s PT, a simplified version of which is presented below. If the following conditions are met: service providers have a system for internal control; supervision is carried out by independent and competent national supervisors; enough resources are provided for effective supervision. And if the following activities are carried out: SSV is centralised in two steps ending up with an independent supervision agency; information about how and why SSV is conducted is provided; national assessment criteria are developed and supervisors are trained; guidelines, supervision policy and routines for the supervision process are developed; the new supervision model is legitimatised; un- and preannounced inspections based on risk analysis and complaints are carried out; demands for action to resolve deficiencies found in supervision are clear and evaluable; feedback to the supervised is communicated respectfully; supervision decisions are followed up; sanctions are imposed on service providers and municipalities if needed. The following short- and long-term effects should occur: (Activities 1–4) a clear division between the control and advice roles; (Activities 3 and 4) competent supervisors use adequate methods, standardised routines and make the same assessments of similar cases; (Activities 2–10) the supervision process is transparent; (Activities 2–4) supervisees take action before inspection to comply with laws and regulations and alleviate any bad conditions identified; (Activity 6) unhealthy conditions detected; (Activities 7–10) action is taken to resolve detected problems and deficiencies; a minimum level of eldercare service is ensured according to national guidelines; elderly men and women are provided with safe and high-quality eldercare. There are activities (in parentheses) for all intended effects except for the last two, because there are no activities developed to achieve these effects. The problems that SSV is to address (Q4) are described in the introduction and substantiated in the referred investigations (SOU 2007:82; Prop. 2008/09:160). The intended effects are clearly stated above under ‘short- and long-term effects’ (Q5). Assessment of the PT’s internal and external validity The PT is somewhat consistent, as there are reasonable activities to achieve or partly achieve the first six intended effects (Q6). Instrumental compliance with the relevant statutes and alleviation of bad conditions can be expected shortly before and after inspections, though the effects will likely not be long-lasting because local actors can be expected to give equal attention to unsupervised national demands after a while. A minimum acceptable level of safe and high-quality eldercare service is unlikely to be achieved just because SSV supervises a selection of eldercare providers and municipalities, focusing on monitoring compliance with certain demands expressed in laws and regulations. Furthermore, extra resources to resolve deficiencies are not provided by the state. Service providers are likely to reallocate resources within the same budget, but it is unlikely that municipalities will shift resources from other departments or that private service providers will reduce their profits. Hence, SSV cannot be expected to enhance the overall quality of eldercare. The assumption that SSV can improve eldercare quality has some support in supervision and evaluation system research in that state supervision has a temporary effect on the aspects of eldercare being supervised in the supervisees, particularly if sanctions are imposed (Q7). Similarly, the assumption that SSV can improve eldercare quality by conducting various supervisions and by adopting citizen and user perspectives is limited to ensuring a minimum standard of eldercare in the particular aspects of care being supervised. SSV cannot be expected to have positive effects on all national eldercare objectives. The quality-enhancing assumption finds little support in caring research, because SSV focuses on compliance with national guidelines related to structural quality, and on documentation and routines. The most important aspects of eldercare are not supervised—that is, individual user-centred care, the relationship between care-giver and care-recipient, and mutuality and trust (Waerness, 1984; Johansson, 2011; Szebehely, 2011). Assessment of intended and unintended effects of SSV This section summarises our assessment of the intended and unintended effects of Swedish SSV based on interviews with twelve chairs of Social Welfare Committees and twelve eldercare managers in fifteen municipalities. All intended effects (Table 1) except two were observed by the interviewees: SSV has not increased ‘awareness of national regulative demands in eldercare’ or contributed to ‘general quality improvement in eldercare’. This indicates some empirical support for the intended effects (Q8). The interviewees also noted four unintended effects of SSV. Caring activities not being supervised were less prioritised, and SSV promoted ‘ritualism’ in that care providers prioritised getting the documents and routines right. Interviews indicate that professionals expressed resistance and questioned how SSV defines quality in eldercare, and that some supervisees distrusted SSV and eldercare governance. It is beyond the scope of this article to comment on the assessment of each of the criteria, but it is nevertheless important to highlight several examples. When inspectors visited, care unit managers checked that documentation was in good order, exemplifying one intended effect. Instrumental compliance with national regulative demands was an immediate effect when night-time staffing was found to be below the national demands. Staff were reallocated to comply with the standard, but this had a crowding-out effect: less care was provided for other clients at other times and places—an unintended effect referred to as output distortion (Hood, 2006). How does this result correspond to what is known about SSV from other sources? The NBHW requested corrective action from seven of ten eldercare providers during the 2010–12 period and required written descriptions of the action taken to alleviate unhealthy conditions. In most cases, the NBHW judged the actions promised by the supervisees to be sufficient and the cases were then closed. However, when the NBHW followed up the implementation of the decided-on measures, it found that, in some 20 per cent of randomly selected supervision cases (of 205 in total), 26 per cent of the public and private eldercare providers had not, or only partly, implemented the decided-on measures (IVO, 2013). This implies that some of the interviewees may have given overly positive descriptions of instrumental compliance. Discussion The core of the Swedish SSV model is supervision of compliance with national regulations in order to ensure that citizens obtain the services they have a right to and that eldercare is of high quality. The research used to explore the soundness of the SSV model does not confirm that supervision is an effective tool for improving the overall quality of eldercare. Notably, SSV did not supervise the most important qualities of eldercare—that is, personal and genuine care, the care-giver–caretaker relationship and trust. The results from interviewing chairs of Social Welfare Committees and care unit managers together with NBHW’s follow-up of the implementation of decided-on actions to alleviate deficiencies also indicate that SSV has not achieved its most important intended effects in practice. In addition, the fact that SSV has had some unintended effects should also be taken into account. The SSV model is based on what Braithwaite et al. (2007, p. 318) call the ‘regulatory enforcement pyramid’—that is, the understanding that risk assessment, fear and prompt response before problem escalation will raise standards above a minimum standard. This rationale has not encouraged improvement of eldercare quality beyond what was perceived as the minimum standard, however. As demonstrated, Swedish SSV does not comprehensively assess whether Swedish eldercare exceeds minimum national regulative demands, as it covers only certain aspects of eldercare. The failure of Swedish SSV to achieve its ultimate objectives is associated with the intrinsic limitations of the regulatory enforcement pyramid, which promotes various kinds of ‘regulatory ritualism’ (Braithwaite et al., 2007, p. 11)—that is, care unit managers prioritise completing forms on outputs and providing other required information instead of deciding to ‘pick important problems and fix them’. Similarly, none of nineteen private care home managers in one region in England associated inspection with service improvement, the most obvious effect of inspection being increased documentation (Furness, 2009). This article investigated the SSV model when the NBHW was responsible for eldercare supervision. Since 2013, this responsibility has moved to the Health and Social Care Inspectorate (IVO), but the core of the supervision model has not changed except that the role of providing advice to the supervised is now part of IVO’s commission (www.ivo.se). A main weakness of the studied (and current) SSV model is its limited ability to support the ongoing improvement of eldercare quality—an inherent problem with the regulatory enforcement model. The director of IVO (DN, 2015) is aware of some of the negative consequences of IVO’s supervision, such as that IVO’s requests can result in the production of policy documents and guidelines without actually solving the problems, exemplifying documentation and legal ritualism. Whether IVO is moving towards a new ‘regulatory strength-based pyramid or model’ by identifying and promoting the supervised organisations’ strengths (Braithwaite et al., 2007) and can succeed in managing the negative effects of the SSV model, only time will tell. An inherent problem with the current SSV model is that it provides little information on what works in eldercare. The share of supervised residential homes and home-help services is rather small and is based on a selection of care providers and municipalities with unhealthy conditions, which does not reflect the general status of eldercare in the country. The Swedish SSV model cannot serve as an example when it comes to ensuring quality in eldercare. Before drawing far-reaching conclusions about the model’s ability to enhance eldercare quality, we need to follow up changes made in the current IVO model and more knowledge of how it affects supervisees and eldercare practice. The main advantage of PT analysis is that it critically examines the assumptions underlying the SSV model, though it does overlook unintended effects. Combining it with an assessment of unintended effects, as is done here, provides a more valid account of the SSV model. The methodology can be applied on other countries’ SSV models to generate knowledge of how a key reactive policy instrument functions in eldercare governance in different political systems. Further studies will tell whether other countries’ models are better in contributing to improve quality in eldercare and achieve intended effects without major unintended effects. Conclusion This PT analysis of the Swedish SSV system demonstrates that most intended effects are likely to be achieved or partly achieved. However, the most important intended effect, namely ensuring a minimum acceptable level of safe and high-quality eldercare, is unlikely to be achieved. Supervision and evaluation system research supports the existence of a temporary effect on the supervisees, specifically in the aspects of eldercare being supervised. The quality-enhancing assumption finds little support in caring research because SSV focuses on compliance with national demands for structural quality, documentation and routines instead of on individual user-centred quality of care and on whether the relationship between care-giver and care-recipient is based on mutuality and trust. In this respect, the Swedish SSV can be claimed to fail to achieve its intended effects as a result of ‘theory failure’ (Weiss, 1972). The article provides empirical support, based on twenty-four interviews, for the partial achievement of all intended effects of SSV except two: SSV has not increased ‘awareness of national regulative demands in eldercare’ or contributed to ‘general quality improvement in eldercare’. The present SSV model has had four unintended effects: caring activities not being supervised are less prioritised; care providers prioritise getting the documents and routines right instead of providing actual care; professionals question how SSV defines quality in eldercare; and some actors express distrust of SSV and eldercare governance. How these flaws and unintended effects influence social work and care practices is yet to be investigated in detail. Still, we can identify a risk that weakened incentives to improve quality may lead staff to focus less on the core values of care work. Our study confirms the findings of international studies about the emphasis on routines and documentation over care quality, but our PT analysis of intended and unintended effects additionally provides a more elaborate judgement of SSV in eldercare than previous studies. Increased state supervision of eldercare cannot be justified as an effective tool or mechanism for enhancing eldercare quality. However, some kind of supervision is necessary for transparency and to hold local governments and public and private service providers to account for compliance with the relevant statutes. Hypothetically, the absence of SSV would leave eldercare users with less influence and higher risks of unequal services levels. Supervision reinforces the national objectives and regulations that should govern and guide Social Welfare Service Committees and public and private service providers—a value in its own right. References Ayres I., Braithwaite J. ( 1992) Responsive Regulation, Transcending the Deregulation Debate , Oxford and New York, Oxford University Press. Beddoe L. ( 2010) ‘ Surveillance or reflection: Professional supervision in “the risk society”’, British Journal of Social Work , 40, pp. 1279– 96. Google Scholar CrossRef Search ADS   Beddoe L. ( 2012) ‘ External supervision in social work: Power, space, risk, and the search for safety’, Australian Social Work , 65( 2), pp. 197– 213. Google Scholar CrossRef Search ADS   Bemelmans-Videc M.-L., Rist R., Vedung E. (eds) ( 1998) Carrots, Sticks and Sermons: Policy Instruments and their Evaluation , 4th edn, New Brunswick, NJ, and London, Transaction Publishers. Bevan R. G., Hood C. ( 2006) ‘ What’s measured is what matters: Targets and gaming in healthcare in the English public health care system’, Public Administration , 84( 3), pp. 517– 38. Bishop V. ( 1998) ‘What is clinical supervision?’, in Bishop V. (ed.), Clinical Supervision in Practice: Some Questions, Answers and Guidelines , Houndmills, Basingstoke, Hampshire and London, Macmillan Press Ltd, pp. 1– 20. Google Scholar CrossRef Search ADS   Blom B., Morén S. ( 2012) ‘ Evaluation of quality in social-work practice’, Nordic Journal of Social Research , 3, pp. 1– 17. Google Scholar CrossRef Search ADS   Blomgren M., Waks C. ( 2011) ‘ Öppna jämförelser inom hälso- och sjukvården—en utmaning mot den nya professionella byråkratin?’ [Open Comparison in health care- a challenge for the new professional bureaucracy], Arbetsmarknad & Arbetsliv , 17( 4), pp. 95– 108. Bouckaert G., Halligan J. ( 2006) ‘Performance and performance management’, in Peters G., Pierre J. (eds), Handbook of Public Policy , London, Sage. Bradley G., Höjer S. ( 2009) ‘ Supervision reviewed: Reflections on two different social work models in England and Sweden’, European Journal of Social Work , 12, pp. 71– 85. Google Scholar CrossRef Search ADS   Braithwaite J., Makkai T., Braithwaite V. ( 2007) Regulating Aged Care: Ritualism and the New Pyramid , Cheltenham, UK, and Northampton, MA, E. Elgar. Google Scholar CrossRef Search ADS   Cooper L. ( 2006) ‘ Clinical supervision: Private arrangement or managed process?’, Social Work Review , 18, pp. 21– 30. Donabedian A. ( 1988/1997) ‘ The quality of care: How can it be assessed?’, Archives of Pathology & Laboratory Medicine , 121( 11), pp. 1743– 8. DN ( 2015) Dagens nyheter, 20 January. Ek E. ( 2012) ‘De granskade—om hur offentliga verksamheter görs granskningsbara’ [The auditees. On how public organisations are made auditable], dissertation, Göteborg, Förvaltningshögskolan, Göteborgs universitet. Fish D., Twinn S. ( 1997) Quality Clinical Supervision in the Health Care Profession: Principled Approaches to Practice , Guildford and Kings Lynn, Biddles Ltd. Fisher B., Tronto J. ( 1990) ‘Toward a feminist theory of caring’, in Abel E. K., Nelson M. K. (eds), Circles of Care , Albany, NY, State University of New York Press. Furness S. ( 2009) ‘ A hindrance or a help? The contribution of inspection to the quality of care in homes for older people’, British Journal of Social Work , 39, pp. 488– 505. Google Scholar CrossRef Search ADS   Green D. ( 2007) ‘ Risk and social work practice’, Australian Social Work , 60, pp. 395– 409. Google Scholar CrossRef Search ADS   Hämberg E. ( 2013) ‘ Supervision as control system: The design of supervision as a regulatory instrument in the social services sector in Sweden’, Scandinavian Journal of Public Administration , 17( 3), pp. 45– 64. Hanberger A. ( 2011) ‘ The real functions of evaluation and response systems’, Evaluation , 17( 4), pp. 327– 49. Google Scholar CrossRef Search ADS   Hanberger A. ( 2013) ‘ Framework for exploring the interplay of governance and evaluation’, Scandinavian Journal of Public Administration , 16( 3), pp. 9– 28. Hanberger A. ( 2014) ‘ What PISA intends to and can possibly achieve: A critical programme theory analysis’, European Educational Research Journal , 13( 2), pp. 167– 80. Google Scholar CrossRef Search ADS   Hood C. ( 2006) ‘ Gaming in Targetworld: The targets approach to managing British public services’, Public Administration Review , 66( 4), pp. 515– 21. Google Scholar CrossRef Search ADS   Hood C. ( 2012) ‘ Public management by numbers as a performance-enhancing drug: Two hypotheses’, Public Administration Review , 72( s1), pp. S85– 92. Google Scholar CrossRef Search ADS   Hood C., Oliver J., Scott C., Travers T. ( 1999) Regulation inside Government: Waste-Watchers, Quality Police, and Sleazebusters , Oxford, Oxford University Press. Google Scholar CrossRef Search ADS   Hsieh H., Shannon S. ( 2005) ‘ Three approaches to qualitative content analysis’, Qualitative Health Research , 15( 9), pp. 1277– 88. Google Scholar CrossRef Search ADS PubMed  Huber M., Rodrigues R., Hoffman F., Gasior K., Marin B. ( 2009) Facts and Figures on Long-Term Care: Europe and North America , Vienna, European Center. Hyrkäs K., Paunonen-Ilmonen M. ( 2001) ‘ The effects of clinical supervision on the quality of care: Examining the results of team supervision’, Journal of Advanced Nursing , 33( 4), pp. 492– 502. Google Scholar CrossRef Search ADS PubMed  IVO ( 2013) Äldre efterfrågar kontinuitet: Nationell tillsyn av vård och omsorg om äldre—slutrapport 2013 [The elderly requests continuity: National supervision of care for elderly - final report 2013] , Stockholm, Inspektionen för vård och omsorg (the Health and Social Care Inspectorate). Johansson S. ( 2011) ‘Utvärdering av äldreomsorg—exemplet Öppna jämförelser’ [Evaluation of eldercare - Open Comparison as example], in Blom B., Morén S., Nygren L. (eds), Utvärdering i socialt arbete [Evaluation in social work] , Stockholm, Natur & Kultur. Johansson V. ( 2006) Tillsyn och effektivitet: statliga inspektörers yrkesroller och strategival [Supervision and effectivity: state inspectors' professional roles and strategic choices] , Umeå, Boréa. Kagan R. ( 1994) ‘Regulatory enforcement’, in Rosenbloom D. H., Schwartz R. D. (eds), Handbook of Regulation and Administrative Law , New York, Marcel Decker. Kusek J., Rist R. ( 2004) Ten Steps to a Results-Based Monitoring and Evaluation System: A Handbook for Development Practitioners , Washington, DC, World Bank Publications. Google Scholar CrossRef Search ADS   Leeuw F. ( 2003) ‘ Reconstructing program theories: Methods available and problems to be solved’, American Journal of Evaluation , 24( 1), pp. 5– 20. Google Scholar CrossRef Search ADS   Leeuw F. L., Furubo J.-E. ( 2008) ‘ Evaluation systems: What are they and why study them?’, Evaluation, 14( 2), pp. 157– 69. Livni D., Crowe T., Gonsalvez C. ( 2012) ‘ Effects of supervision modality and intensity on alliance and outcomes for the supervisee’, Rehabilitation Psychology , 57( 2), pp. 178– 86. Google Scholar CrossRef Search ADS PubMed  Löfstedt R. E. ( 2011) Reclaiming Health and Safety for All: An Independent Review of Health and Safety Legislation , London, Health and Safety Executive (HSE). May P., Wood P. ( 2003) ‘ At the regulatory front lines: Inspectors’ enforcement styles and regulatory compliance’, Journal of Public Administration Research and Theory , 13( 2), pp. 117– 39. Google Scholar CrossRef Search ADS   Meagher G., Szebehely M. (eds) ( 2013) Marketization in Nordic Eldercare: A Research Report on Legislation, Oversight, Extent and Consequences, Stockholm , Stockholm University. Noddings N. ( 1984) Caring: A Feminine Approach to Ethics and Moral Education , Berkeley, University of California Press. Nygren L., Hanberger A. ( 2011) ‘Socialt arbete i granskningssamhället [Social work in the audit society]‘, in Blom B., Nygren L., Morén S. (eds), Utvärdering i socialt arbete [Evaluation in social work] , Stockholm, Natur & kultur. Power M. ( 1997) The Audit Society: Rituals of Verification , Oxford, Oxford University Press. Prop. 2008/09:160 (Government Bill) Samordnad och tydlig tillsyn av socialtjänsten [Coordinated and clear supervison of social work]. Socialstyrelsen ( 2011a) Nationell tillsyn av vård och omsorg om äldre: Delrapport [National supervision of eldercare: Interim report] , Stockholm, Socialstyrelsen. Socialstyrelsen ( 2011b) Socialstyrelsens tillsynspolicy för operativ tillsyn [The National Board of Health and Welfare's supervision policy for operational supervision]  (revised 13 October 2013), Stockholm, Socialstyrelsen. Socialstyrelsen ( 2012a) Nationell tillsyn av vård och omsorg om äldre: Delrapport II [National supervision of eldercare: interim report II] , Stockholm, Socialstyrelsen. Socialstyrelsen ( 2012b) Nationella bedömningskriterier för kvalitet och rättssäkerhet inom socialtjänstens omsorg om äldre [National assessment criteria for quality and rule of law in eldercare] , Stockholm, Socialstyrelsen. Socialstyrelsen ( 2012c) Tillsynsplan [Supervision plan] , Stockholm, Socialstyrelsen. SOU 2004:100 Tillsyn: förslag om en tydligare och effektivare offentlig tillsyn. Slutbetänkande/Tillsynsutredningen [Supervision:suggestations for more clear and effective state supervision. Final report] , Stockholm, Fritzes. SOU 2007:82 Samordnad och tydlig tillsyn av socialtjänsten, slutbetänkande/Utredningen om tillsynen inom socialtjänsten [Coordinated and clear supervison of social work. Final report] , Stockholm, Fritzes. Sparrow M. ( 2000) The Regulatory Craft: Controlling Risks, Solving Problems, and Managing Compliance , Washington, DC, Brookings Institution. Statskontoret ( 2012) Tänk till om tillsynen: om utformningen av statlig tillsyn [Rethink supervision: on the design of state supervision] , Stockholm, Statskontoret. Szebehely M. ( 2011) ‘Insatser för äldre och funktionshindrade i privat regi [Measures for elderly and handicapped in private care provision]’, in Hartman L. (ed.), Konkurrensens konsekvenser: Vad händer med svensk välfärd? [Consequences of competition: What is happening with Swedish welfare?] , Stockholm, SNS Förlag. Szebehely M., Trydegård G.-B. ( 2012) ‘ Home care for older people in Sweden: A universal model in transition’, Health and Social Care in the Community , 20( 3), pp. 300– 9. Google Scholar CrossRef Search ADS PubMed  Waerness K. ( 1984) ‘ The rationality of caring’, Economic and Industrial Democracy , 5, pp. 185– 211. Google Scholar CrossRef Search ADS   Weiss C. H. ( 1972) Evaluation Research: Methods of Assessing Program Effectiveness , Englewood Cliffs, NJ, Prentice-Hall. © The Author 2017. Published by Oxford University Press on behalf of The British Association of Social Workers. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Tong, Catherine E;Franke, Thea;Larcombe, Karen;Gould, Joanie Sims

doi: 10.1093/bjsw/bcx044pmid: N/A

Abstract Organisations promoting health and social outcomes are grappling with two concurrent realities: dwindling budgets and declining state support; and a rapidly ageing population. This is true for all levels of government, non-governmental organisations and non-profits. This study assessed the process and extent to which four non-profit organisations collaborated to meet service objectives related to older adults in a local area. A collaboration survey and semi-structured interviews with nine stakeholders from four community-based public-sector organisations were conducted annually for three years. Interviews were transcribed and data were analysed using topic and analytic coding. Successful inter-agency collaborations involved: (i) shared vision; (ii) effective communication; (iii) time to build relationships; (iv) shared expertise and resources; and (v) strong leadership. Factors that jeopardised inter-agency collaboration included: (i) misinformed understanding of goals; (ii) meetings seen as a waste of time; (iii) not sharing resources; and (iv) lack of organisational resources. This paper makes two distinct contributions. We highlight that successful collaborations are about a process that includes relationship building, sharing of resources and establishing a shared vision; and we offer a method for those involved in the establishment and assessment of collaborations to provide appropriate, accessible and timely assessments of collaborative efforts. Collaboration, voluntary and non-profit organisations (VNPOs), qualitative methods, older adults, community-based social work Introduction Organisations that work to promote health and social outcomes are grappling with two concurrent realities: a climate of austerity (Cottarelli, 2012) and, in many countries, a rapidly ageing population (United Nations, 2013). This is true for all levels of government, non-governmental organisations (NGOs), and voluntary and non-profit organisations (VNPOs). Within the context of dwindling budgets and declining state support, community-based health promotion is increasingly coming under the purview of NGOs and VNPOs. These organisations are expected to do more with less. As the population ages, older adults are increasingly remaining in their own homes and community environments (Office of Policy Development and Research, 2013). As they age, older adults often have increased health and social care needs (International Federation of Social Workers, 2012). No one single agency has the staff or resources to meet all the needs of the ageing population within the community. The purpose of this study is to assess the process and extent to which nine stakeholders from four community-based public-sector organisations collaborated to meet broad service objectives related to older adults in the local area. Inter-agency collaboration: background Many organisations, both public and private, have developed inter-agency collaborations as a result of deficiences existing in service systems. Deficiences such as shortage of funds or resources, lack of qualified or trained personnel, governmental priorities or mandates, duplicated services or the need for additional services (Grubbs, 2000) have been reduced because of inter-agency collaborations. Inter-agency collaborations are recognised as an effective way to: reduce costs (Gulzar and Henry, 2005); create a means for managers to share their responsibilities, and reduce organisational stress (Van Eyk and Baum, 2002); increase ability to reach under-served communities and improve community health (Teufel et al., 2009); and align to advocacy work and sustain funding (Osborne and Murray, 2000). Inter-agency collaboration has been defined in many ways (Frey et al., 2006). Inter-agency collaboration can occur at the governmental and local organisation level, and may involve public, private and/or faith-based sectors as partners. We employ the following definition: inter-agency collaboration is ‘a mutually beneficial relationship between two or more parties who work towards common goals by sharing responsibilities, authority, and accountability for achieving results’ (Chrislip and Larson, 1994, p. 5). This definition delineates not only a shared goal and a mutually beneficial relationship, but also a shared responsibility for the process in establishing the collaboration and the work of the collaboration. Inter-agency collaboration: success There is extensive literature on inter-agency collaboration (Van Eyk and Baum, 2002) that identifies the elements necessary for success. Known factors include: mutual goals and mutual trust (Osborne and Murray, 2000); shared decision making (Van Eyk and Baum, 2002); high-quality inter-personal relationships (Gulzar and Henry, 2005; Cross et al., 2009); and clearly defined governance, roles and leadership (Weiss et al., 2010). Studies have demonstrated that the following processes are needed to implement a successful collaboration: developing a powerful vision, a strongly committed team, trustworthy relationships and stakeholder involvement (Kubisch et al., 2010; Packard et al., 2013). Inter-agency collaboration: challenges Often, many collaborations do not generate efficiencies (e.g. cost savings, increase ability to reach, reduce organisational stress and align to advocacy work) as they are ‘complicated and difficult to manage’ (Cheadle et al., 2005, p. 639) and face a myriad of challenges (Grubbs, 2000). Cited challenges include: a lack of time and resources (Gulzar and Henry, 2005); funding issues; lack of knowledge and poor information flow; lack of personal links and trust; and staff turnover and organisational changes (Van Eyk and Baum, 2002; Weiss et al., 2010). Additionally, outcomes extend beyond the overall efficiency goal for collaborating (e.g. reduce costs, reduce organisational stress). They can be more proximal outcomes (e.g. cognitions, motivations, affect and values), such as staff satisfaction (Gulzar and Henry, 2005) and trust (Salas et al., 2005) and other affective and cognitive processes that are the result of the collaborative process. Collaboration efforts are also hindered by process-oriented challenges such as not recognising the need to accommodate culture, values and goals of existing facilities (Weil, 2010). Differences in organisational culture can act as a significant barrier to bringing organisations together (Fulop et al., 2002). Rather than restricting collaborations to a particular outcome, more focus needs to be placed instead on the process of creating successful collaborations. Seikkula and Arnkil demonstrate the need for a movement beyond a focus on efficiencies. A collaboration: … is not merely a cognitive process but is, instead, an embodied emotional experience. It is not only ‘seeing’ or ‘understanding,’ but also becoming touched as a human being. The new understanding is generated in a shared emotional experience, which means that people become connected with each other in a new, active way (Seikkula and Arnkil, 2006, p. 92). Inter-agency collaboration: serving the needs of older adults Only a handful of studies have examined inter-agency collaboration amongst older adult service providers (e.g. Alkema et al., 2003; Vogel et al., 2007; Ford et al., 2010). While these articles provide descriptions of inter-agency collaborations, none of them provides any sort of data, quantitative or qualitative, from those individuals involved in the partnerships and the processes behind inter-agency collaborations. Ongoing evaluations and data collection at multiple time points, from the perspectives of the individuals involved in the collaboration, may provide greater insights into the processes that make or break collaborative initiatives. Funding for the development of partnerships and collaborative actions is finite, but the efficiency rationale is that well-established collaborations will outlive their funding streams (Frey et al., 2006). While many authors have explored the behaviours and conditions that help to explain why collaborations succeed or fail, there is a need to better understand the processes by which individuals representing various organisations work in concert (Phillips et al., 2000) and, in particular, solutions to challenges for those servicing the ageing population. With the ageing of the population and the growing demand and expectation that organisations should and will collaborate, effort must be made to examine and report on not only the outcomes, but also the processes which allow collaborations to flourish, hence the focus of this study. Methods Setting and context: South Vancouver Neighbourhood House South Vancouver Neighbourhood House (SVNH) acknowledges the need to cultivate strategic collaborations in order to meet its broad community service objectives. SVNH is also aware that other organisations working in their area of Vancouver, British Columbia, have aligned advocacy and community service objectives. Their programmes are diverse, ranging from immigrant settlement support to day-care services and Tai Chi classes; programmes are run by volunteers and staff, including community-based social workers. In an effort to better serve the needs of older adults in the South Vancouver area, SVNH was awarded three years of pilot funding, from a group of private and public funders, to develop a Hub. A Hub is a ‘consortium’ of local partners who coordinate their seniors’ programmes and resources, and work collaboratively on key programming and advocacy issues. SVNH partnered with our university-based research team to design and implement an evaluation of the Seniors’ Hub (www.theseniorshub.org), herein referred to as the ‘Hub’. Collectively, we worked with SVNH, older adult volunteers and the Hub’s community partners to develop and implement a mixed-methods evaluation plan. The three-year evaluation plan covered the duration of the pilot funding (2011–14). A core component of the evaluation plan is tracking the progress and process of inter-agency collaborations. Evaluation approach The ongoing evaluation of the Hub takes a post-positivist, ‘Fourth Generation’ evaluation approach (Guba and Lincoln, 1989), in which the stakeholders and evaluators work in concert to develop and implement an evaluation of a given programme. The evaluation team consists of scientists from the University of British Columbia, Centre for Hip Health and Mobility, older adult volunteers from the Hub, SVNH staff and representatives from the funders. The team meets monthly. Evaluation results are shared through monthly and annual reports, plain-language summaries for older adult volunteers and community presentations. Measurement/assessment of collaboration Collaboration has been examined many ways, including: network analyses, focus groups, interviews, document analysis and survey instruments. For the evaluation of the inter-agency collaboration outcomes, we draw on two tools: annual in-depth interviews with the Hub partners and the ‘Levels of Collaboration Survey’ (Frey et al., 2006), also completed by the partners. We selected the Frey tool, in consultation with the Hub partners, because it is easy and quick to administer, provides concrete definitions and has been thoroughly tested, with published protocols. Partner participation Staff members who represent the partner agencies and SVNH were eligible to participate in the interviews and complete the survey. Recruitment procedures At a monthly inter-agency partner meeting, the Hub coordinator introduced the interview and survey component of the evaluation and provided attendees with a one-page summary. Upon receipt of ethics approval from the University of British Columbia, introduction letters and consent forms were circulated to each inter-agency staff representative. All of the Hub partners agreed to participate. The partners include Southside Community Centre (CC), Union CC and Winston CC. Per our ethics application, the names of the partner organisation and participants have been anonymised. The name of the lead agency, SVNH, has not been blinded. The nine staff members interviewed had an average of sixteen years’ experience in the community development sector (range: four to ‘more than forty’ years) and speak many of the languages used in their ethnically diverse neighbourhood. Their job titles included: volunteer coordinator, seniors outreach worker, recreation/fitness programmer, programme manager/assistant and community development coordinator. This present paper draws on the in-depth, qualitative interviews conducted with the inter-agency partners in the formative years of the organisation (Years 1–3). We conducted seventeen in-depth, semi-structured interviews with nine staff members employed by the Hub partners. We completed interviews at three time points (January 2012, 2013 and 2014), speaking with the same staff members annually. In cases of leave or staff turnover (n = 4), the interim or new employee in the same role was interviewed. Data generation and procedures: interviews In-depth semi-structured interviews ranged from forty-five to ninety minutes each. Interviews were completed by the lead author and conducted at a private location and time of the participant’s choosing. The interview guide asked questions about the development and evolution of the Hub, successes and challenges. Data generation and procedures: ‘Levels of Collaboration Survey’ To track the development of the collaborations, we also used the ‘Levels of Collaboration Survey’ (Frey et al., 2006)—a brief survey that asks partners to rate their perceived level of collaboration on a scale of 0 to 5. Each of the ratings (1–5) is named and defined on the survey instrument (see Table 1). We then used the survey results to develop ‘collaboration maps’—a visual representation of the partners and their perceived relationships between one another. Partners completed the survey at an inter-agency interview meeting shortly after the round of annual interviews. The survey instrument asks the participants to identify all of their perceived partners, which may include both formal partners and other local organisations. Table 1 Sample Levels of Collaboration Survey Please review the descriptions of the different levels of collaborations. Using the 0-5 scale provided, please indicate what ‘level’ you are currently at with each partner of the HUB. All HUB partners should fill this out independently. NAME: ______________________________________________ORGANIZATION___________________________________________DATE____________________________________ Five Levels of Collaboration     Relationship  No interaction  Networking  Cooperation  Coordination  Coalition  Collaboration  Characteristics  0  1  2  3  4  5  • no interaction, but identified as a potential partner  • Aware of organization • Loosely defined roles • Little communication • All decisions made independently  • Provide information to each other • Somewhat defined roles • Formal communication • All decisions are made independently  • Share information and resources • Defined roles • Frequent communication • Some shared decision making  • Share ideas • Share resources • Frequent and prioritized communication • All members have a vote in decision making  • Members belong to one system • Frequent communication is characterized by mutual trust • Consensus is reached on all decisions    Partner Organization              e.g. Union CC              Five Levels of Collaboration     Relationship  No interaction  Networking  Cooperation  Coordination  Coalition  Collaboration  Characteristics  0  1  2  3  4  5  • no interaction, but identified as a potential partner  • Aware of organization • Loosely defined roles • Little communication • All decisions made independently  • Provide information to each other • Somewhat defined roles • Formal communication • All decisions are made independently  • Share information and resources • Defined roles • Frequent communication • Some shared decision making  • Share ideas • Share resources • Frequent and prioritized communication • All members have a vote in decision making  • Members belong to one system • Frequent communication is characterized by mutual trust • Consensus is reached on all decisions    Partner Organization              e.g. Union CC              Data analysis: interviews Interview audio files were transcribed verbatim by a professional agency, and spot-checked for quality and completeness. Transcripts were blinded and uploaded into NVivo 10 (qualitative data analysis software). Each transcript was read twice and a preliminary coding structure was developed in team meetings to facilitate the composition of a two-page summary report for participants. Inductive topic coding (Richards and Morse, 2012) was used to organise the data. Transcripts were coded each year, with new nodes to reflect new themes and topics. Transcripts were coded for themes constant across all three time points (e.g. node: examples of working collaboratively) and themes unique to a given year (e.g. node: barrier: troubles with meetings and time-keeping: Years 1 and 2). A final round of coding and analysis was conducted after all three rounds of data collection, collapsing nodes and refining themes. Data analysis: ‘Levels of Collaboration Survey’ The development of the ‘collaboration maps’ followed the protocol outlined by Frey and colleagues (2006). Partners are represented with circles and the lines of varying widths represent the levels of collaboration, with a thicker line indicating a higher level of collaboration. Arrows indicate the direction of the relationship. For example, if SVNH rates Southside CC as a 3 (Coordination), a line with the appropriate width would originate from the SVNH circle, with the arrow pointing towards the Southside circle. Relationships with a rating of zero (no relationship) or 1 (networking) are not represented in the map. In the instances where partners identify collaboration with organisations that are not formal partners, and therefore would not have been provided with a survey in which to respond, we use a circle with dashed lines. Collaboration maps were shared with stakeholders annually. Figure 1 provides the baseline ‘collaboration map’ for the Hub. Figure 1 View largeDownload slide Collaboration map: Baseline Figure 1 View largeDownload slide Collaboration map: Baseline Strategies for rigour The validity of a qualitative study cannot be ensured through the application of post-hoc strategies (Morse et al., 2008), but rather researchers must employ a number of procedures throughout the research study. Validity is defined as ‘how accurately the account represents participants’ realities of the social phenomena and is credible to them’ (Creswell and Miller, 2000, p. 125). Validity refers not to the data, but to the inferences drawn from them. Strategies to reinforce the validity of this study included: cross-checking the full transcripts against the original audio files; reflexive memoing; and the identification and thoughtful examination of outliers. We also maintained an audit trail (Koch and Kralik, 2006). Through a three-year engagement with participants, we ensured the validity of this study by providing annual summary reports, revising the interview guide and maintaining communication with the participants. Participants provided feedback on the preliminary analysis and one staff member from SVNH reviewed several drafts of this manuscript. Findings Through three rounds of annual interviews, participants provide reflection about the progress and factors related to the specific successes and problems that are part of cultivating a high level of inter-agency collaboration. Successful inter-agency collaborations were found to involve: (i) shared vision; (ii) effective communication; (iii) time to build relationships; (iv) shared expertise, ideas and resources; and (v) a strong leadership role. Factors that jeopardised inter-agency collaboration and areas each group would change in future collaborative efforts included: (i) misinformed understanding of collaboration goals; (ii) meetings seen as a waste of time; (iii) not sharing resources; and (iv) lack of organisational resources. Solutions that the participants implemented are discussed throughout. Successful factors Shared vision Establishing a shared vision and understanding for the Hub was a key factor in setting the foundation for success. Participants were asked during their initial interviews what attracted them to participating in the Hub (Figure 2), and they all noted a shared understanding that more had to be done to coordinate efforts and support local seniors: Figure 2 View largeDownload slide Collaboration map: Year 1 Figure 2 View largeDownload slide Collaboration map: Year 1 I think that a lot of our facilities we would love to do more … . We’re limited in terms of resources we can offer. But the idea of creating hubs of opportunities and recreation and programs in the South Vancouver area, really set in my mind, that at least it would create a greater, broad-based resource for older adults to access (Bob, Year 1). The Hub partners demonstrated a shared vision to making their local area a better place for seniors by sharing resources and providing greater opportunities: Jennifer explained the importance and benefits of working together in order to better serve local seniors: … [before we started meeting as a Hub]… we didn’t really know what was happening in those communities. So now that we know that was happening, what the resources that we have in place, we can bring all that information back to the committees and we can work from that (Jennifer, Year 1). Challenges (misinformed understanding of collaboration goals): Without establishing a shared vision and understanding of the goals of the Hub, efforts to maintain the inter-agency collaboration could be jeopardised. Susan shares her difficulties in joining the Hub and understanding its definition and purpose: I had no idea what the Hub was for the first three months. I think it’s actually conceptually very hard to understand. And I think it’s a weakness of the Hub … . But I do see, even in the six months that I’ve been here, that there’s been more actual partnership between the community centres and the Neighbourhood House (Year 2). Solution: Developing a shared definition of the Hub and governance structures that focus on visioning, strategic planning and practice changes takes time and effort. This process can often be perceived as confusing or ‘conceptually hard to understand’, in particular for individuals who are keen to see tangible results immediately. This is especially true when annual reports focus on outcomes, and not necessarily the processes that need to take place allowing such outcomes to be achieved. The collaboration process requires participants to accept a level of uncertainty as the discussion of the process and establishment of structures develop. Effective communication Central to the success of the Hub was the process of establishing strategies for effective communication to minimise confusion and to ensure that all partners could provide answers to questions such as: Who we are? What are we trying to do? How are we going to do it? Participants spoke at length about the importance of establishing effective communication in the initial phases so that all partners continue to remain enthusiastic about the collaboration. Participants highlighted the importance of strategies such as ‘face-to-face meetings’ that bring partners together to learn from one another and enhance each individuals’ role: I’ve gained a lot of knowledge towards programming, towards collaborating, … working with other groups. If you surround yourself with people of so many different skills, different knowledge, just think what you could collaborate with that. Susan has been programming for many more years than I have and she’s got a great wealth of information behind her (Lynn, Year 2). More experienced staff discussed learning about one another’s working and communication styles. Similarly, junior staff appreciated the opportunity to learn from the more experienced partners. Jennifer explained: If you know that there is something that involves, I don’t know, let’s say Union Community Centre, and we know Susan already and we can talk to her, and she is the one working with seniors … she will for sure know how I can do that better. She will guide me. So you feel more confident and you feel more like you are not alone in the community anymore (Year 1). Dialogue and communication can be the star agents of transformation in collaborative efforts (Seikkula and Arnkil, 2006). Challenges (meetings seen as a waste of time): Communicating the importance of bringing the partners together is a key component to the future success of the collaboration. Without demonstrating that regular meetings help to establish connections, understandings and learnings, partners can view the meetings as a ‘waste’ of time and reduce their enthusiasm throughout the process: When you start something new up it’s like, okay, why am I here? I felt like I just went to a two-hour meeting and really didn’t get anything accomplished. That’s when you walk out going, well, this was a waste of my time. So I think we’re a little further along now where we’re utilizing our times better. We’re really understanding how we can build off of each other as well (Year 2). In their review, Packard and colleagues (2013) found that community groups usually understood that reorganisation or collaboration was a fait accompli, but their acceptance of the change and enthusiasm for implementation depended on their involvement in real-time planning. Solutions: Participants noted that flexibility and reflection, or adaptability and learning as Kubisch and colleagues (2010) wrote, are key elements that can facilitate the collaborative process and establish the importance of ongoing communication and effort for partners: I think you have to have an open mind. I think you have to be flexible. And you have to be open to trying things, that’s the key. Because we found that with some stuff that we try, they don’t work. Is there any way we can change it to make it work? So it’s really going through that evaluation process. And then if it’s not, let’s move on (Lynn, Year 2). Time to build relationships While articulating and defining the Hub initially proved challenging, the lengthy process of establishing this foundation was key to the success of collaborations developing. Susan shared her reflections about how through ongoing communication and time relationships were built and strengthened: For the second year, I see more information sharing and trust among the partners, well we are built stronger than before. Like the shared projects or seeing how frequently we share the information and we promote partners activities or events, I think it reflects the trust and the partnership (Year 2). The collaboration coalescing is captured in the collaboration map for Year 2, in which the mean level of collaboration increased from 3.0 in Year 1 to 4.4 (see Figure 3). Figure 3 View largeDownload slide Collaboration map: Year 2 Figure 3 View largeDownload slide Collaboration map: Year 2 While Figure 3 provides a visual representation of a strengthening Hub, what it does not show is the amount of time, effort and work that goes into creating and widening each line in the collaboration map. Each line can potentially represent hundreds of e-mails, phone calls and hours spent in meetings in order to establish and strengthen partner relationships. The ongoing flourishing of the Hub collaborations is visually represented in the collaboration map for Year 3 (see Figure 4). Figure 4 View largeDownload slide Collaboration map: Year 3 Figure 4 View largeDownload slide Collaboration map: Year 3 Taking the time to establish a foundation upon which trustworthy and effective relationships can be built is a key factor to the success of a collaboration. The final round of interviews were replete with examples of successful collaborative processes and outcomes. In terms of processes, staff members continued to discuss learning from one another. For example, Susan is creating a board of older adult volunteers to guide the seniors’ programming at her community centre; in order to do so, she has relied heavily on the partners: Judith and Anna are really familiar with that area. So they’ve offered to come in and help to facilitate meetings. They’ve been very supportive with ideas when we’ve sat at the table for our meetings … [they are] a good resource and a support team (Year 3). Challenges (lack of time): In spite of the many successes of the collaboration partnerships over time, it is precisely ‘time’ that can be seen as a challenge in that building of relationships and collaborations take time. The initial phases (Years 1 and 2) involved challenges that needed to be overcome with time. Solutions: Participants highlighted how the initial phases of establishing a shared goal and cohesive commitment was essential to the success of the Hub and that time is essential to the process. This finding is consistent with Kubisch and colleagues (2010), who highlight that ‘too often, the goals sought by community change intervention are poorly defined at the outset … having agreement among all stakeholders about the work that will be approached is key to creating and maintaining focus’ (p. 11). The strength of the commitment that was built over time was what brought partners together and helped to maintain the ongoing relationships and new relationships that needed to develop: ‘We’re all there for the same cause. And I see the passion in everybody around the table, wanting to get to our goal’ (Judith, Year 3). Shared expertise, ideas and resources Throughout the previous themes is the underlying importance of partners’ willingness to share expertise, ideas and resources. All of the partners talked about the successful outcomes they experienced by sharing expertise, ideas and resources. Raj is a half-time employee who often attends Hub meetings on personal time. He explained how the Hub saves him time: Because we have collaborative events I don’t have to program more new events or things on my own. I’ll take one event from Union Community Centre, one event from Winston Community Centre and then offer that in my brochure whereas before I’d have to think of three other programs from my own. It helps, coming up with ideas (Year 2). Participants shared examples of creating events together, promoting one another’s event in their brochures, sharing equipment (e.g. newer bingo machines, Nordic walking poles) and sharing the cost of transportation to help their older adults participate in events at all of the partner facilities. The partners felt that cross-promotion and shared resources helped increase their enrolment. Susan explained one initiative to get more seniors attending an event at one of the CCs: We used one of Winston Community Centre’ buses and we provided the driver. We invited some of their members to come and take part in the event, so they had another bus. It was mutually beneficial where their members got to come for free and we got to access another bus to get more of our community members from this area over to Southside Community Centre (Year 3). The cross-promotion of some seniors’ events (e.g. a walking group field trip) has been so successful that the partners had to come to an agreement on how to split the profits. Profits are rolled back into the partners’ organisations to support future seniors’ programming. Challenges (not sharing resources): Prior to launching the Hub, the partners recognised that they were failing to share resources and duplicating efforts. One frequently cited example was that two partner agencies, geographically close to one another, planned a Strawberry Tea—a theme-based gathering—on the same afternoon, targeting the same cohort of local seniors. Solutions: This tea, which immediately preceded the launch of the Hub, was later a shared event and was an example of how they could move forward, as partners, potentially doing more for seniors by coordinating their efforts. Recognising this potential resulted in an enthusiastic buy-in from all of the partner agencies. A strong leadership role Our participants spoke strongly about the essential role of the Hub coordinator, the individual who plays the central role in bringing the partners together, and keeping them working together. The Hub coordinator undertook much of the e-mails, phone calls and hours spent organising meetings that were vital to the success of supporting and facilitating the growth of the collaboration relationships. As the Hub continued to grow, several participants, like Anna, recognised the sheer volume of work that the Hub coordinator oversaw in her efforts to launch the Hub: It’s essential for the long term success of the Hub … I don’t see anybody being able to have the full-time hours to do the amount of meeting and coordinating that there is … . A lot can be done by volunteers, but there’s a coordinating role that needs remuneration. One person’s going to have to do it (Year 2). While it is often assumed that collaborations can occur within the hours and existing funded positions, Anne emphasised the need for a funded coordination position to successfully launch the Hub. Challenge (lack of organisational resources): Concern was expressed as to the amount of work required of the Hub coordinator. The Hub coordinator was a paid, full-time position and assisted by a part-time seniors’ outreach coordinator. The Hub coordinator functioned as the cog at the centre of the Hub, and invested a significant amount of time and energy developing and supporting each line in the collaborations maps (Figures 1–4). The Hub coordinator participated in approximately 350 meetings per year; for the vast majority of these, the Hub coordinator was charged with calling, scheduling, drafting the agenda and chairing the meeting: I’m in awe of what [the Hub coordinator] is able to do, you know. Because they’re human beings, they only have certain hours per week to do the job, right … one person has been maximized all over the map. And how she can keep track of all the things, you know, that’s—she’s doing … . But definitely for me it’s kind of like a concern (Daniel, Year 2). Solution: The Hub coordinator role was seen as essential for the success of the Hub and therefore the only solution highlighted was the value placed on funding this role. The Hub and partner agencies submitted additional grant applications in order to maintain ongoing funding for this role. Discussion Through several years of relationship- and collaboration-building, the Hub has come to better and more efficiently serve local seniors through the process of learning to share activities, share resources and more widely promote events. As a group, they moved from uncoordinated efforts, duplication of events and inefficient use of time and resources to a flourishing coordinated effort. The unwillingness to share resources is often cited as a reason for the failure of many collaborations (Grubbs, 2000). However, this was not an issue for the Hub partners due to the time spent in the initial phase of establishing a shared vision that allowed the willingness to share and try different approaches to grow. Consistently with community-based social work, a key step in the Hub development and resultant inter-agency collaboration included ‘agreeing on core values, common goals and strategic plans allows partners to develop a common language, appreciate and acknowledge the experience of others’ and respect diverse perspectives (Child Welfare, 2008). Simultaneously, establishing effective communication at the beginning allowed the participants to report learning from one another, and having found ‘a broader network of support’ (Raj, Year 2). The collaboration literature emphasises having the right personnel (Gulzar and Henry, 2005; Weiss et al., 2010), yet it is not just about bringing the right people around the table. As noted by our participants, the real work happens once everyone around the table establishes and strengthens the relational ties necessary to function collaboratively. One of the key successes of the early years of the Hub was establishing this foundational ‘shared vision’. Our findings from the interviews and collaboration maps demonstrate that this process requires an investment of both energy (effective communication) and time (time to build relationships). The participants spoke enthusiastically about getting to know one another, sharing their vision of service for seniors and learning from one another. While initiatives such as the Hub are often supported with a year of seed funding to get up and running, both the interviews and collaboration maps demonstrate the real collaborative efforts begin flourishing after the initial years of relationship- and collaboration-building take place. The programming outcomes achieved principally in Year 3 were predicated on the relationship- and collaboration-building that occurred in the first two years. Our findings highlight the essential role of longer-term funding to allow the processes necessary for collaborations to be built and become more established. We see a high concordance between the annual collaboration maps and the annual interviews. Overall, the maps and interviews show a period of working together as a small group (Year 1) with challenges similar to those experienced by many collaborative efforts (e.g. Gulzar and Henry, 2005). Before the Hub could achieve many of their service outcomes, the group required two years to establish themselves, learn from one another, and develop a shared understanding of who they were as a group and what they were going to achieve together. This underscores the importance of outcomes beyond efficiencies to more affective, motivational outcomes such as staff satisfaction and trust that result in the collaborative process. Figures 2 and 3 demonstrate this period of strengthening, which preceded the period of expansion, evident in Figure 4. Collectively, our findings show a story of success, of organisations and individuals overcoming the oft-cited challenges associated with collaboration, and learning the process of accommodating culture, values and goals of existing facilities in order to more effectively work together to serve their local ageing population. In a context of dwindling budgets and an ageing population, their efforts provide a model worthy of emulation (the Hub’s governance manual is available here: http://www.theseniorshub.org/uploads/1/1/4/2/11425458/governance-manual.pdf). In addition, our findings demonstrate that sustainable funding for inter-agency collaborations must also include funding of a Hub coordinator position to ensure the management and organisation of strengthening partnership relationships over time. Our findings demonstrate the importance of having different types of staff and individuals involved in the collaboration process. Participants cited examples of growth by pairing junior and senior staff, and by working with partners with different areas of expertise. Team diversity is essential when trying to reach under-served older adults (Alkema et al., 2003). Social workers are particularly well positioned to take on a leadership role in facilitating this type of collaboration. Longer-term evaluation plans must also be in place to track the evolution of collaboration efforts. Indeed, this was the vision of the group of organisations that funded the Hub: rather than supporting numerous smaller, short-term initiatives, they pooled their funds to create sustainable funding for one three-year project, which included a core evaluation component. Cross and colleagues (2009) have suggested that one of the reasons collaborations fail is because participants are not provided with appropriate, accessible and timely assessments of their collaborative efforts, thereby allowing for refinements and course-corrections. Our evaluation provided participants with ongoing, accessible feedback regarding their growing collaboration. Repeat interviews provided important discussion on participants’ collaborative experience; they allowed the partners to reflect on their work and compare their experiences from year to year. For example, under the theme time to build relationships, Susan compares her experience from Year 2 with Year 1 and highlights the important role that time played in order to establish trust and partnership. Additionally, repeat interviews allowed timely feedback on the collaborative process; findings were shared in an anonymised, plain-language summary report and facilitated the collaborative process. While the interviews provide the deep contextual information, the maps provide a strong visual representation which is easy to share and communicate with stakeholders. Over the years, the partners expressed their appreciation for the tool. They found the maps to be easily interpreted and useful in sharing information visually with all of their stakeholders, including older adult volunteers. Their funders also enjoyed seeing actual, tangible results of their collaborative efforts, through the reports that contained both the visual graphs and quotes from the interviews. Additionally, they appreciated seeing the links that weakened over time, with poorer ratings, because they then knew where to intervene or increase their efforts. However, the collaboration maps do not highlight the intricacies of humans and organisations learning to work together that are more fully and appropriately captured in the interviews. We suggest the use of both repeat interviews and collaborations maps in order to evaluate an inter-agency collaboration. Stahl and Shdaimah (2008) maintain that collaboration between academic researchers and community-based groups is, more generally, an effective way to study social problems. Limitations Although the university ethics committee approved the recruitment protocol, the recruitment procedures that we used could be construed as a form of coercion by having the Hub coordinator recruit participants. Prior to being recruited for interviews, however, it was the participants themselves who had requested the evaluation techniques that we employed. While we aimed to interview all participants at each stage, we were unable to do so due to administrative changes in roles and new hires. Additionally, the mapping can provide an effective way to represent the collaborative activities of the Hub and its partners; the findings are compromised by limitations including the assumption that the person completing the survey has all of the knowledge of how the agency collaborated with others. The key benefit of this tool, however, is that it is able to capture how each individual, be they a programme assistant or manager, perceives the development of the collaboration. It can be argued that assessing participants’ perceptions of the collaboration is just as important and a more global, macro assessment of the Hub. Conclusion Our paper makes two distinct contributions. First, we highlight that successful collaborations are about a process that includes relationship building, sharing of resources and establishing a shared vision. This extends previous literature that focuses primarily on outcomes of collaborations as being the marker of success. Based on these findings, social workers, with training in community development/practice, are particularly well positioned to undertake the work of establishing and supporting collaborations like the Hub. Second, we offer a method for those involved in the establishment and assessment of collaborations to provide appropriate, accessible and timely assessments of collaborative efforts. This method includes the use of repeat interviewing and collaborative mapping. Taken together, the two provide a powerful narrative and visual record to communicate both to those involved in the collaboration and external stakeholders such as government or other potential funders. Our findings show that inter-agency collaboration can indeed produce efficiencies, reduce overlap and help streamline services so that agencies can try to do more with less (Frey et al., 2006) or simply do better by learning from one another. These efficiencies, however, do not absolve governments and funders from supporting those central coordination and evaluation positions that facilitate the collaborative efforts. Inter-agency collaborations to improve services for older adults in the community have been shown to have great potential. In order for them to fulfil that potential, we need to continue to improve the knowledge base about how to how to implement community strategies, how to assess what is working and why, and, finally, how to ensure that all of the key actors make use of and apply that knowledge. Acknowledgements Funding for the evaluation was provided by the Vancouver Foundation (Grant# 20R07558). Catherine Tong’s work is supported by the Canadian Institutes of Health Research Doctoral Award and the University of British Columbia’s Four Year Fellowships. Thea Franke’s work is also supported by the Canadian Institutes of Health Research Doctoral Award and the University of British Columbia’s Four Year Fellowships. Dr Sims Gould is supported by a Canadian Institutes of Health Research New Investigator award and a Michael Smith Foundation for Health Research Scholar award. Our sincerest thanks to SVNH, the Seniors Hub Council and all of the Hub partners for allowing us to work alongside you; it was such a privilege. We also recognise research assistant Suman Auluck for her contributions to the evaluation. References Alkema G. E., Shannon G. R., Wilber K. H. ( 2003) ‘ Using interagency collaboration to serve older adults with chronic care needs: The Care Advocate Program’, Family & Community Health , 26( 3), pp. 221– 9. Google Scholar CrossRef Search ADS PubMed  Cheadle A., Senter S., Solomon L., Beery W. L., Schwartz P. M. ( 2005) ‘ A qualitative exploration of alternative strategies for building community health partnerships: Collaboration-versus issue-oriented approaches’, Journal of Urban Health , 82( 4), pp. 638– 52. Google Scholar CrossRef Search ADS PubMed  Child Welfare ( 2008) Interagency Collaboration, available online at https://www.childwelfare.gov/pubPDFs/interagency.pdf (accessed 11 May 2017). Chrislip D. D., Larson C. E. ( 1994) Collaborative Leadership: How Citizens and Civic Leaders Can Make a Difference , San Francisco, CA, Jossey-Bass. Cottarelli C. ( 2012) ‘The austerity debate: Make haste slowly’, in Austerity: Too Much of a Good Thing?, availble online at http://voxeu.org/sites/default/files/file/austerity_ecollection.pdf (accessed 11 May 2017), pp. 39–44. Creswell J. W., Miller D. L. ( 2000) ‘ Determining validity in qualitative inquiry’, Theory into Practice , 39( 3), pp. 124– 30. Google Scholar CrossRef Search ADS   Cross J. E., Dickmann E., Newman-Gonchar R., Fagan J. M. ( 2009) ‘ Using mixed-method design and network analysis to measure development of interagency collaboration’, American Journal of Evaluation , 30( 3), pp. 310– 29. Google Scholar CrossRef Search ADS   Ford C. R., Henderson J., Handley D. M. ( 2010) ‘ Enhancing long-term care for older adults: An exploration of interagency collaboration within geriatric education centers’, Journal of Health and Human Services Administration , 32( 4), pp. 447– 85. Google Scholar PubMed  Frey B. B., Lohmeier J. H., Lee S. W., Tollefson N. ( 2006) ‘ Measuring collaboration among grant partners’, American Journal of Evaluation , 27( 3), pp. 383– 92. Google Scholar CrossRef Search ADS   Fulop N., Protopsaltis G., Hutchings A., King A., Allen P., Normand C., Walters R. ( 2002) ‘ Process and impact of mergers of NHS trusts: Multicentre case study and management cost analysis’, BMJ , 325( 7358), pp. 246– 53. Google Scholar CrossRef Search ADS PubMed  Grubbs J. W. ( 2000) ‘ Can agencies work together? Collaboration in public and nonprofit organizations’, Public Administration Review , 60( 3), pp. 275– 280. Google Scholar CrossRef Search ADS   Guba E., Lincoln Y. ( 1989) Fourth Generation Evaluation , Thousand Oaks, CA, Sage Publications. Gulzar L., Henry B. ( 2005) ‘ Interorganizational collaboration for health care between nongovernmental organizations (NGOs) in Pakistan’, Social Science & Medicine , 61( 9), pp. 1930– 43. Google Scholar CrossRef Search ADS   International Federation of Social Workers ( 2012) Ageing and Older Adults, available online at http://ifsw.org/policies/ageing-and-older-adults/ (accessed 11 May 2017). Koch T., Kralik D. ( 2006) Participatory Action Research in Health Care , Oxford, UK, Blackwell Publishing. Kubisch A. C., Auspos P., Brown P., Dewar T. ( 2010) ‘ Community change initiatives from 1990–2010: Accomplishments and implications for future work’, Community Investments , 22( 1), pp. 8– 12. Morse J. M., Barrett M., Mayan M., Olson K., Spiers J. ( 2008) ‘ Verification strategies for establishing reliability and validity in qualitative research’, International Journal of Qualitative Methods , 1( 2), pp. 13– 22. Google Scholar CrossRef Search ADS   Office of Policy Development and Research ( 2013) Aging in Place: Facilitating Choice and Independence, available online at https://www.huduser.gov/portal/periodicals/em/fall13/highlight1.html (accessed 11 May 2017). Osborne S. P., Murray V. ( 2000) ‘ Collaboration between non-profit organizations in the provision of social services in Canada: Working together or falling apart?’, International Journal of Public Sector Management , 13( 1), pp. 9– 19. Google Scholar CrossRef Search ADS   Packard T., Patti R., Daly D., Tucker-Tatlow J. ( 2013) ‘ Implementing services integration and interagency collaboration: Experiences in seven counties’, Administration in Social Work , 37( 4), pp. 356– 71. Google Scholar CrossRef Search ADS   Phillips N., Lawrence T. B., Hardy C. ( 2000) ‘ Inter‐organizational collaboration and the dynamics of institutional fields’, Journal of Management Studies , 37( 1), pp. 22– 43. Richards L., Morse J. M. ( 2012) Read Me First for a User’s Guide to Qualitative Methods , Thousand Oaks, CA, Sage Publications. Salas E., Sims D. E., Burke C. S. ( 2005) ‘ Is there a “Big Five” in teamwork?’, Small Group Research , 36( 5), pp. 555– 99. Google Scholar CrossRef Search ADS   Seikkula J., Arnkil T. E. ( 2006) Dialogical Meetings in Social Networks , London, Karnac Books. Stahl R., Shdaimah C. ( 2008) ‘ Collaboration between community advocates and academic researchers: Scientific advocacy or political research?’, British Journal of Social Work , 38( 8), pp. 1610– 29. Google Scholar CrossRef Search ADS   Teufel J. A., Brown S. L., Thorne W., Goffinet D. M., Clemons L. ( 2009) ‘ Process and impact evaluation of a legal assistance and health care community partnership’, Health Promotion Practice , 10( 3), pp. 378– 85. Google Scholar CrossRef Search ADS PubMed  United Nations, Department of Economic and Social Affairs, Population Division ( 2013) World Population Ageing 2013, Report No. ST/ESA/SER.A/348, available online at http://www.un.org/en/development/desa/population/publications/pdf/ageing/WorldPopulationAgeing2013.pdf (accessed 11 May 2017). Van Eyk H., Baum F. ( 2002) ‘ Learning about interagency collaboration: Trialling collaborative projects between hospitals and community health services’, Health & Social Care in the Community , 10( 4), pp. 262– 9. Google Scholar CrossRef Search ADS PubMed  Vogel A., Ransom P., Wai S., Luisi D. ( 2007) ‘ Integrating health and social services for older adults: A case study of interagency collaboration’, Journal of Health and Human Services Administration , 30( 2), pp. 199– 228. Google Scholar PubMed  Weil T. ( 2010) ‘ Hospital mergers: A panacea?’, Journal of Health Services Research & Policy , 15( 4), pp. 251– 3. Google Scholar CrossRef Search ADS PubMed  Weiss E. S., Taber S. K., Breslau E. S., Lillie S. E., Li Y. ( 2010) ‘ The role of leadership and management in six southern public health partnerships: A study of member involvement and satisfaction’, Health Education & Behavior , 37( 5), pp. 737– 52. Google Scholar CrossRef Search ADS   © The Author 2017. Published by Oxford University Press on behalf of The British Association of Social Workers. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Shaw,, Julia

doi: 10.1093/bjsw/bcx036pmid: N/A

Abstract This paper considers the post-secular turn in critical theory as enabling vocational social work to re-engage with public religion within deliberative and participatory democratic practices in the interests of social welfare. In particular, by introducing a theoretical context for post-secular social work, spiritual-but-not-religious rhetoric which characterises modern liberal social work is challenged whilst congruency between social work practice and faith-based social action is affirmed through a mutual critique of neo-liberalism and sympathy with post-liberal ethics of care. Accordingly, post-secular social work recognises faith-based social action as representing an alternative and supplementary care paradigm to the bureaucratic cultures and consumerist models purported by both the private and public sectors. Consequently, rather than privatising faith as a personal lifestyle choice, post-secular social work is indicative of a new politics of social work which subverts former secular, liberal and neo-liberal hegemonies within the profession through a neo-communitarian cultivation of the common good. Ethics of care, faith-based social action, Habermas, post-liberal, post-secular, public religion Introduction As a vocation, social work operates within statutory parameters whereby religion subverts the presumptive secular paradigms upon which modern social work functions given the observation that ‘everyone has the right to the freedom of religion or belief, either alone or in community with others and in public and or private, to manifest religion or belief in teaching, practice, worship and observance’ (Article 9 of the Human Rights Act 1998 and Article 18 of the United Nations Declaration of Human Rights). This statute, which defends religious freedom, therefore paradoxically confounds liberal secular perspectives which have historically marginalised religion from the public sphere on the grounds that faith represents subjective belief not universally accessible and therefore inadmissible as a form of reasoning within modern society (Graham, 2013). In comparison to typically liberal secular views on religion, however, this paper proposes that social work is instead critically obliged to engage with public religion and religious citizenship in the interests of deliberative and participatory democracy which resultantly requires congruent theoretical perspectives especially as the Equality Act 2010 additionally identifies religion, rather than spirituality, as a protected characteristic. In particular, by engaging with public religion, post-secular social work recognises correlations between social work, faith-based social action and post-liberal ethics of care whereby liberal and neo-liberal paradigms are subverted in the interests of social welfare. Critiquing secularism, liberalism and neo-liberalism within social work With regard to theoretical context, post-secular social work may therefore be conceptualised as a development of post-secular theory which represents a considerable challenge to secularism within the profession given that, in recent years, theoretical consideration of an evolving post-secular society has emerged as an academic phenomenon which continues to subvert secularist assumptions concerning the inevitable withdrawal of religion from the public sphere (Nynas et al., 2012; Braidotti et al., 2014). For, although post-secular theory initially developed from philosophical and theological critique as typified by Connolly (1999), Milbank (2006) and Taylor (2007), post-secularism continues to attract significant intellectual currency within the social sciences particularly as a development of postmodern and poststructuralist discourse whereby: … postsecularism, arriving after multiple transcending endeavours such as postmodernism, poststructuralism and postcolonialism is commonly invoked to question the Enlightenment’s handling of religion. It specifically questions the ways in which secularism’s imperatives have consigned religion to a space within presumably neutral forms of modern life in liberal-democratic polities (Furani, 2016, p. 1). The privileging of secularism, which is identified within the intellectual limitations of liberalism, logical positivism and empiricism, has therefore become highly contested (Fountain, 2013). For example, secularism from a post-secular perspective is considered to have marginalised religion as too irrelevant, irrational and inconsequential to warrant legitimacy within an expedient public sphere conditioned more towards the rationalisation of procedural and instrumental bureaucracy in the construction of modern civil society (Williams, 2012), whilst programmatic secularism is implicated in the eradication of the manifestation of religious citizenship through the implementation of an apparently neutral yet secular public sphere wherein any residual notion of religion is banished to the private realm of personal belief (Williams, 2008, 2012). Consequently, secularism from a post-secular perspective is typically considered as a disenchanted utilitarianism which erases the religious Other (Graham, 2013). To date, the implementation of secularism within social work has therefore involved a tacit imposition of practical atheism as identified by Whiting (2008) whereby transgressive faith-based discourses remain unarticulated albeit that the residual traces of a Christian heritage within British social work is discernible within the atypical work of Bowpitt (1998, 2000). As such, despite social workers demonstrating a greater affinity with religious ethics than secular educators generally realise, religious literacy remains an unmet need on social work training programmes mainly due to concerns regarding non-rational approaches to problem solving as well as exposure to proselytising (Furman et al., 2004; Gilligan and Furness, 2006; Furness and Gilligan, 2014), this uneasy relationship between social work and its religious origins having remained problematic within a dominant secular narrative which has come to prevail (Vanderwoerd, 2011). However, despite this entrenched ambivalence, social work authors have begun to appreciate that religion and spirituality within inter-faith society are significant to the profession as illustrated by Derezotes (2006), Holloway (2007), Moss and Thompson (2007), Mathews (2009), Crisp (2008, 2010, 2014), Furness and Gilligan (2010, 2012), Moss (2005, 2012) and Starnino et al. (2014). Nevertheless, in the absence of any rigorous engagement with post-secular theory other than brief references by Neagoe (2013), Garde (2015) and Pandya (2016), social work has tended to focus upon equality and diversity concerns in that concession to religion, through an emphasis upon spirituality, is tolerated within an overarching liberal paradigm. For example, liberal spiritual-but-not-religious narratives typically characterise social work publications on religion such as those authored by Henery (2003), Wong and Vinsky (2008) and Holloway and Moss (2010), this new visibility of spirituality within social work being further evident in the work of Gray (2008) and Crisp (2008, 2010). With reference to liberal social theory, the emergence of spirituality within social work literature has therefore entailed resistance to religious traditions along with the alienating, demystifying and disenchanting outcomes of Western capitalism associated with the secularisation of the profession (Gray, 2008). Consequently, learning from non-Western cultures is advocated as an antidote to oppressive religious canon as well as secular imperialism in the promotion of spiritualities which are identified with indigenous, global and new religious movements (Gray, 2005). Accordingly, liberal spiritual-but-not-religious rhetoric within social work recognises spirituality within the context of non-oppressive practice in affirming reflexive consciousness which resists religious dogma as illustrated by Henery (2003) and Moss (2012). Correlatively, Gray (2008) locates social work literature on spirituality within the broad theoretical parameters of late modernity which observes the rise of liberal individualism and its culmination in reflexive modernity. As such, the emergence of spirituality within social work is considered as an attempt to address the alienating effects of modernity which includes a growing emphasis upon empiricism and evidence-based practice within the profession. Indeed, from a Weberian perspective, Gray (2008) acknowledges that social work was required to negate its religious and moralistic beginnings in order to embrace a secular and bureaucratic modern identity characterised by social scientific and empirical epistemology. However, an increased interest in reflexive spirituality is understood as a counter-cultural movement to re-establish qualitative meaning and purpose within the profession. Late modernity is therefore observed to be characterised by a crisis in religious identity whereby individuals find themselves in a constant state of self-questioning in the absence of any traditional foundations to knowledge. In a modern liberal secular society, whilst religious values are eroded, spirituality resultantly becomes synonymous with a pluralistic culture in which citizens are free to select from a number of privatised lifestyle options. The associated decline in organised religion and emergence of new forms of spiritual expression is consequently regarded as releasing communities from religiously controlled sanctions as faith becomes a matter of choice and reflection of autonomous identity rather than an obligation towards society (Gray, 2008)—this shift of authority from religious institutions to individual choice thereby representing a post-traditional religious eclecticism in which spirituality becomes privatised, solipsistic and more concerned with self-empowerment, self-actualisation, mindfulness and identity politics rather than a collective act of service or commitment towards the common good. To this effect, Maslow (1971) observes peak spiritual experiences as a route to achieving individual growth, integration and fulfilment rather than a contributory factor towards community generation. Within modern liberal secular society, whilst religion is viewed as irrelevant to public and civil discourse, spirituality as a reflexive and privatised lifestyle choice is therefore celebrated (Gray, 2008). Accordingly, spirituality within social work is observed to be ‘a movement of the human spirit that refuses to be limited by, or to be defined by, economic restrictions, consumerist pressures, political constraints or ideologies’ (Moss, 2012, p. 598). Consequently, increasing recognition of religion within social work is primarily framed within liberal paradigms in which spirituality is privileged within culturally competent and identity affirming contexts as illustrated by Moss and Thompson (2007) and Nelson-Becker et al. (2015). Nevertheless, despite efforts towards the integration of religion and spirituality by social work authors such as Moss (2005), Holloway and Moss (2010) and Furness and Gilligan (2010), this paper suggests that such preoccupation with privatised spirituality has eroded the significance of public religion to social work resulting in the wider context of secular hegemony within the profession having remained minimally critiqued. Moreover, further to the above account of liberalism as having marginalised religious citizenship from addressing collective social welfare concerns, this paper additionally observes that concurrent social work critiques of neo-liberalism have also omitted to acknowledge the contribution of faith perspectives to exposing neo-liberal hegemony as an outcome of secular laissez-faire modernity. In response, post-secular social work additionally engages with critiques of neo-liberalism through the deployment of post-secular perspectives which affirm religious citizenship as congruent with social work vocation. In particular, solidarity between social workers and faith-based organisations in the interests of social welfare is recognised as especially timely given the gathering evidence within social work of the detrimental effect that neo-liberalism continues to have upon public services. Such neo-liberalism being defined as (i) uncritical focus upon the efficacy of the unfettered free market as a mechanism for solving social problems by reforming citizens into customers and consumers with a concurrent attempt to eliminate any criticism regarding the social consequences (Goode, 2006), (ii) governance which endorses market fundamentalism, competitiveness, self-interest, performance-related quantifiable targets and economic determinism as typified by New Public Management principles (Steger and Ravi, 2010; Davies, 2014) and (iii) expedient roll-back processes involving the retraction and privatisation of state interests concurrent to roll out marketisation strategies with an increasing emphasis upon a surveillance rather than regulatory role for the state (Peck et al., 2009; Peck, 2010; Gane, 2012). Indeed, the profound alienation experienced by the profession as an outcome of the imposition of technocratic managerialism and authoritarian neo-liberal agendas into practice contexts is observed by a range of social work authors including Carey (2008, 2015), Ferguson (2007, 2008), Houston (2008), Jordan (2010), Ferguson and Lavalette (2013), Gray and Webb (2013), Reisch (2013), Mearns (2014), Harris (2014), McKendrick and Webb (2014) and Singh and Cowden (2015). For example, McDonald et al. (2008) recognise the demoralising effect that neo-liberal managerialism has wrought upon the profession, whilst Ferguson (2007, 2008) notes the erosion of solidarity within social work following the imposition of neo-liberal agendas which have induced atomisation within public services. Correlatively, Houston (2008) critiques the encroachment of neo-liberalism into social work as precipitating an insidious, reductive and distorted view of citizens as operating solely within homo economicus parameters wherein service users are considered as mere rational maximisers of their own self-interest. An associated culture of contempt which maligns a fragmented and disempowered working class is further observed as contributing towards social injustice by Jordan (2010), whilst Garret (2013a, 2013b) acknowledges that the neglect of socialist critique within social work education has undermined strategies of resistance to neo-liberalism. Indeed, social work education is considered to have acquiesced in the face of a ‘liberal onslaught towards narrow technicist competency-based models focusing upon efficiency and outcomes’ (Pease, 2013, p. 32) which, in addition to anti-intellectualism, has desensitised the profession to the effects of neo-liberalism, including the impact of poverty upon vulnerable citizens. Moreover, a climate of fear is identified with the imposition of neo-liberal programmes and austerity measures into social work contexts whilst an increasing intensification of neo-liberalism is observed to have undermined social democracy within the profession (Ferguson and Lavalette, 2013; Singh and Cowden, 2015). Resultantly, as Gray and Webb acknowledge, ‘the neoliberal rationality of competition, consumption and the marketisation of everything, including personal relationships and State-owned enterprises is the very opposite of social work’s core values’ (2013, p. 210). Consequently, a return to the political is advocated by Gray and Webb (2009, 2013) despite the recognition that: … we are persuaded, often by ourselves that radical politics is futile. So we tend towards compromise, resignation and indifference. Mainstream liberal social work discourse has a tendency to limit or even dislodge our experience of what is important or urgent. It tries to persuade that social work is politically neutral. Thus it can take over our voice and regulate our feelings into ones of apathy or disinterest (McKendrick and Webb, 2014, p. 359). Nevertheless, professional resilience is also affirmed by Gray and Webb (2013) through the recognition that social work continues to think politically in times of challenge and accordingly post-secular social work considers that liberal secular paradigms within the profession have obscured the radical potential of religious perspectives to subvert neo-liberalism. As such, in comparison to liberal secular discourse, this paper suggests that the emergence of post-secular theory represents an opportunity for social work to reconfigure religious citizenship as a marginalised and counter-cultural Other which potentially signifies a critical and emancipatory political ontology which is conducive to restoring, reviving and revitalising public commitments towards the common good. To this effect: … postsecularism accords to religion an enduring value, a place at the table in politics, a voice in the public sphere and an abiding role in private life. Postsecularism recognises the persistence of religion and marks an acknowledgement of religious and secular pluralisms. It recognises the ethical resources and community-building efficacy that religious systems and practices can offer and acknowledges the function of religion in constructing and defending cultural identities (Rectenwald and Almeida, 2015, p. 8). In particular, the post-secular turn in the work of Habermas (2006a, 2006b, 2010a, 2010b, 2013) illustrates a re-engagement with public religion within deliberative and participatory democratic practices which are congruent with social work vocation. Indeed, Habermas (1994, 1996) deploys the term ‘deliberative democracy’ to denote the virtues of reasoned participation in democratic decision making. As such, the public sphere is privileged, preserved and sustained as the ideal space for communicative discourse ethics which is structurally different from the competitive nature of the free market and more responsive to social welfare concerns than the instrumental and disenchanted state. Correlatively, religion as a cultural hermeneutic wherein society is reinterpreted through the deployment of post-secular perspectives enables faith communities to engage with the public sphere in ways that are distinct from both the state and market (Ward and Hoelzl, 2008). Resultantly, whilst secularism and associated empiricist methodologies render religion invisible within the public sphere, post-secular thinkers such as Habermas affirm the new presence and fresh relevance of religion to politics. Consequently, post-secular theory repoliticises religion by engaging with the public sphere, thereby developing post-liberal and post-neo-liberal perspectives which affirm public religion as significant to addressing social work concerns. Habermas and the new visibility of post-secular social work Indeed, to date, the deployment of the ideas of Habermas has proved fruitful to social work concerned not only with a critique of neo-liberalism, but also the relevance of communicative action and discourse ethics to practice contexts as illustrated by Gray and Webb (2009, 2013) and Houston (2009, 2013). In particular, Habermasian thought engages social work with critical theory in the defence of democratic and virtuous discourse ethics within a preserved and sustained public sphere. For example, Habermas compares the lifeworld which incorporates a reservoir of collective meanings that integrate our personal and community identities with the system which is organised by the state and dominated by finance, rationality, bureaucracy and instrumentality (Finlayson, 2005). As such, Habermas identifies the authoritarian and colonising potential of the state system to invade the lifeworld thereby displacing modes of integration based upon communicative reason. In advocating communicative action, Habermas therefore affirms discourse ethics in public life as incorporating an inherent mutuality of reason which is observed by social work authors as reaffirming the sine qua non of the profession through the preservation of dignity and integrity between citizens (Houston, 2007a, 2007b, 2009, 2013; Garrett, 2008, 2013b; William and Wilson, 2009). Furthermore, a dialogical paradigm is developed by William and Wilson (2009) based upon shared reason which upholds Habermasian claims that moral goodness requires forms of communication beyond instrumental concerns. Although early Habermasian thought is indebted to Marxism, his revival of virtue ethics is therefore recognised as directly applicable to social work theory and practice (Gray and Lovat, 2007, 2008a). In particular, social work theorists have drawn upon Habermasian ideas by focusing upon the efficacy of communicative action as an exemplar of virtue ethics whereby Habermas is read as a neo-Aristotelian. As such, Habermasian proportionate ethics are proposed in response to denotological social work paradigms, litigious proof practice contexts and the intractable challenge of difference within pluralistic society (Gray and Lovat, 2006, 2008b). Moreover, virtuous communicative action is considered to have occurred when consensus and social integration are achieved through discursive deliberation which subsequently replenishes the lifeworld and also resists the encroachment of the system into every aspect of social life (Garrett, 2008). Resultantly, the public sphere as the ideal space for reasoned communicative exchanges, deliberative democracy and reciprocal discourse ethics is sustained by the value and significance placed upon open dialogue and transparent collective understanding. Consequently, the public sphere is conceived as the discursive counterpart of civil society which in turn informs the political culture of the public sector as a forum which is structurally different from and politically resistant to the competitive nature of the free market (Maffettone, 2010), this virtuous preservation of the public sphere from dominance by the state system and market fundamentalism therefore being entirely congruent with social work ethics. Significantly, however, it is the recent post-secular turn in the work of Habermas which presents social work with an opportunity to further cultivate virtuous discourse ethics within the public sphere with particular regard to religious citizenship. Previously, Habermas abstained from judging the truth of religion in comparison to other critical theorists, such as Horkhiemer and Bloch, who assigned a more substantive significance to religion in the interests of social welfare (Duvenage, 2010). Given his earlier intellectual debt to Marx, Hegel, Weber and Durkhiem, Habermas thereby rehearsed familiar yet reductive structural functionalist, historical materialist and post-metaphysical critiques of religion which assumed a decline in religious belief in the rational progress of society towards public reason. As such, although Habermas observed the sacred as an ancestral source of motivation, performative action and collective consciousness in the foundation of society, the subsequent rationalisation of religion was considered as an inevitable outcome of the modern secular state which privileges procedural rather than substantive ethics in the maintenance of civil society (Habermas, 2002). In early Habermasian thought, religion therefore only remains relevant to the extent that it can be translated or assimilated into discursive language through methodological agnosticism (Duvenage, 2010). Resultantly, early Habermasian discourse on religion remained cautious of theology in defending an inherently secular public forum within which religious diversity was tolerated yet subordinated to public reason (Adams, 2006). However, subsequent to an engagement with previously resisted postmodern thought, Habermas has more recently asserted a post-secular perspective whereby a pragmatic dialogical relationship between public reason and religious faith is proposed (Borradori, 2003; Sheedy, 2016). In response to the coercive exclusion of religion from the public sphere in assertively secular societies, Habermas therefore advocates that religious views ought to be admitted into the deliberation process in order that the public sphere is rejuvenated by the inclusion of the complete experiences of humanity (Lanczi, 2010). Consequently, Habermas’s self-proclaimed goal has involved the ‘self-reflexive overcoming of a rigid and exclusive secularist self-understanding of modernity’ (2010b, p. 145). Accordingly, by affirming deliberative and participatory democracy, Habermas emphasises the redemptive inclusion of religious meaning into generally accessible language in considering that religious perspectives, as part of the public sphere, should be heeded when negotiating controversial normative questions given that faith is an important source of motivation and justification for citizens to participate in democratic practices (Reder and Schmidt, 2010). Moreover, in response to religious pluralism, Habermas (2006b) asserts that it is in the interests of the state to treat with care all sources of conscience upon which citizens draw and observes that the uncontrolled secularisation of society risks jeopardising the project of deliberative and participatory democracy (Reder and Schmidt, 2010). Habermas (2006b) therefore proposes that politically liberal cultures should endeavour to engage secular citizens in efforts to translate relevant contributions from faith traditions into publicly intelligible language. In corresponding political debates, Habermas (2010a) additionally considers religious speech as a potential conveyor of truth which nevertheless requires translation into language that is reasonably accessible. As such, Habermas advocates for dialogue between faith and reason in upholding moral and ethical notions of the Good which directly relate to public concerns regarding social welfare. For example, in dialoguing with the papacy, Habermas reaffirms Christianity as the nourishing source of conscience and human dignity within Western society (Habermas and Ratzinger, 2006). Consequently, Habermas cultivates a socially inclusive perspective which signifies an opportunity for social work to become more accommodating towards public religion. The post-secular turn in the work of Habermas as outlined by Harrington (2007) therefore incorporates a pragmatic counter-critique of a previously confident yet coercive secularist outlook which operates as an important source of authority for contemporary critiques of secularism within social work. Indeed, Habermas (2010a) acknowledges that the recognition of post-secular society by secular citizens, which is epistemologically attuned to the co-existence of religious communities, involves a change in consciousness that is no less cognitively exacting than the adaptation of religious citizens to assertively secular cultures. As such, post-secular social work subverts any presumption of progress from religious belief to secular reason and is especially indebted to the latter work of Habermas which retrieves faith from the private realm of personal belief by restoring religion within the public sphere, thereby repoliticising religious citizenship within deliberative processes. In particular, by demonstrating a sympathy between public religion and social welfare, Habermas revives commitment towards public service by resisting the colonising influence of neo-liberalism into public life, thereby negating any secular notions of public religion as problematic, dogmatic or divisive to social cohesion. Resultantly, given the new visibility of religion within social and political thought, the paradoxical illiberal legacy of liberalism in presuming a default secularism within the public sphere is resisted. For example, Sweeney (2008) acknowledges secular perspectives as intellectually under-resourced to fully comprehend the implications of religion in either domestic or international contexts, whilst developments within post-secular sociology by McLennan (2007, 2010) critique the philosophical foundations of secularism which predisposes a hermeneutic of suspicion, incredulity and curiosity in sociological approaches to religious subjects. In comparison to the limitations of secularised sociology, post-secular sociology is therefore advocated as a restorative project of social and political renewal (Keenan, 2002, 2003). The challenge to social work is therefore not for religious citizens to prove their rational and worthy Otherness to an apparently neutral yet altogether secular public sphere, but rather to deploy reflexivity, rather than description, in examining the dynamics of secularism in marginalising public religion from addressing social welfare concerns. Consequently, given the increasing prominence of post-secular theory within contemporary social and political thought, a more collaborative approach between social, political and religious perspectives is required within social work if epistemological resources are to be shared in ways which cultivate respect for human dignity. Post-secular social work, faith-based organisations and post-liberal ethics of care To this effect, post-secular social work acknowledges religious institutions and faith-based organisations as intermediate associations which, by fostering communitarian ethics within civil society, typically differentiate from a neo-liberal market convergence with the bureaucratic state. Furthermore, by critiquing liberal secularism, post-secular social work invariably distinguishes an ethics of care from the abstract, contractual and universal principles of liberal social theory. Indeed, care ethicists such as Barnes et al. (2015) recognise correlations between neo-communitarianism and faith-based approaches to social welfare. In comparison to liberalism and neo-liberalism, which typically negate the realities of social suffering by determining subjectivity as independent, autonomous and free as critiqued by Frost and Hoggett (2008), post-secular social work therefore recognises alternative post-liberal responses to social welfare concerns. Indeed, ethical perspectives concerned with inter-dependence, embodiment, relationality, proximity, mutuality and trust typically characterise post-liberal ethics and are therefore especially relevant to contemporary social work theory and practice. Accordingly, by outlining developments within critical and radical social work, Pease (2013) acknowledges post-liberal subjectivity as integral to recognising, defining and addressing the personal, qualitative and social particularities of poverty. As such, post-liberal ethics of care represent a substantive premise for further political, religious and social coalescence given that post-liberal relationality emphasises inter-subjective dialogue and reconciliation. For example, the post-liberal apologetics of Milbank and Pabst (2016) restore human dignity through an optimistic post-secular anthropology and virtuous social economy that countervail transactional, economistic and utilitarian cultures which alienate, commodify and impoverish people. Post-liberal humanism therefore orientates subjectivity towards social and economic relationality whilst post-liberal critiques as illustrated by Goodhart (2014, 2017) appreciate the entangled particularities of negotiated common life and purpose. Similar to the post in post-secularism, the post in post-liberalism rather than contentiously anti-liberal therefore exposes the inherent limitations within liberalism to establish a common good which is especially emphasised when an ethics of care is acknowledged as essential to social welfare. Consequently, by subverting former liberal and secular hegemonies within the profession, post-secular social work is sympathetic with emergent neo-communitarian and post-liberal ethics which challenge the presumptions of liberalism and duress of neo-liberalism. Within the context of democratic, collaborative and hybrid governance, post-secular social work therefore recognises faith-based social action as especially characterised by neo-communitarian and post-liberal ethics of care which is indicative of an alternative and supplementary social welfare paradigm to the bureaucratic cultures and consumerist models purported by both the private and public sectors (Shaw, 2016). Correlatively, faith-based organisations, by providing non-profit and holistic services within local communities, are considered as credible alternatives to the profiteering and cherry-picking cultures of privatised monopolies (Noyes and Blond, 2013). As such, faith-based social action is observed to be resistant to neo-liberal directives whilst paradoxically the expansion of markets within a mixed economy of social welfare provision is acknowledged as a dynamic threshold of potential reconciliation between previously disparate religious and secular groups (Cloke et al., 2012). Indeed, congruency between social work and religious citizenship is demonstrated in practice by a range of ecumenical initiatives, including (i) Feeding Britain which as an all-party parliamentary inquiry and coordinating strategy aims to alleviate food poverty across the UK (Shaw, 2016), (ii) missionary approaches to homeless provision whereby receptive generosity, fellowship and agape is a common ethos (Bowpitt et al., 2013), (iii) interdenominational street pastor schemes that appeal to secular, civic and faith values alike (Middleton and Yardwood, 2013), (iv) localised credit unions which deliver sustainable alternatives to high-cost pay-day loan monopolies (Timms, 2014), (v) modern slavery and human-trafficking projects which provide safe houses and specialist assistance (Salvation Army, 2016) and (vi) drug rehabilitation programmes that demonstrate caritas and post-secular practices (Williams, 2015). By pursuing alternative philosophies of care in meeting the needs of those both eligible and ineligible for state support, the performance requirements and technologies indicative of neo-liberalism, as well as the cultures of individualism associated with secular humanism, are therefore modified by faith-based counter-narratives which resist pernicious notions of the undeserving poor that characterise governmental welfare retrenchment (Cloke et al., 2012). Resultantly, faith-based social action, by privileging ethical perspectives, subverts the utilitarian underpinnings of mainstream liberal social policy as well as the prerequisites of neo-liberal welfare reform (Jawed, 2012). Consequently, with regard to post-secular reflexivity and social work education, faith-based organisations are acknowledged to be ‘critical non-State players in the social welfare arena’ (Pandya, 2016, p. 13). Conclusion By introducing post-secular social work, this article therefore responds to the speculative reflection ‘after neoliberalism, new managerialism and postmodernism, what next for social work?’ (Noble and Henrickson, 2011, p. 128). In particular, post-secular social work represents a return to the political in the development of a transformative post-liberal and post-neoliberal new politics of social work. Within the context of post-secular society, post-secular social work therefore invites the profession to re-engage with public religion within deliberative and participatory democratic practices in the interests of social welfare. Accordingly, post-secular social work recognises correlations between social work, faith-based social action and post-liberal ethics of care as indicative of an alternative and supplementary social welfare paradigm to the bureaucratic and consumerist models purported by both the public and private sectors. Consequently, by subverting former secular, liberal and neo-liberal hegemonies within the profession, post-secular social work represents a hopeful vision for the future of social work whereby solidarity between social workers and faith-based organisations is enhanced. References Adams N. ( 2006 ) Habermas and Theology , Cambridge , Cambridge University Press . Google Scholar Crossref Search ADS Google Preview WorldCat COPAC Barnes M. , Brannelly T. , Ward L. , Ward N. ( 2015 ) Ethics of Care: Critical Advances in International Perspective , Bristol, Policy Press . Google Scholar Crossref Search ADS Google Preview WorldCat COPAC Borradori G. ( 2003 ) Philosophy in Time of Terror: Dialogues with Jurgen Habermas and Jacques Derrida , Chicago, University of Chicago Press . Google Preview WorldCat COPAC Bowpitt G. ( 1998 ) ‘ Evangelical Christianity, secular humanism and the genesis of British social work’, British Journal of Social Work , 28 , pp. 675 – 93 . Google Scholar Crossref Search ADS WorldCat Bowpitt G. ( 2000 ) ‘ Working with creative creature: Towards a Christian paradigm for social workers with some practical considerations’, British Journal of Social Work , 30 , pp. 349 – 64 . Google Scholar Crossref Search ADS WorldCat Bowpitt G. , Dwyer P. , Sundin E. , Weinstein M. ( 2013 ) ‘ Places of sanctuary for the undeserving? Homeless people’s day centres and the problem of conditionality ’, British Journal of Social Work , 44 ( 5 ), pp. 1251 – 67 . Google Scholar Crossref Search ADS WorldCat Braidotti R. , Blaagard B. , Graauw T. , Midden E. ( 2014 ) Transformations of Religion and the Public Sphere: Postsecular Publics , London, Palgrave MacMillan . Google Preview WorldCat COPAC Carey M. ( 2008 ) ‘ Everything must go? The privatisation of state social work’, British Journal of Social Work , 38 , pp. 918 – 35 . Google Scholar Crossref Search ADS WorldCat Carey M. ( 2015 ) ‘ The fragmentation of social work: Some ramifications and a critique’, British Journal of Social Work , 45 , pp. 2406 – 22 . Google Scholar Crossref Search ADS WorldCat Cloke P. , Thomas S. , Williams A. ( 2012 ) ‘ Co-constituting neoliberalism: Faith based organisations, co-option and resistance in the UK’, Environment and Planning , 44 , pp. 1479 – 501 . Google Scholar Crossref Search ADS WorldCat Connolly W. ( 1999 ) Why I Am Not a Secularist , Minnesota, University of Minnesota Press . Google Preview WorldCat COPAC Crisp B. ( 2008 ) ‘ Social work and spirituality in a secular society ’, British Journal of Social Work , 8 ( 4 ), pp. 363 – 74 . Google Scholar Crossref Search ADS WorldCat Crisp B. ( 2010 ) Spirituality and Social Work , London , Ashgate . Google Preview WorldCat COPAC Crisp B. ( 2014 ) Social Work and Faith Based Organisations , Oxon , Routledge . Google Preview WorldCat COPAC Davies W. ( 2014 ) ‘ Neoliberalism: A bibliographic review’, Theory, Culture Society , 31 ( 7 ), pp. 309 – 17 . Google Scholar Crossref Search ADS WorldCat Derezotes D. ( 2006 ) Spirituality Orientated Social Work Practice , Boston , Allyn & Bacon . Google Preview WorldCat COPAC Duvenage P. ( 2010 ) ‘ Communicative reason and religion: The case of Habermas’, Sophia , 49 , pp. 343 – 57 . Google Scholar Crossref Search ADS WorldCat Ferguson I. ( 2007 ) ‘ Increasing user choice or privatising risk? The antinomies of personalisation’, British Journal of Social Work , 37 , pp. 387 – 403 . Google Scholar Crossref Search ADS WorldCat Ferguson I. ( 2008 ) Reclaiming Social Work: Challenging Neo-Liberalism and Promoting Social Justice , London , Sage . Google Preview WorldCat COPAC Ferguson I. , Lavalette M. ( 2013 ) ‘ Crisis, austerity and the future(s) of social work in the UK’, Critical and Radical Social Work , 1 ( 1 ), pp. 95 – 110 . Google Scholar Crossref Search ADS WorldCat Finlayson G. ( 2005 ) Habermas: A Very Short Introduction , Oxford , Oxford University Press . Google Scholar Crossref Search ADS Google Preview WorldCat COPAC Fountain P. ( 2013 ) ‘ Towards a postsecular anthropology ’, Australian Journal of Anthropology , 24 ( 3 ), pp. 310 – 28 . Google Scholar Crossref Search ADS WorldCat Frost L. , Hoggett P. ( 2008 ) ‘ Human agency and social suffering’, Critical Social Policy , 28 ( 4 ), pp. 438 – 60 . Google Scholar Crossref Search ADS WorldCat Furani K. ( 2016 ) ‘ Is there a postsecular?’, Journal of the American Academy of Religion , 83 ( 1 ), pp. 1 – 26 . Google Scholar Crossref Search ADS WorldCat Furman L. , Benson P. , Grimwood C. , Canda E. ( 2004 ) ‘ Religion and spirituality in social work education and direct practice at the Millennium: A survey of UK social workers’, British Journal of Social Work , 34 , pp. 767 – 92 . Google Scholar Crossref Search ADS WorldCat Furness S. , Gilligan P. ( 2010 ) Religion, Belief and Social Work: Making a Difference , Bristol , Polity Press . Google Preview WorldCat COPAC Furness S. , Gilligan P. ( 2012 ) ‘ Faith based organisations and UK welfare services: Exploring some ongoing dilemmas’, Social Policy & Society , 11 ( 4 ), pp. 601 – 12 . Google Scholar Crossref Search ADS WorldCat Furness S. , Gilligan P. ( 2014 ) ‘ It never came up: Encouragement and discouragement to addressing religion and belief in professional practice—What do social work students have to say?’, British Journal of Social Work , 44 ( 3 ), pp. 763 – 81 . Google Scholar Crossref Search ADS WorldCat Gane N. ( 2012 ) ‘ The governmentalities of neoliberalism: Panopticism, post-panopticism and beyond’, The Sociological Review , 60 , pp. 611 – 34 . Google Scholar Crossref Search ADS WorldCat Garde J. ( 2015 ) ‘ Shrinking religious communities and thriving inter-religious social work in postsecular Sweden ’, Journal of Religion & Spirituality in Social Work , 34 ( 1 ), pp. 1 – 23 . Google Scholar Crossref Search ADS WorldCat Garrett P. ( 2008 ) ‘ Questioning Habermasian social work: A note on some alternative theoretical resources’, British Journal of Social Work , 39 ( 5 ), pp. 867 – 83 . Google Scholar Crossref Search ADS WorldCat Garrett P. ( 2013a ) ‘Mapping the theoretical and political terrain of social work’, in Gray M. , Webb S. (eds), The New Politics of Social Work , London, Palgrave MacMillan , pp. 44 – 63 . Google Scholar Crossref Search ADS Google Preview WorldCat COPAC Garrett P. ( 2013b ) Social Work and Social Theory: Making Connections , Bristol, Policy Press, University of Bristol . Google Scholar Crossref Search ADS Google Preview WorldCat COPAC Gilligan P. , Furness S. ( 2006 ) ‘ The role of spirituality in social work practice: Views and experiences of social workers and students ’, British Journal of Social Work , 36 , pp. 617 – 37 . Google Scholar Crossref Search ADS WorldCat Goode J. ( 2006 ) ‘ Faith based organisations in Philadelphia: Neoliberal ideology and the decline of political activism ’, Urban Anthropology , 35 ( 2–3 ), pp. 203 – 36 . WorldCat Goodhart D. ( 2014 ) A Postliberal Future ?, London, Demos Quarterly . Google Preview WorldCat COPAC Goodhart D. ( 2017 ) The Road to Somewhere: The Populist Revolt and the Future of Politics , London , Hurst & Company . Google Preview WorldCat COPAC Graham E. ( 2013 ) Between a Rock and a Hard Place: Public Theology in a Post-Secular Age , Norwich , SCM Press . Google Preview WorldCat COPAC Gray M. ( 2005 ) ‘ Dilemmas of international social work: Paradoxical processes in indigenisation, universalism and imperialism’, International Journal of Social Welfare , 14 , pp. 231 – 8 . Google Scholar Crossref Search ADS WorldCat Gray M. ( 2008 ) ‘ Viewing spirituality in social work through the lens of contemporary social theory’, British Journal of Social Work , 38 , pp. 175 – 96 . Google Scholar Crossref Search ADS WorldCat Gray M. , Lovat T. ( 2006 ) ‘ The shaky high moral ground of postmodern ethics ’, Journal of Social Work , 42 ( 3–4 ), pp. 201 – 12 . WorldCat Gray M. , Lovat T. ( 2007 ) ‘ Horse and carriage: Why Habermas’s discourse ethics gives virtue a praxis in social work’, Ethics and Social Welfare , 1 ( 3 ), pp. 310 – 28 . Google Scholar Crossref Search ADS WorldCat Gray M. , Lovat T. ( 2008a ) ‘ Practical mysticism, Habermas and social work praxis ’, Journal of Social Work , 8 ( 92 ), pp. 149 – 62 . Google Scholar Crossref Search ADS WorldCat Gray M. , Lovat T. ( 2008b ) ‘ Towards a proportionist social work ethics: A Habermasian perspective ’, British Journal of Social Work , 38 , pp. 1100 – 14 . Google Scholar Crossref Search ADS WorldCat Gray M. , Webb S. ( 2009 ) Social Work Theories and Methods , London , Sage . Google Preview WorldCat COPAC Gray M. , Webb S. ( 2013 ) The New Politics of Social Work , London, Palgrave Macmillan . Google Scholar Crossref Search ADS Google Preview WorldCat COPAC Habermas J. ( 1994 ) ‘ Three normative models of democracy’, Constellations , 1 ( 1 ), pp. 1 – 10 . Google Scholar Crossref Search ADS WorldCat Habermas J. ( 1996 ) Between Facts and Norms , Cambridge , MTI Press . Google Preview WorldCat COPAC Habermas J. ( 2002 ) Religion and Rationality: Essays on Reason, God and Modernity , Cambridge , Polity . Google Preview WorldCat COPAC Habermas J. ( 2006a ) ‘On the relations between the secular liberal state and religion’ in de Vries H. , Sullivan L. (eds), Political Theologies: Public Religions in a Post Secular World , New York , Fordham University Press . Google Preview WorldCat COPAC Habermas J. ( 2006b ) ‘ Religion in the public sphere ’, European Journal of Philosophy , 14 , pp. 1 – 25 . Google Scholar Crossref Search ADS WorldCat Habermas J. ( 2010a ) An Awareness of What Is Missing: Faith and Reason in a Postsecular Age , Cambridge , Polity Press . Google Preview WorldCat COPAC Habermas J. ( 2010b ) Between Naturalism and Religion Polity , Cambridge , Polity Press . Google Preview WorldCat COPAC Habermas J. ( 2013 ) ‘A reply to my critics’, in Calhoun C. , Mendieta E. , VanAntwerpen J. (eds), Habermas and Religion , Cambridge , Polity Press . Google Preview WorldCat COPAC Habermas J. , Ratzinger J. ( 2006 ) The Dialectics of Secularisation: On Reason and Religion , San Francisco , Ignatius Press . Google Preview WorldCat COPAC Harrington A. ( 2007 ) ‘ Habermas and the postsecular society’, European Journal of Social Theory , 10 ( 4 ), pp. 543 – 60 . Google Scholar Crossref Search ADS WorldCat Harris J. ( 2014 ) ‘ Against neoliberal social work’, Critical and Radical Social Work , 2 ( 1 ), pp. 7 – 22 . Google Scholar Crossref Search ADS WorldCat Henery N. ( 2003 ) ‘ The reality of visions: Contemporary theories of spirituality in social work’, British Journal of Social Work , 33 , pp. 1105 – 13 . Google Scholar Crossref Search ADS WorldCat Holloway M. ( 2007 ) ‘ Spiritual need and the core business of social work’, British Journal of Social Work , 37 , pp. 265 – 80 . Google Scholar Crossref Search ADS WorldCat Holloway M. , Moss B. ( 2010 ) Spirituality and Social Work , London, Palgrave Macmillian . Google Scholar Crossref Search ADS Google Preview WorldCat COPAC Houston S. ( 2007a ) ‘Jurgen Habermas’, in Gray M. , Webb S. (eds), Social Work Theories and Methods , London , Sage . Google Preview WorldCat COPAC Houston S. ( 2007b ) ‘Re-affirming the face to face: Habermas’ critical theory and social work’, in Gray M. , Webb S. (eds), Critical Perspectives in Social Work , London, Sage Publications Ltd . Google Preview WorldCat COPAC Houston S. ( 2008 ) ‘ Beyond homo economicus: Recognition, self-realisation and social work ’, British Journal of Social Work , 40 ( 3 ) pp. 841 – 57 . Google Scholar Crossref Search ADS WorldCat Houston S. ( 2009 ) ‘ Communication, recognition and social work: Aligning the ethical theories of Habermas and Honneth’, British Journal of Social Work , 39 ( 7 ), pp. 1274 – 90 . Google Scholar Crossref Search ADS WorldCat Houston S. ( 2013 ) ‘Social work and the politics of recognition’, in Gray M. , Webb S. (eds), The New Politics of Social Work , London, Palgrave MacMillan, pp. 63 – 77 Google Scholar Crossref Search ADS Google Preview WorldCat COPAC Jawed R. ( 2012 ) ‘ Thinking about religious welfare and rethinking social policy in the British context’, Social Policy and Society , 11 ( 4 ), pp. 613 – 24 . Google Scholar Crossref Search ADS WorldCat Jordan B. ( 2010 ) Why the Third Way Failed: Economics, Morality and the Origins of the Big Society , Bristol , Policy Press . Google Preview WorldCat COPAC Keenan W. ( 2002 ) ‘ Postsecular sociology: Effusions of religion in late modern settings’, European Journal of Social Theory , 5 ( 2 ), pp. 279 – 90 . Google Scholar Crossref Search ADS WorldCat Keenan W. ( 2003 ) ‘ Rediscovering the theological in sociology: Foundation and possibilities’, Theory, Culture & Society , 20 ( 1 ), pp. 19 – 42 . Google Scholar Crossref Search ADS WorldCat Lanczi A. ( 2010 ) ‘Democracy and moral relativism in a post-secular world: Reclaiming obligation’, in Losonczi P. , Singh A. (eds), From Political Theory to Political Theology: Religious Challenges and the Prospects of Democracy , London , Continuum . Google Preview WorldCat COPAC Maffettone S. ( 2010 ) ‘Religion and liberalism: Public reason, public sphere and cultural pluralism’, in Losonczi P. , Singh A. (eds), From Political Theory to Political Theology: Religious Challenges and the Prospects of Democracy , London , Continuum . Google Preview WorldCat COPAC Maslow A. ( 1971 ) The Farther Reaches of Human Nature , New York , Viking Press . Google Preview WorldCat COPAC Mathews I. ( 2009 ) Social Work and Spirituality , Exeter , Learning Matters Ltd . Google Preview WorldCat COPAC McDonald A. , Postle K. , Dawson C. ( 2008 ) ‘ Barriers to retaining and using professional knowledge in local authority social work practice with adults in the UK ’, British Journal of Social Work , 38 ( 7 ), pp. 1370 – 87 . Google Scholar Crossref Search ADS WorldCat McKendrick D. , Webb S. ( 2014 ) ‘ Taking a political stance in social work’, Critical and Radical Social Work , 2 ( 3 ), pp. 357 – 69 . Google Scholar Crossref Search ADS WorldCat McLennan G. ( 2007 ) ‘ Towards postsecular sociology ’, Sociology , 41 ( 5 ), pp. 857 – 70 . Google Scholar Crossref Search ADS WorldCat McLennan G. ( 2010 ) ‘ The postsecular turn’, Theory, Culture & Society , 27 ( 4 ), pp. 3 – 20 . Google Scholar Crossref Search ADS WorldCat Mearns G. ( 2014 ) ‘ The neoliberal agenda and social work values: Using critical discourse analysis to explore the role of language in processes of cultural hegemony’, Critical and Radical Social Work , 2 ( 2 ), pp. 217 – 33 . Google Scholar Crossref Search ADS WorldCat Middleton J. , Yardwood R. ( 2013 ) ‘ Christians out here? Encountering street pastors in postsecular spaces of the UK’s night time economy ’, Urban Studies , 33 ( 3 ), pp. 1 – 16 . WorldCat Milbank J. ( 2006 ) Theology and Social Theory , Oxford , Blackwell Publishing . Google Scholar Crossref Search ADS Google Preview WorldCat COPAC Milbank J. , Pabst A. ( 2016 ) The Politics of Virtue: Post-Liberalism and the Human Future , London, Rowman & Littlefield International . Google Preview WorldCat COPAC Moss B. ( 2005 ) Religion and Spirituality , Lyme Regis, Russell House Publishing . Google Preview WorldCat COPAC Moss B. ( 2012 ) ‘ The pedagogic challenge of spirituality: A co-creative response’, Journal of Social Work , 12 ( 6 ), pp. 595 – 613 . Google Scholar Crossref Search ADS WorldCat Moss B. , Thompson N. ( 2007 ) ‘ Spirituality and equality ’, Social & Public Policy Review , 1 ( 1 ), pp. 1 – 12 . WorldCat Neagoe A. ( 2013 ) ‘ Ethical dilemmas of the social work professional in a (post-)secular society with special reference to the Christian social worker ’, International Social Work , 56 ( 3 ), pp. 310 – 25 . Google Scholar Crossref Search ADS WorldCat Nelson-Becker H. , Ai A. , Hopp F. , McCormick T. , Schlueter J. , Camp J. ( 2015 ) ‘ Spirituality and religion in end-of-life care ethics: The challenge of interfaith and cross-generational matters’, British Journal of Social Work , 45 , pp. 104 – 19 . Google Scholar Crossref Search ADS WorldCat Noble C. , Henrickson M. ( 2011 ) ‘ Editorial: After neoliberalism, new managerialism and postmodernism, what next for social work?’, Journal of Social Work , 11 ( 2 ), pp. 128 – 31 . Google Scholar Crossref Search ADS WorldCat Noyes J. , Blond P. ( 2013 ) Holistic Mission: Social Action and the Church of England , London , Res Publica . Google Preview WorldCat COPAC Nynas P. , Lassander M. , Utriainen T. ( 2012 ) Post-Secular Society , Piscataway, NJ , Transaction Publishers . Google Preview WorldCat COPAC Pandya S. ( 2016 ) ‘ Teaching about faith-based organisations in the social work curriculum: Perspectives of social work educators’, Journal of Teaching in Social Work , 36 ( 2 ), pp. 123 – 39 . Google Scholar Crossref Search ADS WorldCat Pease B. ( 2013 ) ‘A history of critical and radical social work’, in Gray M. , Webb S. (eds), The New Politics of Social Work , London, Palgrave MacMillan , pp. 21 – 44 Google Scholar Crossref Search ADS Google Preview WorldCat COPAC Peck J. ( 2010 ) Constructions of Neoliberal Reason , Oxford , Oxford University Press . Google Scholar Crossref Search ADS Google Preview WorldCat COPAC Peck J. , Theodore N. , Brenner N. ( 2009 ) ‘ Postneoliberalism and its malcontents ’, Antipode , 1 , pp. 94 – 116 . WorldCat Rectenwald M. , Almeida R. ( 2015 ) Global Secularisms in a Postsecular Age , Berlin/Boston , De Gruyter . Google Scholar Crossref Search ADS Google Preview WorldCat COPAC Reder M. , Schmidt S. ( 2010 ) ‘Habermas and religion’, in Habermas, J. (ed.), An Awareness of What is Missing: Faith and Reason in a Postsecular Society , Cambridge , Cambridge University Press , pp. 3 – 7 . Google Preview WorldCat COPAC Reisch M. ( 2013 ) ‘ What is the future of social work?’, Critical and Radical Social Work , 1 ( 1 ), pp. 67 – 85 . Google Scholar Crossref Search ADS WorldCat Salvation Army ( 2016 ) ‘Human trafficking: Committed to supporting victims of modern slavery, available online at www.salvationarmy.org.uk (accessed 29 April 2017). Shaw J. ( 2016 ) ‘Socialism by grace: The new visibility of postsecular social work’, DSW thesis, UK, University of Sussex. Sheedy M. ( 2016 ) ‘ Habermas and religion’, Religion Compass , 10 ( 3 ), pp. 58 – 73 . Google Scholar Crossref Search ADS WorldCat Singh G. , Cowden S. ( 2015 ) ‘ The intensification of neoliberalism and the commodification of human need—a social work perspective’, Critical and Radical Social Work , 3 ( 3 ), pp. 375 – 87 . Google Scholar Crossref Search ADS WorldCat Starnino V. , Gomi S. , Canda E. ( 2014 ) ‘ Spiritual strengths assessment in mental health practice’, British Journal of Social Work , 44 , pp. 849 – 67 . Google Scholar Crossref Search ADS WorldCat Steger M. , Ravi R. ( 2010 ) Neoliberalism: A Very Short Introduction , Oxford , Oxford University Press . Google Scholar Crossref Search ADS Google Preview WorldCat COPAC Sweeney J. ( 2008 ) ‘Revising secularisation theory’, in Ward G. , Hoelzl M. (eds), The New Visibility of Religion: Studies in Religion and Cultural Hermeneutics , London , Continuum . Google Preview WorldCat COPAC Taylor C. ( 2007 ) A Secular Age , Cambridge, MA, Harvard University Press . Google Preview WorldCat COPAC Timms S. ( 2014 ) Restoring Faith in Social Security: The Future of Welfare: A Theos Collection , London , Theos . Google Preview WorldCat COPAC Vanderwoerd J. ( 2011 ) ‘ Reconsidering secularisation and recovering Christianity in social work history ’, Social Work & Christianity , 38 ( 3 ), pp. 244 – 66 . WorldCat Ward G. , Hoelzl M. ( 2008 ) The New Visibility of Religion: Studies in Religion and Cultural Hermeneutics , London, Continuum International Publishing Group Ltd . Google Preview WorldCat COPAC Whiting R. ( 2008 ) ‘No room for religion or spirituality or cooking tips: Exploring practical atheism as an unspoken consensus in the development of social work values in England’, Ethics and Social Welfare , 2 ( 1 ), pp. 67 – 83 . Google Scholar Crossref Search ADS WorldCat William D. , Wilson D. ( 2009 ) ‘ Sentimentality, communicative action and the social self: Adam Smith meets Jurgen Habermas’, History of Human Science , 22 ( 3 ), pp. 75–7 . WorldCat Williams A. ( 2015 ) ‘ Postsecular geographies: Theo-ethics, rapprochement and neoliberal governance in a faith-based drug programme’, Transactions of the Institute of British Geographers , 40 ( 2 ), pp. 192 – 208 . Google Scholar Crossref Search ADS WorldCat Williams R. ( 2008 ) ‘Secularisation, faith and freedom’, in Ward G. , Hoelzl M. (eds), The New Visibility of Religion: Studies in Religion and Cultural Hermeneutics , London , Continuum . Google Preview WorldCat COPAC Williams R. ( 2012 ) Faith in the Public Square , London , Bloomsbury . Google Preview WorldCat COPAC Wong Y. , Vinsky J. ( 2008 ) ‘ Speaking from the margins: A critical reflection on the spiritual-but-not-religious discourse in social work ’, British Journal of Social Work , 39 ( 7 ), pp. 1343 – 59 . Google Scholar Crossref Search ADS WorldCat © The Author 2017. Published by Oxford University Press on behalf of The British Association of Social Workers. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Tafvelin, Susanne;Isaksson, Kerstin;Westerberg, Kristina

doi: 10.1093/bjsw/bcx038pmid: N/A

Abstract In this longitudinal interview study, we have strived to advance the understanding of how organisational factors may hinder the emergence of transformational leadership among first line managers in social service organisations. By interviewing managers in a Swedish social service organisation during their first year of leadership, we first identified leadership ideals and then asked them to identify factors that hinder the performance of this leadership. Qualitative content analysis was used to analyse the data and the results revealed that the managers strived for a transformational leadership, but several factors in the organisation made it difficult to lead in the way they intended. Hindering factors were identified both at the organisational level, such as ‘top-down management’, ‘financial strain’ and ‘continuous change’, and in the managers’ own working environment in terms of ‘no support’, ‘high work-load’, ‘limited influence’, ‘administrative tasks’ and ‘distance to employees’. This study contributes to our understanding of organisational antecedents of transformational leadership as well as the premises of transformational leadership in social service organisations. Transformational leadership, organisational antecedents, social service organisations, qualitative study, content analysis Introduction A considerable amount of empirical research on transformational leadership has produced a consistent pattern of positive relationships between transformational leadership and the performance and effectiveness of organisations as well as the health and well-being of employees (Judge and Piccolo, 2004; Nielsen et al., 2008). However, although the prevalence and consequences of transformational leadership are well documented, few studies have investigated its antecedents (Dóci and Hofmans, 2015). In particular, little is known about the organisational conditions that may facilitate or hinder the emergence of transformational leadership behaviour (Wright and Pandey, 2009). Efforts to determine antecedents to transformational leadership are essential to advance the transformational leadership field and enhance our understanding how transformational leadership behaviours are manifested in organisations (Barbuto and Burbach, 2006). In this longitudinal interview study, we investigate organisational antecedents that may hinder the emergence of transformational leadership by studying newly recruited managers in social service organisation during their first year of leadership. Although meta-analyses have shown that transformational leadership is just as common in the public sector (Lowe et al., 1996), some studies indicate that the size and structure of these organisations may hinder leaders’ possibilities for exhibiting transformational behaviour (Lowe et al., 1996; Wright and Pandey, 2009). It is therefore of added importance to further our understanding of the antecedents of transformational leadership in public organisations such as the social services. Transformational leadership in social service organisations Bass and Riggio (2006) have characterised transformational leadership as including four theoretically distinct components: idealised influence, inspirational motivation, intellectual stimulation and individualised consideration. Idealised influence includes serving as a role model, while stressing the importance of a collective sense of mission. Inspirational motivation involves providing meaning and challenge to followers’ work, including inculcating visions of the future. Intellectual stimulation entails encouraging followers to question assumptions, reframe problems and approach old situations in new ways. Finally, leaders who provide individualised consideration pay attention to followers’ individual needs for achievement and growth. Transformational leadership is part of a model called the Full Range of Leadership model which also includes transactional leadership, where the leader promotes compliance in his or her followers through both rewards and punishments, and includes the two distinct components of contingent reward and management by exception. Finally, the model includes laissez-faire leadership which represents a passive non-leadership. Even though meta-analyses have found that transformational leadership is just as common in the public sector as in the private sector (e.g. Lowe et al., 1996), many scholars of public management are still reluctant about the usefulness of transformational leadership in social service organisations (Wright and Pandey, 2009). Instead, transformational leadership is expected to be less common and less effective in social service organisations. One argument is that public organisations such as the social services rely heavily on bureaucratic control mechanisms such as centralisation, formalisation and routinisation (Bass and Riggio, 2006; Pawar and Eastman, 1997), which provide institutional substitutes for leadership (Lowe et al., 1996). In other words, the structure of public organisations in terms of a set of very formal rules, guidelines and policies can either neutralise the positive effect of a transformational leader or even lower the need for leadership in general. Further, Alvesson (2001) argues that public leaders are too burdened with administrative tasks to exhibit transformational leadership behaviours. This is in line with Karlsson (2006), who found that the leadership role in social service organisations took an administrative turn after the implementation of new public management, as managers received an increased responsibility for budget and staff. On the other hand, when asking social service managers themselves about which kind of leadership they consider suitable in social service organisations, Rank and Hutchison (2000) found that managers described behaviours that are very much in line with transformational leadership, including having a vision, promoting the values of the profession, motivating and stimulating employees, and facilitating change. Empirical research gives mixed support for the benefits of transformational leadership in social service organisations. On one hand, meta-analysis has shown that women are more often perceived as being a transformational leader than men (Eagly et al., 2003) which speaks in favour of social services organisation that often have a larger proportion of female managers than the private sector. On the other hand, large size (Lowe et al., 1996) and hierarchical decision making and communication (Wright and Pandey, 2009) in the organisation have proven to have an adverse effect on transformational leadership ratings, which speaks against social service organisations that are associated with large, bureaucratic organisations. However, some recent studies indicate that the social sector organisations is not as bureaucratic as earlier assumed (Pandey and Wright, 2006), and that bureaucratic control mechanisms per se may not adversely affect transformational leadership (Wright and Pandey, 2009). In addition, an increasing number of empirical studies of transformational leadership in social service organisations suggest that transformational leadership is related to employees’ satisfaction with their job and their manager (Gellis, 2001; Mary, 2005), employee commitment and role clarity (Tafvelin et al., 2014) as well as service effectiveness (Yoo and Brooks, 2005). However, more research is needed to enhance our understanding of how the context of social service organisations influence the emergence of transformational leadership (Wright and Pandey, 2009). Antecedents of transformational leadership Previous studies on the antecedents of transformational leadership can be divided into three different lines of research: antecedents related to the leader, the follower or the situation. Most studies have focused on the influence of leaders’ personal characteristics in relation to transformational leadership and demonstrated that factors such as personality (Deinert et al., 2015), intelligence (Barbuto and Burbach, 2006) and mood (Chi et al., 2011) all play a role in the likelihood of a leader displaying transformational leadership behaviour. Other studies have examined the impact followers have on leaders’ abilities to perform transformational leadership behaviour, and showed that followers’ well-being (Nielsen et al., 2008) and personality (Bono et al., 2012) influence the leader’s display of transformational leadership behaviours. One limitation in both of these lines of study that focus on either the leader or the follower is that they capture individual traits or attitudes which represent factors that may be difficult to change (Nielsen and Cleal, 2011). The third line of research focuses on the situation or context in which transformational leadership takes place, which has been an under-studied area in leadership research (Avolio and Bass, 1995). The importance of context has been theorised for some time, suggesting that the structure of the organisation ought to play a role in the emergence of transformational leadership (Pawar and Eastman, 1997). One of the first empirical investigations found that hierarchical decision making and communication adversely affected transformational leadership ratings in the organisation (Wright and Pandey, 2009). In a related study, Walter and Bruch (2010) found that centralisation and the size of the organisations were negatively related to transformational leadership climate. Formalisation in terms of written rules, procedures and instructions was, on the other hand, positively associated with transformational leadership climate. Further, a study by Nielsen and Cleal (2011) shows that both situations and the leaders’ own work environment may facilitate the display of transformational leadership behaviours. Situations that were cognitively challenging and situations where the leaders felt they were in control were positively related to self-ratings of transformational leadership as well as working conditions characterised by high cognitive demands and feelings of having a meaningful job. These studies indicate that the organisational context in terms of structure, situations and work environment plays an important role in the emergence of transformational leadership but, given the small number of studies in the field, more research is needed. Our study expands existing knowledge by focusing on how leaders’ experience of the organisation and their working environment may hinder the emergence of transformational leadership behaviour. The present study The purpose of this longitudinal interview study is to explore organisational antecedents to leadership behaviour among first line managers in social service organisations. Using transformational leadership theory as a framework, we first investigate what kind of leadership the managers are striving to achieve in terms of leadership ideals in their first year of service. By following managers over time, we then aim to capture organisational and working conditions that may hinder managers from performing this leadership ideal in the realm of their everyday life. Method Procedure and participants We conducted repeated interviews with newly recruited first line managers in their first years as managers in a Swedish social service organisation. This specific social service organisation was situated in the north part of Sweden had at the time around 2,700 employees. The organisation provided services in social welfare, elderly care and care of the disabled. The tasks of first line managers included responsibilities for staffing, scheduling, budget and the meeting of organisational goals. With help from the organisation’s HR-department, all newly recruited managers during the study period (twelve months) were invited to participate in the study. All invited first line managers agreed to participate in the study. Interviews were scheduled during the first month, after six months and after one year, which means that three interviews were conducted with those managers who stayed for the whole year. The participants included eight first line managers, all women, with an average age of 40.6 years (sd = 12.1). They all held university degrees and had previous work experience from a diversity of areas including psychiatric care, school counselling, elderly care and administrative work in the private sector. Five of the interviewees had no previous experience of being a manager, while three of them had worked as managers in other organisations such as in hospitals or in the private sector. The participants worked in social welfare (n = 3), elderly care (n = 2) and care of the disabled (n = 3). Five managers stayed the whole year and were therefore interviewed on all three occasions. One manager moved to another city after the first interview and was therefore only interviewed once. Two managers were interviewed only twice, as one of them had chosen to leave the organisation because of bad working conditions and the other manager experienced work-related stress and was on sick leave by the time of the final interview. We followed the ethical guidelines of the Swedish research council which is aligned with the ethical principles in the Declaration of Helsinki. All managers were informed that their participation was voluntary and anonymous, and they could withdraw at any time. All managers gave verbal informed consent to participate. The interviews We used semi-structured interview guides consisting of open-ended questions to both frame the interview and permit probing for additional information (Miles and Huberman, 1994). Interviews were conducted in-person and lasted forty-five to seventy-five minutes, with an average duration of one hour, consistently with the methods suggested for in-depth interviews (McCracken, 1988). The first interview focused on their expectations for the first year of leadership and their leadership ideals (‘What kind of leadership are your striving for?’, ‘Do you have any leadership ideal?’). The second interview focused on leadership ideals and hindering factors in the organisation (‘What kind of leadership do you think is appropriate at this juncture?’, ‘What do you think that leadership in this organisation should be like?’, ‘Have you experienced any hindering factors in the organisation for performing this leadership?’). The third interview followed up on their leadership ideals from the first interview, explored whether new ideals had emerged during the first year and also hindering factors in the organisation (‘This was your leadership ideal a year ago, does it still apply?’, ‘Have any new ideals emerged during this year?’, ‘Are there any hinders in the organisation to lead in this way?’). All interviews, except two, were done by the first author and tape-recorded with the consent of the interviewees. All interviews were audio taped and transcribed by a third clerical party. The third-party transcription ensured a complete and unbiased recording of interview data (Waldman et al., 1998). Data analysis The large field of qualitative methods makes it important to choose a suitable approach to answer the research question. Qualitative content analysis was chosen as ‘an approach of empirical, methodological controlled analysis of texts within their context of communication, following content analytic rules and step-by-step models’ (Mayring, 2000, p. 5). This approach combines the strengths of the grounded theory approach in the discovery of natural categories (Glaser and Strauss, 1967; Strauss and Corbin, 1998) with the strategies and standards (transparency, objectivity, reliability) from traditional content analysis (Krippendorff, 1980). Qualitative content analysis appreciates the idea that qualitative and quantitative methods should not be regarded as opposing, but complementary approaches (Van Maanen, 1979). The process of qualitative content analysis involved the following steps, described earlier by Schilling (2006): (i) transcription of the tapes and anonymisation of the resulting texts; (ii) reducing the material while preserving the essential contents, by condensing the text to only convey the dimensions of interest (leadership ideals, hindering factors and consequences); (iii) structuring the material into different dimensions (i.e. leadership ideal, hindering factors, consequences); (iv) step-by-step formulation of inductive categories (open coding; Strauss and Corbin, 1998), which included the basic processes of naming and comparing the data fragments to develop categories for multiple statements and—if necessary—subsuming old or formulating new categories (Conger, 1998); (v) revising the categories after 10–50 per cent of the material had been coded; (vi) checking the category codes—about two-thirds of the way through the material, the developing category system was checked to prevent drifting into an idiosyncratic sense of what the codes mean (Miles and Huberman, 1994); and (vii) final revision of the material (selective coding; Locke, 2002). The data-set was re-examined to make sure that the categories are fully described. Random samples of the material were controlled by a second researcher for the appropriateness of the data reduction and coding. The second researcher coded approximately 20 per cent of the material and the agreement of different raters was checked and cases of doubt and problems with scope and overlapping of the categories were discussed within the research team (Schilling, 2006). As Maxwell (1998) points out, qualitative research is not completely limited to inductive category development, but should also take categories from theory as the starting point for analysis. In our study, the development of categories of hindering factors were inductive, while the coding of leadership ideals started with a system of categories based on the Full Range of Leadership model. Results Leadership ideals during the first year of service At the first interview, we asked what kind of leadership they were striving for in order to capture their leadership ideals. At the second interview, six months later, we asked what kind of leadership they now found appropriate. At the third interview, after one year, we followed up the ideals they had at the first interview and asked whether they were still striving for the same kind of leadership or whether new ideals were present. All questions were open-ended and not formulated by any leadership theory. Afterwards, all statements were categorised using the Full Range of Leadership model as a framework. At the first interview, out of fifty-eight statements regarding leadership ideals, fifty-one represented transformational leadership behaviour and seven statements were categorised as transactional leadership behaviour. The categorisation is presented in Table 1, where the ideals are sorted into different transformational and transactional leadership behaviour with representative quotes illustrating each behaviour. The managers described ideals such as being trustworthy, creating confidence, having a common goal and telling employees that they will succeed—all of which were categorised as idealised influence behaviour. Examples of behaviour categorised as inspirational motivation were having a vision, motivating employees, showing enthusiasm for and enjoyment of work, and creating meaningfulness. Intellectually stimulating ideals were mentioned in terms of empowerment, getting the employees involved and developing independent workgroups. Individualised consideration was described in terms of being supportive, listening, seeing the individual, being present and having a dialogue with employees. Transactional leadership behaviour were not commonly described, but the few statements regarding transactional leadership reflected contingent reward behaviour such as being very clear and straightforward about their expectations. No laissez-faire behaviour was described among leadership ideals. Table 1 Leadership ideals: categories, codes and data Leadership behaviour  Representative data  Transformational leadership    Idealised influence    Trustworthy  ‘It is important that I am trustworthy. If I am not they won’t listen to what I say’ (Respondent 1)  Creating confidence  ‘I want them to have faith in me’ (Respondent 2)  Common goal  ‘I may say that now we have this issue to solve and how can we solve this together? At the end of the day we have the same goal’ (Respondent 2)  We will make it  ‘I want them to feel like a group and give them a feeling that they will make it’ (Respondent 2)  Inspirational motivation    Vision  ‘Until autumn we will be the best personal assistants in the municipally … that is the vision’ (Respondent 3)  Motivating employees  ‘It is important to get the employees with you, to meet them where they are and to increase their motivation’ (Respondent 3)  Enthusiasm and enjoyment  ‘I want employees to enjoy work, to feel appreciated and that they do something that they like’. (Respondent 4)  Meaningfulness  ‘To feel needed and that they contribute, to feel in their hearts that this is my thing and I can handle this … this I think is very important’ (Respondent 4)  Intellectual stimulation    Empowerment and involvement  ‘I solve problems together with my personal assistants; I ask them how they see it and if anyone has a solution to the problem. This usually gives a good result, when they get to think on their own’ (Respondent 3)  Independent workgroups  ‘Employees are quite capable, I want to give them tools so that they can manage on their own’ (Respondent 2)  Individualised consideration    Being supportive  ‘Being a manager is a position to be in a relationship. I want to be in a relationship with my employees … so I can support them. Sometimes leading the way but above all supporting them to find their way’ (Respondent 5)  Listening  ‘I try to listen, trying to understand what is going on and to understand their situation’ (Respondent 6)  Seeing the individual  ‘I think it is important to see the individual, everyone should felt seen. It is important that I know everyone by name’ (Respondent 7)  Being present  ‘I think it is crucial to be present, to meet the work groups also between staff meetings’ (Respondent 8)  Having a dialogue  ‘I hope employees feel free to have an open dialogue with me’ (Respondent 3)  Transactional leadership    Contingent reward    Having clear expectations  ‘It is important that I am clear with what I want from them and that they are straight with me’ (Respondent 2)  Leadership behaviour  Representative data  Transformational leadership    Idealised influence    Trustworthy  ‘It is important that I am trustworthy. If I am not they won’t listen to what I say’ (Respondent 1)  Creating confidence  ‘I want them to have faith in me’ (Respondent 2)  Common goal  ‘I may say that now we have this issue to solve and how can we solve this together? At the end of the day we have the same goal’ (Respondent 2)  We will make it  ‘I want them to feel like a group and give them a feeling that they will make it’ (Respondent 2)  Inspirational motivation    Vision  ‘Until autumn we will be the best personal assistants in the municipally … that is the vision’ (Respondent 3)  Motivating employees  ‘It is important to get the employees with you, to meet them where they are and to increase their motivation’ (Respondent 3)  Enthusiasm and enjoyment  ‘I want employees to enjoy work, to feel appreciated and that they do something that they like’. (Respondent 4)  Meaningfulness  ‘To feel needed and that they contribute, to feel in their hearts that this is my thing and I can handle this … this I think is very important’ (Respondent 4)  Intellectual stimulation    Empowerment and involvement  ‘I solve problems together with my personal assistants; I ask them how they see it and if anyone has a solution to the problem. This usually gives a good result, when they get to think on their own’ (Respondent 3)  Independent workgroups  ‘Employees are quite capable, I want to give them tools so that they can manage on their own’ (Respondent 2)  Individualised consideration    Being supportive  ‘Being a manager is a position to be in a relationship. I want to be in a relationship with my employees … so I can support them. Sometimes leading the way but above all supporting them to find their way’ (Respondent 5)  Listening  ‘I try to listen, trying to understand what is going on and to understand their situation’ (Respondent 6)  Seeing the individual  ‘I think it is important to see the individual, everyone should felt seen. It is important that I know everyone by name’ (Respondent 7)  Being present  ‘I think it is crucial to be present, to meet the work groups also between staff meetings’ (Respondent 8)  Having a dialogue  ‘I hope employees feel free to have an open dialogue with me’ (Respondent 3)  Transactional leadership    Contingent reward    Having clear expectations  ‘It is important that I am clear with what I want from them and that they are straight with me’ (Respondent 2)  Six months later, at the second interview, the managers were asked what kind of leadership they now found appropriate, with no reference to the leadership ideals they mentioned at the first interview. Compared to the first interview, there was now more of a balance between transformational and transactional leadership ideals, with nineteen out of thirty-four statements reflecting transformational leadership behaviour and the remaining fifteen reflecting transactional leadership. Transformational leadership ideals centred on individualised consideration such as being present and available, but also on idealised influence behaviour such as staying calm and consistent. Surprisingly, no statements regarding inspirational motivation were mentioned. Statements regarding transactional leadership focused on contingent reward behaviour such as being clear, organising, structuring and setting clear boundaries. At the final interview, one year later, managers were reminded of their initial leadership ideals at the first interview, and were asked whether they still agree or whether new ideals have emerged. Most managers still agreed with their initial ideals, even though hindering factors made them hard to live up to. New ideals had also emerged, mostly in terms of contingent reward behaviour such as being very clear about what the manager wanted the employees to do. In sum, twenty-six out of thirty-three statements at the final interview reflected transformational leadership behaviour and seven statements reflected transactional contingent reward behaviour. Hindering factors in the organisation At the second and third interviews, the managers were asked whether they experienced any hinders in the organisation that prevented them from performing the leadership they were striving for. After analysing all statements using the procedures of content analysis earlier described in the methods section, eight categories of organisational hinders emerged from the data. These were: ‘top-down management’, ‘continuous change’, ‘financial strain’, ‘distance to employees’, ‘administrative tasks’, ‘limited influence’, ‘no support’ and ‘high work-load’. In Figure 1, a concept map (Schilling, 2006) with a summary of the perceived hinders are presented. Categories of hindering factors are presented at the centre of the figure, and the codes belonging to each category are presented further out in the figure. The closer the code is to the category, the more statements it included, meaning that the graphical order of codes within one category equals their ranking based on their frequency. Figure 1 View largeDownload slide Concept map of hindering factors in the organisation Figure 1 View largeDownload slide Concept map of hindering factors in the organisation The category ‘top-down management’ included statements that described a hierarchical organisation with centralised power where decisions takes time, where co-operation between different units is rare, and where there upper management gives first line managers assignments and problems to solve within a short time frame. While related, the category ‘continuous change’ focuses on issues related to an organisation in constant change, with reorganisations which may lead to new work groups for the manager, but also to new decisions that influence the managers’ daily work. The category ‘financial strain’ addresses problems due to an organisation in constant financial crises, where budget cuts and conflicts between employee and financial interests make it difficult to lead. In contrast to the above categories, which reflect organisational conditions that in our interviewees’ opinion hinder them from leading the way they intended, the remaining categories were more related to the managers’ own working conditions. For example, the category ‘administrative tasks’ focused on obstacles in terms of work tasks such as recruiting, paying salaries and reporting back to the organisation via different computer programs, all of which took valuable time away from interacting with employees. Similarly, the category ‘distance to employees’ described obstacles due to the fact that the manager did not have their office close to their employees or that employees themselves were scattered, which made it difficult for the manager to get in touch with employees and to detect problems or knowing what was going on in their work groups. ‘Limited influence’ included statements related to a leadership situation characterised by an unclear mandate, not knowing what kind of authority they had to make decisions. Other statements addressed the experience of little say in the organisation and knowing that meeting performance standards does not matter. ‘High work-load’ included reports of too much to do and work tasks of very different kind while having responsibility for too many clients. Finally, the category ‘no support’ described problems in their leadership due to lack of support and help from managers, colleagues and support functions in the organisation such as the Human Resources. The perceived relationship between hindering factors and leadership ideals over time Some of the hindering factors were described in general terms; others were related to specific leadership behaviours. A summary of how managers expressed that initial ideals at Time 1 was hindered at Time 2 and Time 3 by organisational factors are presented in Table 2. Idealised influence was experienced as difficult to pursue in the presence of hierarchical decision making. The managers found it hard to keep promises to employees when the organisation needed a large amount of time to come to a decision. Although not an organisational factor, they also mentioned their own inexperience as a hindering factor, as they found it difficult to be trustworthy with little experience and knowledge about the organisation. Inspirational motivation was hindered when performance was not rewarded and by scattered work tasks. The managers found it difficult to promote a vision when they knew that living up to that vision would not matter to the organisation and that the high performance of employees did not lead to any reward from the organisation. They also experienced that their scattered work tasks hindered the long-term planning that a vision needs. Intellectual stimulation was hampered by too many administrative tasks and lack of influence. When managers experienced that they themselves had no influence in the organisation and in their own work, they found it harder to empower and stimulate employees to take their own initiatives. They also struggled with employees who resisted development, and with finding time for development activities in light of all the administrative tasks. Individualised consideration was hindered by distance to employees, which meant that they only met their employees occasionally and had no chance of knowing what was going on or whether problems had arisen. Transactional leadership in terms of contingent reward was difficult to pursue when new decisions continuously arrived from higher up in the organisation, which meant that expectations on and deals with the employees had to be constantly renegotiated. Again, the lack of leadership experience was mentioned as a hindering factor which made it hard to know what kind of expectations it was reasonable to have on employees. Please note that a summary of these findings is presented in the doctoral theses of the first author (see Tafvelin, 2013). Table 2 Longitudinal analysis of leadership behaviour and related hindering factors Leadership ideals at Time 1  Hindering factors perceived at Times 2 and 3  Idealised influence  Top-down management (hierarchical decision making)  Inspirational motivation  Limited influence (performance not rewarded), high work-load (scattered work tasks)  Intellectual stimulation  Administrative tasks, limited influence  Individualised consideration  Distance to employees  Contingent reward  Continuous change (new decisions)  Leadership ideals at Time 1  Hindering factors perceived at Times 2 and 3  Idealised influence  Top-down management (hierarchical decision making)  Inspirational motivation  Limited influence (performance not rewarded), high work-load (scattered work tasks)  Intellectual stimulation  Administrative tasks, limited influence  Individualised consideration  Distance to employees  Contingent reward  Continuous change (new decisions)  Discussion The present study focused on organisational antecedents that may hinder the emergence of transformational leadership in social service organisations. By identifying leadership ideals of newly recruited managers, we then captured factors in the organisations that hindered them from performing this ideal. Our results revealed that the managers strived for a transformational leadership, but they experienced a number of hindering factors that were associated with both the organisation and their own working environment. The hindering factors identified by our respondents demonstrated a great variety and originated from different levels in the organisation which we believe illustrates the advantage of our qualitative design. Some factors experienced by our respondents confirm findings in previous research, such as ‘top-down management’ in terms of hierarchical decision making (Wright and Pandey, 2009) and the need for influence or control (Nielsen and Cleal, 2011); others are in line with suggestions in the literature, such as the burden of too many administrative tasks (Alvesson, 2001). Interestingly, two categories have similarities with factors suggested to act as substitutes for leadership (Kerr and Jermier, 1978) such as ‘distance to employees’ that is similar to spatial distance, which is suggested to act as a substitute for leadership, and ‘limited influence’ shares grounds with no control over awards, which is another substitute for leadership. Previous studies have tried to establish substitutes for leadership as a moderator between transformational leadership and outcomes, but more or less failed to do so (Dionne et al., 2002). Our study indicates that substitutes for leadership may act as hindering antecedents, not taking the positive effects of transformational leadership away, but rather inhibiting transformational leadership behaviour altogether. The factors ‘high work-load’, ‘limited influence’ and ‘no support’ are closely connected to the well-known demand–control–support model that represents what is known to be a healthy working environment (de Lange et al., 2003; Karasek and Theorell, 1990). This indicates that good working conditions are important for managers in order to engage in transformational leader behaviours. Further, previous research has demonstrated that organisational change affects employees’ work climate, work attitudes and ill-health (Falkenberg, 2010). Our study indicates that ‘continuous change’ has consequences for leadership behaviour as well, in the sense that transformational leadership behaviour may be hindered when the organisation is in constant change. This may seem contradictory, as transformational leadership is suggested to be a suitable leadership for achieving organisational change (Bass and Riggio, 2006). However, managers are employees affected by organisational change themselves and may be caught in the middle, both trying to implement change ordered by senior management as well as ensuring the well-being of their employees (DeWitt et al., 2003). Further, if organisations are in constant change, this may for example affect the trust employees have in their manager, which makes role-modelling behaviour harder to pursue. Finally, the impact of a harsh economy and cutbacks has previously been connected with employee outcomes such as increased work-load and job insecurity (e.g. Hertting et al., 2005). Our study shows that ‘financial strain’ may also affect the extent to which managers engage in transformational leadership behaviour. All in all, the hindering factors identified by the respondents in this study offer a framework to use as a starting point in future research. The leadership ideals at the first interview were related to specific hindering factors described in later interviews, meaning that our interviewees experienced that specific transformational leadership behaviours were prevented by specific hindering factors. This gives a unique insight into how specific transformational leadership behaviours are related to different hindering factors. For example, in line with Wright and Pandey (2009), who found that performance measures in the public sector were positively correlated with transformational leadership, our analysis showed that inspirational motivation was experienced as hindered by performance not being awarded—a code under ‘limited influence’. Due to the highly correlated transformational factors, this information on sub-dimensions of transformational leadership is rarely seen in quantitative research. The finding that the managers in our study strived for a transformational leadership aligns well with the core attributes of social work leadership proposed by Holosko (2009), who found that leadership in social work is described in terms of vision, influencing others, team work, problem solving and the creating of positive change. Interestingly, at the second interview six months into their service, the managers in our study reported increasingly striving for transactional leadership behaviours such as being clear, organising, structuring and setting clear boundaries. Transactional leadership behaviours are considered an important complement to transformational leadership behaviours that helps the daily work to progress effectively (Bass and Riggio, 2006) and we suggest that transactional behaviours may be an overlooked aspect of leadership in social service organisations that deserve further research attention. Implications Our study adds to current knowledge of which factors that may hinder transformational leadership behaviour, and the results point towards factors at both the organisational level, such as ‘top-down management’, ‘financial strain’ and ‘continuous change’, and the managers’ own working environment in terms of ‘no support’, ‘high work-load’, ‘limited influence’, ‘administrative tasks’ and ‘distance to employees’. These results have implications for transformational leadership theory and touch on a sensitive issue within this theory, namely the role of context. Bass and Riggio (2006) quite explicitly expresses that transformational leadership theory is not a contingency model of leadership and that transformational leadership is expected to work equally well in all contexts, without regard to culture, country, company or sector. Our study, on the other hand, points towards the importance of context in terms of organisational factors that may hinder the emergence of transformational leadership, and lends support for the other opinion that stresses the need to advance our understanding of how the organisational context may hinder or facilitate the emergence of transformational leadership. This may be of particular importance in the public sector, where the usefulness of transformational leadership theory has been much disputed (Wright and Pandey, 2009). Even though many of the hindering factors were general in nature, such as a good working environment, others seems more specific to the public sector, such as ‘top-down management’ and ‘financial strain’. However, more research is needed to determine which factors are public-sector-specific or general to all organisations. Although the results of this study need to be confirmed by future research, there are some practical implications to be drawn from this study. If organisations are interested in implementing transformational leadership, it may be helpful to consider the hindering factors identified in this study in order to avoid organisational and work environmental factors that prevent managers from exhibiting transformational leadership behaviours. Trying to remove these hindering factors may be important in order to avoid ineffective leadership. Some of the hindering factors may be easier to change than others, such as creating a good working environment for the managers in terms of work-load, support and influence. Other factors, such as continuous change and financial strain, may be important for managers higher up in the organisation to be aware of to understand the consequences of constant organisational change and budget cuts. It may also be worthwhile to reconsider the administrative leader role and all the administrative tasks first line leaders in social service organisations undertake. Our study shows that these tasks prevent managers from interacting with employees, and valuable time and possibilities to influence employees in a positive way are lost. Limitations and future directions There are some limitations in the study that should be carefully regarded when interpreting the results. The sample of managers might have biased the results. We recruited managers from a single social service organisation, and the newly recruited managers invited to participate in the study were all female. It remains unclear whether men would stress different hindering factors and experience other consequences as a result of these hindering factors or whether managers in other types of public organisations share the experience of our respondents. It may be argued that the fact that all managers were newly recruited could imply that some of the hindering factors may be due to their inexperience as a manager and that more experienced managers may have overcome these hurdles already. Nevertheless, the inexperience of our sample is also a strength, as it makes it possible to capture the initial ideals before getting too acquainted with the organisation. This would have been difficult with a sample of experienced managers who may have already adjusted their leadership ideals and accepted the hindering factors in the organisation. Although our study included all newly recruited first line managers in the organisation, it still only included eight managers. Also, three managers did not attend all interviews, and we cannot be sure whether the managers that dropped out of the study perceived hindering factors in the same way as those who completed all three interviews. Our results should be interpreted with these limitations in mind. Another limitation to consider is that the system of categories developed in this study might be criticised in such a way that other solutions would have been possible. However, as pointed out by Locke (2002), a theoretical framework based on a qualitative study should be regarded as a starting point in the development of thinking of a phenomenon. The usefulness of the categories elicited in the present study should be evaluated based on whether they have something important to say about the research topic and can advance our knowledge on organisational antecedents of transformational leadership. The need for more research on organisational antecedents of transformational leadership has been pointed out by many scholars, as previous research has tended to focus on the leader as an individual (Wright and Pandey, 2009), and future studies may use our categories as a starting point to further elaborate which organisational factors hinder the emergence of transformational leadership. To conclude, this study reveals the great variety of organisational antecedents that may hinder the emergence of transformational leadership in social service organisations. The results seem especially promising in stimulating new directions for theory and research on organisational antecedents of transformational leadership as well as providing organisations with insights on which factors may hinder their managers from exhibiting transformational leadership behaviour. References Alvesson M. ( 2001) Organisationskultur och Ledning [Organizational Culture and Management], Malmo, Liber AB. Avolio B. J., Bass B. ( 1995) ‘ Individual consideration viewed at multiple levels of analysis: A multi-level framework for examining the diffusion of transformational leadership’, Leadership Quarterly , 6, pp. 199− 218. Google Scholar CrossRef Search ADS   Barbuto J. E., Burbach M. E. ( 2006) ‘ The emotional intelligence of transformational leaders: A field study of elected officials’, Journal of Social Psychology , 146( 1), pp. 51– 64. Google Scholar CrossRef Search ADS PubMed  Bass B. M., Riggio R. E. ( 2006) Transformational Leadership , 2nd edn, Mahwah, NJ, Erlbaum. Bono J. E., Hooper A. C., Yoon D. J. ( 2012) ‘ Impact of rater personality on transformational and transactional leadership ratings’, Leadership Quarterly , 23( 1), pp. 132– 45. Google Scholar CrossRef Search ADS   Chi N. W., Chung Y. Y., Tsai W. C. ( 2011) ‘ How do happy leaders enhance team success? The mediating roles of transformational leadership, group affective tone, and team processes’, Journal of Applied Social Psychology , 41( 6), pp. 1421– 54. Google Scholar CrossRef Search ADS   Conger J. A. ( 1998) ‘ Qualitative research as the cornerstone methodology for understanding leadership’, Leadership Quarterly , 9( 1), pp. 107– 21. Google Scholar CrossRef Search ADS   De Lange A. H., Taris T. W., Kompier M. J., Houtman I. L. D., Bongers P. M. ( 2003) ‘ The very best of the millennium: longitudinal research and the demand-control-(support) model’, Journal of Occupational Health Psychology , 8( 4), pp. 282– 305. Google Scholar CrossRef Search ADS PubMed  Deinert A., Homan A., Boer D., Voelpel S., Gutermann D. ( 2015) ‘ Transformational leadership sub-dimensions and their link to leaders’ personality and performance’, Leadership Quarterly , 26( 6), pp. 1095– 120. Google Scholar CrossRef Search ADS   DeWitt R., Trevino L., Mollica K. ( 2003) ‘ Stuck in the middle: A control-based model of mangers’ reactions to their subordinates’ layoff’, Journal of Managerial Issues , 15, pp. 32– 49. Dionne S., Yammarino F. J., Atwater L., James L. ( 2002) ‘ Neutralizing substitutes for leadership theory: Leadership effects and common-source bias’, Journal of Applied Psychology , 87( 3), pp. 454– 64. Google Scholar CrossRef Search ADS PubMed  Dóci E., Hofmans J. ( 2015) ‘ Task complexity and transformational leadership: The mediating role of leaders’ state core self-evaluations’, Leadership Quarterly , 26( 3), pp. 436– 47. Google Scholar CrossRef Search ADS   Eagly A. H., Johannesen-Schmidt M. C., van Engen M. L. ( 2003) ‘ Transformational, transactional, and laissez-faire leadership styles: A meta-analysis comparing women and men’, Psychological Bulletin , 129( 4), pp. 569– 91. Google Scholar CrossRef Search ADS PubMed  Falkenberg H. ( 2010) ‘How privatization and corporatization affect healthcare employees’ work climate, work attitudes and ill-health: Implications of social status’, doctoral thesis, Stockholm University, Sweden, available online at http://su.diva-portal.org/smash/get/diva2:298968/FULLTEXT01 (accessed 4 May 2017). Gellis Z. ( 2001) ‘ Social work perceptions of transformational and transactional leadership in health care’, Social Work Research , 25( 1), pp. 17– 25. Google Scholar CrossRef Search ADS   Glaser B., Strauss A. ( 1967) The Discovery of Grounded Theory: Strategies for Qualitative Research , New York, Aldine de Gryter. Hertting A., Nilsson K., Theorell T., Larsson U. S. ( 2005) ‘ Assistant nurses in the Swedish healthcare sector during the 1990s: A hard-hit occupational group with a tough job’, Scandinavian Journal of Public Health , 33( 2), pp. 107– 13. Google Scholar CrossRef Search ADS PubMed  Holosko M. J. ( 2009) ‘ Social work leadership: Identifying core attributes’, Journal of Human Behavior in the Social Environment , 19( 4), 448– 59. Google Scholar CrossRef Search ADS   Judge T., Piccolo R. F. ( 2004) ‘ Transformational and transactional leadership: A meta-analytic test of their relative validity’, Journal of Applied Psychology , 89( 5), pp. 755– 68. Google Scholar CrossRef Search ADS PubMed  Karasek R., Theorell T. ( 1990) Healthy Work: Stress, Productivity, and the Reconstruction of Working Life , New York, Basic Books. Karlsson I. ( 2006) ‘Att leda i kommunal äldreomsorg: Om arbetsledares handlingsutrymme, handlingsfrihet och skilda lojaliteter—exemplet Halland’ [‘Leading in elderly care: On supervisors’ autonomy and separated loyalties—Halland as an example’], Social Work Reports , 1652– 8573, Växjö University. Kerr S., Jermier J. ( 1978) ‘ Substitutes for leadership: Their meaning and measurement’, Organizational Behavior and Human Performance , 403, pp. 375– 403. Google Scholar CrossRef Search ADS   Krippendorff K. ( 1980) Content Analysis: An Introduction to Its Methodology , Beverly Hills, CA, Sage. Locke K. ( 2002) ‘The grounded theory approach to qualitative research’, in Drasgow F., Schmitt N. (eds), Measuring and Analyzing Behavior in Organizations: Advances in Measurement and Data Analysis , San Francisco, Jossey-Bass, pp. 17– 43. Lowe K. B., Kroeck K. G., Sivasubramaniam N. ( 1996) ‘ Effectiveness correlates of transformational and transactional leadership: A meta-analytic review of the literature’, Leadership Quarterly , 7, pp. 385– 425. Google Scholar CrossRef Search ADS   Mary L. ( 2005) ‘ Transformational leadership in human service organizations’, Administration in Social Work , 29( 2), pp. 105– 18. Google Scholar CrossRef Search ADS   Maxwell J. A. ( 1998) ‘Designing a qualitative study’, in Bickman L., Rog D. J. (eds), Handbook of Applied Social Research Methods , Thousand Oaks, CA, Sage, pp. 69– 100. Mayring P. ( 2000) ‘ Qualitative content analysis’, Forum: Qualitative Social Research , 2, pp. 1– 28, available online at http://www.qualitative-research.net/index.php/fqs/article/view/1089/2385 (accessed 12 May 2017). McCracken G. ( 1988) The Long Interview , Thousand Oaks, CA, Sage. Google Scholar CrossRef Search ADS   Miles M. B., Huberman A. M. ( 1994) Qualitative Data Analysis: An Expanded Sourcebook , London. Sage. Nielsen K., Cleal B. ( 2011) ‘ Under which conditions do middle managers exhibit transformational leadership behaviors? An experience sampling method study on the predictors of transformational leadership behaviors’, Leadership Quarterly , 22( 2), pp. 344– 52. Google Scholar CrossRef Search ADS   Nielsen K., Randall R., Yarker J., Brenner S.-O. ( 2008) ‘ The effects of transformational leadership on followers’ perceived work characteristics and psychological well-being: A longitudinal study’, Work and Stress , 22( 1), pp. 16– 32. Google Scholar CrossRef Search ADS   Pandey S., Wright B. E. ( 2006) ‘ Connecting the dots in public management: Political environment, organizational goal ambiguity, and the public manager’s role ambiguity’, Journal of Public Administration Research and Theory , 16( 4), pp. 511– 32. Google Scholar CrossRef Search ADS   Pawar B. S., Eastman K. K. ( 1997) ‘ The nature and implications of contextual influences on transformational leadership: A conceptual examination’, Academy of Management Review , 22( 1), p. 80. Rank M. G., Hutchison W. S. ( 2000) ‘ An analysis of leadership within the social work profession’, Journal of Social Work Education , 36( 3), pp. 487– 502. Google Scholar CrossRef Search ADS   Schilling J. ( 2006) ‘ On the pragmatics of qualitative assessment’, European Journal of Psychological Assessment , 22( 1), pp. 28– 37. Google Scholar CrossRef Search ADS   Strauss A., Corbin J. ( 1998) Basics of Qualitative Research: Techniques and Procedures for Developing Grounded Theory , Thousand Oaks, CA, Sage. Tafvelin S. ( 2013) ‘The transformational leadership process: Antecedents, mechanisms, and outcomes in the social services’, doctoral thesis, Umeå University, Sweden, available online at https://www.diva-portal.org/smash/get/diva2:640843/FULLTEXT01.pdf (accessed 4 May 2017). Tafvelin S., Hyvönen U., Westerberg K. ( 2014) ‘ Transformational leadership in the social work context: The importance of leader continuity and co-worker support’, British Journal of Social Work , 44( 4), pp. 886– 904. Google Scholar CrossRef Search ADS   Van Maanen J. ( 1979) ‘ Reclaiming qualitative methods for organizational research: A preface’, Administrative Science Quarterly , 24, pp. 520– 6. Google Scholar CrossRef Search ADS   Waldman D. A., Lituchy T., Gopalakrishnan M., Laframboise K., Galperin B., Kaltsounakis Z. ( 1998) ‘ A qualitative analysis of leadership and quality improvement’, Leadership Quarterly , 9( 2), pp. 177– 201. Google Scholar CrossRef Search ADS   Walter F., Bruch H. ( 2010) ‘ Structural impacts on the occurrence and effectiveness of transformational leadership: An empirical study at the organizational level of analysis’, Leadership Quarterly , 21( 5), pp. 765– 82. Google Scholar CrossRef Search ADS   Wright B. E., Pandey S. K. ( 2009) ‘ Transformational leadership in the public sector: Does structure matter?’, Journal of Public Administration Research and Theory , 20( 1), pp. 75– 89. Google Scholar CrossRef Search ADS   Yoo J., Brooks D. ( 2005) ‘ The role of organizational variables in predicting service effectiveness: An analysis of a multilevel model’, Research on Social Work Practice , 15( 4), pp. 267– 77. Google Scholar CrossRef Search ADS   © The Author 2017. Published by Oxford University Press on behalf of The British Association of Social Workers. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)