Inter-Personal and Critical-Thinking Capabilities in Those about to Enter Qualified Social Work: A Six-Centre Study

Inter-Personal and Critical-Thinking Capabilities in Those about to Enter Qualified Social Work:... Abstract The ‘process’ of intervention is understood to be fundamental to social work—evident in, for example, the literature on reflexivity. Little work, though, has focused on the detailed excavation of the cognitive processes of reasoning in decision making. This is widely recognised as requiring considerable analytic and critical abilities. Although this is long established, its importance is contemporarily apparent at the policy level from the rationale underlying current initiatives such as Frontline. However, it is also long understood that the reasoning capabilities underlying these processes cannot be considered in isolation from the inter-personal–emotional, encapsulated in a long-term theoretical concern for both Heart and Head. Furthermore, terms like ‘capability’ or ‘proficiency’ in professional qualification imply some standard to be reached in practice. This invites measurement. This novel study seeks to bring together three dimensions of the (i) measurement of (ii) the inter-personal–emotional and (iii) critical thinking—measurement of key facets of both Head and Heart. A six-centre, six-university collaboration, it focuses on those at a crucial point: where individuals are about to enter qualified practice. The findings show they score highly on most inter-personal measures (with room for improvement), but show huge variability in critical-thinking capabilities. The implications of this are discussed. Critical thinking, inter-personal skills, expertise Introduction How social workers think—the cognitive processes of decision making—is an issue at least as important as what they should know. The latter is the subject of the much more prominent focus on evidence-based/informed practice (Scurlock-Evans and Upton, 2015). How practitioners make sense, though, is an issue of meaning, but that ‘making of meaning’ is also a cognitive-reasoning process (thinking about something logically): What are the key facets of reasoning involved? What are the mental processes (rather than, for example, situational characteristics) by which they decide a course of action? How can they ensure the maximum rigour when conducting practice? How can error be minimised? and so on. These are all—from a cognitive perspective—issues of reasoning rather than knowledge. This focus on process leads us towards the long-established (theoretically) fundamental element of social work of ‘Heart’ (the emotional-inter-personal) and ‘Head’ (reasoning/intellectual). By way of introduction, we shall focus on these two key issues—which underlie judgement and decision making in practice—alongside that of their implicit and explicit measurement, which have accompanied much recent discourse on policy development in social work education and training. Their fundamental theoretical status, centrality in establishing competence or capability and key position in current and perennial policy issues relating to social work make them a critical focus for our attention. An absence of focused, measured research on this matter makes this focus still more urgent. This focus on reasoning processes is at the heart of judgement and decision making. However, compared with, for example, medicine (Patel et al., 2005), there has been little focus on these reasoning processes in social work research. This does not mean that process as a whole has not been of interest in social work—reflexivity, with its myriad meanings, multiple theoretical positions and, at times, combination of situational factors with process, in part does (D’Cruz et al., 2007). But the contribution—specifically—of process reasoning has been little examined. Where it has, this has primarily been through theoretical considerations or critical analysis rather than research-based (Munro, 1999; Taylor and White, 2006). Equally, the issue of measurement is at the heart of practice—not only in the sense of, for instance, evaluating outcomes or levels of need, but in the quality of practice performed. Measurement is, however, implicit, hidden by its qualitative nature. If we look at the most obvious measures of quality—those of capabilities (BASW, 2016) or proficiencies (HCPC, 2017)—we see standards to be achieved, expressed linguistically. Those qualifying should be competent, capable or proficient (depending on which adjective is in vogue). A myriad of expectations are encapsulated within standards for qualification. Each of these involves reaching a certain level. That level, however, is one qualitatively measured—a matter of judgement and left to a wide range of practice educators to decide. Who is to say that practice educators in Carlisle make judgements on the same grounds as those in Camborne, or those in Berwick the same as Bromley? However, although linguistic, they are as much measures, albeit implicit, as would be the case if they were numerical. As a caveat, we might expect greater exactitude—and potentially consistency—with (rigorously developed) numerical measures. Clearly, we are measuring practice quality when we use criteria, even linguistic, to distinguish that which is good enough from that which is not. This reasoning process, or level of criticality, and implicit measurement are, furthermore, of central policy interest. This is reflected most obviously in initiatives like Frontline (2014). This emerged with the explicit purpose of focusing on improving the quality of social work force, particularly its leadership, through encouraging high-calibre entrants to social work. In an early bulletin, it advertised the huge increase in Oxbridge and Russell Group graduates applying for Frontline, when compared to conventional social work qualifying routes, reflecting, it stated, the need of social work for ‘attracting the best and brightest’ and calling social work ‘a highly skilled leadership profession’ (Frontline, 2014) . It is fast-track, carried out separately from mainstream social work education, trains a minority of starting professionals and costs considerably more. It is the subject of some controversy, some criticising it as misdirected and politically motivated (Cooper, 2017). Both the process of practical reasoning underlying judgement and decision making and its implicit measurement are key to social work, yet subject to limited research. These, though, cannot be considered without reference to the inter-personal-emotional. Their necessarily interwoven nature is encapsulated in the long-standing notion of social work being a matter of ‘Heart and Head’ (Bosanquet, 1901; Hardiker, 1981). These may be considered foundation elements underlying much, if not all, aspects of social work (certainly that related to work with service users). Their status as key foundations for social work is evident from the separate literatures devoted to the intellectual–critical thinking and the inter-personal relationships (Coleman et al., 2002; Brown and Rutter, 2009; Hennessy, 2011; Ruch et al., 2010) and indeed the proficiencies required of qualifying practitioners (HCPC, 2017). Their centrality in social work is widely recognised. Coleman et al. (2002, p. 583) comment—regarding critical thinking—on its practical importance: ‘Critical thinking is an important ability for social workers to have.’ Brown and Rutter (2009, p. 29) indicate why: ‘… the skills of critical thinking allow the best quality decisions or actions for the situations … [social workers] encounter.’ Likewise, on relationships, Howe (1998, p. 45) comments ‘that relationship skills … are necessary in … social work, … not an add-on luxury …. Rather, they are integral and essential to the theory and practices of the occupation’. The features, in other words, are necessary conditions for social work to be taking place at all. Decision making in social work, then, is a process: developmental, with Head and Heart as foundation elements. Some social work research has addressed decision making but not generally in the tradition of cognitive processes analysis (Helm, 2011; Whittaker, 2011). One approach has focused on the examination and explication of cognitive processes (the processes involved in thinking and reasoning). Sheppard’s (1995a, 1995b) theoretical work on ‘Emergent Understanding’ drew on three foci: construction of meaning, process of reasoning and accuracy of judgement (Sheppard, 1995a, 1995b). Subsequently, empirical research conceptualised a range of reasoning processes undertaken. However, the data suggested considerable variation between social workers in precision and capability in this reasoning process (Sheppard et al., 2000, 2001). This tantalising small-scale study was undertaken, however, without formal measurement or overt reference to the ‘Heart’ element of Head and Heart. That Head and Heart, however, have an uneasy, even contradictory, relationship has become evident in a more recent study (Sheppard and Charles, 2015). This demonstrated the intellectual and inter-personal remain highly distinct domains in social work—capabilities in the one at times actively contrary to those in the other. It provided a welcome demonstration of the exceptional capabilities required by social workers—there is evidence in literature beyond social work that proficiencies tend to be in one or the other (cognitive or empathetic) rather than both (Gillin et al., 2013). However, it also indicated evidence of the fundamental demands of social work, before the complexity of cases is even considered. Their fundamental place within social work; the urgency bestowed by current policy and practice developments in social work; their key role in judgement and decision making; and the growing importance of precision in determining performance levels suggest an urgent need for overt measurement of the intellectual–critical thinking and emotional–inter-personal (Head and Heart) in tandem. This research seeks for the first time to undertake a substantial study measuring these two key dimensions jointly at a crucial point by carrying out a six-centre study involving twelve cohorts on the verge of entering qualifying practice. It has the following aims: to identify inter-personal and critical-thinking capabilities in social work graduands; to contextualise these measures through comparison with relevant normative samples; to examine variations in these capabilities according to university (pre/post-92) status and degree (master’s/bachelor’s level). Methods This study was of a cross-sectional survey design. It was a six-university collaboration involving twelve cohorts equally divided between bachelor’s and master’s programmes from geographically diverse universities in England and Wales. These institutions constituted just under one-fifth of the thirty-four universities providing both master’s and bachelor’s social work provision, enabling a comparison by degree taken. The sample was broadly stratified to reflect the fifty/fifty division between pre- and post-92 universities, with three former and three latter institutions. Pre-92 universities are older establishments, while post-92 universities have been established after 1992. The former generally have higher status, applications and entrance standards. A pilot study was undertaken prior to the main research, to ensure both the viability of the study and its efficacy (Sheppard and Charles, 2014). Instruments Two measures were used: the NEOPI-R to measure inter-personal (IP) and the Watson-Glaser Critical Thinking Appraisal (WGCTA) to measure critical-thinking (CT) capabilities. These are the longest established, most widely used and comprehensively tested instruments in their fields, with high levels of reliability and validity, confirmed internationally (Costa and McCrae, 2006; Loom and Thorpe, 1999). They also both possess normative sample data in relation to which our sample could be compared, enabling findings to be placed in context. The NEOPI-R examines the five domains of personality (neuroticism, extraversion, openness, agreeableness and conscientiousness). These domains contain thirty facets, each of which is sufficiently robust to be examined in its own right (Lord, 2007). Our focus was on six facets which have been identified to be of enduring importance to social work. These were (with references to relevant exemplar social work literature in brackets): Altruism (Ngai and Cheung, 2009); Warmth (Koprowska, 2010); Compassion/tender mindedness (Radley and Figley, 2007; Kinman and Grant, 2011); Feeling Insight (personal receptivity) (Kinman and Grant, 2011); Deliberation (Koprowska, 2010); Assertiveness (McCabe, 1998; Hardcastle et al., 2011, Chapter 8). These are areas often associated with emotional intelligence. However, this has been subject to extensive criticism in psychology for: the misapplication of the term ‘intelligence’, failing to differentiate between personality traits and emotional states attitudes and values, and confusion with established personality traits for which there is considerable evidence without recourse to emotional intelligence as a term (Ingram, 2013; Locke, 2005; Schulte et al., 2004; Smith et al., 2008) and may best be understood as a metaphor. Rather than ‘throw the baby out with the bathwater’, we identified key IP areas long established in social work, employing the extensively validated and reliable NEOPI-R for the purpose of measurement. Although we focused on six facets, it was necessary, nevertheless, for the NEO PI-R to be completed as a whole (thence to extract data from the relevant facets). Hence, participants responded to the 240 items from which data from the five domains and thirty constituent facets (each of which contained eight items) may be extracted. The WGCTA measures CT, ‘defined as the ability to identify and analyse problems as well as seek and evaluate relevant information in order to reach appropriate conclusions’ (TalentLens, 2011, p. 3). Although general, as a description of key elements of reasoning and thinking for social work, this can hardly be bettered. However, WGCTA nevertheless focuses on generic CT capabilities, enabling contextualisation of findings/scores. The WGCTA has three domains, measuring reasoning in Inference, Recognising Assumptions and Drawing Conclusions (containing Deduction, Interpretation and Evaluating Arguments) (TalentLens, 2011). Inference involves ‘rating the probability of truth of inferences [deducing or reasoning something from evidence or reasoning rather than explicit statements]… based on given information’; Recognising Assumptions is ‘identifying unstated assumptions or propositions underlying given statements’; Drawing Conclusions encapsulated three facets—Deduction: ‘Determining whether conclusions follow logically from given information’; Interpretation: ‘weighing evidence and deciding if generalisations or conclusions based on data are warranted’; and Evaluating Arguments: ‘evaluating the strength of arguments with respect to a particular question or issue’ (TalentLens, 2011, pp. 3–4). It contains forty items, multiple-choice questions, each of which has a correct answer and overall, therefore, yields a maximum score of 40 (and minimum of 0). Measurement and scores for each domain of CT may also be extrapolated. Ethics and data collection The data were collected during the final term before qualification after, or close to, the end of final placement. Data collection was strategically placed at that point because of its crucial stage in professionalisation—immediately prior to the point of qualification, enabling us to have a ‘window’ into standards close to the point of starting qualified practice—and because this approach enabled us to capture a diverse and substantial sample. The study initially ethically approved by the lead institution received multi-site approval. Before undertaking these tasks, a number of facets were explained to the participants both verbally and in writing, including the nature and purpose of the study, freedom to choose whether or not to participate, confidentiality and the right to withdraw. They were given the opportunity to ask questions (generally taken up) which were answered. Written consent was provided by all participants. Only once this process was complete were the tasks undertaken. A pilot study focused on its efficacy and ethical features, including in particular the need for consent confidentiality and right to withdraw following initial consent (Sheppard and Charles, 2014). The instruments were themselves explained to participants, although clear instructions are also included in the instruments themselves. Participants completed the instruments in groups. They generally each take less than thirty minutes to complete, but participants were given as long as required. Each instrument was completed separately in order to prevent mental fatigue. A member of the research team was available throughout and all finished instruments were checked to ensure they were fully completed. Analysis The instruments produce interval data for both individual items and domains (facets in the case of NEOPI-R). Tests of significance employed, depending on data, the t-test or Analysis of Variance. However, with larger samples, significance (the likelihood that findings are a matter of chance or represent some underlying systematic difference) can be achieved where the scale of that difference is relatively trivial (i.e. the difference may indeed be systematic, but it is not very great). In these circumstances, additional calculations, including where appropriate, effect size (Cohen’s d) has been used. The coefficient of variation was used to examine the degree of homogeneity amongst social work graduands in the CT and IP dimensions examined. Comparison was undertaken according to university status (pre-/post-92), degree level (undergraduate/postgraduate); between different facets/dimensions; and against normative samples. Results The total cohort was 407, of whom 301 (74 per cent) participated—a highly satisfactory response rate. Eighty-six per cent were women; two-thirds (67 per cent) were white British, 23 per cent were black or mixed race (including black British) African or Caribbean; 3 per cent were Indian Pakistan or other Asian; the rest classified themselves as ‘other white’ (including white Irish) or preferred not to say. Mean age was thirty-one (minimum twenty, maximum fifty-nine). IP and CT measures Table 1 focuses on the graduands’ IP scores, in the context of the UK normative sample (Costa and McCrae, 2006). The UK Normative Group comprised 1,301 people: 795 men and 353 women (153 unknown) mean age forty-three. The authors sought to sample as broadly as possible with respect to industry sector, job title and seniority (Costa and McCrae, 2006). It is clear from its make-up that it was an opportunity sample and the implications of this will be made clear later when discussing the results. Table 1 Inter-personal (IP) scores compared with UK normative sample Social work (n=301) UK normative (n=1,301) % mean score difference (Social work versus UK) Mean (sd) min Max Coefficient of variation Mean (sd) Altruism 25.0 (3.4) 14 32 13.6 24.0 (3.4) 3.2 Warmth 24.2 (3.9) 11 32 16.0 23.9 (3.9) 1.0 Compassion 23.3 (3.2) 14 31 13.6 19.7 (3.6) 12.1 Feelings Insight 22.9 (4.1) 11 32 17.9 21.9 (4.2) 3.4 Deliberation 18.3 (4.7) 5 30 26.0 18.9 (4.4) −2.1 Assertiveness 16.6 (4.7) 3 32 28.2 21.0 (4.4) −14.5 Max score 32 32 Mean IP CoV 19.1 95% conf. interval t Df Mean diff. SED Lower Upper p d Altruism 4.6 1,600 1.0 0.22 0.57 1.43 <0.001 0.29* Compassion 16.0 1,600 3.6 0.23 3.16 4.04 <0.001 1.06** Insight 3.74 1,600 1.0 0.27 0.48 1.53 <0.001 0.24* Deliberation 2.10 1,600 0.6 0.29 0.04 1.16 <0.05 0.13 Assertiveness 15.43 1,600 –4.4 0.29 –4.96 –3.84 <0.001 0.97** Social work (n=301) UK normative (n=1,301) % mean score difference (Social work versus UK) Mean (sd) min Max Coefficient of variation Mean (sd) Altruism 25.0 (3.4) 14 32 13.6 24.0 (3.4) 3.2 Warmth 24.2 (3.9) 11 32 16.0 23.9 (3.9) 1.0 Compassion 23.3 (3.2) 14 31 13.6 19.7 (3.6) 12.1 Feelings Insight 22.9 (4.1) 11 32 17.9 21.9 (4.2) 3.4 Deliberation 18.3 (4.7) 5 30 26.0 18.9 (4.4) −2.1 Assertiveness 16.6 (4.7) 3 32 28.2 21.0 (4.4) −14.5 Max score 32 32 Mean IP CoV 19.1 95% conf. interval t Df Mean diff. SED Lower Upper p d Altruism 4.6 1,600 1.0 0.22 0.57 1.43 <0.001 0.29* Compassion 16.0 1,600 3.6 0.23 3.16 4.04 <0.001 1.06** Insight 3.74 1,600 1.0 0.27 0.48 1.53 <0.001 0.24* Deliberation 2.10 1,600 0.6 0.29 0.04 1.16 <0.05 0.13 Assertiveness 15.43 1,600 –4.4 0.29 –4.96 –3.84 <0.001 0.97** * Small effect size. **Large effect size. Source for normative sample: Costa and McCrae (2006). t-test, equal variance not assumed. Table 1 Inter-personal (IP) scores compared with UK normative sample Social work (n=301) UK normative (n=1,301) % mean score difference (Social work versus UK) Mean (sd) min Max Coefficient of variation Mean (sd) Altruism 25.0 (3.4) 14 32 13.6 24.0 (3.4) 3.2 Warmth 24.2 (3.9) 11 32 16.0 23.9 (3.9) 1.0 Compassion 23.3 (3.2) 14 31 13.6 19.7 (3.6) 12.1 Feelings Insight 22.9 (4.1) 11 32 17.9 21.9 (4.2) 3.4 Deliberation 18.3 (4.7) 5 30 26.0 18.9 (4.4) −2.1 Assertiveness 16.6 (4.7) 3 32 28.2 21.0 (4.4) −14.5 Max score 32 32 Mean IP CoV 19.1 95% conf. interval t Df Mean diff. SED Lower Upper p d Altruism 4.6 1,600 1.0 0.22 0.57 1.43 <0.001 0.29* Compassion 16.0 1,600 3.6 0.23 3.16 4.04 <0.001 1.06** Insight 3.74 1,600 1.0 0.27 0.48 1.53 <0.001 0.24* Deliberation 2.10 1,600 0.6 0.29 0.04 1.16 <0.05 0.13 Assertiveness 15.43 1,600 –4.4 0.29 –4.96 –3.84 <0.001 0.97** Social work (n=301) UK normative (n=1,301) % mean score difference (Social work versus UK) Mean (sd) min Max Coefficient of variation Mean (sd) Altruism 25.0 (3.4) 14 32 13.6 24.0 (3.4) 3.2 Warmth 24.2 (3.9) 11 32 16.0 23.9 (3.9) 1.0 Compassion 23.3 (3.2) 14 31 13.6 19.7 (3.6) 12.1 Feelings Insight 22.9 (4.1) 11 32 17.9 21.9 (4.2) 3.4 Deliberation 18.3 (4.7) 5 30 26.0 18.9 (4.4) −2.1 Assertiveness 16.6 (4.7) 3 32 28.2 21.0 (4.4) −14.5 Max score 32 32 Mean IP CoV 19.1 95% conf. interval t Df Mean diff. SED Lower Upper p d Altruism 4.6 1,600 1.0 0.22 0.57 1.43 <0.001 0.29* Compassion 16.0 1,600 3.6 0.23 3.16 4.04 <0.001 1.06** Insight 3.74 1,600 1.0 0.27 0.48 1.53 <0.001 0.24* Deliberation 2.10 1,600 0.6 0.29 0.04 1.16 <0.05 0.13 Assertiveness 15.43 1,600 –4.4 0.29 –4.96 –3.84 <0.001 0.97** * Small effect size. **Large effect size. Source for normative sample: Costa and McCrae (2006). t-test, equal variance not assumed. Table 1 shows, relative to the normative sample, that social work graduands scored particularly highly on Compassion, and higher in relation to Feelings Insight and Altruism, in all of which p < 0.001 (t-test). More significant is the scale of difference (answering the question: how much difference was there between two samples?). This is evident from calculating Cohen’s d, where a large effect size ≥ 0.8; a medium ≥0.5; and small≥ 0.2. This yielded a large effect size to Compassion and a small effect size to Feelings Insight and Altruism. They scored considerably lower, however, on Assertiveness. This was also significant and the effect size was large. The coefficient of variation, which measures the degree of variability around the mean, shows a much greater degree of homogeneity furthermore, amongst the ‘positive affect’ measures than those of Deliberation and Assertiveness. These graduands, in other words, were, relatively, much more alike on those former measures than latter measures where some were far more liable to be assertive and to deliberate than others. Table 2 focuses on CT scores. The (UK) General Population comparison group was chosen because of its general application; Law/Business because of the professional/occupational nature of their qualification and because they were applicants for public-sector jobs—characteristics largely shared with the social work sample. In both cases, furthermore, the focus was on UK candidates (TalentLens, 2011). However, we should again note that these were opportunity samples (of job applicants taking these tests) and there is little information on the social and demographic make-up of these groups (e.g. we are not given gender make-up of the Law/Business group, though we do of the General Population, in that case only one-third being female). Again, the implications of this will be discussed later. Table 2 Critical-thinking scores in context Between-group differences (Social Work, Law and public managers) Social Work graduands (n=301) General Population (fifty occupations) (n=1,546) Public-sector Law/Business job applicants (n=504) % mean score difference (GP) % mean score difference (Law) Mean (sd) Mean (sd) Mean (sd) WCCTT mean 20.7 (6.3) 26.5 (6.0) 27.0 (6.0) –14.5 –15.8 Max. score 40 40 40 Within-group (Social Work) differences Mean % Standard deviation Coefficient of variation Min. Max. Recognise Assumptions 43.3 25.4 58.7 0 100 Evaluate Arguments 53.7 17.8 33.2 8 100 Draw Conclusions 56.4 18.0 32.0 19 100 Total 51.7 15.7 30.4 20 95 Between-group differences (Social Work, Law and public managers) Social Work graduands (n=301) General Population (fifty occupations) (n=1,546) Public-sector Law/Business job applicants (n=504) % mean score difference (GP) % mean score difference (Law) Mean (sd) Mean (sd) Mean (sd) WCCTT mean 20.7 (6.3) 26.5 (6.0) 27.0 (6.0) –14.5 –15.8 Max. score 40 40 40 Within-group (Social Work) differences Mean % Standard deviation Coefficient of variation Min. Max. Recognise Assumptions 43.3 25.4 58.7 0 100 Evaluate Arguments 53.7 17.8 33.2 8 100 Draw Conclusions 56.4 18.0 32.0 19 100 Total 51.7 15.7 30.4 20 95 Source for Law/Business normative sample: TalentLens (2011). Table 2 Critical-thinking scores in context Between-group differences (Social Work, Law and public managers) Social Work graduands (n=301) General Population (fifty occupations) (n=1,546) Public-sector Law/Business job applicants (n=504) % mean score difference (GP) % mean score difference (Law) Mean (sd) Mean (sd) Mean (sd) WCCTT mean 20.7 (6.3) 26.5 (6.0) 27.0 (6.0) –14.5 –15.8 Max. score 40 40 40 Within-group (Social Work) differences Mean % Standard deviation Coefficient of variation Min. Max. Recognise Assumptions 43.3 25.4 58.7 0 100 Evaluate Arguments 53.7 17.8 33.2 8 100 Draw Conclusions 56.4 18.0 32.0 19 100 Total 51.7 15.7 30.4 20 95 Between-group differences (Social Work, Law and public managers) Social Work graduands (n=301) General Population (fifty occupations) (n=1,546) Public-sector Law/Business job applicants (n=504) % mean score difference (GP) % mean score difference (Law) Mean (sd) Mean (sd) Mean (sd) WCCTT mean 20.7 (6.3) 26.5 (6.0) 27.0 (6.0) –14.5 –15.8 Max. score 40 40 40 Within-group (Social Work) differences Mean % Standard deviation Coefficient of variation Min. Max. Recognise Assumptions 43.3 25.4 58.7 0 100 Evaluate Arguments 53.7 17.8 33.2 8 100 Draw Conclusions 56.4 18.0 32.0 19 100 Total 51.7 15.7 30.4 20 95 Source for Law/Business normative sample: TalentLens (2011). Between-group effects (Social Work compared to (a) General Population (b) Law/Business) 95% conf. interval t Df Mean diff. SED Lower Upper p d General Population 15.22 1,845 –5.8 0.38 –6.6 –5.1 <0.001 1.02** Law/Business 14.15 803 –6.3 0.45 –7.2 –5.4 <0.001 0.94** Between-group effects (Social Work compared to (a) General Population (b) Law/Business) 95% conf. interval t Df Mean diff. SED Lower Upper p d General Population 15.22 1,845 –5.8 0.38 –6.6 –5.1 <0.001 1.02** Law/Business 14.15 803 –6.3 0.45 –7.2 –5.4 <0.001 0.94** ** Large effect size. Between-group effects (Social Work compared to (a) General Population (b) Law/Business) 95% conf. interval t Df Mean diff. SED Lower Upper p d General Population 15.22 1,845 –5.8 0.38 –6.6 –5.1 <0.001 1.02** Law/Business 14.15 803 –6.3 0.45 –7.2 –5.4 <0.001 0.94** Between-group effects (Social Work compared to (a) General Population (b) Law/Business) 95% conf. interval t Df Mean diff. SED Lower Upper p d General Population 15.22 1,845 –5.8 0.38 –6.6 –5.1 <0.001 1.02** Law/Business 14.15 803 –6.3 0.45 –7.2 –5.4 <0.001 0.94** ** Large effect size. Greenhouse-Geisser test of within-subject effects Type 111 sum of squares Df Mean square F Sig. Partial eta squared Critical thinking 28,913 1.8 16,292 54.51 <0.001 0.15 Greenhouse-Geisser test of within-subject effects Type 111 sum of squares Df Mean square F Sig. Partial eta squared Critical thinking 28,913 1.8 16,292 54.51 <0.001 0.15 Greenhouse-Geisser test of within-subject effects Type 111 sum of squares Df Mean square F Sig. Partial eta squared Critical thinking 28,913 1.8 16,292 54.51 <0.001 0.15 Greenhouse-Geisser test of within-subject effects Type 111 sum of squares Df Mean square F Sig. Partial eta squared Critical thinking 28,913 1.8 16,292 54.51 <0.001 0.15 Table 2 shows these social work cohorts scored markedly lower in CT than either a General Population group of multiple occupations or public-sector job applicants with Law or Business degrees. The mean percentage score for these social workers was 50.7 per cent compared with 65.2 per cent for the General Population group and 66.5 per cent for the Law/Business group. In both cases, the differences (between social work graduands and the normative samples) was significant. The effect size, furthermore, was large in both cases. This is also clear from the difference in mean scores shown in Table 2. The ‘within-group’ comparison (of performance at different areas of CT) shows scores as a percentage of potential maximum. This approach was taken because there are different maximum scores for different facets of CT. The first noticeable feature is the considerable difference between the highest and lowest scores—individuals were qualifying with utterly different levels of CT capabilities. This diversity in CT is reflected in its coefficient of variation. If we compare this, for example, with the positive affect measures of Table 1, we find they were a far more homogeneous group in positive affect than they were in CT. Table 2 also compares the three facets of CT, showing the capacity to recognise assumptions was significantly lower than the capacity to evaluate arguments or draw conclusions. As interesting, however, was the relative diversity of capabilities in Recognising Assumptions in this sample, compared with Evaluating Arguments and Drawing Conclusions. Comparing sectors and levels: pre- and post-92 and master’s and bachelor’s levels Table 3 shows variations in CT according to university ‘sector’ (pre- and post-1992 universities) and degree level (bachelor’s and master’s). The pre-92 universities did not have graduands with higher CT capabilities. Although the post-92 universities (ex-polytechnics) had a mean score lower than pre-92 universities, this difference was not significant. However, there was a highly significant difference between mean scores for master’s-level compared with bachelor’s-level cohorts. Marked as a percentage out of forty (maximum), the master’s cohort’s mean score was 10 per cent higher than the bachelor’s group. There was, in Cohen’s terms, a ‘medium’ effect size (on its upper reaches) where d = 0.66. Table 3 Variations in critical thinking and Compassion according to degree level and university sector Number Mean Standard deviation Critical-thinking scores Master’s 132 22.9 6.9 Bachelor’s 169 18.9 5.1 Pre-92 121 21.1 6.8 Post-92 180 20.4 5.9 Compassion/tender Mindedness scores Pre-92 121 22.8 3.0 Post-92 180 23.7 3.3 Number Mean Standard deviation Critical-thinking scores Master’s 132 22.9 6.9 Bachelor’s 169 18.9 5.1 Pre-92 121 21.1 6.8 Post-92 180 20.4 5.9 Compassion/tender Mindedness scores Pre-92 121 22.8 3.0 Post-92 180 23.7 3.3 Table 3 Variations in critical thinking and Compassion according to degree level and university sector Number Mean Standard deviation Critical-thinking scores Master’s 132 22.9 6.9 Bachelor’s 169 18.9 5.1 Pre-92 121 21.1 6.8 Post-92 180 20.4 5.9 Compassion/tender Mindedness scores Pre-92 121 22.8 3.0 Post-92 180 23.7 3.3 Number Mean Standard deviation Critical-thinking scores Master’s 132 22.9 6.9 Bachelor’s 169 18.9 5.1 Pre-92 121 21.1 6.8 Post-92 180 20.4 5.9 Compassion/tender Mindedness scores Pre-92 121 22.8 3.0 Post-92 180 23.7 3.3 95% conf. interval t Df SED Lower Upper p d Master’s/bachelor’s 5.6 244.6 0.69 2.6 5.4 <0.001 0.66* Pre-/Post-92 Not significant Equal variance not assumed for critical thinking. 95% conf. interval t Df SED Lower Upper p d Master’s/bachelor’s 5.6 244.6 0.69 2.6 5.4 <0.001 0.66* Pre-/Post-92 Not significant Equal variance not assumed for critical thinking. * Medium effect size. 95% conf. interval t Df SED Lower Upper p d Master’s/bachelor’s 5.6 244.6 0.69 2.6 5.4 <0.001 0.66* Pre-/Post-92 Not significant Equal variance not assumed for critical thinking. 95% conf. interval t Df SED Lower Upper p d Master’s/bachelor’s 5.6 244.6 0.69 2.6 5.4 <0.001 0.66* Pre-/Post-92 Not significant Equal variance not assumed for critical thinking. * Medium effect size. 95% conf. interval t Df SED Lower Upper p Master’s/bachelor’s –2.3 299 0.37 –1.6 –1.3 <0.05 Equal variance may be assumed for Compassion 95% conf. interval t Df SED Lower Upper p Master’s/bachelor’s –2.3 299 0.37 –1.6 –1.3 <0.05 Equal variance may be assumed for Compassion 95% conf. interval t Df SED Lower Upper p Master’s/bachelor’s –2.3 299 0.37 –1.6 –1.3 <0.05 Equal variance may be assumed for Compassion 95% conf. interval t Df SED Lower Upper p Master’s/bachelor’s –2.3 299 0.37 –1.6 –1.3 <0.05 Equal variance may be assumed for Compassion The IP scores showed no significant differences, either between pre-/post-92 universities or between master’s and bachelor’s cohorts except, in relation to the former division, with Compassion. However, we should remember this was one of only twelve measures (six IP for each of the pre-/post-92 and master’s/bachelor’s divisions)—fourteen if we include CT measures. Where significance is set at p < 0.05 (a one-in-twenty likelihood the result will be a matter of chance), finding one significant difference may not be too surprising. The main impression inter-personally is of homogeneity. Discussion We should first note the limits to this study. Although a proportionately significant sample (of England and Wales as a whole), this was nevertheless only one year of graduands. They were, furthermore, a sample at one particular point in their career—becoming qualified practitioners. We do not know, for example, that a study would generate similar findings if, for instance, it concentrated on more experienced practitioners. It is cross-sectional—it does not provide a sense of the possibility for growth and change as might happen where a focus were longitudinally on, for example, students through the course of their study, social workers over a period of growing experience or post-qualifying study. It is focused on locations in England and Wales, rather than the UK as a whole. Nevertheless, there is much to recommend it. In an undeveloped field—the forensic examination of cognitive processes in judgement and decision making—the issue of whether key elements of these processes can be measured is very new indeed. Furthermore, where there is a major concern about standards, the issue of measurement becomes of acute importance. The sample is substantial and covers diverse locations within the UK. It is also structured in terms of key features liable to influence and make representative—the findings: the division between ‘old’ (pre-92) and ‘new’ (post 92) universities and between the two major social work levels of qualification—master’s and bachelor’s. Its focus, moreover, on social workers at a common point in career development lends coherence and focus to the study. That moment furthermore—at the point of qualification—is crucial. Finally, it complements its ‘sister’ three-year longitudinal study of the impact on these factors of social work education (Sheppard and Charles, 2017). Although our focus is on general implications of these findings for social work generally and education in particular, they are underpinned by anti-discriminatory assumptions/principles in two respects: we should be seeking to promote the entrance of candidates with the requisite qualities fairly and consistently in a way that ensures unintended bias does not hinder that entrance; and we should be seeking to promote amongst those selected the capacity to manifest these key professional qualities through educational and development opportunities. The use of these instruments may help in these respects, but also the following/subsequent comments—in terms of both concerns and suggestions—may be considered as intending to promote these two principles. IP capabilities provide a bedrock for social work—a sin qua non of the profession. In this respect, the data on positive emotions such as Compassion, Altruism and personal Feelings and Insight go some way to confirming what may be expected of these capabilities in those about to enter practice. We should be careful, however, about over-interpreting this. The normative group was an opportunity sample and we cannot state that it is representative, in the way characterised by a probability sample, of the adult UK population. Indeed, the gender make-up suggests this is not the case (women make up 47 per cent of the UK workforce, but a third of the normative sample) (Office of National Statistics, 2016). Deliberation and Assertiveness do not perhaps reflect those expectations to the same degree. Deliberation can be highly significant in ensuring decisions are thoughtful and considered, yet these graduands scored markedly less than the normative sample. Likewise, there are many occasions—working with dangerous situations, advocating and ensuring the rights of vulnerable clients with outside agencies and services—where Assertiveness is an important asset. It is, indeed, in the balance between features like Assertiveness and Deliberation and the positive emotions that social work may be most effectively manifested. These data suggest that this balance has perhaps not been most efficaciously reached. This may reflect the stage of these social workers’ career—as starting practitioners, they may not have encountered the situations which demand high levels of Assertiveness. Data on social workers further into their career may settle this issue. The homogeneity in most of these IP measures indicates people with similar IP characteristics. Master’s and bachelor’s levels, as well as both pre- and post-92 university cohorts, were quite similar. Variability indicates some were quite assertive while others were relatively unassertive. This is a feature in the development of early-career social workers that looks to require some focus, with some markedly better equipped than others to manage situations requiring Assertiveness. Measurement of CT raises some concerns. We should note first that, like the normative IP samples, the CT comparison groups are opportunity samples. Furthermore, for example in the case of Law/Business graduates, collection of data occurred in the context of job applications. In a competitive environment, there is an imperative to perform at the optimum level. It makes it more likely, furthermore, that these candidates had practised for the WGCTA (practice tests are available, e.g. Clifford-Chance, 2016). Where candidates prepared themselves, it is reasonable to suppose that their performance would have been enhanced. These points may help explain some of the gap in mean performance levels. Having stated this, over 200 of the completed WGCTAs contained some corrections, suggesting participants were thinking carefully about their responses. Other samples of those attending university—in these cases, unlike the normative samples from outside the UK—do not always show such wide divergence from these social workers in WGCTA forty-item scores. For example, Turkish teachers-in-training had a mean score of 18 (Sendag and Odabsi, 2009), nurses 20 (Walsh and Seldomridge, 2006)—both comparable to these social workers—and American respiratory care trainees scored 23 (Wettstein et al., 2011). Others, though, scored more highly, such as American philosophers and psychologists (25) (Burke et al., 2014) and Canadian management and nursing candidates (27) (Loom and Thorpe, 1999). However, it would be unwise to ignore these differences. In particular, the mean score of bachelor’s-level social workers (18.9) was hardly better than chance (18.5) in a suitably sized sample. In other words, had all the bachelor’s graduands guessed all the answers, we would have expected their mean score would have been practically the same. This shows room for considerable improvement. In a profession where CT capabilities are so fundamental to best practice, we might expect a mean score to be rather higher. If, indeed, these CT capabilities were embedded in the learning undertaken in the social work courses, we might expect this to have been reflected in the WGCTA scores. This raises two possible issues: Are CT capabilities all they should be in social work? And are they sufficiently a focus on social work courses? In relation to the first, there is clearly room for improvement. On the second: is there a tension in the learning process? Recent research has demonstrated convincingly that intellectual and IP domains are highly differentiated in social work (Sheppard and Charles, 2015). This exclusivity implies social work is highly demanding in terms of the expected aptitude of its practitioners—quite different talents are being expected and we may expect these differences to create a tension in the learning processes required. These tensions can be traced to fundamental distinctions in cognitive processes. We can draw on the distinction between System 1 and System 2 thinking. Recognition of affect, creativity and intuitive judgement are all aspects of System 1 thinking—the process of thinking which is fast, direct and spontaneous, and operates automatically. Much of social work emphasises these particular qualities. System 2 thinking is consciously engaged and involves deliberative thinking—it covers the capacity for reasoning which underlies CT, is effortful and intentional (Evans, 2008). Difference, of course, can reflect innate individual differences in reasoning capacity (pre-dispositions). However, while recognising the reservations about the normative samples mentioned earlier, if there is an underlying difference between, for instance, law and social work, this may reflect in part the learning process. The close relationship between performance in law and WGCTA suggests some of the law-learning processes may encapsulate reasoning processes underlying CT more consistently (TalentLens, 2011). Such processes could involve overt and close examination of argumentation, underlying assumptions and the process of drawing conclusions. In an adversarial system, such processes would be strongly highlighted in case-based work and, for example, moots. It may be that social work learning processes do not emphasise these underlying reasoning capabilities to the same extent as, for instance, law. This is understandable. Few would question the importance of intuition, creativity and awareness of affect in social work. However, these are System 1 capabilities, not System 2, which is the focus for reasoning, and it is the former in this case being encouraged rather than the latter. This process exclusivity enhances chances that one may be nurtured more than the other. None of this implies that learning processes designed for CT enhancement do not occur. It does suggest, however, that this could be undertaken more overtly and systematically. More concern may be expressed, however, at the huge diversity of CT capabilities amongst these graduands. The best scored close to 100 per cent—similar to the best of other samples, but the lowest scores raised questions about CT capabilities for practice. Of course, we might suggest that the tests were themselves abstract, and that a focus of similar reasoning processes on social work exemplars may improve scores, and this might have some legitimacy. On the other hand, the critical nature of speculation of possible alternatives in social work practice (What is likely to happen? What alternatives are there? What might be the result if we did this?) indicates that the capacity for such abstract thinking is of considerable importance. These data reflect empirically some of the concerns expressed politically about standards within social work. In particular, Frontline is an education/training development aiming at ‘boosting the image of social work and getting the very best people into the profession’ by recruiting ‘top graduates on to its programmes’ (Frontline, 2014). While, however, the data provide evidence to support the concerns, these data suggest (given very high and very low scores) a focus should also be on the performance levels of those with low CT capabilities. One-quarter of the sample scored 15 or less—substantially below the chance mean. This is a serious concern in a profession where CT is considered to be at the heart of practice. It indicates that, in addition to augmenting social work entrants with a number of high-achiever ‘leaders’, as with Frontline, attention be given to improving the CT of a significant minority of qualifying social workers whose performance is poor. Some of this is about increasing the quality of those entering social work. This is clearly a more significant issue—though not exclusively—for bachelor’s than master’s cohorts. If reliance is placed on formal educational qualifications, then their reliability at indicating CT capabilities needs to be established. Alternatively, psychometric tests additional to these qualifications—rather in the manner established by medicine, screening entrants—may be efficacious. A further radical solution might be to confine social work qualification to master’s level, although, while this would substantially improve mean CT capabilities, there would remain room for considerable improvement at the lowest levels. The second solution lies in the curriculum and perhaps practice itself. It is well beyond the scope of this article to examine in detail possible developments. We may reasonably ask: How, in principle, can CT be enhanced in learning about social work? Some common elements clearly potentially contribute to enhanced CT. Many researchers have maintained that CT is dependent on pre-dispositions (an issue of gate-keeping as above) and purposeful reflection that requires logic (Behar-Hornestein and Niu, 2011). Certainly, purposeful reflection is embedded in the social work curriculum—and is expected in subsequent qualified practice—both in university and practice-based learning, focused furthermore on social work issues and cases. It is a central feature of practice standards (Proficiency 11: ‘be able to reflect on and review practice’ (including Critical Reflection); HCPC, 2017). It is, however, the logic requirement that has been highlighted by these findings. It is not, in other words, sufficient to reflect on issues and cases, even to be imaginative and creative in considering actions, but to do so in a way placing a premium on logic. Broadly, this focuses on the rigour with which the thinking-about-practice (reasoning) processes is undertaken. This is apparent in earlier research which highlighted the importance of ‘forward thinking’ in social work, rather than the backward-focused nature of reflection (Sheppard et al., 2000). This took the form of ‘speculative hypotheses’—‘if–then’ statements, often of a highly complex nature—focusing on alternative future possibilities, their likelihood and consequences, in terms of service users’ thoughts, actions and reactions, the actions and interventions of social workers and others, the interaction and outcome of these actions and the likely possible results of all these interactions. This points to the overt and precise formulation of hypothesis-based thinking as an underlying principle of the learning and practice processes of social work, whichever particular emphasis is placed on teaching–learning processes chosen. The appropriateness of this as an underlying principle is apparent in a particular approach to the teaching of science ‘hypothesis-based learning’: Students are presented with common, simple materials …. Each student … then … begin(s) making observations. When the student finds something interesting, the student should attempt to explain the observation. Given an explanation, he or she needs to think of a way to test the explanation and make a prediction. Of course, the test is next. Then the student must analyze test results to decide if the prediction is supported or unsupported (VanDorn 2005, p. 57; see also VanDorn et al., 2017). This is, of course, a simplistic presentation in terms of social work materials or cases. It also refers to a particular approach, though it is here being used as an indicator of how an underlying principle might operate (in both the curriculum and subsequent practice). However, it is apparent how this approach could, in principle, both develop cognitive processes appropriate to the complex developmental and changing nature of cases and practice situations, and at the same time provide a framework through which the logic of CT may be developed. It is in the provision of a framework to enable the enhancement of reasoning capabilities that such an approach may be efficacious, when linked, for example, with social work-relevant exercises ‘Heart and Head’ have long been seen as the central features of practice, the combination going some way to defining what social work is. The findings here show great strength in positive emotions and great diversity in Assertiveness and Deliberation, the lower levels of which require nurturing. They also indicate serious concerns regarding CT, particularly regarding lower performers upon which an increased intensity of focus needs to be placed. References Behar-Hornestein L. , Niu L. ( 2011 ) ‘ Teaching critical thinking in higher education: A review of the literature ’, Journal of College Teaching and Learning , 8 ( 2 ) pp. 25 – 41 . Bosanquet B. ( 1901 ) ‘ The meaning of social work’, International Journal of Ethics , 11 ( 3 ), pp. 291 – 306 . Google Scholar Crossref Search ADS British Association of Social Workers (BASW) ( 2016 ) Professional Capabilities Framework, available online at https://www.basw.co.uk/pcf/capabilities/?level=7&domain=1 (accessed 5 December 2017). Brown K. , Rutter L. ( 2009 ) Critical Thinking for Social Work , London , Macmillan . Burke B. , Sears S. , Kraus S. ( 2014 ) ‘ Critical analysis: A comparison of critical thinking changes in psychology and philosophy classes’, Teaching of Psychology , 41 ( 1 ), pp. 28 – 36 . Clifford-Chance ( 2016 ) Watson Glaser Practice Tests, available online at https://www.practiceaptitudetests.com/top-employer-profiles/clifford-chance-assessments/ (accessed 12 December 2017). Coleman H. , Rogers G. , King J. ( 2002 ) ‘ Using portfolios to stimulate critical thinking in social work education ’, Social Work Education , 21 ( 2 ), pp. 583 – 95 . Google Scholar Crossref Search ADS Cooper J. ( 2017 ) ‘Social work fast-track schemes’, Community Care, available online at http://www.communitycare.co.uk/2016/02/03/social-work-fast-track-schemes-dont-know/ (accessed 5 December 2017). Costa P. , McCrae R. ( 2006 ) Revised NEO Personality Inventory Manual (UK Edition) , Oxford , HOGREFE . D’Cruz H. , Gillingham P. , Melendez S. ( 2007 ) ‘ Reflexivity: Its meaning and relevance for social work: A critical review of the literature ’, British Journal of Social Work , 37 ( 1 ), pp. 73 – 90 . Evans J. ( 2008 ) ‘ Dual process accounts of reasoning thinking and social cognition’, Annual Review of Psychology , 59 , pp. 255–78. Frontline ( 2014 ) ‘Thousands of top graduates apply to new social work programme’, available online at https://thefrontline.org.uk/2014/03/17/thousands-of-top-graduates-apply-to-new-social-work-programme/ (accessed 12 December 2017). Gillin D. , Maddux W. W. , Carpenter J. , Galinsky A. D. ( 2013 ) ‘ When to use your heart and when to use your head: The different value of perspective taking versus empathy’, Personality and Social Psychology Bulletin , 39 ( 3), pp. 3 – 16 . Hardcastle D. A. , Powers P. R. , Wenocour S. ( 2011 ) Community Practice: Theories and Skills for Social Workers , 3 rd edn, Oxford , Oxford University Press . Hardiker P. ( 1981 ) ‘ Heart or head: The function and role of knowledge in social work ’, Issues in Social Work Education , 1 ( 2 ), pp. 85 – 111 . HCPC (Health and Care Professionals Council) ( 2017 ) Standards of Proficiency: Social Workers in England , available online at http://www.hpc-uk.org/assets/documents/10003B08Standardsofproficiency-SocialworkersinEngland.pdf (accessed 12 December 2017). Helm D. ( 2011 ) ‘ Judgements or assumptions? The role of analysis in assessing children and young people's needs ’, British Journal of Social Work , 41 , 894 – 911 . Hennessy R. ( 2011 ) Relationship Skills in Social Work , London, Sage . Howe D. ( 1998 ) ‘ Relationship based thinking in social work’, Journal of Social Work Practice , 12 , pp. 45 – 56 . Google Scholar Crossref Search ADS Ingram R. ( 2013 ) ‘ Locating emotional intelligence at the heart of social work practice’, British Journal of Social Work , 43 ( 5 ), pp. 987 – 1004 . Google Scholar Crossref Search ADS Kinman G. , Grant L. ( 2011 ) ‘ Exploring stress resilience in trainee social workers: The role of emotional and social competencies’, British Journal of Social Work , 41 ( 2 ), pp. 261 – 75 . Google Scholar Crossref Search ADS Koprowska J. ( 2010 ) Communication and Interpersonal Skills in Social Work , Exeter , Learning Matters . Locke E. A. ( 2005 ) ‘ Why emotional intelligence is an invalid concept’, Journal of Organisational Behaviour , 26 , pp. 411 – 24 . Google Scholar Crossref Search ADS Loom R. , Thorpe K. ( 1999 ) ‘ A psychometric investigation of scores on the Watson Glaser Critical Thinking Appraisal new form S’, Educational and Psychological Measurement , 59 , pp. 995 – 1002 . Google Scholar Crossref Search ADS Lord W. ( 2007 ) NEO PI R: A Guide to Interpretation and Feedback in a Work Context , Oxford , HOGREFE . McCabe P. ( 1998 ) The Assertive Social Worker , Aldershot , Ashgate . Munro E. ( 1999 ) ‘ Common errors of reasoning in child protection work’, Child Abuse and Neglect , 23 ( 8 ), pp. 745 – 58 . Google Scholar Crossref Search ADS PubMed Ngai S. S. , Cheung C. ( 2009 ) ‘ Idealism, altruism career orientation and emotional exhaustion amongst social work undergraduates’, Journal of Social Work Education , 45 ( 1 ), pp. 105 – 21 . Google Scholar Crossref Search ADS Office of National Statistics ( 2016 ) Employment and Employee Types, available online at https://www.ons.gov.uk/employmentandlabourmarket/peopleinwork/employmentandemployeetypes (accessed 5 December 2017). Patel V. L. , Arocha J. F. , Zhang J. ( 2005 ) ‘Thinking and reasoning in medicine’, in Holyoak K. , Morrison R. G. (eds), Handbook of Thinking and Reasoning , Cambridge , Cambridge University Press , pp. 298 – 358 . Radley M. , Figley C. R. ( 2007 ) ‘ The social psychology of compassion’, Clinical Social Work Journal , 35 ( 3 ), pp. 207 – 14 . Google Scholar Crossref Search ADS Ruch G. , Turney D. , Ward A. ( 2010 ) Relationship Based Social Work , London , Kingsley . Scurlock-Evans L. , Upton D. ( 2015 ) ‘ The role and rating of evidence: A systematic review of social workers’ Evidence Based Practice orientation, attitudes and implementation ’, Journal of Evidence-Informed Social Work , 12 ( 4 ), available online at http://www.tandfonline.com/doi/full/10.1080/15433714.2013.853014 (accessed 5 December 2017). Sendag S. , Odabsi H. ( 2009 ) ‘ Effect of an online problem based course on content knowledge acquisition and critical thinking skills’, Computers and Education , 53 , pp. 132 – 41 . Sheppard M. ( 1995a ) Care Management and the New Social Work , London , Whiting and Birch . Sheppard M. ( 1995b ) ‘ Social work social science and practice wisdom ’, British Journal of Social Work , 25 , pp. 65 – 93 . Sheppard M. , Charles M. ( 2014 ) ‘ Critical thinking and interpersonal dispositions in those commencing social work training’, British Journal of Social Work , 44 , pp. 2057 – 66 . Google Scholar Crossref Search ADS Sheppard M. , Charles M. ( 2015 ) ‘ Head and Heart: An examination of the relationship between the intellectual and interpersonal in social work’, British Journal of Social Work , 45 , pp. 1837 – 54 . Google Scholar Crossref Search ADS Sheppard M. , Charles M. ( 2017 ) ‘ A longitudinal comparative study of the impact of the experience of social work education on interpersonal and critical thinking capabilities’, Social Work Education , 36 , pp. 745 – 57 . Google Scholar Crossref Search ADS Sheppard M. , Newstead S. , DiCaccavo A. , Ryan K. ( 2000 ) ‘ Reflexivity and the development of process knowledge in social work’, British Journal of Social Work , 30 , pp. 465 – 88 . Google Scholar Crossref Search ADS Sheppard M. , Newstead S. , DiCaccavo A. , Ryan K. ( 2001 ) ‘ Comparative hypothesis assessment and quasi triangulation as process knowledge assessment strategies in social work practice’, British Journal of Social Work , 31 , pp. 863 – 85 . Google Scholar Crossref Search ADS Schulte M. , Ree M. , Carretta T. ( 2004 ) ‘ Emotional intelligence: Not much more than g and personality’, Personality and Individual Difference , 37 ( 5 ), pp. 1059 – 68 . Google Scholar Crossref Search ADS Smith L. , Ciarrochi J. , Heaven P. C. L. ( 2008 ) ‘ The stability and change of trait emotional intelligence, conflict communication patterns, and relationship satisfaction: A one-year longitudinal study’, Personality and Individual Differences , 45 , pp. 738 – 43 . Google Scholar Crossref Search ADS TalentLens ( 2011 ) Watson Glaser Supervised UK Edition , Appendix E, Business and Law, Pearson . Taylor C. , White S. ( 2006 ) ‘ Knowledge and reasoning in social work: Educating for humane judgement’, British Journal of Social Work , 36 ( 6 ), pp. 937 – 54 . Google Scholar Crossref Search ADS VanDorn K. ( 2005 ) ‘ Hypothesis-based learning ’, Science Scope , 29 ( 3 ), p. 57 . VanDorn K. , Montes L. , Ackerson B. , Rockley M. ( 2017 ) ‘Hypothesis based learning’, NTSA News Digest, available online at http://www.nsta.org/publications/news/story.aspx?id=48919 (accessed 6 December 2017). Walsh C. , Seldomridge L. ( 2006 ) ‘ Measuring critical thinking’, Nurse Educator , 31 ( 4 ), pp. 159 – 62 . Google Scholar Crossref Search ADS PubMed Wettstein R. B. , Wilkins R. L. , Gardner D. D. , Restrepo R. D. ( 2011 ) ‘ Critical-thinking ability in respiratory care students and its correlation with age, educational background, and performance on national board examinations’, Respiratory Care , 56 , 284 – 9 . Google Scholar Crossref Search ADS PubMed Whittaker A. ( 2011 ) ‘ Social defences and organisational culture in a local authority child protection setting: challenges for the Munro review? ’, Journal of Social Work Practice , 25 , 481 – 95 . 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/open_access/funder_policies/chorus/standard_publication_model) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The British Journal of Social Work Oxford University Press

Inter-Personal and Critical-Thinking Capabilities in Those about to Enter Qualified Social Work: A Six-Centre Study

Loading next page...
 
/lp/ou_press/inter-personal-and-critical-thinking-capabilities-in-those-about-to-oZY2Exd3eJ
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of The British Association of Social Workers. All rights reserved.
ISSN
0045-3102
eISSN
1468-263X
D.O.I.
10.1093/bjsw/bcx143
Publisher site
See Article on Publisher Site

Abstract

Abstract The ‘process’ of intervention is understood to be fundamental to social work—evident in, for example, the literature on reflexivity. Little work, though, has focused on the detailed excavation of the cognitive processes of reasoning in decision making. This is widely recognised as requiring considerable analytic and critical abilities. Although this is long established, its importance is contemporarily apparent at the policy level from the rationale underlying current initiatives such as Frontline. However, it is also long understood that the reasoning capabilities underlying these processes cannot be considered in isolation from the inter-personal–emotional, encapsulated in a long-term theoretical concern for both Heart and Head. Furthermore, terms like ‘capability’ or ‘proficiency’ in professional qualification imply some standard to be reached in practice. This invites measurement. This novel study seeks to bring together three dimensions of the (i) measurement of (ii) the inter-personal–emotional and (iii) critical thinking—measurement of key facets of both Head and Heart. A six-centre, six-university collaboration, it focuses on those at a crucial point: where individuals are about to enter qualified practice. The findings show they score highly on most inter-personal measures (with room for improvement), but show huge variability in critical-thinking capabilities. The implications of this are discussed. Critical thinking, inter-personal skills, expertise Introduction How social workers think—the cognitive processes of decision making—is an issue at least as important as what they should know. The latter is the subject of the much more prominent focus on evidence-based/informed practice (Scurlock-Evans and Upton, 2015). How practitioners make sense, though, is an issue of meaning, but that ‘making of meaning’ is also a cognitive-reasoning process (thinking about something logically): What are the key facets of reasoning involved? What are the mental processes (rather than, for example, situational characteristics) by which they decide a course of action? How can they ensure the maximum rigour when conducting practice? How can error be minimised? and so on. These are all—from a cognitive perspective—issues of reasoning rather than knowledge. This focus on process leads us towards the long-established (theoretically) fundamental element of social work of ‘Heart’ (the emotional-inter-personal) and ‘Head’ (reasoning/intellectual). By way of introduction, we shall focus on these two key issues—which underlie judgement and decision making in practice—alongside that of their implicit and explicit measurement, which have accompanied much recent discourse on policy development in social work education and training. Their fundamental theoretical status, centrality in establishing competence or capability and key position in current and perennial policy issues relating to social work make them a critical focus for our attention. An absence of focused, measured research on this matter makes this focus still more urgent. This focus on reasoning processes is at the heart of judgement and decision making. However, compared with, for example, medicine (Patel et al., 2005), there has been little focus on these reasoning processes in social work research. This does not mean that process as a whole has not been of interest in social work—reflexivity, with its myriad meanings, multiple theoretical positions and, at times, combination of situational factors with process, in part does (D’Cruz et al., 2007). But the contribution—specifically—of process reasoning has been little examined. Where it has, this has primarily been through theoretical considerations or critical analysis rather than research-based (Munro, 1999; Taylor and White, 2006). Equally, the issue of measurement is at the heart of practice—not only in the sense of, for instance, evaluating outcomes or levels of need, but in the quality of practice performed. Measurement is, however, implicit, hidden by its qualitative nature. If we look at the most obvious measures of quality—those of capabilities (BASW, 2016) or proficiencies (HCPC, 2017)—we see standards to be achieved, expressed linguistically. Those qualifying should be competent, capable or proficient (depending on which adjective is in vogue). A myriad of expectations are encapsulated within standards for qualification. Each of these involves reaching a certain level. That level, however, is one qualitatively measured—a matter of judgement and left to a wide range of practice educators to decide. Who is to say that practice educators in Carlisle make judgements on the same grounds as those in Camborne, or those in Berwick the same as Bromley? However, although linguistic, they are as much measures, albeit implicit, as would be the case if they were numerical. As a caveat, we might expect greater exactitude—and potentially consistency—with (rigorously developed) numerical measures. Clearly, we are measuring practice quality when we use criteria, even linguistic, to distinguish that which is good enough from that which is not. This reasoning process, or level of criticality, and implicit measurement are, furthermore, of central policy interest. This is reflected most obviously in initiatives like Frontline (2014). This emerged with the explicit purpose of focusing on improving the quality of social work force, particularly its leadership, through encouraging high-calibre entrants to social work. In an early bulletin, it advertised the huge increase in Oxbridge and Russell Group graduates applying for Frontline, when compared to conventional social work qualifying routes, reflecting, it stated, the need of social work for ‘attracting the best and brightest’ and calling social work ‘a highly skilled leadership profession’ (Frontline, 2014) . It is fast-track, carried out separately from mainstream social work education, trains a minority of starting professionals and costs considerably more. It is the subject of some controversy, some criticising it as misdirected and politically motivated (Cooper, 2017). Both the process of practical reasoning underlying judgement and decision making and its implicit measurement are key to social work, yet subject to limited research. These, though, cannot be considered without reference to the inter-personal-emotional. Their necessarily interwoven nature is encapsulated in the long-standing notion of social work being a matter of ‘Heart and Head’ (Bosanquet, 1901; Hardiker, 1981). These may be considered foundation elements underlying much, if not all, aspects of social work (certainly that related to work with service users). Their status as key foundations for social work is evident from the separate literatures devoted to the intellectual–critical thinking and the inter-personal relationships (Coleman et al., 2002; Brown and Rutter, 2009; Hennessy, 2011; Ruch et al., 2010) and indeed the proficiencies required of qualifying practitioners (HCPC, 2017). Their centrality in social work is widely recognised. Coleman et al. (2002, p. 583) comment—regarding critical thinking—on its practical importance: ‘Critical thinking is an important ability for social workers to have.’ Brown and Rutter (2009, p. 29) indicate why: ‘… the skills of critical thinking allow the best quality decisions or actions for the situations … [social workers] encounter.’ Likewise, on relationships, Howe (1998, p. 45) comments ‘that relationship skills … are necessary in … social work, … not an add-on luxury …. Rather, they are integral and essential to the theory and practices of the occupation’. The features, in other words, are necessary conditions for social work to be taking place at all. Decision making in social work, then, is a process: developmental, with Head and Heart as foundation elements. Some social work research has addressed decision making but not generally in the tradition of cognitive processes analysis (Helm, 2011; Whittaker, 2011). One approach has focused on the examination and explication of cognitive processes (the processes involved in thinking and reasoning). Sheppard’s (1995a, 1995b) theoretical work on ‘Emergent Understanding’ drew on three foci: construction of meaning, process of reasoning and accuracy of judgement (Sheppard, 1995a, 1995b). Subsequently, empirical research conceptualised a range of reasoning processes undertaken. However, the data suggested considerable variation between social workers in precision and capability in this reasoning process (Sheppard et al., 2000, 2001). This tantalising small-scale study was undertaken, however, without formal measurement or overt reference to the ‘Heart’ element of Head and Heart. That Head and Heart, however, have an uneasy, even contradictory, relationship has become evident in a more recent study (Sheppard and Charles, 2015). This demonstrated the intellectual and inter-personal remain highly distinct domains in social work—capabilities in the one at times actively contrary to those in the other. It provided a welcome demonstration of the exceptional capabilities required by social workers—there is evidence in literature beyond social work that proficiencies tend to be in one or the other (cognitive or empathetic) rather than both (Gillin et al., 2013). However, it also indicated evidence of the fundamental demands of social work, before the complexity of cases is even considered. Their fundamental place within social work; the urgency bestowed by current policy and practice developments in social work; their key role in judgement and decision making; and the growing importance of precision in determining performance levels suggest an urgent need for overt measurement of the intellectual–critical thinking and emotional–inter-personal (Head and Heart) in tandem. This research seeks for the first time to undertake a substantial study measuring these two key dimensions jointly at a crucial point by carrying out a six-centre study involving twelve cohorts on the verge of entering qualifying practice. It has the following aims: to identify inter-personal and critical-thinking capabilities in social work graduands; to contextualise these measures through comparison with relevant normative samples; to examine variations in these capabilities according to university (pre/post-92) status and degree (master’s/bachelor’s level). Methods This study was of a cross-sectional survey design. It was a six-university collaboration involving twelve cohorts equally divided between bachelor’s and master’s programmes from geographically diverse universities in England and Wales. These institutions constituted just under one-fifth of the thirty-four universities providing both master’s and bachelor’s social work provision, enabling a comparison by degree taken. The sample was broadly stratified to reflect the fifty/fifty division between pre- and post-92 universities, with three former and three latter institutions. Pre-92 universities are older establishments, while post-92 universities have been established after 1992. The former generally have higher status, applications and entrance standards. A pilot study was undertaken prior to the main research, to ensure both the viability of the study and its efficacy (Sheppard and Charles, 2014). Instruments Two measures were used: the NEOPI-R to measure inter-personal (IP) and the Watson-Glaser Critical Thinking Appraisal (WGCTA) to measure critical-thinking (CT) capabilities. These are the longest established, most widely used and comprehensively tested instruments in their fields, with high levels of reliability and validity, confirmed internationally (Costa and McCrae, 2006; Loom and Thorpe, 1999). They also both possess normative sample data in relation to which our sample could be compared, enabling findings to be placed in context. The NEOPI-R examines the five domains of personality (neuroticism, extraversion, openness, agreeableness and conscientiousness). These domains contain thirty facets, each of which is sufficiently robust to be examined in its own right (Lord, 2007). Our focus was on six facets which have been identified to be of enduring importance to social work. These were (with references to relevant exemplar social work literature in brackets): Altruism (Ngai and Cheung, 2009); Warmth (Koprowska, 2010); Compassion/tender mindedness (Radley and Figley, 2007; Kinman and Grant, 2011); Feeling Insight (personal receptivity) (Kinman and Grant, 2011); Deliberation (Koprowska, 2010); Assertiveness (McCabe, 1998; Hardcastle et al., 2011, Chapter 8). These are areas often associated with emotional intelligence. However, this has been subject to extensive criticism in psychology for: the misapplication of the term ‘intelligence’, failing to differentiate between personality traits and emotional states attitudes and values, and confusion with established personality traits for which there is considerable evidence without recourse to emotional intelligence as a term (Ingram, 2013; Locke, 2005; Schulte et al., 2004; Smith et al., 2008) and may best be understood as a metaphor. Rather than ‘throw the baby out with the bathwater’, we identified key IP areas long established in social work, employing the extensively validated and reliable NEOPI-R for the purpose of measurement. Although we focused on six facets, it was necessary, nevertheless, for the NEO PI-R to be completed as a whole (thence to extract data from the relevant facets). Hence, participants responded to the 240 items from which data from the five domains and thirty constituent facets (each of which contained eight items) may be extracted. The WGCTA measures CT, ‘defined as the ability to identify and analyse problems as well as seek and evaluate relevant information in order to reach appropriate conclusions’ (TalentLens, 2011, p. 3). Although general, as a description of key elements of reasoning and thinking for social work, this can hardly be bettered. However, WGCTA nevertheless focuses on generic CT capabilities, enabling contextualisation of findings/scores. The WGCTA has three domains, measuring reasoning in Inference, Recognising Assumptions and Drawing Conclusions (containing Deduction, Interpretation and Evaluating Arguments) (TalentLens, 2011). Inference involves ‘rating the probability of truth of inferences [deducing or reasoning something from evidence or reasoning rather than explicit statements]… based on given information’; Recognising Assumptions is ‘identifying unstated assumptions or propositions underlying given statements’; Drawing Conclusions encapsulated three facets—Deduction: ‘Determining whether conclusions follow logically from given information’; Interpretation: ‘weighing evidence and deciding if generalisations or conclusions based on data are warranted’; and Evaluating Arguments: ‘evaluating the strength of arguments with respect to a particular question or issue’ (TalentLens, 2011, pp. 3–4). It contains forty items, multiple-choice questions, each of which has a correct answer and overall, therefore, yields a maximum score of 40 (and minimum of 0). Measurement and scores for each domain of CT may also be extrapolated. Ethics and data collection The data were collected during the final term before qualification after, or close to, the end of final placement. Data collection was strategically placed at that point because of its crucial stage in professionalisation—immediately prior to the point of qualification, enabling us to have a ‘window’ into standards close to the point of starting qualified practice—and because this approach enabled us to capture a diverse and substantial sample. The study initially ethically approved by the lead institution received multi-site approval. Before undertaking these tasks, a number of facets were explained to the participants both verbally and in writing, including the nature and purpose of the study, freedom to choose whether or not to participate, confidentiality and the right to withdraw. They were given the opportunity to ask questions (generally taken up) which were answered. Written consent was provided by all participants. Only once this process was complete were the tasks undertaken. A pilot study focused on its efficacy and ethical features, including in particular the need for consent confidentiality and right to withdraw following initial consent (Sheppard and Charles, 2014). The instruments were themselves explained to participants, although clear instructions are also included in the instruments themselves. Participants completed the instruments in groups. They generally each take less than thirty minutes to complete, but participants were given as long as required. Each instrument was completed separately in order to prevent mental fatigue. A member of the research team was available throughout and all finished instruments were checked to ensure they were fully completed. Analysis The instruments produce interval data for both individual items and domains (facets in the case of NEOPI-R). Tests of significance employed, depending on data, the t-test or Analysis of Variance. However, with larger samples, significance (the likelihood that findings are a matter of chance or represent some underlying systematic difference) can be achieved where the scale of that difference is relatively trivial (i.e. the difference may indeed be systematic, but it is not very great). In these circumstances, additional calculations, including where appropriate, effect size (Cohen’s d) has been used. The coefficient of variation was used to examine the degree of homogeneity amongst social work graduands in the CT and IP dimensions examined. Comparison was undertaken according to university status (pre-/post-92), degree level (undergraduate/postgraduate); between different facets/dimensions; and against normative samples. Results The total cohort was 407, of whom 301 (74 per cent) participated—a highly satisfactory response rate. Eighty-six per cent were women; two-thirds (67 per cent) were white British, 23 per cent were black or mixed race (including black British) African or Caribbean; 3 per cent were Indian Pakistan or other Asian; the rest classified themselves as ‘other white’ (including white Irish) or preferred not to say. Mean age was thirty-one (minimum twenty, maximum fifty-nine). IP and CT measures Table 1 focuses on the graduands’ IP scores, in the context of the UK normative sample (Costa and McCrae, 2006). The UK Normative Group comprised 1,301 people: 795 men and 353 women (153 unknown) mean age forty-three. The authors sought to sample as broadly as possible with respect to industry sector, job title and seniority (Costa and McCrae, 2006). It is clear from its make-up that it was an opportunity sample and the implications of this will be made clear later when discussing the results. Table 1 Inter-personal (IP) scores compared with UK normative sample Social work (n=301) UK normative (n=1,301) % mean score difference (Social work versus UK) Mean (sd) min Max Coefficient of variation Mean (sd) Altruism 25.0 (3.4) 14 32 13.6 24.0 (3.4) 3.2 Warmth 24.2 (3.9) 11 32 16.0 23.9 (3.9) 1.0 Compassion 23.3 (3.2) 14 31 13.6 19.7 (3.6) 12.1 Feelings Insight 22.9 (4.1) 11 32 17.9 21.9 (4.2) 3.4 Deliberation 18.3 (4.7) 5 30 26.0 18.9 (4.4) −2.1 Assertiveness 16.6 (4.7) 3 32 28.2 21.0 (4.4) −14.5 Max score 32 32 Mean IP CoV 19.1 95% conf. interval t Df Mean diff. SED Lower Upper p d Altruism 4.6 1,600 1.0 0.22 0.57 1.43 <0.001 0.29* Compassion 16.0 1,600 3.6 0.23 3.16 4.04 <0.001 1.06** Insight 3.74 1,600 1.0 0.27 0.48 1.53 <0.001 0.24* Deliberation 2.10 1,600 0.6 0.29 0.04 1.16 <0.05 0.13 Assertiveness 15.43 1,600 –4.4 0.29 –4.96 –3.84 <0.001 0.97** Social work (n=301) UK normative (n=1,301) % mean score difference (Social work versus UK) Mean (sd) min Max Coefficient of variation Mean (sd) Altruism 25.0 (3.4) 14 32 13.6 24.0 (3.4) 3.2 Warmth 24.2 (3.9) 11 32 16.0 23.9 (3.9) 1.0 Compassion 23.3 (3.2) 14 31 13.6 19.7 (3.6) 12.1 Feelings Insight 22.9 (4.1) 11 32 17.9 21.9 (4.2) 3.4 Deliberation 18.3 (4.7) 5 30 26.0 18.9 (4.4) −2.1 Assertiveness 16.6 (4.7) 3 32 28.2 21.0 (4.4) −14.5 Max score 32 32 Mean IP CoV 19.1 95% conf. interval t Df Mean diff. SED Lower Upper p d Altruism 4.6 1,600 1.0 0.22 0.57 1.43 <0.001 0.29* Compassion 16.0 1,600 3.6 0.23 3.16 4.04 <0.001 1.06** Insight 3.74 1,600 1.0 0.27 0.48 1.53 <0.001 0.24* Deliberation 2.10 1,600 0.6 0.29 0.04 1.16 <0.05 0.13 Assertiveness 15.43 1,600 –4.4 0.29 –4.96 –3.84 <0.001 0.97** * Small effect size. **Large effect size. Source for normative sample: Costa and McCrae (2006). t-test, equal variance not assumed. Table 1 Inter-personal (IP) scores compared with UK normative sample Social work (n=301) UK normative (n=1,301) % mean score difference (Social work versus UK) Mean (sd) min Max Coefficient of variation Mean (sd) Altruism 25.0 (3.4) 14 32 13.6 24.0 (3.4) 3.2 Warmth 24.2 (3.9) 11 32 16.0 23.9 (3.9) 1.0 Compassion 23.3 (3.2) 14 31 13.6 19.7 (3.6) 12.1 Feelings Insight 22.9 (4.1) 11 32 17.9 21.9 (4.2) 3.4 Deliberation 18.3 (4.7) 5 30 26.0 18.9 (4.4) −2.1 Assertiveness 16.6 (4.7) 3 32 28.2 21.0 (4.4) −14.5 Max score 32 32 Mean IP CoV 19.1 95% conf. interval t Df Mean diff. SED Lower Upper p d Altruism 4.6 1,600 1.0 0.22 0.57 1.43 <0.001 0.29* Compassion 16.0 1,600 3.6 0.23 3.16 4.04 <0.001 1.06** Insight 3.74 1,600 1.0 0.27 0.48 1.53 <0.001 0.24* Deliberation 2.10 1,600 0.6 0.29 0.04 1.16 <0.05 0.13 Assertiveness 15.43 1,600 –4.4 0.29 –4.96 –3.84 <0.001 0.97** Social work (n=301) UK normative (n=1,301) % mean score difference (Social work versus UK) Mean (sd) min Max Coefficient of variation Mean (sd) Altruism 25.0 (3.4) 14 32 13.6 24.0 (3.4) 3.2 Warmth 24.2 (3.9) 11 32 16.0 23.9 (3.9) 1.0 Compassion 23.3 (3.2) 14 31 13.6 19.7 (3.6) 12.1 Feelings Insight 22.9 (4.1) 11 32 17.9 21.9 (4.2) 3.4 Deliberation 18.3 (4.7) 5 30 26.0 18.9 (4.4) −2.1 Assertiveness 16.6 (4.7) 3 32 28.2 21.0 (4.4) −14.5 Max score 32 32 Mean IP CoV 19.1 95% conf. interval t Df Mean diff. SED Lower Upper p d Altruism 4.6 1,600 1.0 0.22 0.57 1.43 <0.001 0.29* Compassion 16.0 1,600 3.6 0.23 3.16 4.04 <0.001 1.06** Insight 3.74 1,600 1.0 0.27 0.48 1.53 <0.001 0.24* Deliberation 2.10 1,600 0.6 0.29 0.04 1.16 <0.05 0.13 Assertiveness 15.43 1,600 –4.4 0.29 –4.96 –3.84 <0.001 0.97** * Small effect size. **Large effect size. Source for normative sample: Costa and McCrae (2006). t-test, equal variance not assumed. Table 1 shows, relative to the normative sample, that social work graduands scored particularly highly on Compassion, and higher in relation to Feelings Insight and Altruism, in all of which p < 0.001 (t-test). More significant is the scale of difference (answering the question: how much difference was there between two samples?). This is evident from calculating Cohen’s d, where a large effect size ≥ 0.8; a medium ≥0.5; and small≥ 0.2. This yielded a large effect size to Compassion and a small effect size to Feelings Insight and Altruism. They scored considerably lower, however, on Assertiveness. This was also significant and the effect size was large. The coefficient of variation, which measures the degree of variability around the mean, shows a much greater degree of homogeneity furthermore, amongst the ‘positive affect’ measures than those of Deliberation and Assertiveness. These graduands, in other words, were, relatively, much more alike on those former measures than latter measures where some were far more liable to be assertive and to deliberate than others. Table 2 focuses on CT scores. The (UK) General Population comparison group was chosen because of its general application; Law/Business because of the professional/occupational nature of their qualification and because they were applicants for public-sector jobs—characteristics largely shared with the social work sample. In both cases, furthermore, the focus was on UK candidates (TalentLens, 2011). However, we should again note that these were opportunity samples (of job applicants taking these tests) and there is little information on the social and demographic make-up of these groups (e.g. we are not given gender make-up of the Law/Business group, though we do of the General Population, in that case only one-third being female). Again, the implications of this will be discussed later. Table 2 Critical-thinking scores in context Between-group differences (Social Work, Law and public managers) Social Work graduands (n=301) General Population (fifty occupations) (n=1,546) Public-sector Law/Business job applicants (n=504) % mean score difference (GP) % mean score difference (Law) Mean (sd) Mean (sd) Mean (sd) WCCTT mean 20.7 (6.3) 26.5 (6.0) 27.0 (6.0) –14.5 –15.8 Max. score 40 40 40 Within-group (Social Work) differences Mean % Standard deviation Coefficient of variation Min. Max. Recognise Assumptions 43.3 25.4 58.7 0 100 Evaluate Arguments 53.7 17.8 33.2 8 100 Draw Conclusions 56.4 18.0 32.0 19 100 Total 51.7 15.7 30.4 20 95 Between-group differences (Social Work, Law and public managers) Social Work graduands (n=301) General Population (fifty occupations) (n=1,546) Public-sector Law/Business job applicants (n=504) % mean score difference (GP) % mean score difference (Law) Mean (sd) Mean (sd) Mean (sd) WCCTT mean 20.7 (6.3) 26.5 (6.0) 27.0 (6.0) –14.5 –15.8 Max. score 40 40 40 Within-group (Social Work) differences Mean % Standard deviation Coefficient of variation Min. Max. Recognise Assumptions 43.3 25.4 58.7 0 100 Evaluate Arguments 53.7 17.8 33.2 8 100 Draw Conclusions 56.4 18.0 32.0 19 100 Total 51.7 15.7 30.4 20 95 Source for Law/Business normative sample: TalentLens (2011). Table 2 Critical-thinking scores in context Between-group differences (Social Work, Law and public managers) Social Work graduands (n=301) General Population (fifty occupations) (n=1,546) Public-sector Law/Business job applicants (n=504) % mean score difference (GP) % mean score difference (Law) Mean (sd) Mean (sd) Mean (sd) WCCTT mean 20.7 (6.3) 26.5 (6.0) 27.0 (6.0) –14.5 –15.8 Max. score 40 40 40 Within-group (Social Work) differences Mean % Standard deviation Coefficient of variation Min. Max. Recognise Assumptions 43.3 25.4 58.7 0 100 Evaluate Arguments 53.7 17.8 33.2 8 100 Draw Conclusions 56.4 18.0 32.0 19 100 Total 51.7 15.7 30.4 20 95 Between-group differences (Social Work, Law and public managers) Social Work graduands (n=301) General Population (fifty occupations) (n=1,546) Public-sector Law/Business job applicants (n=504) % mean score difference (GP) % mean score difference (Law) Mean (sd) Mean (sd) Mean (sd) WCCTT mean 20.7 (6.3) 26.5 (6.0) 27.0 (6.0) –14.5 –15.8 Max. score 40 40 40 Within-group (Social Work) differences Mean % Standard deviation Coefficient of variation Min. Max. Recognise Assumptions 43.3 25.4 58.7 0 100 Evaluate Arguments 53.7 17.8 33.2 8 100 Draw Conclusions 56.4 18.0 32.0 19 100 Total 51.7 15.7 30.4 20 95 Source for Law/Business normative sample: TalentLens (2011). Between-group effects (Social Work compared to (a) General Population (b) Law/Business) 95% conf. interval t Df Mean diff. SED Lower Upper p d General Population 15.22 1,845 –5.8 0.38 –6.6 –5.1 <0.001 1.02** Law/Business 14.15 803 –6.3 0.45 –7.2 –5.4 <0.001 0.94** Between-group effects (Social Work compared to (a) General Population (b) Law/Business) 95% conf. interval t Df Mean diff. SED Lower Upper p d General Population 15.22 1,845 –5.8 0.38 –6.6 –5.1 <0.001 1.02** Law/Business 14.15 803 –6.3 0.45 –7.2 –5.4 <0.001 0.94** ** Large effect size. Between-group effects (Social Work compared to (a) General Population (b) Law/Business) 95% conf. interval t Df Mean diff. SED Lower Upper p d General Population 15.22 1,845 –5.8 0.38 –6.6 –5.1 <0.001 1.02** Law/Business 14.15 803 –6.3 0.45 –7.2 –5.4 <0.001 0.94** Between-group effects (Social Work compared to (a) General Population (b) Law/Business) 95% conf. interval t Df Mean diff. SED Lower Upper p d General Population 15.22 1,845 –5.8 0.38 –6.6 –5.1 <0.001 1.02** Law/Business 14.15 803 –6.3 0.45 –7.2 –5.4 <0.001 0.94** ** Large effect size. Greenhouse-Geisser test of within-subject effects Type 111 sum of squares Df Mean square F Sig. Partial eta squared Critical thinking 28,913 1.8 16,292 54.51 <0.001 0.15 Greenhouse-Geisser test of within-subject effects Type 111 sum of squares Df Mean square F Sig. Partial eta squared Critical thinking 28,913 1.8 16,292 54.51 <0.001 0.15 Greenhouse-Geisser test of within-subject effects Type 111 sum of squares Df Mean square F Sig. Partial eta squared Critical thinking 28,913 1.8 16,292 54.51 <0.001 0.15 Greenhouse-Geisser test of within-subject effects Type 111 sum of squares Df Mean square F Sig. Partial eta squared Critical thinking 28,913 1.8 16,292 54.51 <0.001 0.15 Table 2 shows these social work cohorts scored markedly lower in CT than either a General Population group of multiple occupations or public-sector job applicants with Law or Business degrees. The mean percentage score for these social workers was 50.7 per cent compared with 65.2 per cent for the General Population group and 66.5 per cent for the Law/Business group. In both cases, the differences (between social work graduands and the normative samples) was significant. The effect size, furthermore, was large in both cases. This is also clear from the difference in mean scores shown in Table 2. The ‘within-group’ comparison (of performance at different areas of CT) shows scores as a percentage of potential maximum. This approach was taken because there are different maximum scores for different facets of CT. The first noticeable feature is the considerable difference between the highest and lowest scores—individuals were qualifying with utterly different levels of CT capabilities. This diversity in CT is reflected in its coefficient of variation. If we compare this, for example, with the positive affect measures of Table 1, we find they were a far more homogeneous group in positive affect than they were in CT. Table 2 also compares the three facets of CT, showing the capacity to recognise assumptions was significantly lower than the capacity to evaluate arguments or draw conclusions. As interesting, however, was the relative diversity of capabilities in Recognising Assumptions in this sample, compared with Evaluating Arguments and Drawing Conclusions. Comparing sectors and levels: pre- and post-92 and master’s and bachelor’s levels Table 3 shows variations in CT according to university ‘sector’ (pre- and post-1992 universities) and degree level (bachelor’s and master’s). The pre-92 universities did not have graduands with higher CT capabilities. Although the post-92 universities (ex-polytechnics) had a mean score lower than pre-92 universities, this difference was not significant. However, there was a highly significant difference between mean scores for master’s-level compared with bachelor’s-level cohorts. Marked as a percentage out of forty (maximum), the master’s cohort’s mean score was 10 per cent higher than the bachelor’s group. There was, in Cohen’s terms, a ‘medium’ effect size (on its upper reaches) where d = 0.66. Table 3 Variations in critical thinking and Compassion according to degree level and university sector Number Mean Standard deviation Critical-thinking scores Master’s 132 22.9 6.9 Bachelor’s 169 18.9 5.1 Pre-92 121 21.1 6.8 Post-92 180 20.4 5.9 Compassion/tender Mindedness scores Pre-92 121 22.8 3.0 Post-92 180 23.7 3.3 Number Mean Standard deviation Critical-thinking scores Master’s 132 22.9 6.9 Bachelor’s 169 18.9 5.1 Pre-92 121 21.1 6.8 Post-92 180 20.4 5.9 Compassion/tender Mindedness scores Pre-92 121 22.8 3.0 Post-92 180 23.7 3.3 Table 3 Variations in critical thinking and Compassion according to degree level and university sector Number Mean Standard deviation Critical-thinking scores Master’s 132 22.9 6.9 Bachelor’s 169 18.9 5.1 Pre-92 121 21.1 6.8 Post-92 180 20.4 5.9 Compassion/tender Mindedness scores Pre-92 121 22.8 3.0 Post-92 180 23.7 3.3 Number Mean Standard deviation Critical-thinking scores Master’s 132 22.9 6.9 Bachelor’s 169 18.9 5.1 Pre-92 121 21.1 6.8 Post-92 180 20.4 5.9 Compassion/tender Mindedness scores Pre-92 121 22.8 3.0 Post-92 180 23.7 3.3 95% conf. interval t Df SED Lower Upper p d Master’s/bachelor’s 5.6 244.6 0.69 2.6 5.4 <0.001 0.66* Pre-/Post-92 Not significant Equal variance not assumed for critical thinking. 95% conf. interval t Df SED Lower Upper p d Master’s/bachelor’s 5.6 244.6 0.69 2.6 5.4 <0.001 0.66* Pre-/Post-92 Not significant Equal variance not assumed for critical thinking. * Medium effect size. 95% conf. interval t Df SED Lower Upper p d Master’s/bachelor’s 5.6 244.6 0.69 2.6 5.4 <0.001 0.66* Pre-/Post-92 Not significant Equal variance not assumed for critical thinking. 95% conf. interval t Df SED Lower Upper p d Master’s/bachelor’s 5.6 244.6 0.69 2.6 5.4 <0.001 0.66* Pre-/Post-92 Not significant Equal variance not assumed for critical thinking. * Medium effect size. 95% conf. interval t Df SED Lower Upper p Master’s/bachelor’s –2.3 299 0.37 –1.6 –1.3 <0.05 Equal variance may be assumed for Compassion 95% conf. interval t Df SED Lower Upper p Master’s/bachelor’s –2.3 299 0.37 –1.6 –1.3 <0.05 Equal variance may be assumed for Compassion 95% conf. interval t Df SED Lower Upper p Master’s/bachelor’s –2.3 299 0.37 –1.6 –1.3 <0.05 Equal variance may be assumed for Compassion 95% conf. interval t Df SED Lower Upper p Master’s/bachelor’s –2.3 299 0.37 –1.6 –1.3 <0.05 Equal variance may be assumed for Compassion The IP scores showed no significant differences, either between pre-/post-92 universities or between master’s and bachelor’s cohorts except, in relation to the former division, with Compassion. However, we should remember this was one of only twelve measures (six IP for each of the pre-/post-92 and master’s/bachelor’s divisions)—fourteen if we include CT measures. Where significance is set at p < 0.05 (a one-in-twenty likelihood the result will be a matter of chance), finding one significant difference may not be too surprising. The main impression inter-personally is of homogeneity. Discussion We should first note the limits to this study. Although a proportionately significant sample (of England and Wales as a whole), this was nevertheless only one year of graduands. They were, furthermore, a sample at one particular point in their career—becoming qualified practitioners. We do not know, for example, that a study would generate similar findings if, for instance, it concentrated on more experienced practitioners. It is cross-sectional—it does not provide a sense of the possibility for growth and change as might happen where a focus were longitudinally on, for example, students through the course of their study, social workers over a period of growing experience or post-qualifying study. It is focused on locations in England and Wales, rather than the UK as a whole. Nevertheless, there is much to recommend it. In an undeveloped field—the forensic examination of cognitive processes in judgement and decision making—the issue of whether key elements of these processes can be measured is very new indeed. Furthermore, where there is a major concern about standards, the issue of measurement becomes of acute importance. The sample is substantial and covers diverse locations within the UK. It is also structured in terms of key features liable to influence and make representative—the findings: the division between ‘old’ (pre-92) and ‘new’ (post 92) universities and between the two major social work levels of qualification—master’s and bachelor’s. Its focus, moreover, on social workers at a common point in career development lends coherence and focus to the study. That moment furthermore—at the point of qualification—is crucial. Finally, it complements its ‘sister’ three-year longitudinal study of the impact on these factors of social work education (Sheppard and Charles, 2017). Although our focus is on general implications of these findings for social work generally and education in particular, they are underpinned by anti-discriminatory assumptions/principles in two respects: we should be seeking to promote the entrance of candidates with the requisite qualities fairly and consistently in a way that ensures unintended bias does not hinder that entrance; and we should be seeking to promote amongst those selected the capacity to manifest these key professional qualities through educational and development opportunities. The use of these instruments may help in these respects, but also the following/subsequent comments—in terms of both concerns and suggestions—may be considered as intending to promote these two principles. IP capabilities provide a bedrock for social work—a sin qua non of the profession. In this respect, the data on positive emotions such as Compassion, Altruism and personal Feelings and Insight go some way to confirming what may be expected of these capabilities in those about to enter practice. We should be careful, however, about over-interpreting this. The normative group was an opportunity sample and we cannot state that it is representative, in the way characterised by a probability sample, of the adult UK population. Indeed, the gender make-up suggests this is not the case (women make up 47 per cent of the UK workforce, but a third of the normative sample) (Office of National Statistics, 2016). Deliberation and Assertiveness do not perhaps reflect those expectations to the same degree. Deliberation can be highly significant in ensuring decisions are thoughtful and considered, yet these graduands scored markedly less than the normative sample. Likewise, there are many occasions—working with dangerous situations, advocating and ensuring the rights of vulnerable clients with outside agencies and services—where Assertiveness is an important asset. It is, indeed, in the balance between features like Assertiveness and Deliberation and the positive emotions that social work may be most effectively manifested. These data suggest that this balance has perhaps not been most efficaciously reached. This may reflect the stage of these social workers’ career—as starting practitioners, they may not have encountered the situations which demand high levels of Assertiveness. Data on social workers further into their career may settle this issue. The homogeneity in most of these IP measures indicates people with similar IP characteristics. Master’s and bachelor’s levels, as well as both pre- and post-92 university cohorts, were quite similar. Variability indicates some were quite assertive while others were relatively unassertive. This is a feature in the development of early-career social workers that looks to require some focus, with some markedly better equipped than others to manage situations requiring Assertiveness. Measurement of CT raises some concerns. We should note first that, like the normative IP samples, the CT comparison groups are opportunity samples. Furthermore, for example in the case of Law/Business graduates, collection of data occurred in the context of job applications. In a competitive environment, there is an imperative to perform at the optimum level. It makes it more likely, furthermore, that these candidates had practised for the WGCTA (practice tests are available, e.g. Clifford-Chance, 2016). Where candidates prepared themselves, it is reasonable to suppose that their performance would have been enhanced. These points may help explain some of the gap in mean performance levels. Having stated this, over 200 of the completed WGCTAs contained some corrections, suggesting participants were thinking carefully about their responses. Other samples of those attending university—in these cases, unlike the normative samples from outside the UK—do not always show such wide divergence from these social workers in WGCTA forty-item scores. For example, Turkish teachers-in-training had a mean score of 18 (Sendag and Odabsi, 2009), nurses 20 (Walsh and Seldomridge, 2006)—both comparable to these social workers—and American respiratory care trainees scored 23 (Wettstein et al., 2011). Others, though, scored more highly, such as American philosophers and psychologists (25) (Burke et al., 2014) and Canadian management and nursing candidates (27) (Loom and Thorpe, 1999). However, it would be unwise to ignore these differences. In particular, the mean score of bachelor’s-level social workers (18.9) was hardly better than chance (18.5) in a suitably sized sample. In other words, had all the bachelor’s graduands guessed all the answers, we would have expected their mean score would have been practically the same. This shows room for considerable improvement. In a profession where CT capabilities are so fundamental to best practice, we might expect a mean score to be rather higher. If, indeed, these CT capabilities were embedded in the learning undertaken in the social work courses, we might expect this to have been reflected in the WGCTA scores. This raises two possible issues: Are CT capabilities all they should be in social work? And are they sufficiently a focus on social work courses? In relation to the first, there is clearly room for improvement. On the second: is there a tension in the learning process? Recent research has demonstrated convincingly that intellectual and IP domains are highly differentiated in social work (Sheppard and Charles, 2015). This exclusivity implies social work is highly demanding in terms of the expected aptitude of its practitioners—quite different talents are being expected and we may expect these differences to create a tension in the learning processes required. These tensions can be traced to fundamental distinctions in cognitive processes. We can draw on the distinction between System 1 and System 2 thinking. Recognition of affect, creativity and intuitive judgement are all aspects of System 1 thinking—the process of thinking which is fast, direct and spontaneous, and operates automatically. Much of social work emphasises these particular qualities. System 2 thinking is consciously engaged and involves deliberative thinking—it covers the capacity for reasoning which underlies CT, is effortful and intentional (Evans, 2008). Difference, of course, can reflect innate individual differences in reasoning capacity (pre-dispositions). However, while recognising the reservations about the normative samples mentioned earlier, if there is an underlying difference between, for instance, law and social work, this may reflect in part the learning process. The close relationship between performance in law and WGCTA suggests some of the law-learning processes may encapsulate reasoning processes underlying CT more consistently (TalentLens, 2011). Such processes could involve overt and close examination of argumentation, underlying assumptions and the process of drawing conclusions. In an adversarial system, such processes would be strongly highlighted in case-based work and, for example, moots. It may be that social work learning processes do not emphasise these underlying reasoning capabilities to the same extent as, for instance, law. This is understandable. Few would question the importance of intuition, creativity and awareness of affect in social work. However, these are System 1 capabilities, not System 2, which is the focus for reasoning, and it is the former in this case being encouraged rather than the latter. This process exclusivity enhances chances that one may be nurtured more than the other. None of this implies that learning processes designed for CT enhancement do not occur. It does suggest, however, that this could be undertaken more overtly and systematically. More concern may be expressed, however, at the huge diversity of CT capabilities amongst these graduands. The best scored close to 100 per cent—similar to the best of other samples, but the lowest scores raised questions about CT capabilities for practice. Of course, we might suggest that the tests were themselves abstract, and that a focus of similar reasoning processes on social work exemplars may improve scores, and this might have some legitimacy. On the other hand, the critical nature of speculation of possible alternatives in social work practice (What is likely to happen? What alternatives are there? What might be the result if we did this?) indicates that the capacity for such abstract thinking is of considerable importance. These data reflect empirically some of the concerns expressed politically about standards within social work. In particular, Frontline is an education/training development aiming at ‘boosting the image of social work and getting the very best people into the profession’ by recruiting ‘top graduates on to its programmes’ (Frontline, 2014). While, however, the data provide evidence to support the concerns, these data suggest (given very high and very low scores) a focus should also be on the performance levels of those with low CT capabilities. One-quarter of the sample scored 15 or less—substantially below the chance mean. This is a serious concern in a profession where CT is considered to be at the heart of practice. It indicates that, in addition to augmenting social work entrants with a number of high-achiever ‘leaders’, as with Frontline, attention be given to improving the CT of a significant minority of qualifying social workers whose performance is poor. Some of this is about increasing the quality of those entering social work. This is clearly a more significant issue—though not exclusively—for bachelor’s than master’s cohorts. If reliance is placed on formal educational qualifications, then their reliability at indicating CT capabilities needs to be established. Alternatively, psychometric tests additional to these qualifications—rather in the manner established by medicine, screening entrants—may be efficacious. A further radical solution might be to confine social work qualification to master’s level, although, while this would substantially improve mean CT capabilities, there would remain room for considerable improvement at the lowest levels. The second solution lies in the curriculum and perhaps practice itself. It is well beyond the scope of this article to examine in detail possible developments. We may reasonably ask: How, in principle, can CT be enhanced in learning about social work? Some common elements clearly potentially contribute to enhanced CT. Many researchers have maintained that CT is dependent on pre-dispositions (an issue of gate-keeping as above) and purposeful reflection that requires logic (Behar-Hornestein and Niu, 2011). Certainly, purposeful reflection is embedded in the social work curriculum—and is expected in subsequent qualified practice—both in university and practice-based learning, focused furthermore on social work issues and cases. It is a central feature of practice standards (Proficiency 11: ‘be able to reflect on and review practice’ (including Critical Reflection); HCPC, 2017). It is, however, the logic requirement that has been highlighted by these findings. It is not, in other words, sufficient to reflect on issues and cases, even to be imaginative and creative in considering actions, but to do so in a way placing a premium on logic. Broadly, this focuses on the rigour with which the thinking-about-practice (reasoning) processes is undertaken. This is apparent in earlier research which highlighted the importance of ‘forward thinking’ in social work, rather than the backward-focused nature of reflection (Sheppard et al., 2000). This took the form of ‘speculative hypotheses’—‘if–then’ statements, often of a highly complex nature—focusing on alternative future possibilities, their likelihood and consequences, in terms of service users’ thoughts, actions and reactions, the actions and interventions of social workers and others, the interaction and outcome of these actions and the likely possible results of all these interactions. This points to the overt and precise formulation of hypothesis-based thinking as an underlying principle of the learning and practice processes of social work, whichever particular emphasis is placed on teaching–learning processes chosen. The appropriateness of this as an underlying principle is apparent in a particular approach to the teaching of science ‘hypothesis-based learning’: Students are presented with common, simple materials …. Each student … then … begin(s) making observations. When the student finds something interesting, the student should attempt to explain the observation. Given an explanation, he or she needs to think of a way to test the explanation and make a prediction. Of course, the test is next. Then the student must analyze test results to decide if the prediction is supported or unsupported (VanDorn 2005, p. 57; see also VanDorn et al., 2017). This is, of course, a simplistic presentation in terms of social work materials or cases. It also refers to a particular approach, though it is here being used as an indicator of how an underlying principle might operate (in both the curriculum and subsequent practice). However, it is apparent how this approach could, in principle, both develop cognitive processes appropriate to the complex developmental and changing nature of cases and practice situations, and at the same time provide a framework through which the logic of CT may be developed. It is in the provision of a framework to enable the enhancement of reasoning capabilities that such an approach may be efficacious, when linked, for example, with social work-relevant exercises ‘Heart and Head’ have long been seen as the central features of practice, the combination going some way to defining what social work is. The findings here show great strength in positive emotions and great diversity in Assertiveness and Deliberation, the lower levels of which require nurturing. They also indicate serious concerns regarding CT, particularly regarding lower performers upon which an increased intensity of focus needs to be placed. References Behar-Hornestein L. , Niu L. ( 2011 ) ‘ Teaching critical thinking in higher education: A review of the literature ’, Journal of College Teaching and Learning , 8 ( 2 ) pp. 25 – 41 . Bosanquet B. ( 1901 ) ‘ The meaning of social work’, International Journal of Ethics , 11 ( 3 ), pp. 291 – 306 . Google Scholar Crossref Search ADS British Association of Social Workers (BASW) ( 2016 ) Professional Capabilities Framework, available online at https://www.basw.co.uk/pcf/capabilities/?level=7&domain=1 (accessed 5 December 2017). Brown K. , Rutter L. ( 2009 ) Critical Thinking for Social Work , London , Macmillan . Burke B. , Sears S. , Kraus S. ( 2014 ) ‘ Critical analysis: A comparison of critical thinking changes in psychology and philosophy classes’, Teaching of Psychology , 41 ( 1 ), pp. 28 – 36 . Clifford-Chance ( 2016 ) Watson Glaser Practice Tests, available online at https://www.practiceaptitudetests.com/top-employer-profiles/clifford-chance-assessments/ (accessed 12 December 2017). Coleman H. , Rogers G. , King J. ( 2002 ) ‘ Using portfolios to stimulate critical thinking in social work education ’, Social Work Education , 21 ( 2 ), pp. 583 – 95 . Google Scholar Crossref Search ADS Cooper J. ( 2017 ) ‘Social work fast-track schemes’, Community Care, available online at http://www.communitycare.co.uk/2016/02/03/social-work-fast-track-schemes-dont-know/ (accessed 5 December 2017). Costa P. , McCrae R. ( 2006 ) Revised NEO Personality Inventory Manual (UK Edition) , Oxford , HOGREFE . D’Cruz H. , Gillingham P. , Melendez S. ( 2007 ) ‘ Reflexivity: Its meaning and relevance for social work: A critical review of the literature ’, British Journal of Social Work , 37 ( 1 ), pp. 73 – 90 . Evans J. ( 2008 ) ‘ Dual process accounts of reasoning thinking and social cognition’, Annual Review of Psychology , 59 , pp. 255–78. Frontline ( 2014 ) ‘Thousands of top graduates apply to new social work programme’, available online at https://thefrontline.org.uk/2014/03/17/thousands-of-top-graduates-apply-to-new-social-work-programme/ (accessed 12 December 2017). Gillin D. , Maddux W. W. , Carpenter J. , Galinsky A. D. ( 2013 ) ‘ When to use your heart and when to use your head: The different value of perspective taking versus empathy’, Personality and Social Psychology Bulletin , 39 ( 3), pp. 3 – 16 . Hardcastle D. A. , Powers P. R. , Wenocour S. ( 2011 ) Community Practice: Theories and Skills for Social Workers , 3 rd edn, Oxford , Oxford University Press . Hardiker P. ( 1981 ) ‘ Heart or head: The function and role of knowledge in social work ’, Issues in Social Work Education , 1 ( 2 ), pp. 85 – 111 . HCPC (Health and Care Professionals Council) ( 2017 ) Standards of Proficiency: Social Workers in England , available online at http://www.hpc-uk.org/assets/documents/10003B08Standardsofproficiency-SocialworkersinEngland.pdf (accessed 12 December 2017). Helm D. ( 2011 ) ‘ Judgements or assumptions? The role of analysis in assessing children and young people's needs ’, British Journal of Social Work , 41 , 894 – 911 . Hennessy R. ( 2011 ) Relationship Skills in Social Work , London, Sage . Howe D. ( 1998 ) ‘ Relationship based thinking in social work’, Journal of Social Work Practice , 12 , pp. 45 – 56 . Google Scholar Crossref Search ADS Ingram R. ( 2013 ) ‘ Locating emotional intelligence at the heart of social work practice’, British Journal of Social Work , 43 ( 5 ), pp. 987 – 1004 . Google Scholar Crossref Search ADS Kinman G. , Grant L. ( 2011 ) ‘ Exploring stress resilience in trainee social workers: The role of emotional and social competencies’, British Journal of Social Work , 41 ( 2 ), pp. 261 – 75 . Google Scholar Crossref Search ADS Koprowska J. ( 2010 ) Communication and Interpersonal Skills in Social Work , Exeter , Learning Matters . Locke E. A. ( 2005 ) ‘ Why emotional intelligence is an invalid concept’, Journal of Organisational Behaviour , 26 , pp. 411 – 24 . Google Scholar Crossref Search ADS Loom R. , Thorpe K. ( 1999 ) ‘ A psychometric investigation of scores on the Watson Glaser Critical Thinking Appraisal new form S’, Educational and Psychological Measurement , 59 , pp. 995 – 1002 . Google Scholar Crossref Search ADS Lord W. ( 2007 ) NEO PI R: A Guide to Interpretation and Feedback in a Work Context , Oxford , HOGREFE . McCabe P. ( 1998 ) The Assertive Social Worker , Aldershot , Ashgate . Munro E. ( 1999 ) ‘ Common errors of reasoning in child protection work’, Child Abuse and Neglect , 23 ( 8 ), pp. 745 – 58 . Google Scholar Crossref Search ADS PubMed Ngai S. S. , Cheung C. ( 2009 ) ‘ Idealism, altruism career orientation and emotional exhaustion amongst social work undergraduates’, Journal of Social Work Education , 45 ( 1 ), pp. 105 – 21 . Google Scholar Crossref Search ADS Office of National Statistics ( 2016 ) Employment and Employee Types, available online at https://www.ons.gov.uk/employmentandlabourmarket/peopleinwork/employmentandemployeetypes (accessed 5 December 2017). Patel V. L. , Arocha J. F. , Zhang J. ( 2005 ) ‘Thinking and reasoning in medicine’, in Holyoak K. , Morrison R. G. (eds), Handbook of Thinking and Reasoning , Cambridge , Cambridge University Press , pp. 298 – 358 . Radley M. , Figley C. R. ( 2007 ) ‘ The social psychology of compassion’, Clinical Social Work Journal , 35 ( 3 ), pp. 207 – 14 . Google Scholar Crossref Search ADS Ruch G. , Turney D. , Ward A. ( 2010 ) Relationship Based Social Work , London , Kingsley . Scurlock-Evans L. , Upton D. ( 2015 ) ‘ The role and rating of evidence: A systematic review of social workers’ Evidence Based Practice orientation, attitudes and implementation ’, Journal of Evidence-Informed Social Work , 12 ( 4 ), available online at http://www.tandfonline.com/doi/full/10.1080/15433714.2013.853014 (accessed 5 December 2017). Sendag S. , Odabsi H. ( 2009 ) ‘ Effect of an online problem based course on content knowledge acquisition and critical thinking skills’, Computers and Education , 53 , pp. 132 – 41 . Sheppard M. ( 1995a ) Care Management and the New Social Work , London , Whiting and Birch . Sheppard M. ( 1995b ) ‘ Social work social science and practice wisdom ’, British Journal of Social Work , 25 , pp. 65 – 93 . Sheppard M. , Charles M. ( 2014 ) ‘ Critical thinking and interpersonal dispositions in those commencing social work training’, British Journal of Social Work , 44 , pp. 2057 – 66 . Google Scholar Crossref Search ADS Sheppard M. , Charles M. ( 2015 ) ‘ Head and Heart: An examination of the relationship between the intellectual and interpersonal in social work’, British Journal of Social Work , 45 , pp. 1837 – 54 . Google Scholar Crossref Search ADS Sheppard M. , Charles M. ( 2017 ) ‘ A longitudinal comparative study of the impact of the experience of social work education on interpersonal and critical thinking capabilities’, Social Work Education , 36 , pp. 745 – 57 . Google Scholar Crossref Search ADS Sheppard M. , Newstead S. , DiCaccavo A. , Ryan K. ( 2000 ) ‘ Reflexivity and the development of process knowledge in social work’, British Journal of Social Work , 30 , pp. 465 – 88 . Google Scholar Crossref Search ADS Sheppard M. , Newstead S. , DiCaccavo A. , Ryan K. ( 2001 ) ‘ Comparative hypothesis assessment and quasi triangulation as process knowledge assessment strategies in social work practice’, British Journal of Social Work , 31 , pp. 863 – 85 . Google Scholar Crossref Search ADS Schulte M. , Ree M. , Carretta T. ( 2004 ) ‘ Emotional intelligence: Not much more than g and personality’, Personality and Individual Difference , 37 ( 5 ), pp. 1059 – 68 . Google Scholar Crossref Search ADS Smith L. , Ciarrochi J. , Heaven P. C. L. ( 2008 ) ‘ The stability and change of trait emotional intelligence, conflict communication patterns, and relationship satisfaction: A one-year longitudinal study’, Personality and Individual Differences , 45 , pp. 738 – 43 . Google Scholar Crossref Search ADS TalentLens ( 2011 ) Watson Glaser Supervised UK Edition , Appendix E, Business and Law, Pearson . Taylor C. , White S. ( 2006 ) ‘ Knowledge and reasoning in social work: Educating for humane judgement’, British Journal of Social Work , 36 ( 6 ), pp. 937 – 54 . Google Scholar Crossref Search ADS VanDorn K. ( 2005 ) ‘ Hypothesis-based learning ’, Science Scope , 29 ( 3 ), p. 57 . VanDorn K. , Montes L. , Ackerson B. , Rockley M. ( 2017 ) ‘Hypothesis based learning’, NTSA News Digest, available online at http://www.nsta.org/publications/news/story.aspx?id=48919 (accessed 6 December 2017). Walsh C. , Seldomridge L. ( 2006 ) ‘ Measuring critical thinking’, Nurse Educator , 31 ( 4 ), pp. 159 – 62 . Google Scholar Crossref Search ADS PubMed Wettstein R. B. , Wilkins R. L. , Gardner D. D. , Restrepo R. D. ( 2011 ) ‘ Critical-thinking ability in respiratory care students and its correlation with age, educational background, and performance on national board examinations’, Respiratory Care , 56 , 284 – 9 . Google Scholar Crossref Search ADS PubMed Whittaker A. ( 2011 ) ‘ Social defences and organisational culture in a local authority child protection setting: challenges for the Munro review? ’, Journal of Social Work Practice , 25 , 481 – 95 . 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/open_access/funder_policies/chorus/standard_publication_model)

Journal

The British Journal of Social WorkOxford University Press

Published: Oct 1, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off