Ideology and Deliberation: An Analysis of Public Support for Deliberative Practices in Finland

Ideology and Deliberation: An Analysis of Public Support for Deliberative Practices in Finland The introduction of deliberative practices has been proposed as a method for revitalizing representative democracies suffering from decreasing political involvement and increasing fragmentation (Dryzek, 2000; Grönlund, Bächtiger, & Setälä, 2014; Parkinson & Mansbridge, 2012). Some scholars, however, question whether citizens are willing to engage in deliberative practices (Bartels, 2003; Hibbing & Theiss-Morse, 2002; Posner, 2004). A related concern is that deliberative practices may not appeal equally to all groups of citizens, and consequently jeopardize the principle of inclusion of all relevant arguments and experiences in deliberation (Young, 2000). Whether deliberative practices can help revitalize democracy partly depends on the extent to which support for these practices is related to ideological predispositions. Previous studies suggest that ideological predispositions systematically influence individual attitudes toward public involvement in political processes (Bengtsson & Mattila, 2009; Dalton & Welzel, 2014; Inglehart, 1997; Kriesi, 2010). If a corresponding pattern were to be found for deliberative practices, the result could be detrimental to the deliberative focus on inclusiveness. An ideological bias among supporters might translate into overrepresentation of certain ideological perspectives in deliberative events. Furthermore, if support for deliberative practices depends on strong ideological predispositions, the prospective participants in deliberation are likely to behave like activists—eager to pursue their established goals by any means necessary (Mackuen, Wolak, Keele, & Marcus, 2010, p. 440)—rather than deliberative citizens, who are considerate, balanced, open-minded, and willing to compromise (Levine & Nierras, 2007; MacKuen et al., 2010). If the composition of deliberative events is ideologically unrepresentative, it can also result in harmful group composition effects that affect the final outcome (Baek, Wojcieszak, & Delli Carpini, 2012; Barabas, 2004; Gastil, Black, & Moscovitz, 2008). It is therefore important to establish how ideological predispositions shape attitudes toward the use of deliberative practices in society. Based on these considerations, we aim to study how support for deliberative practices is linked to ideological predispositions on the traditional left–right dimension and the nationalist–cosmopolitan values dimension. Jacobs, Cook, & Delli Carpini (2009) and Neblo, Esterling, Kennedy, Lazer, & Sokhey (2010) constitute a couple of pioneering studies that examine the willingness to deliberate in the United States, and show that politically marginalized groups are willing to deliberate (Neblo et al. 2010), and that self-identified liberals are more likely to engage in deliberation than people who consider themselves to be more conservatives (Jacobs et al. 2009). However, these studies are not necessarily applicable outside the American context. Moreover, Jacobs et al. (2009) restrict their analysis to a one-dimensional conceptualization of ideology that fails to capture contemporary ideological richness. We use the Finnish National Election Study from 2015 (FNES2015) to demonstrate that individuals who are left-leaning and/or adhere to cosmopolitan values support deliberative practices. In line with the work of Zaller (1992), we also find that political awareness furthers the ability to formulate opinions that are consistent with ideological predispositions. Deliberation and Ideological Predispositions Important distinctions can be made between different theories of deliberative democracy (Bächtiger, Niemeyer, Neblo, Steenbergen, & Steiner, 2010), but they share an emphasis on discussions as prerequisites for enlightened political decision-making. In this study, we use the concept of deliberative practices to refer to a broad range of measures aiming to improve the deliberative capacity of political systems by promoting discussions and exchanges of arguments in society (Parkinson & Mansbridge, 2012). There are various kinds of public discussions, some of which fail to fulfill the most stringent definitions of deliberation (Levine & Nierras, 2007; Karpowitz & Mendelberg, 2007). Nevertheless, we here use a broad definition of deliberation, as the aim is to include all processes that may enhance the deliberative capacity of political systems, even when they do not completely adhere to the ideal of deliberation.1 1Contrary to others (Christensen, Himmelroos, & Grönlund, 2017; Jacobs et al., 2009), we do not use the term discursive practices to stress that the discussions should involve more than “just talk,” that is, they should be connected to existing political processes and, at least potentially, affect their outcomes, which is in line with the systemic approach to deliberation (Parkinson & Mansbridge, 2012). Our approach corresponds to the focus of Levine and Nierras (2007) on public deliberation that aims to resolve real problems. Successful deliberation ideally should include all relevant arguments and opinions, and be a formative process (Goodin, 2004; Mansbridge et al., 2010; Parkinson, 2006). To ensure that all relevant positions and perspectives are included, the ideological predispositions of participants in deliberative practices ought to reflect the ideological composition of society (Gastil 2008; Gastil et al., 2008; Young, 2000). It is however questionable whether deliberative practices appeal to all segments of society, especially when it comes to the ideological predispositions of presumptive supporters. From a theoretical perspective, deliberation has been accused of harboring an elitist notion of democracy that does not appeal to common people, and of promoting liberal values over more conservative ones (Blattberg 2003; Parvin, 2015). Empirical evidence also suggests that ideological predispositions affect support for deliberative practices. First, it is well established that politically right-leaning individuals are more conservative when it comes to institutional changes (Anderson & Singer, 2008, p. 574). Although deliberation may help preserve the status quo by blocking drastic political reforms, those on the right may consequently oppose such novel methods of decision-making. Furthermore, previous studies have found that people are generally consistent in their preferences of decision-making procedures (Bengtsson & Christensen, 2016; Bengtsson & Mattila, 2009; Font, Wojcieszak, & Navarro, 2015). Some studies even suggest that leftists show stronger support for participatory and direct democratic processes (Bengtsson & Mattila, 2009; Esaiasson, Gilljam, & Persson, 2010), and they may likewise be more inclined toward deliberative practices. Second, the progression of what has been termed postmodernist, libertarian, or cosmopolitan values in Western democracies is tied to support for participatory reforms (Dalton & Welzel, 2014; Inglehart, 1997; Kriesi, 2010). According to Inglehart (1977, 1997), postmaterialist generations crave innovative forms of political participation to complement the traditional representative structures. Third, studies on deliberation show that certain ideological positions dominate deliberative events. Jacobs et al. (2009, p. 53) found that liberals are more likely than conservatives to engage in deliberation, and a study of recruitment to a deliberative mini-public shows that attrition is higher among citizens with anti-immigration attitudes (Karjalainen & Rapeli, 2015).2 2Moreover, Gastil, Bacci, and Dollinger (2010) find that outcomes of deliberation tend to be biased toward cosmopolitan and collectivist beliefs. Political ideology, as we use the term here, departs from dominant issue divides with political relevance (Bartolini, 2011; Deegan-Krause, 2007). Traditionally, ideology has been considered a one-dimensional phenomenon distributed on a left–right continuum (Downs, 1957), which in a European context has its roots in historical class cleavages (Lipset & Rokkan, 1967; Knutsen, 2006) that divided politics into the liberal–conservative right and the social–democratic or socialist left. Although this left–right dimension has largely revolved around issues of redistribution, over time it has developed into a meta-dimension that has structured party systems and the ideological beliefs of citizens (Mair, 2007). However, several works contend that ideology can no longer be considered one dimensional, and that a dimension concerning immaterial values rather than issues of redistribution has complemented the left–right dimension. Scholars here use various labels to denote the poles on closely related value dimensions, such as libertarian/authoritarian (Flanagan & Lee, 2003; Kitschelt, 1994; Kitschelt & McGann, 1995), postmaterialist/materialist (Inglehart, 1977), green–alternative–libertarian (GAL)/traditional–authoritarian–nationalist (TAN) (Hooghe, Marks, & Wilson, 2002) or cosmopolitans/nationalists (Norris & Inglehart, 2009).3 3Other examples of labels used are self-expression/survival (Inglehart & Welzel, 2005), libertarian, universalistic/traditionalist–communitarian (Bornschier, 2010), integration/demarcation (Kriesi, 2010). Despite the different labels and empirical operationalizations, these scholars share a belief that when examining the causes and consequences of ideological predispositions, it should be acknowledged that ideology today is a multidimensional phenomenon. For this reason, we also examine the relationship between the dimension that we label national–cosmopolitan values and support for deliberative practices. The theoretical and empirical reasons presented above suggest that ideological predispositions on both dimensions are likely to influence attitudes toward deliberative practices, as individuals with more leftist and/or cosmopolitan values are likely to support the introduction of such mechanisms to complement representative decision-making. These suggestions are in line with the established finding that general political predispositions function as guides or heuristic devices when people structure their views on more specific issues (Feldman & Zaller, 1992; Goren, Schoen, Reifler, Scotto, & Chittick, 2016; Sniderman, Brody, & Tetlock, 1991). However, the impact of these predispositions may be less straightforward than suggested above. As convincingly demonstrated by Zaller (1992), people need a certain level of political awareness (attention and understanding) to translate their ideological predispositions into opinions on specific issues.4 4We would like to thank one of the anonymous reviewers for this suggestion. Politically aware citizens are more likely to organize their political ideas in an ideologically consistent manner5 5Or to have “politically sophisticated” belief systems (Luskin, 1987). (Gastil & Dillard, 1999), that is, they are more likely to hold political beliefs that are in line with their ideological predispositions. This implies that it is among the politically aware that we would expect to find a stronger link between ideological predispositions and attitudes toward deliberative practices. Based on these considerations, we examine the following hypotheses on the relationship between ideological predispositions and support for deliberative practices in the empirical section: H1: Leftist values on the traditional left–right dimension are positively associated with positive attitudes toward deliberative practices. H2: Cosmopolitan values on the nationalist–cosmopolitan dimension are positively associated with positive attitudes toward deliberative practices. H3: Political awareness moderates the relationship between ideological predispositions and attitudes toward deliberative practices. Data and Operationalizations The data come from the Finnish National Election Study conducted following the national elections on April 19, 2015 (FNES2015), which involved a cross-sectional survey using a two-stage process with introductory face-to-face interviews of 1,602 respondents during April–June 2015.6 6See https://services.fsd.uta.fi/catalogue/FSD3067?lang=en&study_language=en for more information on FNES2015. Although the cross-sectional data do not make it possible to settle causal effects, FNES2015 is ideal for examining associations between ideological predispositions and support for deliberative practices, as the survey includes suitable measures for all variables. Finland constitutes an opportune case, as, like most European democracies, it has an established tradition of representative decision-making complemented by a restrictive use of consultative referendums. However, studies show demand for more citizen involvement in political decision-making (Bengtsson & Christensen, 2016), echoing similar sentiments across the continent. In this sense, even though the generalizability of the results may be limited, Finland presents a paradigmatic case for studying attitudes toward deliberative practices. The general public has a limited familiarity with formal deliberative institutions such as deliberative mini-publics.7 7Deliberative mini-publics have mainly been used in the form of academic experiments in Finland. However, this does not preclude the public from forming opinions on the presumptive use of deliberative practices in the form of public discussions to complement representative decision-making. All variables are coded to vary between 0 and 1, with 1 indicating the highest value of the variable in question. More information on the coding of variables, and descriptive statistics, is found in Table A1. Dependent Variable The dependent variable is the attitude toward the use of deliberative practices as a complement to representative decision-making.8 8We use the same measure as Christensen et al. (2017), who study whether participation in a deliberative mini-public enhances support for arranging public discussions in representative democracy. The measure is based on responses to two statements: (1) Political discussions for ordinary citizens should be arranged in support of representative democracy, and (2) I myself would like to participate in political discussions arranged for ordinary citizens. The responses to both items were indicated using a four-point Likert scale (totally agree to totally disagree), and these scores were subsequently combined to form an additive index (M = 0.53). A closer inspection of the individual questions (not shown) reveals that about 75% tend to agree that discussions should be arranged to support representative democracy, while only about 42% are willing to take part in them.9 9We fitted ordinal regression models examining the ideological impact on the separate questions. The associations with the ideological predispositions were generally similar with the exception that the direct effect of cosmopolitanism is not significant (B = 0.55, p = .084) for the question on introducing political discussions for ordinary citizens. Our operationalization is based on a broad conceptualization of deliberative practices in line with the definition offered above, and may therefore overestimate the extent of support for a more narrowly defined interpretation. Some may even object to the label deliberative practices, as the questions do not refer to genuine deliberative institutions such as mini-publics (Grönlund et al., 2014). However, as most people are unfamiliar with these formal deliberative institutions, it makes little sense to ask respondents about their attitudes toward them directly. Previous survey research on how citizens see public deliberation has also relied on similarly wide understandings of the topic (Baek et al., 2012; Wojcieszak, Baek, & Delli Carpini, 2009). A broad conceptualization is warranted in this case, as the objective is not to identify support for specific arrangements (i.e., mini-publics), but to examine support for the use of talk-centric measures to improve representative decision making. Our conceptualization is closer to what Bächtiger et al. (2010) entitle Type II deliberation, which involves more flexible discourses than the more narrowly defined Type I deliberation, which focuses on rational discourse and procedural aspects. Nevertheless, our operationalization entails a focus on discussions that are connected to formal political decision-making in line with Levine and Nierras (2007). Independent Variables To capture ideological predispositions, we focused on the traditional left–right dimension (Mair, 2007) and the increasingly important dimension of what we here refer to as nationalist–cosmopolitan values. The position on the left–right dimension was measured with a variable, where the respondent indicated his or her placement on an 11-point left–right scale (recoded to vary between 0 and 1 (1 = right). The degree of national–cosmopolitan values was captured with an index based on three questions measuring the attitudes of the respondents toward multiculturalism in general and tolerance toward two out-groups: sexual minorities and immigrants (Cronbach’s alpha = .76, 1 = cosmopolitan).10 10One of the items included in our index concerns the rights of sexual minorities rather than cosmopolitan/nationalist values, but this has been shown to correlate strongly with similar dimensions (Inglehart, 1997). Moderator Variables Political awareness is conceptualized as the understanding of and attentiveness toward politics, and operationalized by two different variables. The first is factual political knowledge, which is in accordance with the operationalization used by Zaller (1992). This was measured with an index, where respondents received a point for each correct answer to five factual questions on international and domestic Finnish political matters (0 = no correct answers). The second variable is political interest, which has been considered a suitable indicator (Zaller 1992, pp. 333–4) and has been used in previous research on similar topics (Verba, Schlozman, & Brady, 1995). We measured this with a common question on the topic, where respondents indicated how interested they were in politics on a four-point scale (0 = lowest interest). Control Variables To ascertain the robustness of the results, we included sociodemographic characteristics (age, gender, education, and full-time employment) and attitudes toward the political system (satisfaction with democracy, political trust, and internal and external political efficacy). All of these variables have been demonstrated to influence the propensity for political participation in general (Bengtsson & Christensen, 2016; Verba et al., 1995) and in regard to deliberation (Karjalainen & Rapeli, 2015; Neblo et al., 2010). Age in years was divided by 100 to approximate the 0–1 coding of all independent variables. In regard to education, the respondents indicated the highest level of education completed on an eight-point scale (0 = lowest level of education). The other two characteristics are dichotomous. For gender, 0 indicates female and 1 male, whereas for employment, 1 corresponds to full-time employment, and all other occupations (student, retired, half-time employment, etc.) are coded 0. For satisfaction with democracy, respondents indicated their extent of satisfaction with how democracy functions in Finland on a four-point scale (0 = lowest satisfaction). We measured internal political efficacy with a question, where respondents indicated the extent to which they agree with the statement, “Sometimes politics seems so complicated that I can’t really understand what is going on,” on a four-point Likert scale (0 = lowest efficacy). External efficacy was measured with an index based on four statements, where respondents indicated their views on the responsiveness of the political system on a four-point Likert scale (0 = lowest efficacy; Cronbach’s alpha = .77). For political trust, we use an index based on five questions, where respondents indicated their level of trust on an 11-point scale (0–10) in regard to the Finnish parliament, politicians, political parties, the Finnish president, and the government (Cronbach’s alpha = .91). Methods of Analysis A Shapiro–Wilk test shows that the dependent variable follows a normal distribution, as the null hypothesis cannot be rejected (W = 0.998; p = .0812). We therefore used linear regression with robust standard errors to examine our hypotheses. To examine H1 and H2 concerning the associations between ideological factors and support for deliberative practices, the first model, M1, only included the two ideological factors, left–right and nationalist–cosmopolitan, whereas M2 also included the measures of political awareness and control variables to verify that the linkages persist when controlling for other factors. To examine H3 and the moderating effect of political awareness, in M3, we included four interaction terms constituted by the two ideological factors and the two factors of political awareness in the regression model. Empirical Analysis Table 1 displays the results of the regression analyses. Table 1 Multivariate Linear Regression Models Examining Support for Deliberative Practices Variables  M1   M2   M3     Coefficient  Robust SE  p  Coefficient  Robust SE  p  Coefficient  Robust SE  p  Independent variables                        Left–right  −0.22  (0.04)  .000  −0.21  (0.04)  .000  0.12  (0.10)  .235      Nationalist–cosmopolitan  0.15  (0.04)  .000  0.12  (0.04)  .002  0.15  (0.13)  .230  Political awareness                        Political knowledge        −0.02  (0.03)  .428  0.14  (0.12)  .245      Political interest        0.39  (0.03)  .000  0.53  (0.11)  .000  Interactions                        Political knowledge × left–right              −0.24  (0.13)  .072      Political interest × left–right              −0.25  (0.12)  .042      Political knowledge × nationalist–cosmopolitan              −0.06  (0.15)  .693      Political interest × nationalist– cosmopolitan              0.00  (0.14)  .977  Control variables                        Age/100        −0.09  (0.05)  .047  −0.10  (0.05)  .026      Gender (1 = male)        −0.03  (0.02)  .066  −0.03  (0.02)  .093      Education        −0.06  (0.03)  .102  −0.05  (0.03)  .117      Full-time employment (1 = yes)        −0.01  (0.02)  .376  −0.02  (0.02)  .300      Satisfaction with democracy        −0.05  (0.05)  .288  −0.05  (0.05)  .269      Internal efficacy        −0.07  (0.03)  .009  −0.08  (0.03)  .006      External efficacy        0.07  (0.04)  .135  0.07  (0.04)  .117      Political trust        −0.09  (0.06)  .118  −0.09  (0.06)  .112      Constant  0.58  (0.03)  .000  0.53  (0.05)  .000  0.33  (0.09)  .000  F  (21,292) = 31.25, p < .001  (121,201) = 28.41, p < .001  (161,197) = 24.21, p < .001  Adjusted R2  0.05  0.19  0.19  N  1,295  1,214  1,214  Variables  M1   M2   M3     Coefficient  Robust SE  p  Coefficient  Robust SE  p  Coefficient  Robust SE  p  Independent variables                        Left–right  −0.22  (0.04)  .000  −0.21  (0.04)  .000  0.12  (0.10)  .235      Nationalist–cosmopolitan  0.15  (0.04)  .000  0.12  (0.04)  .002  0.15  (0.13)  .230  Political awareness                        Political knowledge        −0.02  (0.03)  .428  0.14  (0.12)  .245      Political interest        0.39  (0.03)  .000  0.53  (0.11)  .000  Interactions                        Political knowledge × left–right              −0.24  (0.13)  .072      Political interest × left–right              −0.25  (0.12)  .042      Political knowledge × nationalist–cosmopolitan              −0.06  (0.15)  .693      Political interest × nationalist– cosmopolitan              0.00  (0.14)  .977  Control variables                        Age/100        −0.09  (0.05)  .047  −0.10  (0.05)  .026      Gender (1 = male)        −0.03  (0.02)  .066  −0.03  (0.02)  .093      Education        −0.06  (0.03)  .102  −0.05  (0.03)  .117      Full-time employment (1 = yes)        −0.01  (0.02)  .376  −0.02  (0.02)  .300      Satisfaction with democracy        −0.05  (0.05)  .288  −0.05  (0.05)  .269      Internal efficacy        −0.07  (0.03)  .009  −0.08  (0.03)  .006      External efficacy        0.07  (0.04)  .135  0.07  (0.04)  .117      Political trust        −0.09  (0.06)  .118  −0.09  (0.06)  .112      Constant  0.58  (0.03)  .000  0.53  (0.05)  .000  0.33  (0.09)  .000  F  (21,292) = 31.25, p < .001  (121,201) = 28.41, p < .001  (161,197) = 24.21, p < .001  Adjusted R2  0.05  0.19  0.19  N  1,295  1,214  1,214  Note: Entries are coefficients from linear regression analyses with robust standard errors in parentheses. Table 1 Multivariate Linear Regression Models Examining Support for Deliberative Practices Variables  M1   M2   M3     Coefficient  Robust SE  p  Coefficient  Robust SE  p  Coefficient  Robust SE  p  Independent variables                        Left–right  −0.22  (0.04)  .000  −0.21  (0.04)  .000  0.12  (0.10)  .235      Nationalist–cosmopolitan  0.15  (0.04)  .000  0.12  (0.04)  .002  0.15  (0.13)  .230  Political awareness                        Political knowledge        −0.02  (0.03)  .428  0.14  (0.12)  .245      Political interest        0.39  (0.03)  .000  0.53  (0.11)  .000  Interactions                        Political knowledge × left–right              −0.24  (0.13)  .072      Political interest × left–right              −0.25  (0.12)  .042      Political knowledge × nationalist–cosmopolitan              −0.06  (0.15)  .693      Political interest × nationalist– cosmopolitan              0.00  (0.14)  .977  Control variables                        Age/100        −0.09  (0.05)  .047  −0.10  (0.05)  .026      Gender (1 = male)        −0.03  (0.02)  .066  −0.03  (0.02)  .093      Education        −0.06  (0.03)  .102  −0.05  (0.03)  .117      Full-time employment (1 = yes)        −0.01  (0.02)  .376  −0.02  (0.02)  .300      Satisfaction with democracy        −0.05  (0.05)  .288  −0.05  (0.05)  .269      Internal efficacy        −0.07  (0.03)  .009  −0.08  (0.03)  .006      External efficacy        0.07  (0.04)  .135  0.07  (0.04)  .117      Political trust        −0.09  (0.06)  .118  −0.09  (0.06)  .112      Constant  0.58  (0.03)  .000  0.53  (0.05)  .000  0.33  (0.09)  .000  F  (21,292) = 31.25, p < .001  (121,201) = 28.41, p < .001  (161,197) = 24.21, p < .001  Adjusted R2  0.05  0.19  0.19  N  1,295  1,214  1,214  Variables  M1   M2   M3     Coefficient  Robust SE  p  Coefficient  Robust SE  p  Coefficient  Robust SE  p  Independent variables                        Left–right  −0.22  (0.04)  .000  −0.21  (0.04)  .000  0.12  (0.10)  .235      Nationalist–cosmopolitan  0.15  (0.04)  .000  0.12  (0.04)  .002  0.15  (0.13)  .230  Political awareness                        Political knowledge        −0.02  (0.03)  .428  0.14  (0.12)  .245      Political interest        0.39  (0.03)  .000  0.53  (0.11)  .000  Interactions                        Political knowledge × left–right              −0.24  (0.13)  .072      Political interest × left–right              −0.25  (0.12)  .042      Political knowledge × nationalist–cosmopolitan              −0.06  (0.15)  .693      Political interest × nationalist– cosmopolitan              0.00  (0.14)  .977  Control variables                        Age/100        −0.09  (0.05)  .047  −0.10  (0.05)  .026      Gender (1 = male)        −0.03  (0.02)  .066  −0.03  (0.02)  .093      Education        −0.06  (0.03)  .102  −0.05  (0.03)  .117      Full-time employment (1 = yes)        −0.01  (0.02)  .376  −0.02  (0.02)  .300      Satisfaction with democracy        −0.05  (0.05)  .288  −0.05  (0.05)  .269      Internal efficacy        −0.07  (0.03)  .009  −0.08  (0.03)  .006      External efficacy        0.07  (0.04)  .135  0.07  (0.04)  .117      Political trust        −0.09  (0.06)  .118  −0.09  (0.06)  .112      Constant  0.58  (0.03)  .000  0.53  (0.05)  .000  0.33  (0.09)  .000  F  (21,292) = 31.25, p < .001  (121,201) = 28.41, p < .001  (161,197) = 24.21, p < .001  Adjusted R2  0.05  0.19  0.19  N  1,295  1,214  1,214  Note: Entries are coefficients from linear regression analyses with robust standard errors in parentheses. M1 shows that both ideological variables have the expected relationships with support for deliberative practices, which is also the case for M2 after controlling for other variables. For the left–right dimension, the negative coefficient of −0.21 in M2 (p < .001) shows that being to the left on this ideological dimension is associated with stronger support for deliberative practices. For the nationalist–cosmopolitan dimension, the positive coefficient of 0.12 (p = .002) shows that a higher degree of cosmopolitanism is associated with stronger support for deliberative practices. Figure 1a and b shows what this entails for the associations between ideological factors and support for deliberative practices when holding all other factors at their mean values. Figure 1 View largeDownload slide Predicted support for deliberative practices Figure 1 View largeDownload slide Predicted support for deliberative practices For the left–right dimension, the predicted value of support for deliberative practices is about 0.66 for those furthest to the left, but this decreases to about 0.45 for the most extreme rightist respondents. For the nationalist–cosmopolitan dimension, the most nationalist respondents have a predicted level of support of about 0.47, which increases to 0.59 for the respondents with the most cosmopolitan values. This supports H1 and H2, as both ideological variables have the expected associations with level of support for using deliberative practices. M3 shows that there are no significant effects for the nationalist–cosmopolitan dimension. There is, however, a significant interaction effect between the left–right dimension and political interest and the interaction effect between the left–right dimension and political knowledge is marginally insignificant (B = –0.25, p = .072). As traditional significance tests are unreliable for establishing the substantial importance of interaction effects, we show the implications of both of these in Figure 2 (Bedeian & Mossholder, 1994; Brambor, Clark, & Golder, 2006).11 11The exclusion of the remaining nonsignificant interaction effects is also warranted by their weak coefficients (see M3), meaning their substantial impact is in any case limited. Even if we run separate analyses for all interaction effects, the two nonsignificant interaction terms do not gain significance and the substantive impacts remain limited. We therefore remain confident that the two effects we focus on are the most relevant. To allow for a comprehensive examination of what the plots signify, we included both marginal effects of left–right ideology depending on political awareness and developments in predicted support for deliberative practices for those furthest to the left and those furthest to the right. Figure 2 View largeDownload slide Interaction effects (marginal effects and predicted support) Figure 2 View largeDownload slide Interaction effects (marginal effects and predicted support) The substantive interpretations were similar regardless of whether we measured political awareness with political interest or political knowledge. At lower levels of political awareness, the associations between the left–right dimension and support for deliberative practices are negligible, but they strengthen when respondents become politically aware. Hence, leftist individuals are only supportive of deliberative practices when they also pay attention to what happens in politics. The results thereby partly support H3, as the associations between the left–right dimension and support for deliberative practices depend on the level of political awareness, whereas there are no moderating effects for the nationalist–cosmopolitan dimension. Implications of Results The main finding presented here is that support for deliberative practices is higher among individuals with leftist and/or cosmopolitan values. If this support were to be reflected in actual participation, it would challenge the inclusiveness of deliberative practices emphasized by several deliberative theorists (Mansbridge et al., 2010; Parkinson, 2006; Young, 2000).12 12Even if the results for our control variables suggest there are few significant differences in terms of sociodemographic characteristics, suggesting that support for deliberative practices is fairly egalitarian in terms of descriptive representation (James, 2008). When deliberative practices mainly attract participants who hold certain ideological positions, the result can undermine both the quality and legitimacy of the output of deliberative practices (Dryzek, 2000; Goodin, 2004; James, 2008). Our findings support previous studies that also found ideological predispositions to be associated with positive attitudes toward participatory mechanisms in general (Bengtsson & Mattila, 2009; Dalton & Welzel, 2014; Inglehart, 1997) and deliberative practices in particular (Jacobs et al., 2009). The results thus support the notion that those with leftist/cosmopolitan values favor participatory political processes, albeit other mechanisms specifically related to the deliberative character of the processes under scrutiny may also be in play. Some studies suggest that issues traditionally associated with the right or nationalism are considered less amenable to public scrutiny (Gastil et al., 2010, pp. 4–5), causing individuals who agree with these issues from the outset to be less likely to support the use of deliberative practices. There is therefore a need to ensure that less popular, and maybe even more controversial, viewpoints are represented during prospective deliberative events. This is especially the case since previous studies suggest that group composition affects developments in attitudes during deliberation (Gastil et al., 2008; Karpowitz & Mendelberg, 2007). When participants tend to hold similar ideological predispositions, it is less likely that they are exposed to different ideological perspectives, which some studies suggest is a prerequisite for genuine deliberation (Barabas, 2004). It becomes more likely that participants act as activists rather than deliberative citizens (Levine & Nierras, 2007; MacKuen et al., 2010), which may ultimately undermine the legitimacy of such events. We have also demonstrated that the link between left–right values and support for deliberation is stronger among the politically aware. A greater understanding of, and attentiveness toward, politics hence furthers the ability to formulate opinions that are consistent with ideological predispositions (Zaller 1992). This finding supports the appeal made by Zaller (1992, pp. 308–309), that it is important for researchers to consider that people can vary greatly in their abilities to act on their values and interests. In this case, this entails that individuals with low political awareness may not articulate the attitude toward deliberative practices that their ideological predispositions would suggest. In connection to this, it is noteworthy that political awareness only moderated the relationship between the traditional left–right ideological dimension and support for deliberative practices. Although the implications of this are unclear, it shows that the new ideological dimension emphasizing cultural perspectives has a more straightforward relationship with support for deliberative practices, even for those with low levels of political awareness. It is, however, possible that this finding is because of the character of the operationalization of this ideological dimension. Whereas an established measure exists for the left–right dimension, for measuring the position on the nationalist–cosmopolitan dimension, we had to rely on issue positions in the form of attitudes toward various out-groups. Although common in the literature, it is possible that this approach leads people to think about specific interactions with the mentioned groups, rather than the abstract ideological principles that are of central concern here. This suggests a need to develop a measure of this cultural dimension that is equivalent to the left–right self-placement variable. These findings do not come without limitations. Although the Finnish case is instructive, it cannot be ascertained that similar findings would be found in other settings. Furthermore, it is worth reiterating that these findings do not necessarily entail support for specific deliberative institutions such as mini-publics, as our aim here was to examine support for a broader notion of deliberative practices. To assess whether similar results could be obtained in other contexts and for other practices, more research is needed. References Anderson C. J., Singer M. M. ( 2008). The sensitive left and the impervious right: Multilevel models and the politics of inequality, ideology, and legitimacy in Europe. Comparative Political Studies , 41, 564– 599. doi: 10.1177/0010414007313113 Google Scholar CrossRef Search ADS   Baek Y. M., Wojcieszak M., Delli Carpini M. X. ( 2012). Online versus face-to-face deliberation: Who? Why? What? With what effects? New Media and Society , 14, 363– 383. doi: 10.1177/1461444811413191 Google Scholar CrossRef Search ADS   Barabas J. ( 2004). How deliberation affects policy opinions. American Political Science Review , 98, 687– 701. doi: 10.1017/S0003055404041425 Google Scholar CrossRef Search ADS   Bartels L. ( 2003). Democracy with attitudes. In MacKuen M. B., Rabinowitz G. (Eds.), Electoral democracy  (pp. 48– 82). Ann Arbor, MI: University of Michigan Press. Bartolini S. ( 2011). Cleavages, social and political. In Badie B., Berg-Schlosser D., Morlino L. (Eds.), International encyclopedia of political science  (pp. 276– 282). Thousand Oaks: SAGE Publications. Bedeian A., Mossholder K. ( 1994). Simple question, not so simple answer: Interpreting interaction terms in moderated multiple regression. Journal of Management , 20, 159– 165. doi: 10.1177/014920639402000108 Google Scholar CrossRef Search ADS   Bengtsson Å., Christensen H. S. ( 2016). Ideals and actions: Do citizens’ patterns of political participation correspond to their conceptions of democracy? Government and Opposition , 51, 234– 260. doi: 10.1017/gov.2014.29 Google Scholar CrossRef Search ADS   Bengtsson Å., Mattila M. ( 2009). Direct democracy and its critics: Support for direct democracy and stealth democracy in Finland. West European Politics , 32, 1013– 1048. doi: 10.1080/01402380903065256 Google Scholar CrossRef Search ADS   Blattberg C. ( 2003). Patriotic, not deliberative, democracy. Critical Review of International Social and Political Philosophy , 6, 155– 174. doi: 10.1080/13698230510001702723 Google Scholar CrossRef Search ADS   Bornschier S. ( 2010). The New Cultural Divide and the Two-Dimensional Political Space in Western Europe. West European Politics , 33, 419– 444. Google Scholar CrossRef Search ADS   Brambor T., Clark W. R., Golder M. ( 2006). Understanding interaction models: Improving empirical analyses. Political Analysis , 14, 63– 82. doi: 10.1093/pan/mpi014 Google Scholar CrossRef Search ADS   Bächtiger A., Niemeyer S., Neblo M., Steenbergen M. R., Steiner J. ( 2010). Disentangling diversity in deliberative democracy: Competing theories, their blind spots and complementarities. Journal of Political Philosophy , 18, 32– 63. doi: 10.1111/j.1467-9760.2009.00342.x Google Scholar CrossRef Search ADS   Christensen H. S., Himmelroos S., Grönlund K. ( 2017). Does deliberation breed an appetite for discursive participation? Developments in attitudes towards deliberative practices after first-hand experience. Political Studies , 65, 64– 83. doi: 10.1177/0032321715617771 Google Scholar CrossRef Search ADS   Dalton R. J., Welzel C. (Eds.) ( 2014). The civic culture transformed: from allegiant to assertive citizens . Cambridge, MA: Cambridge University Press. Google Scholar CrossRef Search ADS   Deegan-Krause K. ( 2007). New dimensions of political cleavage. In Dalton R. J., Klingemann H. D. (Eds.), Oxford handbook of political science  (pp. 538– 556). Oxford: Oxford University Press. Downs A. ( 1957). An economic theory of democracy . New York: Harper. Dryzek J. ( 2000). Deliberative democracy and beyond. Liberals, critics, contestations . Oxford: Oxford University Press. Esaiasson P., Gilljam M., Persson M. ( 2010). Medborgarnas demokratiuppfattningar. In Holmberg S., Weibull L., Oscarsson H. (Eds.), Lycksalighetens ö  (pp. 269– 277). Göteborg: SOM-institutet, Göteborgs universitet. Feldman S., Zaller J. ( 1992). The political culture of ambivalence: Ideological responses to the welfare state. American Journal of Political Science , 36, 268– 307. doi: 10.2307/2111433 Google Scholar CrossRef Search ADS   Flanagan S. C., Lee A.-R. ( 2003). The new politics, cultural wars, and the authoritarian libertarian value change in advanced industrial democracies. Comparative Political Studies , 36, 235– 270. doi: 10.1177/0010414002250664 Google Scholar CrossRef Search ADS   Font J., Wojcieszak M., Navarro C. J. ( 2015). Participation, representation and expertise: Citizen preferences for political decision-making processes. Political Studies , 63(1 suppl), 153– 172. doi: 10.1111/1467-9248.12191 Google Scholar CrossRef Search ADS   Gastil J. ( 2008). Political communication and deliberation . London: Sage University Press. Google Scholar CrossRef Search ADS   Gastil J., Bacci C., Dollinger M. ( 2010). Is deliberation neutral? Patterns of attitude change during “The deliberative polls”. Journal of Public Deliberation , 6, article 3. Available at: http://www.publicdeliberation.net/jpd/vol6/iss2/art Gastil J., Dillard J. P. ( 1999). Increasing political sophistication through public deliberation. Political Communication , 16, 3– 23. doi: 10.1080/105846099198749 Google Scholar CrossRef Search ADS   Gastil J., Black L., Moscovitz K. ( 2008). Ideology, attitude change, and deliberation in small face-to-face groups. Political Communication , 25, 23– 46. doi: 10.1080/10584600701807836 Google Scholar CrossRef Search ADS   Goodin R. E. ( 2004). Representing diversity. British Journal of Political Science , 34, 453– 468. doi: 10.1017/S0007123404000134 Google Scholar CrossRef Search ADS   Goren P., Schoen H., Reifler J., Scotto T., Chittick W. ( 2016). A unified theory of value-based reasoning and US public opinion. Political Behavior , 38, 977– 997. doi: 10.1007/s11109-016-9344-x Google Scholar CrossRef Search ADS   Grönlund K., Bächtiger A., Setälä M. ( 2014). Deliberative mini-publics—Involving citizens in the democratic process . Colchester: ECPR Press. Hibbing J. R., Theiss-Morse E. ( 2002). Stealth democracy: Americans' beliefs about how government should work . Cambridge, MA: Cambridge University Press. Google Scholar CrossRef Search ADS   Hooghe L., Marks G., Wilson C. ( 2002). Does left/right structure party position on European integration? Comparative Political Studies , 35, 965– 989. doi: 10.1177/001041402236310 Google Scholar CrossRef Search ADS   Inglehart R. ( 1977). The silent revolution: Changing values and political styles among Western publics . Princeton, NJ: Princeton University Press. Inglehart R. ( 1997). Modernization and postmodernization: Cultural, economic, and political change in 43 societies , Princeton, NJ: Princeton University Press. Inglehart R., Welzel C. ( 2005). Modernization, cultural change, and democracy: The human development sequence . Cambridge, MA: Cambridge University Press. Google Scholar CrossRef Search ADS   Jacobs L. R., Cook F. L., Delli Carpini M. X. ( 2009). Talking together: Public deliberation and political participation in America . Chicago: Chicago University Press. Google Scholar CrossRef Search ADS   James M. R. ( 2008) Descriptive representation in the British Columbia citizens’ assembly. In Warren M. E., Pearse H. (eds.), Designing deliberative democracy: The British Columbia citizens’ assembly  (pp. 106– 126). Cambridge, MA: Cambridge University Press. Google Scholar CrossRef Search ADS   Karjalainen M., Rapeli L. ( 2015). Who will not deliberate? Attrition in a multi-stage citizen deliberation experiment. Quality and Quantity , 49, 407– 422. doi: 10.1007/s11135-014-9993-y Google Scholar CrossRef Search ADS   Karpowitz C. F., Mendelberg T. ( 2007). Groups and deliberation. Swiss Political Science Review , 13, 645– 662. doi: 10.1002/j.1662-6370.2007.tb00092.x Google Scholar CrossRef Search ADS   Kitschelt H. ( 1994). The transformation of European social democracy . Cambridge, MA: Cambridge University Press. Google Scholar CrossRef Search ADS   Kitschelt H. in collaboration with McGann, A. ( 1995). The radical right in Western Europe , Ann Arbor: University of Michigan Press. Knutsen O. ( 2006). Class voting in Western Europe. A comparative longitudinal study . Lanham: Lexington. Kriesi H. ( 2010). Restructuration of partisan politics and the emergence of a new cleavage based on values. West European Politics , 33, 673– 685. doi: 10.1080/01402381003654726 Google Scholar CrossRef Search ADS   Levine P., Nierras R. M. ( 2007). Activists' views of deliberation. Journal of Public Deliberation , 3, article 4. Available at: http://www.publicdeliberation.net/jpd/vol3/iss1/art4 Lipset S. M., Rokkan S. (Eds.) ( 1967). Party systems and voter alignments: Cross-national perspectives . New York, NY: Free Press. Luskin R. C. ( 1987). Measuring political sophistication. American Journal of Political Science , 31, 856– 899. doi: 10.2307/2111227 Google Scholar CrossRef Search ADS   MacKuen M., Wolak J., Keele L., Marcus G. E. ( 2010). Civic engagements: Resolute partisanship or reflective deliberation. American Journal of Political Science , 54, 440– 458. doi: 10.1111/j.1540-5907.2010.00440.x Google Scholar CrossRef Search ADS   Mair P. ( 2007). Left-right orientations. In Dalton R. J., Klingemann D. H. (Eds.), The Oxford handbook of political behaviour  (pp. 206– 222). Oxford: Oxford University Press. Mansbridge J., Bohman J., Chambers S., Estlund D., Føllesdal A., Fung A., Lafont C., Manin B. ( 2010). The place of self‐interest and the role of power in deliberative democracy. Journal of Political Philosophy , 18, 64– 100. doi: 10.1111/j.1467-9760.2009.00344.x Google Scholar CrossRef Search ADS   Neblo M. A., Esterling K. M., Kennedy R. P, Lazer D. M. J., Sokhey A. E. ( 2010). Who wants to deliberate – and why? American Political Science Review , 104, 566– 583. doi: 10.1017/S0003055410000298 Google Scholar CrossRef Search ADS   Norris P., Inglehart R. ( 2009). Cosmopolitan communications. Cultural diversity in a globalized world . Cambridge: Cambridge University Press. Google Scholar CrossRef Search ADS   Parkinson J. ( 2006). Deliberating in the real world: Problems of legitimacy in deliberative democracy . Oxford: Oxford University Press. Google Scholar CrossRef Search ADS   Parkinson J., Mansbridge J. (Eds.) ( 2012). Deliberative systems—Deliberative democracy at the large scale . Cambridge, MA: Cambridge University Press. Google Scholar CrossRef Search ADS   Parvin P. ( 2015). Is deliberative democracy feasible? Political disengagement and trust in liberal democratic states. The Monist , 9, 407– 423. doi: 10.1093/monist/onv021 Google Scholar CrossRef Search ADS   Posner R. ( 2004, January/February). Smooth sailing. Legal affairs. Retrieved from: https://www.legalaffairs.org/issues/January-February-2004/feature_posner_janfeb04.msp Sniderman P. M., Brody R., Tetlock P. E. ( 1991). Reasoning and choice: Explorations in political psychology . New York, NY: Cambridge University Press. Google Scholar CrossRef Search ADS   Verba S., Schlozman K. L., Brady H. E. ( 1995). Voice and equality: Civic voluntarism in American politics . Cambridge, MA: Harvard University Press. Wojcieszak M. E., Baek Y. M., Delli Carpini M. X. ( 2009). What is really going on? Structure underlying face-to-face and online deliberation. Information, Communication and Society , 12, 1080– 1102. doi: 10.1080/13691180902725768 Google Scholar CrossRef Search ADS   Young I. M. ( 2000). Inclusion and democracy , Oxford: Oxford University Press. Zaller J. R. ( 1992). The nature and origins of mass opinion , Cambridge, MA: Cambridge University Press. Biographical Notes Henrik Serup Christensen is an Academy Research Fellow at the Social Science Research Institute (Samforsk) at Åbo Akademi University. His research interests include political behavior and the consequences for democracy. His current research project concerns democratic innovations and their impact on democratic legitimacy. Åsa von Schoultz (née Bengtsson) is a Professor of Political Science at University of Helsinki. She specializes in political behavior, with a specific focus on citizens’ preferences for political decision-making processes and on intraparty competition. She has extensive experience of survey research directed toward citizens and political elites, and she is a member of the steering groups of the Finnish National Election study and the Comparative Candidate Survey. She has coauthored the book The Nordic Voter (ECPR Press) and published in journals, such as European Journal of Political Research, West European Politics, Scandinavian Political Studies, Government and Opposition, and Parliamentary Affairs. Appendix Table A1 Coding of Variables and Descriptive Statistics Variable  Coding  N  Mean  SE  Min  Maximum  VIF  Dependent variable                 Q1  Political discussions for ordinary citizens should be arranged in support of representative democracy; answer on four-point scale (0 = totally disagree)  1,502  1.94  0.83  0.00  3.00  N/A   Q2  I myself would like to participate in political discussions arranged for ordinary citizens; answer on four-point scale (0 = totally disagree)  1,530  1.21  1.05  0.00  3.00  N/A   Support deliberative practices  Additive index based on Q1 and Q2 coded to vary between 0 and 1 (1 = highest extent of support)  1,469  0.53  0.27  0.00  1.00  N/A  Independent variables                 Left–right dimension  Where would you place yourself on a scale where 0 stands for left and 10 stands for right? Coded to vary between 0 and 1 (1 = right)  1,388  0.56  0.21  0.00  1.00  1.13   Nationalist–cosmopolitan dimension  Index based on answers to three proposals concerning the future of Finland: (1) a multicultural Finland where people are tolerant toward people from other countries; (2) a Finland where the rights of sexual minorities are strengthened; (3) a Finland with more immigration. All responses 0–10 (10 = very good proposal). Index coded to vary between 0 and 1 (1 = cosmopolitan; Cronbach’s alpha = .76)  1,587  0.56  0.22  0.00  1.00  1.16  Political awareness (moderator variables)                 Political knowledge  Number of correct answers to five factual questions on political matters; coded to vary between 0 and 1 (1 = 5 correct)  1,587  0.57  0.28  0.00  1.00  1.27   Political interest  How interested are you in politics? Answer on four-graded scale coded to vary between 0 and 1 (1 = very interested)  1,587  0.61  0.29  0.00  1.00  1.31  Control variables                 Age  2015-Birth year, divided by 100  1,587  0.51  0.20  0.18  0.94  1.31   Gender  Respondent’s gender; 0 = female, 1 = male  1,587  0.50  0.50  0.00  1.00  1.08   Education  What is your highest educational attainment? Answer on seven-graded scale coded to vary between 0 and 1 (1 = highest educational attainment)  1,583  0.47  0.26  0.00  1.00  1.37   Employment  Which of the following options best describes your life situation? Answer included 12 possibilities, coded dichotomously with 1 = full-time employment (at least 32 hr per week), 0 = anything else  1,587  0.32  0.46  0.00  1.00  1.16   Satisfaction with democracy  How satisfied are you with the way democracy works in Finland? Answer on four-graded scale coded to vary between 0 and 1 (1 = completely satisfied)  1,568  0.60  0.19  0.00  1.00  1.18   Internal political efficacy  Sometimes politics is so complicated that I do not really understand what is going on. Answer on four-graded scale coded to vary between 0 and 1 (1 = disagree completely)  1,553  0.36  0.32  0.00  1.00  1.24   External political efficacy  Index based on extent of agreement to four questions: (1) politicians do not care about the opinions of ordinary people; (2) I cannot influence what the country’s government and parliament decide; (3) parties are only interested in peoples’ votes, not their opinions; (4) it does not matter what parties are in government, the policies still do not change. All answers on four-graded scales “Totally agree” to “Totally disagree;” index coded to vary between 0 and 1 (1 = highest external efficacy; Cronbach’s alpha = .77)  1,459  0.48  0.24  0.00  1.00  1.53   Political trust  How much do you trust each of the following actors? All responses 0–10 (10 = complete trust). Index based on trust in Finnish President, Political parties, Parliament, Finnish Government, and Politicians; coded to vary between 0 and 1 (1 complete trust; Cronbach’s alpha = .91)  1,587  0.63  0.17  0.00  1.00  1.30  Variable  Coding  N  Mean  SE  Min  Maximum  VIF  Dependent variable                 Q1  Political discussions for ordinary citizens should be arranged in support of representative democracy; answer on four-point scale (0 = totally disagree)  1,502  1.94  0.83  0.00  3.00  N/A   Q2  I myself would like to participate in political discussions arranged for ordinary citizens; answer on four-point scale (0 = totally disagree)  1,530  1.21  1.05  0.00  3.00  N/A   Support deliberative practices  Additive index based on Q1 and Q2 coded to vary between 0 and 1 (1 = highest extent of support)  1,469  0.53  0.27  0.00  1.00  N/A  Independent variables                 Left–right dimension  Where would you place yourself on a scale where 0 stands for left and 10 stands for right? Coded to vary between 0 and 1 (1 = right)  1,388  0.56  0.21  0.00  1.00  1.13   Nationalist–cosmopolitan dimension  Index based on answers to three proposals concerning the future of Finland: (1) a multicultural Finland where people are tolerant toward people from other countries; (2) a Finland where the rights of sexual minorities are strengthened; (3) a Finland with more immigration. All responses 0–10 (10 = very good proposal). Index coded to vary between 0 and 1 (1 = cosmopolitan; Cronbach’s alpha = .76)  1,587  0.56  0.22  0.00  1.00  1.16  Political awareness (moderator variables)                 Political knowledge  Number of correct answers to five factual questions on political matters; coded to vary between 0 and 1 (1 = 5 correct)  1,587  0.57  0.28  0.00  1.00  1.27   Political interest  How interested are you in politics? Answer on four-graded scale coded to vary between 0 and 1 (1 = very interested)  1,587  0.61  0.29  0.00  1.00  1.31  Control variables                 Age  2015-Birth year, divided by 100  1,587  0.51  0.20  0.18  0.94  1.31   Gender  Respondent’s gender; 0 = female, 1 = male  1,587  0.50  0.50  0.00  1.00  1.08   Education  What is your highest educational attainment? Answer on seven-graded scale coded to vary between 0 and 1 (1 = highest educational attainment)  1,583  0.47  0.26  0.00  1.00  1.37   Employment  Which of the following options best describes your life situation? Answer included 12 possibilities, coded dichotomously with 1 = full-time employment (at least 32 hr per week), 0 = anything else  1,587  0.32  0.46  0.00  1.00  1.16   Satisfaction with democracy  How satisfied are you with the way democracy works in Finland? Answer on four-graded scale coded to vary between 0 and 1 (1 = completely satisfied)  1,568  0.60  0.19  0.00  1.00  1.18   Internal political efficacy  Sometimes politics is so complicated that I do not really understand what is going on. Answer on four-graded scale coded to vary between 0 and 1 (1 = disagree completely)  1,553  0.36  0.32  0.00  1.00  1.24   External political efficacy  Index based on extent of agreement to four questions: (1) politicians do not care about the opinions of ordinary people; (2) I cannot influence what the country’s government and parliament decide; (3) parties are only interested in peoples’ votes, not their opinions; (4) it does not matter what parties are in government, the policies still do not change. All answers on four-graded scales “Totally agree” to “Totally disagree;” index coded to vary between 0 and 1 (1 = highest external efficacy; Cronbach’s alpha = .77)  1,459  0.48  0.24  0.00  1.00  1.53   Political trust  How much do you trust each of the following actors? All responses 0–10 (10 = complete trust). Index based on trust in Finnish President, Political parties, Parliament, Finnish Government, and Politicians; coded to vary between 0 and 1 (1 complete trust; Cronbach’s alpha = .91)  1,587  0.63  0.17  0.00  1.00  1.30  VIF = Variance Inflation Factor View Large Table A1 Coding of Variables and Descriptive Statistics Variable  Coding  N  Mean  SE  Min  Maximum  VIF  Dependent variable                 Q1  Political discussions for ordinary citizens should be arranged in support of representative democracy; answer on four-point scale (0 = totally disagree)  1,502  1.94  0.83  0.00  3.00  N/A   Q2  I myself would like to participate in political discussions arranged for ordinary citizens; answer on four-point scale (0 = totally disagree)  1,530  1.21  1.05  0.00  3.00  N/A   Support deliberative practices  Additive index based on Q1 and Q2 coded to vary between 0 and 1 (1 = highest extent of support)  1,469  0.53  0.27  0.00  1.00  N/A  Independent variables                 Left–right dimension  Where would you place yourself on a scale where 0 stands for left and 10 stands for right? Coded to vary between 0 and 1 (1 = right)  1,388  0.56  0.21  0.00  1.00  1.13   Nationalist–cosmopolitan dimension  Index based on answers to three proposals concerning the future of Finland: (1) a multicultural Finland where people are tolerant toward people from other countries; (2) a Finland where the rights of sexual minorities are strengthened; (3) a Finland with more immigration. All responses 0–10 (10 = very good proposal). Index coded to vary between 0 and 1 (1 = cosmopolitan; Cronbach’s alpha = .76)  1,587  0.56  0.22  0.00  1.00  1.16  Political awareness (moderator variables)                 Political knowledge  Number of correct answers to five factual questions on political matters; coded to vary between 0 and 1 (1 = 5 correct)  1,587  0.57  0.28  0.00  1.00  1.27   Political interest  How interested are you in politics? Answer on four-graded scale coded to vary between 0 and 1 (1 = very interested)  1,587  0.61  0.29  0.00  1.00  1.31  Control variables                 Age  2015-Birth year, divided by 100  1,587  0.51  0.20  0.18  0.94  1.31   Gender  Respondent’s gender; 0 = female, 1 = male  1,587  0.50  0.50  0.00  1.00  1.08   Education  What is your highest educational attainment? Answer on seven-graded scale coded to vary between 0 and 1 (1 = highest educational attainment)  1,583  0.47  0.26  0.00  1.00  1.37   Employment  Which of the following options best describes your life situation? Answer included 12 possibilities, coded dichotomously with 1 = full-time employment (at least 32 hr per week), 0 = anything else  1,587  0.32  0.46  0.00  1.00  1.16   Satisfaction with democracy  How satisfied are you with the way democracy works in Finland? Answer on four-graded scale coded to vary between 0 and 1 (1 = completely satisfied)  1,568  0.60  0.19  0.00  1.00  1.18   Internal political efficacy  Sometimes politics is so complicated that I do not really understand what is going on. Answer on four-graded scale coded to vary between 0 and 1 (1 = disagree completely)  1,553  0.36  0.32  0.00  1.00  1.24   External political efficacy  Index based on extent of agreement to four questions: (1) politicians do not care about the opinions of ordinary people; (2) I cannot influence what the country’s government and parliament decide; (3) parties are only interested in peoples’ votes, not their opinions; (4) it does not matter what parties are in government, the policies still do not change. All answers on four-graded scales “Totally agree” to “Totally disagree;” index coded to vary between 0 and 1 (1 = highest external efficacy; Cronbach’s alpha = .77)  1,459  0.48  0.24  0.00  1.00  1.53   Political trust  How much do you trust each of the following actors? All responses 0–10 (10 = complete trust). Index based on trust in Finnish President, Political parties, Parliament, Finnish Government, and Politicians; coded to vary between 0 and 1 (1 complete trust; Cronbach’s alpha = .91)  1,587  0.63  0.17  0.00  1.00  1.30  Variable  Coding  N  Mean  SE  Min  Maximum  VIF  Dependent variable                 Q1  Political discussions for ordinary citizens should be arranged in support of representative democracy; answer on four-point scale (0 = totally disagree)  1,502  1.94  0.83  0.00  3.00  N/A   Q2  I myself would like to participate in political discussions arranged for ordinary citizens; answer on four-point scale (0 = totally disagree)  1,530  1.21  1.05  0.00  3.00  N/A   Support deliberative practices  Additive index based on Q1 and Q2 coded to vary between 0 and 1 (1 = highest extent of support)  1,469  0.53  0.27  0.00  1.00  N/A  Independent variables                 Left–right dimension  Where would you place yourself on a scale where 0 stands for left and 10 stands for right? Coded to vary between 0 and 1 (1 = right)  1,388  0.56  0.21  0.00  1.00  1.13   Nationalist–cosmopolitan dimension  Index based on answers to three proposals concerning the future of Finland: (1) a multicultural Finland where people are tolerant toward people from other countries; (2) a Finland where the rights of sexual minorities are strengthened; (3) a Finland with more immigration. All responses 0–10 (10 = very good proposal). Index coded to vary between 0 and 1 (1 = cosmopolitan; Cronbach’s alpha = .76)  1,587  0.56  0.22  0.00  1.00  1.16  Political awareness (moderator variables)                 Political knowledge  Number of correct answers to five factual questions on political matters; coded to vary between 0 and 1 (1 = 5 correct)  1,587  0.57  0.28  0.00  1.00  1.27   Political interest  How interested are you in politics? Answer on four-graded scale coded to vary between 0 and 1 (1 = very interested)  1,587  0.61  0.29  0.00  1.00  1.31  Control variables                 Age  2015-Birth year, divided by 100  1,587  0.51  0.20  0.18  0.94  1.31   Gender  Respondent’s gender; 0 = female, 1 = male  1,587  0.50  0.50  0.00  1.00  1.08   Education  What is your highest educational attainment? Answer on seven-graded scale coded to vary between 0 and 1 (1 = highest educational attainment)  1,583  0.47  0.26  0.00  1.00  1.37   Employment  Which of the following options best describes your life situation? Answer included 12 possibilities, coded dichotomously with 1 = full-time employment (at least 32 hr per week), 0 = anything else  1,587  0.32  0.46  0.00  1.00  1.16   Satisfaction with democracy  How satisfied are you with the way democracy works in Finland? Answer on four-graded scale coded to vary between 0 and 1 (1 = completely satisfied)  1,568  0.60  0.19  0.00  1.00  1.18   Internal political efficacy  Sometimes politics is so complicated that I do not really understand what is going on. Answer on four-graded scale coded to vary between 0 and 1 (1 = disagree completely)  1,553  0.36  0.32  0.00  1.00  1.24   External political efficacy  Index based on extent of agreement to four questions: (1) politicians do not care about the opinions of ordinary people; (2) I cannot influence what the country’s government and parliament decide; (3) parties are only interested in peoples’ votes, not their opinions; (4) it does not matter what parties are in government, the policies still do not change. All answers on four-graded scales “Totally agree” to “Totally disagree;” index coded to vary between 0 and 1 (1 = highest external efficacy; Cronbach’s alpha = .77)  1,459  0.48  0.24  0.00  1.00  1.53   Political trust  How much do you trust each of the following actors? All responses 0–10 (10 = complete trust). Index based on trust in Finnish President, Political parties, Parliament, Finnish Government, and Politicians; coded to vary between 0 and 1 (1 complete trust; Cronbach’s alpha = .91)  1,587  0.63  0.17  0.00  1.00  1.30  VIF = Variance Inflation Factor View Large © The Author(s) 2018. Published by Oxford University Press on behalf of The World Association for Public Opinion Research. All rights reserved. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png International Journal of Public Opinion Research Oxford University Press

Ideology and Deliberation: An Analysis of Public Support for Deliberative Practices in Finland

Loading next page...
 
/lp/ou_press/ideology-and-deliberation-an-analysis-of-public-support-for-cxDiS0n3FJ
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of The World Association for Public Opinion Research. All rights reserved.
ISSN
0954-2892
eISSN
1471-6909
D.O.I.
10.1093/ijpor/edx022
Publisher site
See Article on Publisher Site

Abstract

The introduction of deliberative practices has been proposed as a method for revitalizing representative democracies suffering from decreasing political involvement and increasing fragmentation (Dryzek, 2000; Grönlund, Bächtiger, & Setälä, 2014; Parkinson & Mansbridge, 2012). Some scholars, however, question whether citizens are willing to engage in deliberative practices (Bartels, 2003; Hibbing & Theiss-Morse, 2002; Posner, 2004). A related concern is that deliberative practices may not appeal equally to all groups of citizens, and consequently jeopardize the principle of inclusion of all relevant arguments and experiences in deliberation (Young, 2000). Whether deliberative practices can help revitalize democracy partly depends on the extent to which support for these practices is related to ideological predispositions. Previous studies suggest that ideological predispositions systematically influence individual attitudes toward public involvement in political processes (Bengtsson & Mattila, 2009; Dalton & Welzel, 2014; Inglehart, 1997; Kriesi, 2010). If a corresponding pattern were to be found for deliberative practices, the result could be detrimental to the deliberative focus on inclusiveness. An ideological bias among supporters might translate into overrepresentation of certain ideological perspectives in deliberative events. Furthermore, if support for deliberative practices depends on strong ideological predispositions, the prospective participants in deliberation are likely to behave like activists—eager to pursue their established goals by any means necessary (Mackuen, Wolak, Keele, & Marcus, 2010, p. 440)—rather than deliberative citizens, who are considerate, balanced, open-minded, and willing to compromise (Levine & Nierras, 2007; MacKuen et al., 2010). If the composition of deliberative events is ideologically unrepresentative, it can also result in harmful group composition effects that affect the final outcome (Baek, Wojcieszak, & Delli Carpini, 2012; Barabas, 2004; Gastil, Black, & Moscovitz, 2008). It is therefore important to establish how ideological predispositions shape attitudes toward the use of deliberative practices in society. Based on these considerations, we aim to study how support for deliberative practices is linked to ideological predispositions on the traditional left–right dimension and the nationalist–cosmopolitan values dimension. Jacobs, Cook, & Delli Carpini (2009) and Neblo, Esterling, Kennedy, Lazer, & Sokhey (2010) constitute a couple of pioneering studies that examine the willingness to deliberate in the United States, and show that politically marginalized groups are willing to deliberate (Neblo et al. 2010), and that self-identified liberals are more likely to engage in deliberation than people who consider themselves to be more conservatives (Jacobs et al. 2009). However, these studies are not necessarily applicable outside the American context. Moreover, Jacobs et al. (2009) restrict their analysis to a one-dimensional conceptualization of ideology that fails to capture contemporary ideological richness. We use the Finnish National Election Study from 2015 (FNES2015) to demonstrate that individuals who are left-leaning and/or adhere to cosmopolitan values support deliberative practices. In line with the work of Zaller (1992), we also find that political awareness furthers the ability to formulate opinions that are consistent with ideological predispositions. Deliberation and Ideological Predispositions Important distinctions can be made between different theories of deliberative democracy (Bächtiger, Niemeyer, Neblo, Steenbergen, & Steiner, 2010), but they share an emphasis on discussions as prerequisites for enlightened political decision-making. In this study, we use the concept of deliberative practices to refer to a broad range of measures aiming to improve the deliberative capacity of political systems by promoting discussions and exchanges of arguments in society (Parkinson & Mansbridge, 2012). There are various kinds of public discussions, some of which fail to fulfill the most stringent definitions of deliberation (Levine & Nierras, 2007; Karpowitz & Mendelberg, 2007). Nevertheless, we here use a broad definition of deliberation, as the aim is to include all processes that may enhance the deliberative capacity of political systems, even when they do not completely adhere to the ideal of deliberation.1 1Contrary to others (Christensen, Himmelroos, & Grönlund, 2017; Jacobs et al., 2009), we do not use the term discursive practices to stress that the discussions should involve more than “just talk,” that is, they should be connected to existing political processes and, at least potentially, affect their outcomes, which is in line with the systemic approach to deliberation (Parkinson & Mansbridge, 2012). Our approach corresponds to the focus of Levine and Nierras (2007) on public deliberation that aims to resolve real problems. Successful deliberation ideally should include all relevant arguments and opinions, and be a formative process (Goodin, 2004; Mansbridge et al., 2010; Parkinson, 2006). To ensure that all relevant positions and perspectives are included, the ideological predispositions of participants in deliberative practices ought to reflect the ideological composition of society (Gastil 2008; Gastil et al., 2008; Young, 2000). It is however questionable whether deliberative practices appeal to all segments of society, especially when it comes to the ideological predispositions of presumptive supporters. From a theoretical perspective, deliberation has been accused of harboring an elitist notion of democracy that does not appeal to common people, and of promoting liberal values over more conservative ones (Blattberg 2003; Parvin, 2015). Empirical evidence also suggests that ideological predispositions affect support for deliberative practices. First, it is well established that politically right-leaning individuals are more conservative when it comes to institutional changes (Anderson & Singer, 2008, p. 574). Although deliberation may help preserve the status quo by blocking drastic political reforms, those on the right may consequently oppose such novel methods of decision-making. Furthermore, previous studies have found that people are generally consistent in their preferences of decision-making procedures (Bengtsson & Christensen, 2016; Bengtsson & Mattila, 2009; Font, Wojcieszak, & Navarro, 2015). Some studies even suggest that leftists show stronger support for participatory and direct democratic processes (Bengtsson & Mattila, 2009; Esaiasson, Gilljam, & Persson, 2010), and they may likewise be more inclined toward deliberative practices. Second, the progression of what has been termed postmodernist, libertarian, or cosmopolitan values in Western democracies is tied to support for participatory reforms (Dalton & Welzel, 2014; Inglehart, 1997; Kriesi, 2010). According to Inglehart (1977, 1997), postmaterialist generations crave innovative forms of political participation to complement the traditional representative structures. Third, studies on deliberation show that certain ideological positions dominate deliberative events. Jacobs et al. (2009, p. 53) found that liberals are more likely than conservatives to engage in deliberation, and a study of recruitment to a deliberative mini-public shows that attrition is higher among citizens with anti-immigration attitudes (Karjalainen & Rapeli, 2015).2 2Moreover, Gastil, Bacci, and Dollinger (2010) find that outcomes of deliberation tend to be biased toward cosmopolitan and collectivist beliefs. Political ideology, as we use the term here, departs from dominant issue divides with political relevance (Bartolini, 2011; Deegan-Krause, 2007). Traditionally, ideology has been considered a one-dimensional phenomenon distributed on a left–right continuum (Downs, 1957), which in a European context has its roots in historical class cleavages (Lipset & Rokkan, 1967; Knutsen, 2006) that divided politics into the liberal–conservative right and the social–democratic or socialist left. Although this left–right dimension has largely revolved around issues of redistribution, over time it has developed into a meta-dimension that has structured party systems and the ideological beliefs of citizens (Mair, 2007). However, several works contend that ideology can no longer be considered one dimensional, and that a dimension concerning immaterial values rather than issues of redistribution has complemented the left–right dimension. Scholars here use various labels to denote the poles on closely related value dimensions, such as libertarian/authoritarian (Flanagan & Lee, 2003; Kitschelt, 1994; Kitschelt & McGann, 1995), postmaterialist/materialist (Inglehart, 1977), green–alternative–libertarian (GAL)/traditional–authoritarian–nationalist (TAN) (Hooghe, Marks, & Wilson, 2002) or cosmopolitans/nationalists (Norris & Inglehart, 2009).3 3Other examples of labels used are self-expression/survival (Inglehart & Welzel, 2005), libertarian, universalistic/traditionalist–communitarian (Bornschier, 2010), integration/demarcation (Kriesi, 2010). Despite the different labels and empirical operationalizations, these scholars share a belief that when examining the causes and consequences of ideological predispositions, it should be acknowledged that ideology today is a multidimensional phenomenon. For this reason, we also examine the relationship between the dimension that we label national–cosmopolitan values and support for deliberative practices. The theoretical and empirical reasons presented above suggest that ideological predispositions on both dimensions are likely to influence attitudes toward deliberative practices, as individuals with more leftist and/or cosmopolitan values are likely to support the introduction of such mechanisms to complement representative decision-making. These suggestions are in line with the established finding that general political predispositions function as guides or heuristic devices when people structure their views on more specific issues (Feldman & Zaller, 1992; Goren, Schoen, Reifler, Scotto, & Chittick, 2016; Sniderman, Brody, & Tetlock, 1991). However, the impact of these predispositions may be less straightforward than suggested above. As convincingly demonstrated by Zaller (1992), people need a certain level of political awareness (attention and understanding) to translate their ideological predispositions into opinions on specific issues.4 4We would like to thank one of the anonymous reviewers for this suggestion. Politically aware citizens are more likely to organize their political ideas in an ideologically consistent manner5 5Or to have “politically sophisticated” belief systems (Luskin, 1987). (Gastil & Dillard, 1999), that is, they are more likely to hold political beliefs that are in line with their ideological predispositions. This implies that it is among the politically aware that we would expect to find a stronger link between ideological predispositions and attitudes toward deliberative practices. Based on these considerations, we examine the following hypotheses on the relationship between ideological predispositions and support for deliberative practices in the empirical section: H1: Leftist values on the traditional left–right dimension are positively associated with positive attitudes toward deliberative practices. H2: Cosmopolitan values on the nationalist–cosmopolitan dimension are positively associated with positive attitudes toward deliberative practices. H3: Political awareness moderates the relationship between ideological predispositions and attitudes toward deliberative practices. Data and Operationalizations The data come from the Finnish National Election Study conducted following the national elections on April 19, 2015 (FNES2015), which involved a cross-sectional survey using a two-stage process with introductory face-to-face interviews of 1,602 respondents during April–June 2015.6 6See https://services.fsd.uta.fi/catalogue/FSD3067?lang=en&study_language=en for more information on FNES2015. Although the cross-sectional data do not make it possible to settle causal effects, FNES2015 is ideal for examining associations between ideological predispositions and support for deliberative practices, as the survey includes suitable measures for all variables. Finland constitutes an opportune case, as, like most European democracies, it has an established tradition of representative decision-making complemented by a restrictive use of consultative referendums. However, studies show demand for more citizen involvement in political decision-making (Bengtsson & Christensen, 2016), echoing similar sentiments across the continent. In this sense, even though the generalizability of the results may be limited, Finland presents a paradigmatic case for studying attitudes toward deliberative practices. The general public has a limited familiarity with formal deliberative institutions such as deliberative mini-publics.7 7Deliberative mini-publics have mainly been used in the form of academic experiments in Finland. However, this does not preclude the public from forming opinions on the presumptive use of deliberative practices in the form of public discussions to complement representative decision-making. All variables are coded to vary between 0 and 1, with 1 indicating the highest value of the variable in question. More information on the coding of variables, and descriptive statistics, is found in Table A1. Dependent Variable The dependent variable is the attitude toward the use of deliberative practices as a complement to representative decision-making.8 8We use the same measure as Christensen et al. (2017), who study whether participation in a deliberative mini-public enhances support for arranging public discussions in representative democracy. The measure is based on responses to two statements: (1) Political discussions for ordinary citizens should be arranged in support of representative democracy, and (2) I myself would like to participate in political discussions arranged for ordinary citizens. The responses to both items were indicated using a four-point Likert scale (totally agree to totally disagree), and these scores were subsequently combined to form an additive index (M = 0.53). A closer inspection of the individual questions (not shown) reveals that about 75% tend to agree that discussions should be arranged to support representative democracy, while only about 42% are willing to take part in them.9 9We fitted ordinal regression models examining the ideological impact on the separate questions. The associations with the ideological predispositions were generally similar with the exception that the direct effect of cosmopolitanism is not significant (B = 0.55, p = .084) for the question on introducing political discussions for ordinary citizens. Our operationalization is based on a broad conceptualization of deliberative practices in line with the definition offered above, and may therefore overestimate the extent of support for a more narrowly defined interpretation. Some may even object to the label deliberative practices, as the questions do not refer to genuine deliberative institutions such as mini-publics (Grönlund et al., 2014). However, as most people are unfamiliar with these formal deliberative institutions, it makes little sense to ask respondents about their attitudes toward them directly. Previous survey research on how citizens see public deliberation has also relied on similarly wide understandings of the topic (Baek et al., 2012; Wojcieszak, Baek, & Delli Carpini, 2009). A broad conceptualization is warranted in this case, as the objective is not to identify support for specific arrangements (i.e., mini-publics), but to examine support for the use of talk-centric measures to improve representative decision making. Our conceptualization is closer to what Bächtiger et al. (2010) entitle Type II deliberation, which involves more flexible discourses than the more narrowly defined Type I deliberation, which focuses on rational discourse and procedural aspects. Nevertheless, our operationalization entails a focus on discussions that are connected to formal political decision-making in line with Levine and Nierras (2007). Independent Variables To capture ideological predispositions, we focused on the traditional left–right dimension (Mair, 2007) and the increasingly important dimension of what we here refer to as nationalist–cosmopolitan values. The position on the left–right dimension was measured with a variable, where the respondent indicated his or her placement on an 11-point left–right scale (recoded to vary between 0 and 1 (1 = right). The degree of national–cosmopolitan values was captured with an index based on three questions measuring the attitudes of the respondents toward multiculturalism in general and tolerance toward two out-groups: sexual minorities and immigrants (Cronbach’s alpha = .76, 1 = cosmopolitan).10 10One of the items included in our index concerns the rights of sexual minorities rather than cosmopolitan/nationalist values, but this has been shown to correlate strongly with similar dimensions (Inglehart, 1997). Moderator Variables Political awareness is conceptualized as the understanding of and attentiveness toward politics, and operationalized by two different variables. The first is factual political knowledge, which is in accordance with the operationalization used by Zaller (1992). This was measured with an index, where respondents received a point for each correct answer to five factual questions on international and domestic Finnish political matters (0 = no correct answers). The second variable is political interest, which has been considered a suitable indicator (Zaller 1992, pp. 333–4) and has been used in previous research on similar topics (Verba, Schlozman, & Brady, 1995). We measured this with a common question on the topic, where respondents indicated how interested they were in politics on a four-point scale (0 = lowest interest). Control Variables To ascertain the robustness of the results, we included sociodemographic characteristics (age, gender, education, and full-time employment) and attitudes toward the political system (satisfaction with democracy, political trust, and internal and external political efficacy). All of these variables have been demonstrated to influence the propensity for political participation in general (Bengtsson & Christensen, 2016; Verba et al., 1995) and in regard to deliberation (Karjalainen & Rapeli, 2015; Neblo et al., 2010). Age in years was divided by 100 to approximate the 0–1 coding of all independent variables. In regard to education, the respondents indicated the highest level of education completed on an eight-point scale (0 = lowest level of education). The other two characteristics are dichotomous. For gender, 0 indicates female and 1 male, whereas for employment, 1 corresponds to full-time employment, and all other occupations (student, retired, half-time employment, etc.) are coded 0. For satisfaction with democracy, respondents indicated their extent of satisfaction with how democracy functions in Finland on a four-point scale (0 = lowest satisfaction). We measured internal political efficacy with a question, where respondents indicated the extent to which they agree with the statement, “Sometimes politics seems so complicated that I can’t really understand what is going on,” on a four-point Likert scale (0 = lowest efficacy). External efficacy was measured with an index based on four statements, where respondents indicated their views on the responsiveness of the political system on a four-point Likert scale (0 = lowest efficacy; Cronbach’s alpha = .77). For political trust, we use an index based on five questions, where respondents indicated their level of trust on an 11-point scale (0–10) in regard to the Finnish parliament, politicians, political parties, the Finnish president, and the government (Cronbach’s alpha = .91). Methods of Analysis A Shapiro–Wilk test shows that the dependent variable follows a normal distribution, as the null hypothesis cannot be rejected (W = 0.998; p = .0812). We therefore used linear regression with robust standard errors to examine our hypotheses. To examine H1 and H2 concerning the associations between ideological factors and support for deliberative practices, the first model, M1, only included the two ideological factors, left–right and nationalist–cosmopolitan, whereas M2 also included the measures of political awareness and control variables to verify that the linkages persist when controlling for other factors. To examine H3 and the moderating effect of political awareness, in M3, we included four interaction terms constituted by the two ideological factors and the two factors of political awareness in the regression model. Empirical Analysis Table 1 displays the results of the regression analyses. Table 1 Multivariate Linear Regression Models Examining Support for Deliberative Practices Variables  M1   M2   M3     Coefficient  Robust SE  p  Coefficient  Robust SE  p  Coefficient  Robust SE  p  Independent variables                        Left–right  −0.22  (0.04)  .000  −0.21  (0.04)  .000  0.12  (0.10)  .235      Nationalist–cosmopolitan  0.15  (0.04)  .000  0.12  (0.04)  .002  0.15  (0.13)  .230  Political awareness                        Political knowledge        −0.02  (0.03)  .428  0.14  (0.12)  .245      Political interest        0.39  (0.03)  .000  0.53  (0.11)  .000  Interactions                        Political knowledge × left–right              −0.24  (0.13)  .072      Political interest × left–right              −0.25  (0.12)  .042      Political knowledge × nationalist–cosmopolitan              −0.06  (0.15)  .693      Political interest × nationalist– cosmopolitan              0.00  (0.14)  .977  Control variables                        Age/100        −0.09  (0.05)  .047  −0.10  (0.05)  .026      Gender (1 = male)        −0.03  (0.02)  .066  −0.03  (0.02)  .093      Education        −0.06  (0.03)  .102  −0.05  (0.03)  .117      Full-time employment (1 = yes)        −0.01  (0.02)  .376  −0.02  (0.02)  .300      Satisfaction with democracy        −0.05  (0.05)  .288  −0.05  (0.05)  .269      Internal efficacy        −0.07  (0.03)  .009  −0.08  (0.03)  .006      External efficacy        0.07  (0.04)  .135  0.07  (0.04)  .117      Political trust        −0.09  (0.06)  .118  −0.09  (0.06)  .112      Constant  0.58  (0.03)  .000  0.53  (0.05)  .000  0.33  (0.09)  .000  F  (21,292) = 31.25, p < .001  (121,201) = 28.41, p < .001  (161,197) = 24.21, p < .001  Adjusted R2  0.05  0.19  0.19  N  1,295  1,214  1,214  Variables  M1   M2   M3     Coefficient  Robust SE  p  Coefficient  Robust SE  p  Coefficient  Robust SE  p  Independent variables                        Left–right  −0.22  (0.04)  .000  −0.21  (0.04)  .000  0.12  (0.10)  .235      Nationalist–cosmopolitan  0.15  (0.04)  .000  0.12  (0.04)  .002  0.15  (0.13)  .230  Political awareness                        Political knowledge        −0.02  (0.03)  .428  0.14  (0.12)  .245      Political interest        0.39  (0.03)  .000  0.53  (0.11)  .000  Interactions                        Political knowledge × left–right              −0.24  (0.13)  .072      Political interest × left–right              −0.25  (0.12)  .042      Political knowledge × nationalist–cosmopolitan              −0.06  (0.15)  .693      Political interest × nationalist– cosmopolitan              0.00  (0.14)  .977  Control variables                        Age/100        −0.09  (0.05)  .047  −0.10  (0.05)  .026      Gender (1 = male)        −0.03  (0.02)  .066  −0.03  (0.02)  .093      Education        −0.06  (0.03)  .102  −0.05  (0.03)  .117      Full-time employment (1 = yes)        −0.01  (0.02)  .376  −0.02  (0.02)  .300      Satisfaction with democracy        −0.05  (0.05)  .288  −0.05  (0.05)  .269      Internal efficacy        −0.07  (0.03)  .009  −0.08  (0.03)  .006      External efficacy        0.07  (0.04)  .135  0.07  (0.04)  .117      Political trust        −0.09  (0.06)  .118  −0.09  (0.06)  .112      Constant  0.58  (0.03)  .000  0.53  (0.05)  .000  0.33  (0.09)  .000  F  (21,292) = 31.25, p < .001  (121,201) = 28.41, p < .001  (161,197) = 24.21, p < .001  Adjusted R2  0.05  0.19  0.19  N  1,295  1,214  1,214  Note: Entries are coefficients from linear regression analyses with robust standard errors in parentheses. Table 1 Multivariate Linear Regression Models Examining Support for Deliberative Practices Variables  M1   M2   M3     Coefficient  Robust SE  p  Coefficient  Robust SE  p  Coefficient  Robust SE  p  Independent variables                        Left–right  −0.22  (0.04)  .000  −0.21  (0.04)  .000  0.12  (0.10)  .235      Nationalist–cosmopolitan  0.15  (0.04)  .000  0.12  (0.04)  .002  0.15  (0.13)  .230  Political awareness                        Political knowledge        −0.02  (0.03)  .428  0.14  (0.12)  .245      Political interest        0.39  (0.03)  .000  0.53  (0.11)  .000  Interactions                        Political knowledge × left–right              −0.24  (0.13)  .072      Political interest × left–right              −0.25  (0.12)  .042      Political knowledge × nationalist–cosmopolitan              −0.06  (0.15)  .693      Political interest × nationalist– cosmopolitan              0.00  (0.14)  .977  Control variables                        Age/100        −0.09  (0.05)  .047  −0.10  (0.05)  .026      Gender (1 = male)        −0.03  (0.02)  .066  −0.03  (0.02)  .093      Education        −0.06  (0.03)  .102  −0.05  (0.03)  .117      Full-time employment (1 = yes)        −0.01  (0.02)  .376  −0.02  (0.02)  .300      Satisfaction with democracy        −0.05  (0.05)  .288  −0.05  (0.05)  .269      Internal efficacy        −0.07  (0.03)  .009  −0.08  (0.03)  .006      External efficacy        0.07  (0.04)  .135  0.07  (0.04)  .117      Political trust        −0.09  (0.06)  .118  −0.09  (0.06)  .112      Constant  0.58  (0.03)  .000  0.53  (0.05)  .000  0.33  (0.09)  .000  F  (21,292) = 31.25, p < .001  (121,201) = 28.41, p < .001  (161,197) = 24.21, p < .001  Adjusted R2  0.05  0.19  0.19  N  1,295  1,214  1,214  Variables  M1   M2   M3     Coefficient  Robust SE  p  Coefficient  Robust SE  p  Coefficient  Robust SE  p  Independent variables                        Left–right  −0.22  (0.04)  .000  −0.21  (0.04)  .000  0.12  (0.10)  .235      Nationalist–cosmopolitan  0.15  (0.04)  .000  0.12  (0.04)  .002  0.15  (0.13)  .230  Political awareness                        Political knowledge        −0.02  (0.03)  .428  0.14  (0.12)  .245      Political interest        0.39  (0.03)  .000  0.53  (0.11)  .000  Interactions                        Political knowledge × left–right              −0.24  (0.13)  .072      Political interest × left–right              −0.25  (0.12)  .042      Political knowledge × nationalist–cosmopolitan              −0.06  (0.15)  .693      Political interest × nationalist– cosmopolitan              0.00  (0.14)  .977  Control variables                        Age/100        −0.09  (0.05)  .047  −0.10  (0.05)  .026      Gender (1 = male)        −0.03  (0.02)  .066  −0.03  (0.02)  .093      Education        −0.06  (0.03)  .102  −0.05  (0.03)  .117      Full-time employment (1 = yes)        −0.01  (0.02)  .376  −0.02  (0.02)  .300      Satisfaction with democracy        −0.05  (0.05)  .288  −0.05  (0.05)  .269      Internal efficacy        −0.07  (0.03)  .009  −0.08  (0.03)  .006      External efficacy        0.07  (0.04)  .135  0.07  (0.04)  .117      Political trust        −0.09  (0.06)  .118  −0.09  (0.06)  .112      Constant  0.58  (0.03)  .000  0.53  (0.05)  .000  0.33  (0.09)  .000  F  (21,292) = 31.25, p < .001  (121,201) = 28.41, p < .001  (161,197) = 24.21, p < .001  Adjusted R2  0.05  0.19  0.19  N  1,295  1,214  1,214  Note: Entries are coefficients from linear regression analyses with robust standard errors in parentheses. M1 shows that both ideological variables have the expected relationships with support for deliberative practices, which is also the case for M2 after controlling for other variables. For the left–right dimension, the negative coefficient of −0.21 in M2 (p < .001) shows that being to the left on this ideological dimension is associated with stronger support for deliberative practices. For the nationalist–cosmopolitan dimension, the positive coefficient of 0.12 (p = .002) shows that a higher degree of cosmopolitanism is associated with stronger support for deliberative practices. Figure 1a and b shows what this entails for the associations between ideological factors and support for deliberative practices when holding all other factors at their mean values. Figure 1 View largeDownload slide Predicted support for deliberative practices Figure 1 View largeDownload slide Predicted support for deliberative practices For the left–right dimension, the predicted value of support for deliberative practices is about 0.66 for those furthest to the left, but this decreases to about 0.45 for the most extreme rightist respondents. For the nationalist–cosmopolitan dimension, the most nationalist respondents have a predicted level of support of about 0.47, which increases to 0.59 for the respondents with the most cosmopolitan values. This supports H1 and H2, as both ideological variables have the expected associations with level of support for using deliberative practices. M3 shows that there are no significant effects for the nationalist–cosmopolitan dimension. There is, however, a significant interaction effect between the left–right dimension and political interest and the interaction effect between the left–right dimension and political knowledge is marginally insignificant (B = –0.25, p = .072). As traditional significance tests are unreliable for establishing the substantial importance of interaction effects, we show the implications of both of these in Figure 2 (Bedeian & Mossholder, 1994; Brambor, Clark, & Golder, 2006).11 11The exclusion of the remaining nonsignificant interaction effects is also warranted by their weak coefficients (see M3), meaning their substantial impact is in any case limited. Even if we run separate analyses for all interaction effects, the two nonsignificant interaction terms do not gain significance and the substantive impacts remain limited. We therefore remain confident that the two effects we focus on are the most relevant. To allow for a comprehensive examination of what the plots signify, we included both marginal effects of left–right ideology depending on political awareness and developments in predicted support for deliberative practices for those furthest to the left and those furthest to the right. Figure 2 View largeDownload slide Interaction effects (marginal effects and predicted support) Figure 2 View largeDownload slide Interaction effects (marginal effects and predicted support) The substantive interpretations were similar regardless of whether we measured political awareness with political interest or political knowledge. At lower levels of political awareness, the associations between the left–right dimension and support for deliberative practices are negligible, but they strengthen when respondents become politically aware. Hence, leftist individuals are only supportive of deliberative practices when they also pay attention to what happens in politics. The results thereby partly support H3, as the associations between the left–right dimension and support for deliberative practices depend on the level of political awareness, whereas there are no moderating effects for the nationalist–cosmopolitan dimension. Implications of Results The main finding presented here is that support for deliberative practices is higher among individuals with leftist and/or cosmopolitan values. If this support were to be reflected in actual participation, it would challenge the inclusiveness of deliberative practices emphasized by several deliberative theorists (Mansbridge et al., 2010; Parkinson, 2006; Young, 2000).12 12Even if the results for our control variables suggest there are few significant differences in terms of sociodemographic characteristics, suggesting that support for deliberative practices is fairly egalitarian in terms of descriptive representation (James, 2008). When deliberative practices mainly attract participants who hold certain ideological positions, the result can undermine both the quality and legitimacy of the output of deliberative practices (Dryzek, 2000; Goodin, 2004; James, 2008). Our findings support previous studies that also found ideological predispositions to be associated with positive attitudes toward participatory mechanisms in general (Bengtsson & Mattila, 2009; Dalton & Welzel, 2014; Inglehart, 1997) and deliberative practices in particular (Jacobs et al., 2009). The results thus support the notion that those with leftist/cosmopolitan values favor participatory political processes, albeit other mechanisms specifically related to the deliberative character of the processes under scrutiny may also be in play. Some studies suggest that issues traditionally associated with the right or nationalism are considered less amenable to public scrutiny (Gastil et al., 2010, pp. 4–5), causing individuals who agree with these issues from the outset to be less likely to support the use of deliberative practices. There is therefore a need to ensure that less popular, and maybe even more controversial, viewpoints are represented during prospective deliberative events. This is especially the case since previous studies suggest that group composition affects developments in attitudes during deliberation (Gastil et al., 2008; Karpowitz & Mendelberg, 2007). When participants tend to hold similar ideological predispositions, it is less likely that they are exposed to different ideological perspectives, which some studies suggest is a prerequisite for genuine deliberation (Barabas, 2004). It becomes more likely that participants act as activists rather than deliberative citizens (Levine & Nierras, 2007; MacKuen et al., 2010), which may ultimately undermine the legitimacy of such events. We have also demonstrated that the link between left–right values and support for deliberation is stronger among the politically aware. A greater understanding of, and attentiveness toward, politics hence furthers the ability to formulate opinions that are consistent with ideological predispositions (Zaller 1992). This finding supports the appeal made by Zaller (1992, pp. 308–309), that it is important for researchers to consider that people can vary greatly in their abilities to act on their values and interests. In this case, this entails that individuals with low political awareness may not articulate the attitude toward deliberative practices that their ideological predispositions would suggest. In connection to this, it is noteworthy that political awareness only moderated the relationship between the traditional left–right ideological dimension and support for deliberative practices. Although the implications of this are unclear, it shows that the new ideological dimension emphasizing cultural perspectives has a more straightforward relationship with support for deliberative practices, even for those with low levels of political awareness. It is, however, possible that this finding is because of the character of the operationalization of this ideological dimension. Whereas an established measure exists for the left–right dimension, for measuring the position on the nationalist–cosmopolitan dimension, we had to rely on issue positions in the form of attitudes toward various out-groups. Although common in the literature, it is possible that this approach leads people to think about specific interactions with the mentioned groups, rather than the abstract ideological principles that are of central concern here. This suggests a need to develop a measure of this cultural dimension that is equivalent to the left–right self-placement variable. These findings do not come without limitations. Although the Finnish case is instructive, it cannot be ascertained that similar findings would be found in other settings. Furthermore, it is worth reiterating that these findings do not necessarily entail support for specific deliberative institutions such as mini-publics, as our aim here was to examine support for a broader notion of deliberative practices. To assess whether similar results could be obtained in other contexts and for other practices, more research is needed. References Anderson C. J., Singer M. M. ( 2008). The sensitive left and the impervious right: Multilevel models and the politics of inequality, ideology, and legitimacy in Europe. Comparative Political Studies , 41, 564– 599. doi: 10.1177/0010414007313113 Google Scholar CrossRef Search ADS   Baek Y. M., Wojcieszak M., Delli Carpini M. X. ( 2012). Online versus face-to-face deliberation: Who? Why? What? With what effects? New Media and Society , 14, 363– 383. doi: 10.1177/1461444811413191 Google Scholar CrossRef Search ADS   Barabas J. ( 2004). How deliberation affects policy opinions. American Political Science Review , 98, 687– 701. doi: 10.1017/S0003055404041425 Google Scholar CrossRef Search ADS   Bartels L. ( 2003). Democracy with attitudes. In MacKuen M. B., Rabinowitz G. (Eds.), Electoral democracy  (pp. 48– 82). Ann Arbor, MI: University of Michigan Press. Bartolini S. ( 2011). Cleavages, social and political. In Badie B., Berg-Schlosser D., Morlino L. (Eds.), International encyclopedia of political science  (pp. 276– 282). Thousand Oaks: SAGE Publications. Bedeian A., Mossholder K. ( 1994). Simple question, not so simple answer: Interpreting interaction terms in moderated multiple regression. Journal of Management , 20, 159– 165. doi: 10.1177/014920639402000108 Google Scholar CrossRef Search ADS   Bengtsson Å., Christensen H. S. ( 2016). Ideals and actions: Do citizens’ patterns of political participation correspond to their conceptions of democracy? Government and Opposition , 51, 234– 260. doi: 10.1017/gov.2014.29 Google Scholar CrossRef Search ADS   Bengtsson Å., Mattila M. ( 2009). Direct democracy and its critics: Support for direct democracy and stealth democracy in Finland. West European Politics , 32, 1013– 1048. doi: 10.1080/01402380903065256 Google Scholar CrossRef Search ADS   Blattberg C. ( 2003). Patriotic, not deliberative, democracy. Critical Review of International Social and Political Philosophy , 6, 155– 174. doi: 10.1080/13698230510001702723 Google Scholar CrossRef Search ADS   Bornschier S. ( 2010). The New Cultural Divide and the Two-Dimensional Political Space in Western Europe. West European Politics , 33, 419– 444. Google Scholar CrossRef Search ADS   Brambor T., Clark W. R., Golder M. ( 2006). Understanding interaction models: Improving empirical analyses. Political Analysis , 14, 63– 82. doi: 10.1093/pan/mpi014 Google Scholar CrossRef Search ADS   Bächtiger A., Niemeyer S., Neblo M., Steenbergen M. R., Steiner J. ( 2010). Disentangling diversity in deliberative democracy: Competing theories, their blind spots and complementarities. Journal of Political Philosophy , 18, 32– 63. doi: 10.1111/j.1467-9760.2009.00342.x Google Scholar CrossRef Search ADS   Christensen H. S., Himmelroos S., Grönlund K. ( 2017). Does deliberation breed an appetite for discursive participation? Developments in attitudes towards deliberative practices after first-hand experience. Political Studies , 65, 64– 83. doi: 10.1177/0032321715617771 Google Scholar CrossRef Search ADS   Dalton R. J., Welzel C. (Eds.) ( 2014). The civic culture transformed: from allegiant to assertive citizens . Cambridge, MA: Cambridge University Press. Google Scholar CrossRef Search ADS   Deegan-Krause K. ( 2007). New dimensions of political cleavage. In Dalton R. J., Klingemann H. D. (Eds.), Oxford handbook of political science  (pp. 538– 556). Oxford: Oxford University Press. Downs A. ( 1957). An economic theory of democracy . New York: Harper. Dryzek J. ( 2000). Deliberative democracy and beyond. Liberals, critics, contestations . Oxford: Oxford University Press. Esaiasson P., Gilljam M., Persson M. ( 2010). Medborgarnas demokratiuppfattningar. In Holmberg S., Weibull L., Oscarsson H. (Eds.), Lycksalighetens ö  (pp. 269– 277). Göteborg: SOM-institutet, Göteborgs universitet. Feldman S., Zaller J. ( 1992). The political culture of ambivalence: Ideological responses to the welfare state. American Journal of Political Science , 36, 268– 307. doi: 10.2307/2111433 Google Scholar CrossRef Search ADS   Flanagan S. C., Lee A.-R. ( 2003). The new politics, cultural wars, and the authoritarian libertarian value change in advanced industrial democracies. Comparative Political Studies , 36, 235– 270. doi: 10.1177/0010414002250664 Google Scholar CrossRef Search ADS   Font J., Wojcieszak M., Navarro C. J. ( 2015). Participation, representation and expertise: Citizen preferences for political decision-making processes. Political Studies , 63(1 suppl), 153– 172. doi: 10.1111/1467-9248.12191 Google Scholar CrossRef Search ADS   Gastil J. ( 2008). Political communication and deliberation . London: Sage University Press. Google Scholar CrossRef Search ADS   Gastil J., Bacci C., Dollinger M. ( 2010). Is deliberation neutral? Patterns of attitude change during “The deliberative polls”. Journal of Public Deliberation , 6, article 3. Available at: http://www.publicdeliberation.net/jpd/vol6/iss2/art Gastil J., Dillard J. P. ( 1999). Increasing political sophistication through public deliberation. Political Communication , 16, 3– 23. doi: 10.1080/105846099198749 Google Scholar CrossRef Search ADS   Gastil J., Black L., Moscovitz K. ( 2008). Ideology, attitude change, and deliberation in small face-to-face groups. Political Communication , 25, 23– 46. doi: 10.1080/10584600701807836 Google Scholar CrossRef Search ADS   Goodin R. E. ( 2004). Representing diversity. British Journal of Political Science , 34, 453– 468. doi: 10.1017/S0007123404000134 Google Scholar CrossRef Search ADS   Goren P., Schoen H., Reifler J., Scotto T., Chittick W. ( 2016). A unified theory of value-based reasoning and US public opinion. Political Behavior , 38, 977– 997. doi: 10.1007/s11109-016-9344-x Google Scholar CrossRef Search ADS   Grönlund K., Bächtiger A., Setälä M. ( 2014). Deliberative mini-publics—Involving citizens in the democratic process . Colchester: ECPR Press. Hibbing J. R., Theiss-Morse E. ( 2002). Stealth democracy: Americans' beliefs about how government should work . Cambridge, MA: Cambridge University Press. Google Scholar CrossRef Search ADS   Hooghe L., Marks G., Wilson C. ( 2002). Does left/right structure party position on European integration? Comparative Political Studies , 35, 965– 989. doi: 10.1177/001041402236310 Google Scholar CrossRef Search ADS   Inglehart R. ( 1977). The silent revolution: Changing values and political styles among Western publics . Princeton, NJ: Princeton University Press. Inglehart R. ( 1997). Modernization and postmodernization: Cultural, economic, and political change in 43 societies , Princeton, NJ: Princeton University Press. Inglehart R., Welzel C. ( 2005). Modernization, cultural change, and democracy: The human development sequence . Cambridge, MA: Cambridge University Press. Google Scholar CrossRef Search ADS   Jacobs L. R., Cook F. L., Delli Carpini M. X. ( 2009). Talking together: Public deliberation and political participation in America . Chicago: Chicago University Press. Google Scholar CrossRef Search ADS   James M. R. ( 2008) Descriptive representation in the British Columbia citizens’ assembly. In Warren M. E., Pearse H. (eds.), Designing deliberative democracy: The British Columbia citizens’ assembly  (pp. 106– 126). Cambridge, MA: Cambridge University Press. Google Scholar CrossRef Search ADS   Karjalainen M., Rapeli L. ( 2015). Who will not deliberate? Attrition in a multi-stage citizen deliberation experiment. Quality and Quantity , 49, 407– 422. doi: 10.1007/s11135-014-9993-y Google Scholar CrossRef Search ADS   Karpowitz C. F., Mendelberg T. ( 2007). Groups and deliberation. Swiss Political Science Review , 13, 645– 662. doi: 10.1002/j.1662-6370.2007.tb00092.x Google Scholar CrossRef Search ADS   Kitschelt H. ( 1994). The transformation of European social democracy . Cambridge, MA: Cambridge University Press. Google Scholar CrossRef Search ADS   Kitschelt H. in collaboration with McGann, A. ( 1995). The radical right in Western Europe , Ann Arbor: University of Michigan Press. Knutsen O. ( 2006). Class voting in Western Europe. A comparative longitudinal study . Lanham: Lexington. Kriesi H. ( 2010). Restructuration of partisan politics and the emergence of a new cleavage based on values. West European Politics , 33, 673– 685. doi: 10.1080/01402381003654726 Google Scholar CrossRef Search ADS   Levine P., Nierras R. M. ( 2007). Activists' views of deliberation. Journal of Public Deliberation , 3, article 4. Available at: http://www.publicdeliberation.net/jpd/vol3/iss1/art4 Lipset S. M., Rokkan S. (Eds.) ( 1967). Party systems and voter alignments: Cross-national perspectives . New York, NY: Free Press. Luskin R. C. ( 1987). Measuring political sophistication. American Journal of Political Science , 31, 856– 899. doi: 10.2307/2111227 Google Scholar CrossRef Search ADS   MacKuen M., Wolak J., Keele L., Marcus G. E. ( 2010). Civic engagements: Resolute partisanship or reflective deliberation. American Journal of Political Science , 54, 440– 458. doi: 10.1111/j.1540-5907.2010.00440.x Google Scholar CrossRef Search ADS   Mair P. ( 2007). Left-right orientations. In Dalton R. J., Klingemann D. H. (Eds.), The Oxford handbook of political behaviour  (pp. 206– 222). Oxford: Oxford University Press. Mansbridge J., Bohman J., Chambers S., Estlund D., Føllesdal A., Fung A., Lafont C., Manin B. ( 2010). The place of self‐interest and the role of power in deliberative democracy. Journal of Political Philosophy , 18, 64– 100. doi: 10.1111/j.1467-9760.2009.00344.x Google Scholar CrossRef Search ADS   Neblo M. A., Esterling K. M., Kennedy R. P, Lazer D. M. J., Sokhey A. E. ( 2010). Who wants to deliberate – and why? American Political Science Review , 104, 566– 583. doi: 10.1017/S0003055410000298 Google Scholar CrossRef Search ADS   Norris P., Inglehart R. ( 2009). Cosmopolitan communications. Cultural diversity in a globalized world . Cambridge: Cambridge University Press. Google Scholar CrossRef Search ADS   Parkinson J. ( 2006). Deliberating in the real world: Problems of legitimacy in deliberative democracy . Oxford: Oxford University Press. Google Scholar CrossRef Search ADS   Parkinson J., Mansbridge J. (Eds.) ( 2012). Deliberative systems—Deliberative democracy at the large scale . Cambridge, MA: Cambridge University Press. Google Scholar CrossRef Search ADS   Parvin P. ( 2015). Is deliberative democracy feasible? Political disengagement and trust in liberal democratic states. The Monist , 9, 407– 423. doi: 10.1093/monist/onv021 Google Scholar CrossRef Search ADS   Posner R. ( 2004, January/February). Smooth sailing. Legal affairs. Retrieved from: https://www.legalaffairs.org/issues/January-February-2004/feature_posner_janfeb04.msp Sniderman P. M., Brody R., Tetlock P. E. ( 1991). Reasoning and choice: Explorations in political psychology . New York, NY: Cambridge University Press. Google Scholar CrossRef Search ADS   Verba S., Schlozman K. L., Brady H. E. ( 1995). Voice and equality: Civic voluntarism in American politics . Cambridge, MA: Harvard University Press. Wojcieszak M. E., Baek Y. M., Delli Carpini M. X. ( 2009). What is really going on? Structure underlying face-to-face and online deliberation. Information, Communication and Society , 12, 1080– 1102. doi: 10.1080/13691180902725768 Google Scholar CrossRef Search ADS   Young I. M. ( 2000). Inclusion and democracy , Oxford: Oxford University Press. Zaller J. R. ( 1992). The nature and origins of mass opinion , Cambridge, MA: Cambridge University Press. Biographical Notes Henrik Serup Christensen is an Academy Research Fellow at the Social Science Research Institute (Samforsk) at Åbo Akademi University. His research interests include political behavior and the consequences for democracy. His current research project concerns democratic innovations and their impact on democratic legitimacy. Åsa von Schoultz (née Bengtsson) is a Professor of Political Science at University of Helsinki. She specializes in political behavior, with a specific focus on citizens’ preferences for political decision-making processes and on intraparty competition. She has extensive experience of survey research directed toward citizens and political elites, and she is a member of the steering groups of the Finnish National Election study and the Comparative Candidate Survey. She has coauthored the book The Nordic Voter (ECPR Press) and published in journals, such as European Journal of Political Research, West European Politics, Scandinavian Political Studies, Government and Opposition, and Parliamentary Affairs. Appendix Table A1 Coding of Variables and Descriptive Statistics Variable  Coding  N  Mean  SE  Min  Maximum  VIF  Dependent variable                 Q1  Political discussions for ordinary citizens should be arranged in support of representative democracy; answer on four-point scale (0 = totally disagree)  1,502  1.94  0.83  0.00  3.00  N/A   Q2  I myself would like to participate in political discussions arranged for ordinary citizens; answer on four-point scale (0 = totally disagree)  1,530  1.21  1.05  0.00  3.00  N/A   Support deliberative practices  Additive index based on Q1 and Q2 coded to vary between 0 and 1 (1 = highest extent of support)  1,469  0.53  0.27  0.00  1.00  N/A  Independent variables                 Left–right dimension  Where would you place yourself on a scale where 0 stands for left and 10 stands for right? Coded to vary between 0 and 1 (1 = right)  1,388  0.56  0.21  0.00  1.00  1.13   Nationalist–cosmopolitan dimension  Index based on answers to three proposals concerning the future of Finland: (1) a multicultural Finland where people are tolerant toward people from other countries; (2) a Finland where the rights of sexual minorities are strengthened; (3) a Finland with more immigration. All responses 0–10 (10 = very good proposal). Index coded to vary between 0 and 1 (1 = cosmopolitan; Cronbach’s alpha = .76)  1,587  0.56  0.22  0.00  1.00  1.16  Political awareness (moderator variables)                 Political knowledge  Number of correct answers to five factual questions on political matters; coded to vary between 0 and 1 (1 = 5 correct)  1,587  0.57  0.28  0.00  1.00  1.27   Political interest  How interested are you in politics? Answer on four-graded scale coded to vary between 0 and 1 (1 = very interested)  1,587  0.61  0.29  0.00  1.00  1.31  Control variables                 Age  2015-Birth year, divided by 100  1,587  0.51  0.20  0.18  0.94  1.31   Gender  Respondent’s gender; 0 = female, 1 = male  1,587  0.50  0.50  0.00  1.00  1.08   Education  What is your highest educational attainment? Answer on seven-graded scale coded to vary between 0 and 1 (1 = highest educational attainment)  1,583  0.47  0.26  0.00  1.00  1.37   Employment  Which of the following options best describes your life situation? Answer included 12 possibilities, coded dichotomously with 1 = full-time employment (at least 32 hr per week), 0 = anything else  1,587  0.32  0.46  0.00  1.00  1.16   Satisfaction with democracy  How satisfied are you with the way democracy works in Finland? Answer on four-graded scale coded to vary between 0 and 1 (1 = completely satisfied)  1,568  0.60  0.19  0.00  1.00  1.18   Internal political efficacy  Sometimes politics is so complicated that I do not really understand what is going on. Answer on four-graded scale coded to vary between 0 and 1 (1 = disagree completely)  1,553  0.36  0.32  0.00  1.00  1.24   External political efficacy  Index based on extent of agreement to four questions: (1) politicians do not care about the opinions of ordinary people; (2) I cannot influence what the country’s government and parliament decide; (3) parties are only interested in peoples’ votes, not their opinions; (4) it does not matter what parties are in government, the policies still do not change. All answers on four-graded scales “Totally agree” to “Totally disagree;” index coded to vary between 0 and 1 (1 = highest external efficacy; Cronbach’s alpha = .77)  1,459  0.48  0.24  0.00  1.00  1.53   Political trust  How much do you trust each of the following actors? All responses 0–10 (10 = complete trust). Index based on trust in Finnish President, Political parties, Parliament, Finnish Government, and Politicians; coded to vary between 0 and 1 (1 complete trust; Cronbach’s alpha = .91)  1,587  0.63  0.17  0.00  1.00  1.30  Variable  Coding  N  Mean  SE  Min  Maximum  VIF  Dependent variable                 Q1  Political discussions for ordinary citizens should be arranged in support of representative democracy; answer on four-point scale (0 = totally disagree)  1,502  1.94  0.83  0.00  3.00  N/A   Q2  I myself would like to participate in political discussions arranged for ordinary citizens; answer on four-point scale (0 = totally disagree)  1,530  1.21  1.05  0.00  3.00  N/A   Support deliberative practices  Additive index based on Q1 and Q2 coded to vary between 0 and 1 (1 = highest extent of support)  1,469  0.53  0.27  0.00  1.00  N/A  Independent variables                 Left–right dimension  Where would you place yourself on a scale where 0 stands for left and 10 stands for right? Coded to vary between 0 and 1 (1 = right)  1,388  0.56  0.21  0.00  1.00  1.13   Nationalist–cosmopolitan dimension  Index based on answers to three proposals concerning the future of Finland: (1) a multicultural Finland where people are tolerant toward people from other countries; (2) a Finland where the rights of sexual minorities are strengthened; (3) a Finland with more immigration. All responses 0–10 (10 = very good proposal). Index coded to vary between 0 and 1 (1 = cosmopolitan; Cronbach’s alpha = .76)  1,587  0.56  0.22  0.00  1.00  1.16  Political awareness (moderator variables)                 Political knowledge  Number of correct answers to five factual questions on political matters; coded to vary between 0 and 1 (1 = 5 correct)  1,587  0.57  0.28  0.00  1.00  1.27   Political interest  How interested are you in politics? Answer on four-graded scale coded to vary between 0 and 1 (1 = very interested)  1,587  0.61  0.29  0.00  1.00  1.31  Control variables                 Age  2015-Birth year, divided by 100  1,587  0.51  0.20  0.18  0.94  1.31   Gender  Respondent’s gender; 0 = female, 1 = male  1,587  0.50  0.50  0.00  1.00  1.08   Education  What is your highest educational attainment? Answer on seven-graded scale coded to vary between 0 and 1 (1 = highest educational attainment)  1,583  0.47  0.26  0.00  1.00  1.37   Employment  Which of the following options best describes your life situation? Answer included 12 possibilities, coded dichotomously with 1 = full-time employment (at least 32 hr per week), 0 = anything else  1,587  0.32  0.46  0.00  1.00  1.16   Satisfaction with democracy  How satisfied are you with the way democracy works in Finland? Answer on four-graded scale coded to vary between 0 and 1 (1 = completely satisfied)  1,568  0.60  0.19  0.00  1.00  1.18   Internal political efficacy  Sometimes politics is so complicated that I do not really understand what is going on. Answer on four-graded scale coded to vary between 0 and 1 (1 = disagree completely)  1,553  0.36  0.32  0.00  1.00  1.24   External political efficacy  Index based on extent of agreement to four questions: (1) politicians do not care about the opinions of ordinary people; (2) I cannot influence what the country’s government and parliament decide; (3) parties are only interested in peoples’ votes, not their opinions; (4) it does not matter what parties are in government, the policies still do not change. All answers on four-graded scales “Totally agree” to “Totally disagree;” index coded to vary between 0 and 1 (1 = highest external efficacy; Cronbach’s alpha = .77)  1,459  0.48  0.24  0.00  1.00  1.53   Political trust  How much do you trust each of the following actors? All responses 0–10 (10 = complete trust). Index based on trust in Finnish President, Political parties, Parliament, Finnish Government, and Politicians; coded to vary between 0 and 1 (1 complete trust; Cronbach’s alpha = .91)  1,587  0.63  0.17  0.00  1.00  1.30  VIF = Variance Inflation Factor View Large Table A1 Coding of Variables and Descriptive Statistics Variable  Coding  N  Mean  SE  Min  Maximum  VIF  Dependent variable                 Q1  Political discussions for ordinary citizens should be arranged in support of representative democracy; answer on four-point scale (0 = totally disagree)  1,502  1.94  0.83  0.00  3.00  N/A   Q2  I myself would like to participate in political discussions arranged for ordinary citizens; answer on four-point scale (0 = totally disagree)  1,530  1.21  1.05  0.00  3.00  N/A   Support deliberative practices  Additive index based on Q1 and Q2 coded to vary between 0 and 1 (1 = highest extent of support)  1,469  0.53  0.27  0.00  1.00  N/A  Independent variables                 Left–right dimension  Where would you place yourself on a scale where 0 stands for left and 10 stands for right? Coded to vary between 0 and 1 (1 = right)  1,388  0.56  0.21  0.00  1.00  1.13   Nationalist–cosmopolitan dimension  Index based on answers to three proposals concerning the future of Finland: (1) a multicultural Finland where people are tolerant toward people from other countries; (2) a Finland where the rights of sexual minorities are strengthened; (3) a Finland with more immigration. All responses 0–10 (10 = very good proposal). Index coded to vary between 0 and 1 (1 = cosmopolitan; Cronbach’s alpha = .76)  1,587  0.56  0.22  0.00  1.00  1.16  Political awareness (moderator variables)                 Political knowledge  Number of correct answers to five factual questions on political matters; coded to vary between 0 and 1 (1 = 5 correct)  1,587  0.57  0.28  0.00  1.00  1.27   Political interest  How interested are you in politics? Answer on four-graded scale coded to vary between 0 and 1 (1 = very interested)  1,587  0.61  0.29  0.00  1.00  1.31  Control variables                 Age  2015-Birth year, divided by 100  1,587  0.51  0.20  0.18  0.94  1.31   Gender  Respondent’s gender; 0 = female, 1 = male  1,587  0.50  0.50  0.00  1.00  1.08   Education  What is your highest educational attainment? Answer on seven-graded scale coded to vary between 0 and 1 (1 = highest educational attainment)  1,583  0.47  0.26  0.00  1.00  1.37   Employment  Which of the following options best describes your life situation? Answer included 12 possibilities, coded dichotomously with 1 = full-time employment (at least 32 hr per week), 0 = anything else  1,587  0.32  0.46  0.00  1.00  1.16   Satisfaction with democracy  How satisfied are you with the way democracy works in Finland? Answer on four-graded scale coded to vary between 0 and 1 (1 = completely satisfied)  1,568  0.60  0.19  0.00  1.00  1.18   Internal political efficacy  Sometimes politics is so complicated that I do not really understand what is going on. Answer on four-graded scale coded to vary between 0 and 1 (1 = disagree completely)  1,553  0.36  0.32  0.00  1.00  1.24   External political efficacy  Index based on extent of agreement to four questions: (1) politicians do not care about the opinions of ordinary people; (2) I cannot influence what the country’s government and parliament decide; (3) parties are only interested in peoples’ votes, not their opinions; (4) it does not matter what parties are in government, the policies still do not change. All answers on four-graded scales “Totally agree” to “Totally disagree;” index coded to vary between 0 and 1 (1 = highest external efficacy; Cronbach’s alpha = .77)  1,459  0.48  0.24  0.00  1.00  1.53   Political trust  How much do you trust each of the following actors? All responses 0–10 (10 = complete trust). Index based on trust in Finnish President, Political parties, Parliament, Finnish Government, and Politicians; coded to vary between 0 and 1 (1 complete trust; Cronbach’s alpha = .91)  1,587  0.63  0.17  0.00  1.00  1.30  Variable  Coding  N  Mean  SE  Min  Maximum  VIF  Dependent variable                 Q1  Political discussions for ordinary citizens should be arranged in support of representative democracy; answer on four-point scale (0 = totally disagree)  1,502  1.94  0.83  0.00  3.00  N/A   Q2  I myself would like to participate in political discussions arranged for ordinary citizens; answer on four-point scale (0 = totally disagree)  1,530  1.21  1.05  0.00  3.00  N/A   Support deliberative practices  Additive index based on Q1 and Q2 coded to vary between 0 and 1 (1 = highest extent of support)  1,469  0.53  0.27  0.00  1.00  N/A  Independent variables                 Left–right dimension  Where would you place yourself on a scale where 0 stands for left and 10 stands for right? Coded to vary between 0 and 1 (1 = right)  1,388  0.56  0.21  0.00  1.00  1.13   Nationalist–cosmopolitan dimension  Index based on answers to three proposals concerning the future of Finland: (1) a multicultural Finland where people are tolerant toward people from other countries; (2) a Finland where the rights of sexual minorities are strengthened; (3) a Finland with more immigration. All responses 0–10 (10 = very good proposal). Index coded to vary between 0 and 1 (1 = cosmopolitan; Cronbach’s alpha = .76)  1,587  0.56  0.22  0.00  1.00  1.16  Political awareness (moderator variables)                 Political knowledge  Number of correct answers to five factual questions on political matters; coded to vary between 0 and 1 (1 = 5 correct)  1,587  0.57  0.28  0.00  1.00  1.27   Political interest  How interested are you in politics? Answer on four-graded scale coded to vary between 0 and 1 (1 = very interested)  1,587  0.61  0.29  0.00  1.00  1.31  Control variables                 Age  2015-Birth year, divided by 100  1,587  0.51  0.20  0.18  0.94  1.31   Gender  Respondent’s gender; 0 = female, 1 = male  1,587  0.50  0.50  0.00  1.00  1.08   Education  What is your highest educational attainment? Answer on seven-graded scale coded to vary between 0 and 1 (1 = highest educational attainment)  1,583  0.47  0.26  0.00  1.00  1.37   Employment  Which of the following options best describes your life situation? Answer included 12 possibilities, coded dichotomously with 1 = full-time employment (at least 32 hr per week), 0 = anything else  1,587  0.32  0.46  0.00  1.00  1.16   Satisfaction with democracy  How satisfied are you with the way democracy works in Finland? Answer on four-graded scale coded to vary between 0 and 1 (1 = completely satisfied)  1,568  0.60  0.19  0.00  1.00  1.18   Internal political efficacy  Sometimes politics is so complicated that I do not really understand what is going on. Answer on four-graded scale coded to vary between 0 and 1 (1 = disagree completely)  1,553  0.36  0.32  0.00  1.00  1.24   External political efficacy  Index based on extent of agreement to four questions: (1) politicians do not care about the opinions of ordinary people; (2) I cannot influence what the country’s government and parliament decide; (3) parties are only interested in peoples’ votes, not their opinions; (4) it does not matter what parties are in government, the policies still do not change. All answers on four-graded scales “Totally agree” to “Totally disagree;” index coded to vary between 0 and 1 (1 = highest external efficacy; Cronbach’s alpha = .77)  1,459  0.48  0.24  0.00  1.00  1.53   Political trust  How much do you trust each of the following actors? All responses 0–10 (10 = complete trust). Index based on trust in Finnish President, Political parties, Parliament, Finnish Government, and Politicians; coded to vary between 0 and 1 (1 complete trust; Cronbach’s alpha = .91)  1,587  0.63  0.17  0.00  1.00  1.30  VIF = Variance Inflation Factor View Large © The Author(s) 2018. Published by Oxford University Press on behalf of The World Association for Public Opinion Research. All rights reserved.

Journal

International Journal of Public Opinion ResearchOxford University Press

Published: Feb 7, 2018

There are no references for this article.

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