An Alternative to Representation: Explaining Preferences for Citizens as Political Decision-Makers

Abstract

Extensive scholarly attention is devoted to citizens’ preferences for alternative models of political decision-making. However, few efforts were made to identify who these citizens are and why they display a certain preference. To address this void in the literature, our article analyzes the determinants of preferences for citizens as decision-makers. It uses individual-level data from a 2014 survey on a probability representative sample in Germany and tests the effects of political attitudes toward institutions of representative democracy, interest in politics, and civic involvement. It controls for consumption of political news, education, and age.

Keywords

decision-makers citizens critique of democracy political interest Germany

Introduction

In recent decades, extensive scholarly attention has been dedicated to citizens’ preferences for alternative models of political decision-making (Bengtsson and Christensen, 2016; Esaiasson et al., 2012; Gherghina and Geissel, 2017; Hibbing and Theiss-Morse, 2001). The findings indicate that some citizens favor political representatives as decision-makers, some are more oriented toward experts (Rapeli, 2015), while others want citizens to have (more often) the final say (Anderson and Goodyear-Grant, 2010; Bengtsson and Mattila, 2009; Bowler et al., 2007; Coffé and Michels, 2014; Font et al., 2015; Gabriel, 2013; Webb, 2013). The latter perspective has recently gained momentum and, in addition to the increasing amount of referendums organized worldwide (Altman, 2011; Qvortrup, 2014), there are important indicators of increasingly positive attitudes toward referendums.¹ For example, according to the World Values Survey (Wave 5, 2015) the majority of people consider the item “people can change the laws in referendums” as an “essential characteristic of democracy.” Also, the European Social Survey (Round 6, 2012) points out that Europeans consider referendums to be as essential as free opposition, media freedom, or minority rights.

In spite of this development toward more participatory preferences, only isolated efforts were made to explain them. So far, research always referred to one survey question applied in (inter-)national surveys asking for preferences without information about how consistent this preference actually is. Looking at explanatory variables, these studies focused mainly on the ways in which factors like dissatisfaction with democracy, education, or political interest may shape preferences for decision-makers (Bengtsson and Mattila, 2009; Coffé and Michels, 2014; Dalton et al., 2001; Hibbing and Theiss-Morse, 2001). Donovan and Karp (2006) have carved out the main question discussed in the following years, whether the dissatisfied, less educated citizens favor direct democratic decision-making more than the politically interested and educated citizens. Apart from not reaching any conclusive results due to the mixed evidence, this body of research ignored several attitudes that may be relevant to determine the preferences for citizens as decision-makers. To partially address this void in the literature, our article tests the extent to which political satisfaction with democracy as well as with different institutions of representative democracy, interest in politics, and civic engagement can explain the consistent preference for citizens as decision-makers. We control for the consumption of news, education, and age. It uses individual-level data from a 2014 survey on a probability representative sample in Germany, where almost one-quarter of the respondents (approximately 700) had a clear preference toward citizens as decision-makers; the others either favor representative/expert democracy or have mixed preferences.

The following section reviews the state of the art on explaining citizens’ preferences, provides the theoretical underpinnings for the main testable hypotheses, and briefly discusses the potential impact of controls. Next, we describe our methodological approach and the data, with emphasis on the case selection and variable operationalization. The fourth section presents and interprets the results of the quantitative analysis, while the conclusion summarizes the main findings and discusses broader implications for the study of public attitudes toward direct democratic decision-making.²

State of the Art and Hypotheses

Scholars have recently started to pay closer attention to citizens’ preferences considering their role as political decision-makers (Font et al., 2015).³ A variety of surveys provide data on certain aspects of citizens’ conceptions: the different barometers conducted around the world (Afrobarometer, Eurobarometer, etc.), the World Values Survey (WVS), the International Social Survey Program (ISSP, particularly the modules on citizenship in 2004 or 2014) or the European Social Survey (Round 6, 2012), as well as country surveys. These studies reveal that citizens all over the world rank referendums as an essential, indispensable characteristic of democracy, which means they hold strong preferences for citizens as political decision-makers (Pew Research Center 2017). Research explaining these attitudes provides mixed evidence (Font et al., 2015). While some studies show that political dissatisfaction is the main explanatory variable, other works see mainly the level of education as a driver or use the interactions between these two variables (Bengtsson and Christensen, 2016; Bengtsson and Mattila, 2009; Bowler et al., 2007; Coffé and Michels, 2014; Dalton et al., 2001; Hibbing & Theiss-Morse, 2001, 2002; Webb, 2013).

Based on existing research in the field, but also inspired from the literature of political attitude formation and participation, we structure the debate along the two main features potentially influencing the preference for citizens as decision-makers. These are the dissatisfaction with the institutions of representative democracy, and political interest and active engagement in society. Political dissatisfaction and political involvement are the two variables—in addition to the socioeconomic status (SES) variables—discussed in most studies on preferences for direct democratic decision-making. Bowler et al. (2007) summarized the debate when they asked whether “enraged” (dissatisfied) or “engaged” citizens prefer “direct citizen participation.” This debate continued to shape empirical analysis until today, most recently in the study by Christensen (2017) focusing on “knowing and distrusting,” by Gabriel (2013) including mainly variables in the context of political trust and political interest, or by Rose and Borz (2013) who examined the political involvement and dissatisfaction as the main variables; for a similar effort, see Schuck and De Vreese (2011). Education is the standard variable supposedly explaining preferences for citizens as decision-makers—in some countries citizens with higher education are less enthusiastic about direct democratic decision-making than citizens with a lower educational background (Coffé and Michels, 2014; Dalton et al., 2001). News consumption is also a factor considered as crucial (Gabriel, 2013). Also, age has turned out to have an impact (Dalton et al., 2001). These variables will now be discussed in more detail.

Political Dissatisfaction

Citizens who are critical toward the institutions of political representation and dissatisfied with the existing mechanisms are likely to favor a change. They dislike how democracy in general and institutions of democracy in particular function and thus strive for a political transformation by involving the citizens in the decision-making process. Hibbing and Theiss-Morse (2001, 2002) considered dissatisfaction with representative democracy as the main rationale behind the citizens’ endorsement of direct democratic decision-making. Further studies strengthened this idea and identified a positive correlation between political dissatisfaction and preference for direct democratic decision-making (Bengtsson and Mattila, 2009; Christensen, 2017; Dalton et al., 2001).

However, other studies show less clear-cut and more mixed findings about the relationship between political dissatisfaction and preference for direct democratic decision-making (Donovan and Karp, 2006). Citizens with preferences for referendums often had a strong commitment for democracy, but little trust in institutions of representative democracy (Geissel, 2011). Other works come even to opposing conclusions. For example, a comparative study of 16 countries found that citizens with trust in institutions of representative democracy (among other variables) support direct democratic decision-making more strongly than citizens with less trust (Bowler et al., 2007). Recently the preference of so-called populist political parties and their voters for direct democracy gained specific scientific attention, especially in the context of political dissatisfaction (Pauwels, 2014; Webb, 2013). Bowler et al. (2017) found out that preferences of right-wing voters in Australia, Canada, and New Zealand have to be explained with different factors. In Australia and New Zealand, politically dissatisfied citizens did not favor referendums, whereas in Canada political disaffection and support for direct democratic decision-making were clearly linked—similar to findings from Norway, Sweden, and Finland (Donovan and Karp, 2006).

These mixed findings reflect a potential relationship between dissatisfaction with representative democracy and preference for direct democratic decision-making. The differences between the surveys questions might be a reason for the inconsistent findings. In order to measure preferences for referendums different questions were applied, for example, “what do you think about referendums” or “do you think that referendums are a good thing.” The same is true for the measurement of “political dissatisfaction,” for example, it was asked whether respondents trust government, believed that their MPs were “out of touch” (Donovan and Karp, 2006: 678–679), or whether they are satisfied with democracy in general. Accordingly, we assume that one of the main reasons behind the contradicting findings is the cacophony of survey questions. And similarly important is the fact that studies lumped attitudes recorded at systemic (“democracy in general”) and institutional levels.

In contrast, we are convinced that it is necessary to distinguish between general (“diffuse”) support (Easton, 1965) for democracy in general and specific support for different political institutions. Discontent with representative democracy in general or with government or with legislating bodies will most likely contribute in different ways to a preference for an alternative form of decision-making. This approach is in line with several empirical studies in the field. For example, there is a difference within the representative system between government and parliament (among other institutions such as parties) (Christensen, 2017) or political dissatisfaction can be differentiated between representative democracy in general and institutions of representative democracy (Arnold et al., 2014).

Accordingly, we expect a relation between satisfaction with representative democracy in general and preference for citizens as political decision-makers (H1). We also expect that discontent with institutions of representation correlates as well and is even stronger than the discontent with democracy in general with participatory preferences (H2 and H3).⁴ To pinpoint the specific attitudes of citizens toward the institutions of representative democracy, we look at their opinions about the performance of the government (H2) and how they assess the key function of the parliament, that is, the right to legislate (H3). We are aware that dissatisfaction with representative democracy in general and dissatisfaction with government and parliament as institutions of representation might correlate. However, in spite of some overlap, we see reasons to differentiate and to test the three factors separately.

H1: Dissatisfaction with representative democracy enhances preference for direct democratic decision-making.

H2: Dissatisfaction with the government enhances preference for direct democratic decision-making.

H3: Dissatisfaction with parliament as legislating body enhances preference for direct democratic decision-making.

Political and Civic Involvement

A complementary argument points out that not the dissatisfied citizens but the politically interested and engaged ones strive for more citizens’ involvement. The argument is “that citizens with greater interest in politics may … desire greater participatory democracy” (Donovan and Karp, 2006: 872). Interested citizens want to be involved and to influence politics directly. The reasoning behind the hypothesized effects is inspired by the literature on political participation. An extensive body of literature indicates that political interest is one of the most powerful predictors of political participation (Christensen, 2017; Norris, 2000; Verba et al., 1995). The argument is straightforward: individuals who are already interested in politics are more likely to participate than the rest of the public.⁵ These considerations point out that political interest might go hand in hand with the preferences for citizens as decision-makers.

Nevertheless, the empirical evidence is quite mixed. Hibbing and Theiss-Morse (2002) show how US citizens without interest in politics endorse the idea of stealth democracy, which is an alternative to representative democracy quite far from that of direct democracy. In Finland, this correlation does not appear to exist (Bengtsson and Mattila, 2009). Christensen (2017) shows also based on Finnish data that the involvement in direct democratic procedures is mainly observable in knowledgeable citizens. In this sense, political knowledge is crucial for involvement in direct democratic procedures. Previous research indicates how political interest is also a good predictor of the willingness to get involved in deliberative practices (Jacobs et al., 2009), another alternative to representative democracy. Dalton et al. (2001: 150) found that popular support for direct democracy “is located among citizens at the periphery of politics – the less interested and the less informed, and the adherents of extreme parties.” The mixed evidence shows positive correlations between preferences for direct democracy and political interest in some countries, but negative correlations in others (Donovan and Karp, 2006). Some differences may be explained by the rough indicators used to measure the independent variables. For example, the political motivation and interest were measured by self-reported voting activity or the frequency of political discussions (Donovan and Karp, 2006: 678–679). In spite of these problems and contradictory findings, the theoretical arguments presented above provide sufficient reasons to test for the existence of a positive effect of political interest on the preference for citizens as decision-makers (H4).

Following a similar line of argument, citizens who are civically engaged may be inclined to have more say in the political arena. Civic engagement has been long considered as a valid explanation for political participation (Putnam, 2000; Rosenstone and Hansen, 1993; Verba et al., 1995; Verba and Nie, 1972), but it may also be a strong predictor for preference formation. The membership in organizations (political or nonpolitical) can enhance the political education and inspire citizens to get more say. People can be socialized with the pro-participatory values and learn specific skills that may lead to aspiring further participation (Leighley, 1996). According to the core argument of Putnam (1993), citizens who get involved in voluntary organizations might increase their interest in societal problems. Consequently, their propensity toward political participation and further involvement of citizens might be higher than that of their fellow inactive citizens (H5). Consequently, we hypothesize that:

H4: Interest in politics enhances preference for direct democratic decision-making.

H5: Civic engagement enhances preference for direct democratic decision-making.

Control Variables

In addition to these main effects, we test for the impact of three factors revealed in the literature as having a potential impact: consumption of political news, education, and age. News consumption has been considered a potential driver for preference formation, especially in the context of referendum campaigns (Donovan et al., 2009; Gabriel, 2013; Schuck and De Vreese, 2011). Media are the primary source for information in general and political information in particular (e.g. Schuck and De Vreese, 2011) and an extensive body of literature identified the existence of an impact of media use on political knowledge (Drew and Weaver, 2006; Scheufele, 2002). In turn, knowledge influences the ways in which citizens wish to get involved in the decision-making process, as discussed above. Empirical evidence has shown that individuals who closely follow the development of public affairs are more involved in comparison to the rest of the citizens (McLeod et al., 1999). On the basis of these mechanisms, it is no surprise that earlier findings pointed to a strong correlation between news interest and consumption, on one hand, and involvement in decision-making, on the other hand (Putnam, 2000).

It can be argued that citizens with a lower education have more reasons to prefer direct democratic decision-making because political representatives more often satisfy the interests of the well-off voters than those of less privileged citizens (Elsässer et al., 2017; Schäfer and Schön, 2013). In this sense, the preferences of lower educated citizens for direct democracy are a rational reaction to politicians skewed responsiveness (Gabriel, 2013). Empirical findings provide a fuzzy picture. Studies on Finland and the Netherlands showed that especially citizens with a lower education are more likely to favor direct democracy (Bengtsson and Mattila, 2009; Bovens and Wille, 2010; Coffé and Michels, 2014). However, in the United States, there is no correlation between education and support for citizens as decision-makers (Bowler et al., 2007). Gabriel (2013) detected that among German citizens the level of education influenced the preference for direct democracy much less than the perception of being represented or not. Citizens, who did not feel represented by the representative system and its institutions, were more inclined to prefer direct democratic decision-making—irrespective of their educational background. In many Latin American countries especially highly educated citizens favor direct democracy, whereas the low educated are less fond of referendums (Pew Research Center, 2017: 22f).

We also checked for the effect of age, since several studies showed that younger people favor direct democracy more strongly than older citizens (Arnold et al., 2014; Dalton et al., 2001; Donovan and Karp, 2006; Rose and Borz, 2013).

Research Design and Descriptive Statistics

To test our hypotheses, we use individual level data from a national survey conducted in August–September 2014 by the GESIS Leibniz Institute for the Social Sciences in Germany. The GESIS Panel is a mixed-mode (online and mail) omnibus access panel representative for the German-speaking population aged between 18 and 70 years. The survey included a specific one-wave battery of questions about preferences for certain political decision-makers proposed by the authors of this article. Around 4000 respondents answered the questions used to operationalize the variables of this study. The “do not know” and “no answer” options were treated as missing values and were excluded from analysis.

Germany is an appropriate context to investigate variables explaining citizens’ preferences for decision-makers for several reasons (Gherghina and Geissel, 2017). In terms of participatory behavior, Germany plays in the middle field compared with other EU member states (European Social Survey 2012).⁶ Looking at subnational popular votes within European countries, Germany is currently one of the leading countries. Popular vote can be launched top-down, that is, by authorities, or bottom-up, that is, by citizens, either to “correct” a decision of the local council or to put a topic on the political agenda.

Variable Operationalization and Descriptive Findings

One methodological choice that could lead to the contradictory results is the use of a standard survey question, for example, whether respondents favor the idea or use of referendums. Such an approach does not take into account the inconsistent preferences of citizens and they might, for example, endorse at the same time citizens, experts, and politicians as decision-makers. Citizens might declare to want more say in politics and, at the same time, to strive for more decision-making power for experts (Gherghina and Geissel, 2017). To explain preferences for citizens as decision-makers, we want to include only consistent preferences. An analysis of citizens’ preferences is more valid, significant, and meaningful if we can make sure that people’s preferences are stable and robust. To avoid the oversimplification of our dependent variable, we use an approach that allows the identification of consistent preferences. Consequently, the dependent variable of this study is dichotomous to make sure that the preferences for decision-makers are mutually exclusive. This measure differs from those usually employed in the literature (Bengtsson, 2012; Bengtsson and Christensen, 2016; Font et al., 2015), in the sense that it ensures the belonging of respondents to a single preference category. In addition, it had been used in one of our previous works and provided useful insights into the preferences of citizens (Gherghina and Geissel, 2017).

The dependent variable is calculated as follows: respondents had to choose on a six-point continuum between three political decision-makers (politicians, citizens, or experts), taken two by two. The respondents were asked the following question: “Who should make important policy decisions? Please indicate the number on the scale from 1 to 6 that is closest to your opinion.” They had to make three choices between (a) Citizens (1) and Politicians (6); (b) Politicians (1) and Politically independent experts (6), and (c) Politically independent experts (1) and Citizens (6).⁷ Respondents are considered as preferring citizens as decision-makers when they favor citizens over both politicians and experts. To measure the preferences for citizens as decision-makers, all respondents who chose 1, 2, or 3 on the question about citizens vs politicians and those who chose 4, 5, or 6 on the question about experts or citizens were coded 1. All the other respondents preferred an alternative or had inconsistent preferences and were coded 0; for an overview of the descriptive statistics for every variable, see Appendix 1. After applying this coding systematically, 24% of the total sample strive for citizens as decision-makers; as a comparison, they are as many as those in favor of representative democracy (who are 25%) and some respondents who have inconsistent preferences (21%).

The satisfaction with democracy (H1) and with government performance (H2) are measured on a 11-point ordinal scale (0 = very unsatisfied, 10 = very satisfied) corresponding to the answers provided to the straightforward questions in the survey. The critique of parliament as the institution to legislate (H3) is measured on a 6-point ordinal scale (1 = totally disagree, 6 = totally agree) as the answer provided to the following statement “Parliament is the best institution to decide on laws.” Interest in politics (H4) is an ordinal variable measured on a 5-point scale as the answers provided to the question about how much interest respondents have in politics (1 = not at all, 5 = very interested). Civic engagement (H5) is a five-point cumulative index of four dummy variables that measure the involvement (coded as 1) of respondents into religious, social, union, and charity organizations.

The consumption of political news is measured as the answer to the question about how much time is spent on average on political news. It is an ordinal scale that ranges from 1 (never) to 7 (daily, for more than 2 hours). Education is a four-point ordinal variable that ranges from basic/primary (1) studies to university degree (4). Age has been recoded into four categories with the respondents up to 30 years old in the first group, those between 31 and 45 in the second, those aged 46 and 60 in the third, and Germans above 61 in the last one.⁸

Analysis and Results

Table 1 presents the results of the bivariate correlation analysis between each of the independent variables and the preference for citizens as decision-makers. There is empirical evidence for the first three hypothesized effects about the dissatisfaction of citizens, all statistically significant at the 0.01 level. The strongest among these is the most particular attitude of discontent with the parliament as a legislative body. This confirms our theoretical argument that the preference for citizens’ decision-making is likely to emerge when respondents perceive particular problems that can be solved through the involvement of citizens. When the parliament is not the appropriate institution to enact legislation, the direct decision of citizens on policies can be a valid alternative. However, this applies to a less extent in the case of dissatisfaction with the government or with democracy, more general processes, or more distant from the citizen’s power to change things through involvement.

Table 1.

Correlations with Preference for Citizens as Political Decision-Makers.

Variable	Coefficient	N
Satisfaction with democracy	−0.15***	3634
Satisfaction with the government	−0.15***	3648
Parliament is the best institution to decide on laws	−0.18***	3621
Political interest	−0.01	3508
Civic engagement	0.02	3371
Political news	−0.06**	3522
Education	−0.05**	3642
Age	−0.05**	3668

Reported coefficients are nonparametric (Spearman).

p < 0.1.

p < 0.05.

***

p < 0.01.

There is no empirical support for H4 and H5, since the values of coefficients are very close to statistical independence and they lack significance. All three control variables have a negative sign, being quite weak but statistically significant at the 0.05 level. They indicate that respondents who consume less political news are poorer educated, and the young are more likely to display a preference for citizens as decision-makers.

Table 2 includes the results of the logistic regression analysis, divided between a model without and one with controls.⁹ The results strengthen the argument according to which the specific discontent with institutions of representative democracy has a stronger effect on the preference for citizens as decision-makers. The critique of parliament, as the most specific one, is considerably stronger than that of the government which, in turn, has an effect that is stronger than that of dissatisfaction with democracy. The latter is significant at the 0.05 level, while the effects in H2 and H3 are statistically significant at the 0.01 level. The respondents who are unhappy with the parliament as a legislating body are 1.33 times more likely (this is the reciprocal of OR = 0.75, expressed like this for a simpler interpretation) to have a preference for citizens deciding when compared to respondents who are satisfied with the parliament. In comparison, respondents who are dissatisfied with the government are 1.09 more likely and those who are dissatisfied with democracy only are 1.05 times more likely to have a similar preference.

Table 2.

Binary Logistic Regression with Preference for Citizens as Political Decision-Makers.

Variable	Model 1	Model 2
Satisfaction with democracy	0.95** (0.02)	0.94** (0.02)
Satisfaction with the government	0.92*** (0.03)	0.91*** (0.03)
Parliament is the best institution to decide on laws	0.75*** (0.04)	0.77*** (0.04)
Political interest	1.07 (0.05)	1.18*** (0.06)
Civic engagement	1.10 (0.06)	1.12* (0.06)
Political news		0.92* (0.05)
Education		0.94 (0.06)
Age		0.88*** (0.05)
N	3248	3192
Pseudo R²	0.07	0.08
−2 Log likelihood	3432.42	3369.44

Reported coefficients are odds-ratios (standard errors in brackets).

p < 0.1.

p < 0.05.

***

p < 0.01.

At the same time, the multivariate regression analysis brings additional information to what was observed in the bivariate correlations. When including all five effects into the same model, political interest and civic engagement have an effect, unlike what is observed in Table 1. The empirical evidence confirms the expectations for both H4 and H5: although not statistically significant, this in the hypothesized direction and its strength is comparable to that of dissatisfaction with government or democracy. These effects get even stronger when the controls are added and they gain statistical significance. In Model 2, the political interest is the second strongest predictor after the critique against parliament. Most of the hypothesized effects get stronger when more variables are added. This happens particularly with interest in politics and civic engagement, both variables become statistically significant in Model 2 (as opposed to Model 1). One possible explanation for this is empirical and lies in the positive correlation between them and some of the controls such as the news consumption or age. When introducing the controls in the statistical model, the strength of the main effects can get higher in comparison with the model without controls. Another possible explanation is methodological: Model 1 has 50 respondents more than Model 2 and the attitudes of those respondents could make a difference.

Figure 1 presents the marginal effects of the critique against the parliament and dissatisfaction with democracy (for comparison) and political interest. The first diagram in the figure shows a strong effect where the criticism against Parliament makes a substantive difference in the way in which respondents shape their preferences toward citizens as decision-makers. At a glance, the direction of effect can be somewhat confusing because it is negative, but this is due to the coding of the variable (see the research design). The diagram shows that when the parliament is not the appropriate institution to enact legislation and the critique against it is high (coded 1), then there is a stronger preference for direct democratic decision-making. The second diagram depicts the marginal effect of satisfaction with democracy and it is weaker. Finally, the political interest appears to be the weakest among the three (as indicated in Table 1, but different in the regression analysis). The interval on which it has an effect on the dependent variable (vertical axis) is considerably smaller compared to the effect of the previous two variables examined here.

Figure 1.

The Marginal Effects of Critique against Parliament, Dissatisfaction with Democracy, and Political Interest.

Among the controls, age has a strong effect indicating that younger respondents are more likely to favor direct democratic decision-making (OR = 0.88, statistically significant at the 0.01 level). The consumption of political news has a weak and negative effect with respondents who spend less time getting political news being more in favor of citizens as decision-makers. While this may appear counterintuitive at a glance, especially when observing the positive effect of political interest, there are two possible explanations. First, the quality of political information and not the quantity may be relevant. Respondents who bear an interest in politics are likely to pay attention to information received in the first place and do not require extensive attention to internalize it. Second, those who have an interest in politics may have alternative sources of information than political news from the media. They need less exposure to political news than other citizens.

Finally, contrary to previous studies, education does not appear to make a difference. It has a very weak effect and it is the only variable that lacks statistical significance in Model 2. This indicates that, on average, highly educated people have an equal preference for direct democratic decision-making as those poorly educated.

Discussion and Conclusion

This article aimed to identify the determinants of preferences for citizens as decision-makers in Germany. Its main three findings are not straightforward, as it is often the case in social science research, but bear relevant implications for the study of political preferences. First, citizens, who are dissatisfied with the institutions of representative democracy are more likely to favor a change toward more direct democratic decision-making. In particular, the citizens rejecting the notion of parliament as the best institution to decide on laws, support the idea of citizens making decisions. We cannot identify whether they dislike the notion of parliament as a decision-maker in theory or in practice, but we observe that they endorse a greater say for citizens. Dissatisfaction with the legislative institution shapes the preference for citizens to a greater extent than general political dissatisfaction. Second, citizens with an interest in politics and involved in societal issues want to gain a stronger influence in the political arena. They strive to get citizens more involved. Third, in contrast to earlier findings that the less educated citizens favor direct democratic decision-making, our study shows that the more interested and involved strata of society have those preferences, irrespective of their level of education. The effect of education “disappears” in the multivariate regression, although it has a rather small effect on the bivariate correlation. This is most likely to happen because the effect of education is mediated through other variables such as dissatisfaction, political interest, and civic engagement and thus education cannot be the main explanatory variable. Coming back to the question asked by Donovan and Karp (2006) about whether direct democratic decision-making is endorsed by dissatisfied or politically interested citizens, we find evidence that respondents who display each of the two features have that preference.

At the theoretical level, we showed that a better understanding of individual attitudes requires a differentiation between general (dis-)satisfaction with democracy and institutions of representative democracy. Our findings reveal that dissatisfaction with political institutions fosters more the preferences for citizens as decision-makers. Citizens who generally support democracy but do not feel represented by government and parliament are more inclined to endorse other forms of decision-making. Future research might differentiate even further, for example, differentiate also dissatisfaction with politicians and political parties. Moreover, it might also be interesting to ask about different ways and mechanisms of will formation and decision-making. This can refer to (preferred) mechanisms of citizens’ involvement, for example, participatory deliberative procedures or consultative popular votes, as well as to the (preferred) mechanisms applied when politicians act as decisions-makers. Do citizens endorse vote aggregation, deliberation, or negotiation? More research, including qualitative approaches, is necessary on these questions.

Since political and civic engagement strongly influence the support for direct democracy, this factor should be included in the future research—with fine-grained differentiation. Whereas the findings considering political interest are clear-cut and straightforward—politically interested citizens favor direct democracy more strongly than their non-interested fellows—the effect of civic engagement requires a closer look. In our study, we were not able to differentiate between different kinds of civic engagement, but it is reasonable to assume that there are differences. Involvement in organizations such as trade unions or environmental groups might correlate more strongly with preferences for direct democracy than involvement in church choirs. Not all forms of civic engagement have positive effects on political interest and some have negative effects (Eliasoph, 1998). Since we now know that civic involvement is positively correlated with preferences for citizens as the decision-maker, future research should take a differentiating look at this correlation.

In terms of methodology, one of the important implications of our study is the use of a different and more sophisticated way to measure the likelihood to support direct democratic decision-making. As opposed to the usual procedure of asking one question about referendums, our measure asked they had to decide between several options and show a consistent preference for direct democratic decision-making. In contrast to other surveys, in which respondents could display preferences for all decision-makers, respondents in our survey made a clear, unambiguous choice. This approach excluded all citizens with mixed preferences. Future research should be more aware that citizens might have consistent or mixed preferences.

At the empirical level, the analysis provides the ground for broader discussions. While our study is limited to the German case, comparisons with existing studies indicate that our findings are generalizable to a certain extent. Similar to our results, Dalton et al. (2001: 148f.) had shown that dissatisfaction with institutions of representative democracy is a crucial factor for endorsing direct democracy in many European countries. On the other hand, as Rose and Borz (2013) have proved, national contexts determine the support for referendums, at least on EU-related referendums, and the studies by Donovan et al. (2009) and Leininger (2015) point to a similar direction. The fundamental differences between explanatory factors in different countries insinuate that individual-level analysis does not suffice to understand citizens’ endorsement for direct democratic decision-making. So far, little is known about the impacts of macro-structural differences such as level of welfare, social homogeneity, corruption, Gini-coefficient, or the average level of education. Such macro-structural differences might not only explain why citizens in some countries endorse direct democracy more strongly in one country than in another country; they might also explain, why the level of education plays a role in some countries, but not in other countries. Comparing, for example, Finland and the United States, two countries with tremendous macro-structural differences, it is intuitively expectable that variables explaining citizenries’ preferences for democratic decision-making procedures differ vastly. Since there is not one decisive predictor for direct democratic preferences, a more complex framework of explanation is necessary. Such a framework would have to combine several micro-level as well as macro-level features.

The availability of direct democratic options might play a role as well, differentiating between availability and use at national, state/region, and local levels. Does it make a difference whether citizens have some experience with direct democracy or not? Does it make a difference whether they have direct democratic options? Does it make a difference whether direct democracy is only allowed at the local level? These and other questions considering the availability and the use of direct democratic instruments at the different levels must be answered in the future research.

The debate about alternative forms of democracy beyond the representative version is on the rise and debated fiercely, not only in Germany. Since citizens are at the center of any democracy, they should be heard in this debate. Consequently, we should be interested in the form of democracy citizens endorse, and try to understand how these preferences can be explained. In the current debate, preferences for citizens as decision-makers are often downgraded and evaluated as “populist tendencies” of low-educated citizens (Bowler et al., 2017; Webb, 2013). However, oversimplified explanations are not helpful, for example, the devaluation of direct democracy as “populist” propaganda for the low educated. Preferences for citizens as decision-makers and more participatory procedures are for many citizens valid and desirable alternatives. Democracy will only survive if it is willing to change and these changes should be in line with citizens’ preferences. Understanding these preferences by applying a sophisticated and comprehensive framework of analysis is a necessary step for political science to pave the way for future democratic developments.

Footnotes

Appendix

Appendix 1.

Descriptive Statistics of the Variables Included in Analysis.

Variables	Min.	Max.	Mean	Standard deviation
Direct democratic decision-making	0	1	0.24	0.43
Satisfaction with democracy	0	10	5.56	2.29
Satisfaction with the government	0	10	4.50	2.13
Parliament is the best institution to decide on laws	1	6	3.67	1.27
Political interest	1	5	3.13	0.89
Civic engagement	0	4	0.41	0.69
Political news	1	7	4.74	1.19
Education	1	4	3.08	0.85
Age	1	4	2.05	0.97

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

ORCID iD

Sergiu Gherghina

Author Biographies

Sergiu Gherghina is a Lecturer in Comparative Politics at the University of Glasgow. His research interests lie in party politics, political participation, democratisation and the use of direct democracy.

Brigitte Geissel is Professor of Political Science in the Department of Political Science at the Goethe-Universität Frankfurt am Main. She is the head of the Research Unit ‘Democratic Innovations’. Her research interests lie in democratic innovations, political participation, new forms of governance and civil society.

References

Altman

(2011) Direct Democracy Worldwide. Cambridge: Cambridge University Press.

Anderson

Goodyear-Grant

(2010) Why Are Highly Informed Citizens Sceptical of Referenda? Electoral Studies 29 (2): 227–238.

Arnold

Freier

Pallauf

et al . (2014) Voting for Direct Democracy: Evidence from a Unique Popular Initiative in Bavaria. Discussion Paper 1435, Deutsches Institut für Wirtschaftsforschung (DIW), Berlin.

Baviskar

Malone

MFT

(2004) What Democracy Means to Citizens—And Why It Matters. Revista Europea de Estudios Latinoamericanos y del Caribe 76 (1): 3–23.

Bengtsson

(2012) Citizens’ Perceptions of Political Processes. A Critical Evaluation of Preference Consistency and Survey Items. Revista Internacional de Sociología 70 (2): 45–64.

Bengtsson

Christensen

(2016) Ideals and Actions: Do Citizens’ Patterns of Political Participation Correspond to Their Conceptions of Democracy? Government and Opposition 51 (2): 234–260.

Bengtsson

Mattila

(2009) Direct Democracy and Its Critics: Support for Direct Democracy and “Stealth” Democracy in Finland. West European Politics 32 (5): 1031–1048.

Bovens

Wille

(2010) The Education Gap in Participation and Its Political Consequences. Acta Politica 45 (4): 393–422.

Bowler

Denemark

Donovan

et al . (2017) Right-Wing Populist Party Supporters: Dissatisfied but Not Direct Democrats. European Journal of Political Research 56 (1): 70–91.

10.

Bowler

Donovan

Karp

(2007) Enraged or Engaged? Preferences for Direct Citizen Participation in Affluent Democracies. Political Research Quarterly 60 (3): 351–362.

11.

Cain

(2015) Democracy More or Less: America’s Political Reform Quandary. Cambridge: Cambridge University Press.

12.

Christensen

(2017) Knowing and Distrusting: How Political Trust and Knowledge Shape Direct-Democratic Participation. European Societies 20: 572–594.

13.

Coffé

Michels

(2014) Education and Support for Representative, Direct and Stealth Democracy. Electoral Studies 35: 1–11.

14.

Dalton

Burklin

Drummond

(2001) Public Opinion and Direct Democracy. Journal of Democracy 12 (4): 141–153.

15.

Donovan

Karp

(2006) Popular Support for Direct Democracy. Party Politics 12 (5): 671–688.

16.

Donovan

Tolbert

Smith

(2009) Political Engagement, Mobilization, and Direct Democracy. Public Opinion Quarterly 73 (1): 98–118.

17.

Drew

Weaver

(2006) Voter Learning in the 2004 Presidential Election: Did the Media Matter? Journalism & Mass Communication Quarterly 83 (1): 25–42.

18.

Easton

(1965) A Systems Analysis of Political Life. New York: John Wiley.

19.

Eliasoph

(1998) Avoiding Politics: How Americans Produce Apathy in Everyday Life. Cambridge: Cambridge University Press.

20.

Elsässer

Hense

Schäfer

(2017) “Dem Deutschen Volke”? Die ungleiche Responsivität des Bundestags. Zeitschrift für Politikwissenschaft 27 (1): 161–180.

21.

Esaiasson

Gilljam

Persson

(2012) Which Decision-Making Arrangements Generate the Strongest Legitimacy Beliefs? Evidence from a Randomized Field Experiment. European Journal of Political Research 51 (6): 785–808.

22.

Ferrin

Kriesi

(eds) (2016) How Europeans View and Evaluate Democracy. Oxford: Oxford University Press.

23.

Font

Wojcieszak

Navarro

(2015) Participation, Representation and Expertise: Citizen Preferences for Political Decision-Making Processes. Political Studies 63 (Suppl. 1): 153–172.

24.

Gabriel

(2013) Repräsentationsschwächen und die zweite Transformation der Demokratie: Wer will in Deutschland direkte Demokratie? Zeitschrift für Parlamentsfragen 44 (3): 592–612.

25.

Geissel

(2011) Politische Kritik—Gefahr oder Ressource für die Demokratie? Berlin: Campus Verlag.

26.

Gherghina

Geissel

(2017) Linking Democratic Preferences and Political Participation: Evidence from Germany. Political Studies 65 (Suppl. 1): 24–42.

27.

Hibbing

Theiss-Morse

(2001) Process Preferences and American Politics: What the People Want Government to Be. American Political Science Review 95 (1): 145–153.

28.

Hibbing

Theiss-Morse

(2002) Stealth Democracy: Americans’ Beliefs about How Government Should Work. Cambridge: Cambridge University Press.

29.

Jacobs

Cook

Carpini

MXD

(2009) Talking Together: Public Deliberation and Political Participation in America. Chicago, IL: University of Chicago Press.

30.

Leighley

(1996) Group Membership and the Mobilization of Political Participation. Journal of Politics 58 (2): 447–463.

31.

Leininger

(2015) Popular Support for Direct Democracy in Europe. Paper Presented at ECPR Joint Sessions of Workshops, Warsaw, 29 March–2 April 2015.

32.

McLeod

Scheufele

Moy

(1999) Community, Communication, and Participation: The Role of Mass Media and Interpersonal Discussion in Local Political Participation. Political Communication 16 (3): 315–336.

33.

Mendelsohn

Parkin

(2001) Referendum Democracy Citizens, Elites and Deliberation in Referendum Campaigns. Basingstoke: Palgrave Macmillan.

34.

Nicholson

(2005) Voting the Agenda: Candidates, Elections, and Ballot Propositions. Princeton, NJ: Princeton University Press.

35.

Norris

(2000) A Virtuous Circle: Political Communications in Postindustrial Societies. New York: Cambridge University Press.

36.

Norris

(2011) Democratic Deficit: Critical Citizens Revisited. Cambridge: Cambridge University Press.

37.

Pauwels

(2014) Populism in Western Europe: Comparing Belgium, Germany and The Netherlands. Abingdon: Routledge.

38.

Pew Research Center (2017) Globally, Broad Support for Representative and Direct Democracy. Available at: http://www.pewglobal.org/2017/10/16/globally-broad-support-for-representative-and-direct-democracy/ (accessed December 2017).

39.

Putnam

(1993) Making Democracy Work: Civic Traditions in Modern Italy. Princeton, NJ: Princeton University Press.

40.

Putnam

(2000) Bowling Alone: The Collapse and Revival of American Community. New York: Simon & Schuster.

41.

Qvortrup

(2014) Referendums around the World: The Continued Growth of Direct Democracy. Basingstoke: Palgrave Macmillan.

42.

Rapeli

(2016) Public Support for Expert Decision-Making: Evidence from Finland. Politics 36: 142–152.

43.

Rose

Borz

(2013) What Determines Demand for European Union Referendums? Journal of European Integration 35 (5): 619–633.

44.

Rosenstone

Hansen

(1993) Mobilization, Participation, and Democracy in America. New York: Macmillan Publishing Company.

45.

Schäfer

Schön

(2013) Mehr Demokratie, aber nur für wenige? Der Zielkonflikt zwischen mehr Beteiligung und politischer Gleichheit. Leviathan 41 (1): 94–120.

46.

Scheufele

(2002) Examining Differential Gains from Mass Media and Their Implications for Participatory Behavior. Communication Research 29 (1): 46–65.

47.

Schuck

ART

De Vreese

(2011) Public Support for Referendums: The Role of the Media. West European Politics 34 (2): 181–207.

48.

Setala

(2009) Introduction. In: Setala

Schiller

(eds) Referendums and Representative Democracy: Responsiveness, Accountability and Deliberation. London and New York: Routledge/ECPR, pp.1–10.

49.

Setala

Schiller

(eds) (2012) Citizens’ Initiatives in Europe: Procedures and Consequences of Agenda-Setting by Citizens. Basingstoke: Palgrave Macmillan.

50.

Verba

Nie

(1972) Participation in America: Political Democracy and Social Equality. New York: Harper & Row Publishers.

51.

Verba

Schlozman

Brady

(1995) Voice and Equality: Civic Voluntarism in American Politics. Cambridge, MA: Harvard University Press.

52.

Webb

(2013) Who Is Willing to Participate? Dissatisfied Democrats, Stealth Democrats and Populists in the United Kingdom. European Journal of Political Research 52 (6): 747–772.