Does Discussion With Disagreement Discourage All Types of Political Participation? Survey Evidence From Hong Kong

Abstract

Past studies have found contradicting evidence regarding the influence of political discussion with disagreeing others on political participation. This study argues that the impact of discussion with disagreement is likely to depend on the type of participatory activities involved. It conceptualizes a distinction between position-taking and nonposition-taking activities. The former refers to activities designed to have potential participants taking up a set position on a specific matter, whereas the latter refers to activities not designed as such. It is argued that discussion with disagreeing others is likely to suppress participation in position-taking activities, since such activities leave lesser room for participants to express the complexities of and possible ambivalence in their views. In contrast, discussion with disagreeing others may encourage participation in nonposition-taking activities. Relevant hypotheses are tested with a survey conducted in Hong Kong examining the impact of discussion and disagreement on voting, protest participation, and calling into talk radio.

Keywords

political conversation disagreement political participation deliberation-participation paradox position-taking versus nonposition-taking activities

Introduction

Inspired by theories of deliberative democracy, empirical research on the influence of interpersonal political discussions has proliferated in the past decade. Although political conversations among common people in everyday settings are by no means instances of “public deliberation” (Schudson, 1997), many theorists see such “ordinary political conversation” (Kim, Wyatt, & Katz, 1999) as part of a society’s deliberative system at large (Mansbridge, 1999; Thompson, 2008). To date, much evidence has shown that interpersonal discussions can lead to heightened levels of tolerance, political information, opinion quality, and sense of efficacy (e.g., Bennett, Flickinger, & Rhine, 2000; Eveland, 2004; Eveland & Thomson, 2006; Lee, 2009; Mutz, 2002a; Pattie & Johnston, 2008; Searing, Solt, Conover, & Crewe, 2007; Wyatt, Katz, & Kim, 2000; but see Hardy & Scheufele, 2009).

However, empirical research has also produced a conundrum regarding the impact of discussion on participation. While many found discussion frequency to relate positively to participation (e.g., McClurg, 2003; Pan, Shen, Paek, & Sun, 2006; Scheufele, 2002), others showed that exposure to differing views, a core component of deliberative discussion, lowers participation (McClurg, 2006a; Mutz, 2002b, 2006; Parsons, 2010). These findings suggest a deliberation-participation paradox: While a benefit of deliberation is the presumably stronger legitimacy of the resulting collective decision, paradoxically fewer people would participate in making the collective decision after deliberation.

But how real and serious is the tension between deliberation and participation? Does disagreement always discourage participation? This article contends that the relationship between discussion, disagreement, and participation is likely to depend on the type of participation involved. More specifically, discussion with disagreeing others is likely to discourage participation in position-taking activities, that is, activities which are designed to have potential participants taking up a set position on a specific matter, but it is likely to encourage participation in nonposition-taking activities. This article aims at contributing to the burgeoning literature on the democratic influence of political discussion by articulating and empirically examining this argument.

The next section further elaborates on the deliberation-participation paradox, explicates the notions of position-taking and nonposition-taking activities, and discusses how the conceptual distinction differs from other ways to categorize types of political activities. The article then briefly introduces the context of the study and sets up the research hypotheses. Data analysis follows, and the implications of the findings are discussed at the end.

Impact of Disagreement on Participation and Type of Participatory Activities

Before discussing the influence of disagreement on participation, it would be appropriate to first clarify the meanings of the word “deliberation” and how it relates to the more concrete notion of “discussion with disagreement.” Theorists of deliberative democracy conceptualize deliberation as discussion of matters of common concerns occurring under a set of ideal conditions, such as equality, inclusiveness, and accountability (e.g., Chambers, 1996; Guttman & Thompson, 1996; Habermas, 1996; Young, 2000). Presence of disagreement, therefore, is only one feature of a deliberative discussion.

However, actually occurring political discussions are necessarily incomplete (Fishkin, 1995). The idealistic character of the concept of deliberation makes a gap between normative theory and empirical research inevitable (Thompson, 2008). But as Mutz (2008) explicates, it is both possible and important to develop “middle-range theories” by analyzing more sharply and narrowly defined concepts or phenomena in order to empirically verify some of the claims made by the deliberative democrats. Discussion with disagreement is one of such narrower concepts amenable to empirical research. Studies on these narrower concepts can inform the theorization and practice of deliberative democracy. If discussion with disagreement is shown to have certain undesirable influence, it does not entail that deliberative democracy is completely wrong. It nonetheless means that theorists and practitioners should pay more attention to the possible downside of actually existing, incomplete deliberation.

The Deliberation-Participation Paradox

With the above general orientation, let us return to the deliberation-participation paradox. Political discussions are often thought to be capable of generating participation because they are occasions for information exchange and opinion articulation. Discussions can lead to intrapersonal reflection, further interpersonal discussions, and further media use (Eveland, 2004; Eveland & Hively, 2009; Scheufele, 2002). Discussions thus enhance political knowledge and lead people into a self-perpetuating participation cycle (Pan et al., 2006).

But when discussion exposes people to contrasting views, the positive impact of discussion may be weakened or even become negative. Feldman and Price (2008), for instance, showed that discussion leads to increases in political knowledge mainly when disagreement within discussion network is low. More directly pertinent to this study, a number of studies have demonstrated the negative impact of disagreement on participation (Belanger & Eagles, 2007; McClurg, 2006a, 2006b; Mutz, 2006).

Theoretically, as Mutz (2006) argued, exposure to disagreement leads to awareness of the rationales behind the opposing view. In fact, to the extent that people can persuade each other through discussion, having a heterogeneous discussion network tends to produce attitudinal ambivalence, that is, people’s cognitions about and affect toward specific objects become less one-sided and polarized (Huckfeldt, Johnson, & Sprague, 2004; Parsons, 2010). As political psychologists have shown, ambivalence tends to reduce attitude strength, opinion certainty, and participation (e.g., McGraw & Bartels, 2005; Meffert, Guge, & Lodge, 2004).

In addition, exposure to disagreement can undermine participation through social accountability, that is, the need to be accountable to disagreeing others in a person’s social network may reduce the person’s willingness to express his or her view through action (Mutz, 2006). The importance of social accountability on opinion expression, of course, can be tied to a long line of research on the spiral of silence (Noelle-Neumann, 1983; Scheufele & Moy, 2000). Due to a fear of isolation or a desire to avoid conflict, people holding a minority view often refrain from expressing their view publicly.

However, is the negative impact of disagreement on participation inevitable? It would be immature to give an affirmative answer. First, different factors may condition the negative relationship between disagreement and participation (Delli Carpini, Cook, & Jacobs, 2004). Kwak, Williams, Wang, and Lee (2005), for instance, showed that network heterogeneity relates negatively to participation mainly when both network size and discussion attention are low. Mutz (2006) herself also showed that the negative relationship between disagreement and participation applies only to people who are conflict avoidant.

Moreover, other studies have actually shown different types of relationships between discussion, disagreement, and participation. Pattie and Johnston (2009) stated that existing research “on the relationship between network disagreement and political participation runs the whole gamut of possible outcomes” (p. 265). For example, both Huckfeldt, Mendez, and Osborn (2004) and Nir (2005) failed to find a significant effect of disagreement on participation. Meanwhile, some studies have shown a positive effect of structural heterogeneity and/or exposure to disagreement on participation (Leighley, 1990; Scheufele, Hardy, Brossard, Waismel-Manor, & Nisbet, 2006; Scheufele, Nisbet, Brossard, & Nisbet, 2004; Wojcieszak, Baek, & Delli Carpini, 2010).

A number of factors, such as the measurement of network disagreement or heterogeneity (Eveland & Hively, 2009), the form of participation being examined (Pattie & Johnston, 2009), and the contexts of the studies, may all have contributed to the variations in findings. This study does not aim at resolving the discrepancies once and for all. It aims more sharply at examining how the influence of disagreement may vary across types of participatory activities. In fact, it is notable that studies which demonstrated the negative influence of disagreement on participation tend to focus on voting and other kinds of electoral participation (e.g., McClurg, 2006a, 2006b; Mutz, 2002), whereas studies showing the positive influence of disagreement tend to include a wider array of activities outside the electoral context (e.g., Pattie & Johnston, 2009; Scheufele et al., 2004; Wojcieszak et al., 2010). But to draw better connections with the theoretical arguments about the deliberation-participation paradox, this study develops the conceptual distinction between position-taking and nonposition-taking participatory activities.

Categorizing Participation: Position-Taking Versus Nonposition-Taking Activities

As explicated above, exposure to disagreement can undermine participation partly because it makes people’s opinions more complex and ambivalent. Directly following this claim, we may expect the negative impact of disagreement to be most pronounced when the participatory activity does invite participants to express a simplistic and one-sided opinion. In contrast, disagreement may not undermine participation if the activity allows people to communicate the complexities of their thoughts and express their ambivalence. Disagreement may even promote participation in this latter type of activities, as discussing with disagreeing others, through which people can understand the rationales behind different views and practice the art of making arguments, may actually help prepare individuals for activities in which they can or have to articulate the complexities of their views.

This study uses the notions of position-taking and nonposition-taking activities to capture the above distinction. More formally, position-taking activities refer to activities designed to have participants taking up a fixed position on a matter. In an ideal-typical position-taking activity, participants are invited to support a stance or a message which is already set and framed by the organizer. Participation has an all-or-nothing character: either a person joins the action to support the stance, or not. Moreover, there is little room within the activity for people to express the specificities and complexities of their individual thoughts on the matter. In empirical reality, petition-signing is a good example of a position-taking activity. A petition campaign usually involves an organizer drafting a message for people to sign onto. Signers do not participate in framing the message. They only choose to sign or not to sign. Voting and protests, which this study actually examines, can also be considered as position-taking activities. They are further discussed in the next section.

Nonposition-taking activities, in contrast, simply refer to activities not designed for people to express support for a fixed position. In an ideal-typical nonposition-taking activity, there is no fixed message for the participants to take up. People can control the articulation and communication of their individual views. There is room for expressing complexities of thoughts and ambivalence in feelings. People can acknowledge the claims held by different sides in a debate. In fact, they do not even need to commit to a side. Writing letters to newspapers or calling into talk radio are good examples of nonposition-taking activities. Although people can write letters or call into talk radio to express a one-sided view, they are not requested to do so. In fact, empirical research has shown that many letter writers and radio callers do express complicated views, and sometimes people may write or call just to share personal experiences (e.g., Cushion, Franklin, & Court, 2006; Leurdijk, 1997).

It should be noted that categorizing an activity as “position-taking” does not entail the total absence of differences among the participants in the activity. People who vote for the same candidate can do so for different reasons, and people supporting a social movement may disagree with each other on matters of strategies (e.g., Heaney & Rojas, 2007). Rather than neglecting such individual differences, the significance of the distinction between position- and nonposition-taking activities resides exactly in the recognition that people supporting the same cause often differ in the strength of their support and on their views regarding more detailed matters. A major difference between the two types of activities is the extent to which individual differences can be communicated in and through the activities. This is pertinent to our concern of the impact of disagreement because, as stated earlier, if disagreement can undermine participation through making people’s views more complicated and less one-sided, then disagreement should undermine participation mainly in activities which tend not to facilitate the communication of the complexities in individuals’ views.

Certainly, the conceptual distinction developed here is not the only way to classify political activities. Political scientists have employed different categorizing schemes when studying participation (e.g., Dalton, 1996; Verba & Nie, 1972). More directly related to the impact of disagreement, Pattie and Johnston (2009) differentiated between high-cost, collective, and low-cost, privatized actions. They contended that the negative influence of disagreement should be stronger on participation in low-cost activities because participants in such activities tend to have smaller personal stakes in the outcomes and thus lower levels of commitments to the cause. Underlying their expectation is the argument that people with different levels of commitments would process disagreeing messages differently: Less committed people are more likely to be persuaded by disagreeing messages, whereas committed people are likely to dismiss the disagreeing messages and even have their views reinforced by such an encounter with disagreeing views.

Another distinction potentially relevant to the present research topic is Scheufele and Eveland’s (2001) distinction between “public” and “non-public” participation. If exposure to disagreement undermines participation partly by enhancing people’s need to be socially accountable, then the negative impact of disagreement should be stronger when the activity is “public”, that is, when their act is visible to others. In contrast, accountability should be a lesser issue when nonpublic activities are concerned.

Space concerns forbid a thorough comparison of the different ways to categorize political activities. Suffice it to state that one major difference among the three distinctions—low- or high-cost, public or nonpublic, position- or nonposition-taking—resides in the theoretical arguments to which each is mainly connected. The pertinence of the low- versus high-cost distinction is based mainly on arguments about how more or less committed people process disagreeing messages; the pertinence of the public versus nonpublic distinction is based on arguments about people’s responses to social pressure; the pertinence of the position- versus nonposition-taking distinction is mainly tied to the presumed impact of disagreement on ambivalence and complexities in individual opinions. At the conceptual level, therefore, there is no necessary contradiction among the three distinctions.

Nevertheless, the three distinctions may generate different expectations in specific cases of empirical analysis. Following Scheufele and Eveland (2001), for instance, disagreement should undermine calling into talk radio, since the latter involves “public” opinion expression. But calling into talk radio is also a nonposition-taking activity, and hence we can also expect disagreement to enhance participation in it. Therefore, although this study does not aim at arguing that the position- versus nonposition-taking distinction is better or more important than the other two, the empirical findings can nonetheless be discussed in terms of their goodness of fit when placed under different conceptual lenses.

Background and Hypotheses

Based on the discussion in the previous section, the main proposition to be examined in this article is: disagreement has a negative impact on participation in position-taking activities but a positive impact on participation in nonposition-taking activities. This proposition can be turned into specific hypotheses by considering the local context of the present study.

Without being a full democracy, Hong Kong is nonetheless undergoing a process of gradual democratization. Not unlike their counterparts in democratic countries, Hong Kong citizens have ample opportunities to express their views and participate in politics. Democratic elections for the District Council (a local consultative body) and half of the seats in the legislature have been established for more than two decades. At the same time, some scholars have claimed that Hong Kong citizens have been quite active in noninstitutionalized political activities such as protests and rallies (Lam, 2004; Lui & Chui, 2000).

The following analysis examines both voting and protest participation. Both can be considered as position-taking activities as conceptualized earlier. Voting involves people choosing a candidate among a range of options without the chance to register their potentially complicated and ambivalent views about the candidates. Admittedly, compared with voting in a two-party system, voting in Hong Kong arguably has a weaker “either-or” character given the multiparty system in place. But on the whole, voting should still approximate the ideal type of position-taking activities. Similarly, protest participation involves giving support to the protest’s theme and organizer. Although there can be some room in protests for individuals to express their own views (e.g., they may carry their own banners), their participation will still be taken by the organizers and others as giving support for the cause of the protest. Protest participation, therefore, also comes close to the ideal-typical position-taking activity.

Comparatively, it is more difficult to identify nonposition-taking activities in Hong Kong. As noted in the previous section, writing letters to newspapers and calling into talk radio can be considered as approximations of the ideal-typical nonposition-taking activity. While most local newspapers do not have letters section, talk radio has obtained high levels of prominence in the city (Lee, 2007; Ma & Chan, 2006). The following analysis thus examines talk radio calling as a nonposition-taking activity.

To belabor the point a bit further, the distinction between position-taking and nonposition-taking activities resides in the “basic design” of the participatory activity. People can express a one-sided stance when calling into talk radio, but they are not requested to do so, and talk radio is not designed only for the expression of one-sided views. Similarly, people are expected to participate in a protest to support a designated cause, despite the fact that individual participants may not have the same level of support for the cause.¹

Past research has followed two approaches to analyze the impact of disagreement: The first is to examine the separate main effects of discussion frequency and disagreement (e.g., Mutz, 2002a, 2002b; Scheufele et al., 2004), whereas the second is to examine the interaction effect between discussion and disagreement (e.g., Kwak et al., 2005; Lee, 2009). The former approach is relatively clear as to what to expect, since it posits only a main effect hypothesis regarding the influence of disagreement. The latter approach postulates disagreement as both an independent variable and a moderating variable. This, nevertheless, also means that the latter provides a more comprehensive test of the impact of discussion and disagreement. Hence this study adopts the interaction effect approach.

Following past research, discussion frequency should relate positively to all kinds of participation. Disagreement is expected to undermine protest and voting, and yet enhance calling into talk radio. Without theoretical reasons to expect otherwise, we posit that the influence of disagreement can be manifested through both a main effect and an interaction effect with discussion frequency. Summarizing the arguments and considerations, the hypotheses are stated as follows:

Hypothesis 1 (H1): Discussion frequency relates positively to protest participation.

Hypothesis 2 (H2): Disagreement in discussion relates negatively to protest participation.

Hypothesis 3 (H3): Disagreement moderates the effect of discussion postulated in H1 such that the effect is weaker when discussion involves higher levels of disagreement.

Hypothesis 4 (H4): Discussion frequency relates positively to voting.

Hypothesis 5 (H5): Disagreement in discussion relates negatively to voting.

Hypothesis 6 (H6): Disagreement moderates the effect of discussion postulated in H4 such that the effect is weaker when discussion involves higher levels of disagreement.

Hypothesis 7 (H7): Discussion frequency relates positively to calling in to talk radio.

Hypothesis 8 (H8): Disagreement in discussion relates positively to calling in to talk radio.

Hypothesis 9 (H9): Disagreement moderates the effect of discussion postulated in H7 such that the effect is stronger when discussion involves higher levels of disagreement.

Method and Data

Data analyzed in the following were obtained from a telephone survey conducted by the center for communication research at the Chinese University of Hong Kong in August 2008. The target respondents were all Chinese-speaking Hong Kong residents aged between 18 and 70.² To derive the sample, telephone numbers were randomly selected from the most updated residential telephone directory. The last digit of the selected numbers was adjusted so as to include nonlisted numbers. When a household was reached, the target respondent was selected by the most recent birthday method. At the end of the survey period, 836 interviews were completed, yielding a response rate of 58.3% following AAPOR RR6 and 19.6% following AAPOR RR1.³

The sample has 55.0% females. About 24.8% of the respondents were aged 30 or below, with 22.4% aged between 31 and 40. University graduates constituted 23.5% of the sample. About 13.8% lived in households with a monthly income of HK$10,000 or less, whereas 25.2% lived in households with a monthly income between HK$10,001 and HK$20,000. According to government statistics, females constituted 52.8% of the Hong Kong population. Among people aged between 15 and 70, 24.3% aged 30 or below and 21.6% aged between 31 and 40. Among people aged above 15, 15.4% had a university degree. About 27.9% had a monthly household income of less than HK$10,000, whereas 27.7% had a monthly household income between HK$10,000 and HK$19,999. The sample, therefore, contains higher proportions of highly educated people and people with higher levels of household income. The sample was weighted according to income in the analysis.⁴

The key variables for the analysis are operationalized as follows:

Discussion frequency

A set of six questions asked the respondents how frequently they discussed with their family and friends about (1) educational issues, (2) social welfare issues, (3) environmental and heritage protection issues, (4) the performance of the Hong Kong government, (5) the performance of political parties and legislators, and (6) democratic development. Answers were recorded with a 4-point scale ranging from 1 (never) to 4 (frequently). The six items constitute one factor in an exploratory factor analysis. Hence they were averaged to form an index for discussion frequency (α = .85, M = 2.49, SD = 0.67).

Disagreement in discussion network

Another set of questions asked the respondents if they think their views on the same six issue areas are generally the same as or different from their family and friends. The answering categories are 1 = largely the same, 2 = about half and half, and 3 = largely different. The six items are all positively correlated with each other (r ranges from .26 to .52). They were therefore averaged to form an index representing amount of disagreement experienced in discussion networks (α = .77). Following the practice of past research (Lee, 2009; Scheufele et al., 2006), a person was treated as experiencing no disagreement if she has a missing value on the items regarding disagreement. The resulting index has a mean score of 1.63 (SD = 0.46). In other words, the majority of Hong Kong people regarded themselves as located in relatively homogeneous networks in terms of people’s opinions concerning public affairs.

Protest participation

The survey asked the respondents three questions regarding their protest and rally participation. The first asked them how frequently they participated in the so-called “July 1 protest” in Hong Kong, which is a series of prodemocracy protests occurring in the city since 2003. The second asked them how frequently they participated in the annual June 4 rallies commemorating the Tiananmen Square incident in 1989. The third asked them how frequently they participated in protests and rallies other than the July 1 and June 4 series. All three questions were recorded with a 3-point scale with 0 = never, 1 = once or twice, and 2 = more than twice. The three items are positively correlated with each other, but the correlations are not very strong since the July 1 and June 4 protest series have their own specific goals and characteristics (r ranges from .24 to .43). Hence the Cronbach’s alpha among the three items is not optimal (α = .57). Nevertheless, they were still averaged to form an index for a more efficient analysis of the hypotheses (M = 0.16, SD = 0.32).

Voting participation

The survey asked the respondents if they have voted in the district council election in 2007 and the legislative council election in 2008. The two items were added to form a 0-to-2 index on voting participation (α = .79, r = .66, M = 1.02, SD = 0.91).

Calling into talk radio

The survey asked the respondents to indicate with a 4-point scale (ranging from 1 = never to 4 = frequently) if they frequently listened to phone-in talk radio about public affairs. Those who listened at least occasionally were then asked if they have ever tried to call into the shows. The answers were recorded with a 3-point scale with 1 = never, 2 = once or twice, and 3 = several times or more (M = 1.12, SD = 0.40).

Control variables include (1) four demographics, (2) talk radio listening frequencies (as mentioned above), (3) news attention (average of two 5-point Likert-type scaled questions about amount of attention paid to political news in newspapers and on television, r = .59, M = 3.38, SD = .74), social connections (average of two 5-point Likert-type scaled statements about whether the respondents have a lot of friends and whether they meet with their friends and relative frequently, r = .53, M = 3.15, SD = .1.12), and internal, external, as well as collective efficacy (each measured by agreement with two 5-point Likert-type scaled statements, r = .57, .50, .58, M = 2.79, 2.63, 3.40, and SD = 1.09, 0.95, 1.02, respectively).

Analysis and Results

Multiple regression analyses were conducted to examine the hypotheses. The model includes the demographics and other control variables, frequency of discussion, disagreement in discussion network, and the interaction term between discussion and disagreement. The interaction term was centered to alleviate the problem of multicollinearity. The core tests of the hypotheses reside in the significance of the regression coefficients of the variables concerned. Hence all factors were added into the model simultaneously.

We can first examine H1 to H6, which are concerned with the impact of discussion and disagreement on participation in protest and voting. Table 1 summarizes the findings of the regression analysis. The first column shows that more educated people, older people, and males are more likely to have participated in various protest activities. Not surprisingly, protest activity is related to a lower sense of external efficacy, that is, a feeling that the political system is not responsive to public opinions. Consistent with past research in Hong Kong (Lee, 2006), protest activity is related positively to a sense of collective efficacy, that is, the feeling that Hong Kong citizens, as a collective actor, can effect social and political change.

Table 1.

The Impact of Discussion and Disagreement on Protest and Voting Turnout

	Protest participation	Voting turnout
Sex	−.08*	.00
Age	.09*	.30***
Education	.21***	.05
Income	−.00	−.01
News attention	−.02	−.01
Talk radio listening	.07	.11**
Social connection	.06	.04
Internal efficacy	.01	.10*
External efficacy	−.17***	−.09*
Collective efficacy	.11**	.03
Discussion	.18***	.13**
Disagreement	−.08*	−.01
Discussion × Disagreement	−.09*	−.01
Adjusted R²	16.4%***	14.3%***

Note: Entries are standardized regression coefficients. Missing cases were deleted pairwise. N = 812 and 803, respectively.

p < .05. **p < .01. ***p < .001.

More important, even after controlling for all the other variables, the three discussion-related variables all obtain statistically significant regression coefficients. The signs of the coefficients are consistent with our expectations. Supporting H1, political discussion is positively related to protest participation. Supporting H2, disagreement is negatively related to protest participation.

Nevertheless, as Hayes, Glynn, and Huge (2008) explicate, when a regression model involves an interaction term, the main effect coefficients of the independent variables constituting the interaction term should be understood as the conditional effect of the independent variables when the interaction term is zero in value. In other words, talks about the “main effects” of discussion frequency and disagreement on protest participation can be misleading when the two have a significant interaction effect on the latter. Indeed, Table 1 shows that the interaction term has a significant negative coefficient. Consistent with H3, the seemingly positive influence of discussion frequency on protest participation would become weaker when level of disagreement increases.

The interaction effect can be presented graphically as in Figure 1.⁵ Among people experiencing high levels of disagreement in political discussion (set at 2 SD above the sample mean score), discussion frequency does not have any observable positive influence on protest participation. People’s score on the protest participation variable only moves from .22 to .26 when discussion frequency moves from low to high (set at 2 SD below and above mean, respectively). In contrast, the relationship between discussion frequency and protest participation is substantially positive among people experiencing low levels of disagreement (set at 1, the lowest value on the disagreement scale, since 2 SD below the sample mean score would be lower than 1). People’s score on the protest participation variable moves from .15 to .50 as discussion frequency moves from low to high.

Figure 1.

The interaction effect of disagreement and discussion frequency on participation in protests

We will further discuss the findings in the concluding section. Here, let us turn to the second column of Table 1. The predictors of voting are somewhat different from the predictors of protest participation. Older people, people with higher levels of internal efficacy, and listeners of radio phone-in talk shows are more likely to vote. There is also a negative relationship between external efficacy and voting—it points to the phenomenon of “protest voting” in the city.

Regarding the three discussion-related variables, only discussion frequency has a significant and positive impact on voting. In other words, H4 is supported, but not H5 and H6. The regression coefficients of the disagreement variable and the interaction term are close to zero. Therefore, on the whole the hypothesized negative impact of political disagreement on participation in position-taking activities is only partly supported.

We now turn to the impact of discussion and disagreement on calling into talk radio. The same multiple regression analysis was conducted. Table 2 summarizes the findings. Calling into talk radio is positively related to talk radio listening itself and to social connections. Calling into talk radio also has a negative relationship with external efficacy, suggesting that the medium serves as the platform for the expression of people’s discontent in the city.

Table 2.

The Impact of Discussion and Disagreement on Calling Into Talk Radio

	Calling into talk radio
Sex	−.01
Age	.07
Education	−.03
Income	−.07
News attention	.03
Talk radio listening	.18***
Social connection	.13**
Internal efficacy	.04
External efficacy	−.10*
Collective efficacy	.02
Discussion	.18***
Disagreement	.09*
Discussion × Disagreement	.09*
Adjusted R²	12.4%***

Note: Entries are standardized regression coefficients. Missing cases were deleted pairwise. N = 571.

p < .05. **p < .01. ***p < .001.

All three discussion-related variables obtain statistically significant coefficients, and the signs of the coefficients are all in the expected direction. Consistent with H7, discussion frequency has a positive relationship with calling into talk radio. H8 is also supported, as disagreement has a positive relationship with calling. Moreover, the interaction effect between discussion and disagreement is again statistically significant. But this time the coefficient is positive in sign. Consistent with H9, disagreement strengthens instead of weakens the influence of discussion frequency on calling into talk radio.

Figure 2 presents the results graphically. Among people experiencing low levels of political disagreement (set at 1 on the disagreement scale), the relationship between discussion frequency and calling into talk radio is only slightly positive, moving from 1.00 to 1.12 when discussion frequency moves from low to high. But among people experiencing high levels of disagreement (set at 2 SD above mean), the relationship between discussion frequency and calling into talk radio is highly positive, moving from 0.92 to 1.45 as discussion frequency moves from low to high.

Figure 2.

The interaction effect of disagreement and discussion frequency on calling into talk radio

Discussion

This article examines the influence of political discussion and discussant disagreement on different forms of participation. It argues that the deliberation-participation paradox is applicable mainly to political activities designed for citizens to take up already fixed positions. It is mainly because disagreement-induced ambivalence and complexities in opinions is likely to make people feel uneasy about taking up a fixed and inflexible position set by someone else. In contrast, when political activities allow citizens to articulate their own views in their own ways, disagreement should not hinder participation. Instead, by enhancing people’s understanding of different viewpoints and by strengthening people’s ability to handle arguments, discussion with disagreement is likely to enhance participation in nonposition-taking activities.

The analysis generally supports the expectations. Political discussion relates positively to all three kinds of participation examined. However, disagreement undermines the positive influence of discussion on protest participation. This finding both supports and challenges the arguments by Mutz (2006) and McClurg (2006a, 2006b). On one hand, disagreement does seem to have some negative influence on protest participation, as both the main effect and the interaction effect terms in Table 1 obtain the expected coefficients. On the other hand, as Pattie and Johnston (2009) argued, the negative influence of disagreement can be mitigated by the positive influence of discussion itself. This is exactly what Figure 1 shows. If disagreement suppresses protest behavior among people who discuss politics frequently, what it does is merely to reduce their participation to a level similar to that of those who do not discuss politics at all. In other words, discussion at most would fail to generate protest participation when disagreement level is high. But the overall impact of discussion and disagreement is rarely suppressive of protest participation. Here, we can see the utility of the “interaction-effect approach” when studying the effect of disagreement and discussion, for it can provide a more precise and fine-grain analysis of the impact of political discussions with different characteristics (e.g., Feldman & Price, 2008; Kwak et al., 2005).

The findings in Figure 2 challenge the argument for a deliberation-participation paradox to an even larger extent. Interpersonal political discussion is generally positively related to calling into talk radio, a finding consistent with past research in Hong Kong (Lee, 2005). The present analysis further shows that disagreement among discussants is a contributory condition (Eveland, 1997) to the impact of discussion: While discussion has a slightly positive relationship with calling into talk radio when disagreement level is low, the relationship becomes more strongly positive when disagreement level is high.

Certainly, not all hypotheses are supported. Disagreement did not negatively affect voting. One possible reason for the null finding is the limitation of the measure. As the survey was conducted in summer 2008, the “previous Legislative Council election” was referring to the election in 2004. The district council, meanwhile, is a local level consultative body. Voting in the District Council elections in Hong Kong may be driven not so much by political views but by recognition of the candidates’ record of community services. It is possible that the impact of political disagreement could not be demonstrated when voting in a not-so-political election and voting in an election 4 years ago are concerned.

More broadly speaking, one limitation of this study is that it examines only three types of participatory activities. The limited number is partly due to concerns with the length of the survey and partly due to considerations of the type of activities prevalent in the local context. But undeniably, the inclusion of a larger number of participatory activities can provide a more robust test of the main theoretical proposition of this study.

Despite the above limitations, there are several considerations which can strengthen this study’s claim about the importance of the conceptual distinction between position-taking and nonposition-taking activities. Theoretically, the concept of position-taking activity allows us to specify the differential influence of discussion and disagreement on political participation, while linking such differential influence to the arguments already developed in the literature. More specifically, the influence of disagreement on participation is often explained in terms of how it generates attitudinal ambivalence on one hand (when the influence is expected to be negative, Mutz, 2002b, 2006), and how it generates political knowledge and experience with handling disagreement on the other (when the influence is expected to be positive, see Scheufele et al., 2004, 2006). Ambivalence, knowledge about the opposing views, and experience with handling disagreement, in turn, do conceivably relate to participation in position-taking and nonposition-taking activities differently.

In relation to the present study’s empirical findings, as stated earlier, other researchers have developed other conceptual distinctions, most notably the high- versus low-cost distinction (Pattie & Johnston, 2009) and the public versus nonpublic distinction (Scheufele & Eveland, 2001), which are potentially relevant to the impact of disagreement on participation. But in the present study, if we follow Pattie and Johnston (2009), we should have expected disagreement to relate positively to protest and negatively to calling into talk radio, since protest is a high-cost, collective activity and talk radio calling is a low-cost, privatized activity. In other words, Pattie and Johnston’s (2009) argument would have led us to expect exactly the opposite of what is found. Similarly, if we follow Scheufele and Eveland (2001), we would expect disagreement to negatively impinge on participation in both protests and calling into talk radio, since both types of activities are public in their own ways. In other words, although the findings of the present study do not support all hypotheses perfectly, the position-taking versus non-position-taking distinction nonetheless remains the more useful conceptual lens when making sense of the current findings.

Moreover, beyond the present study, the expectations generated by the position-taking versus nonposition-taking activities distinction also seem to be largely consistent with the contradictory findings in past research discussed in the theoretical section. To repeat, past research which demonstrated the negative influence of disagreement usually focused on electoral participation (McClurg, 2006a; Mutz, 2006; Parsons, 2010), whereas research which demonstrated the positive influence of disagreement usually focused on participation in a wider and more “civic” context (Scheufele et al., 2004, 2006; Wojcieszak et al., 2010). We can indeed make sense of these findings with the distinction between position-taking versus nonposition-taking activities. Electoral participation, as explicated in this study, is largely position-taking, and hence disagreement is likely to undermine electoral participation. Civic participation in general, comparatively speaking, involves more activities not designed for people to take up fixed positions. Hence it is reasonable to see disagreement and discussion to relate positively to such participation.

To use just one more example, Scheufele and Eveland’s (2001) study on the impact of opinion climate on public and nonpublic participation actually comes up with results contrary to their expectations: Being in the minority relates negatively to the nonpublic activities of voting and contributing money to parties, but positively to public forms of participation such as attending rallies and trying to persuade others to vote for a candidate. These findings, nevertheless, seem to make better sense when the concepts of position-taking and nonposition-taking activities are applied. The experiences of disagreeing with others can be expected to undermine the position-taking activity of voting. Persuading others to vote for a candidate, meanwhile, does also involve people taking up a position. Hence it is not the prototypical nonposition-taking activity. But at the same time, people would have much room for expressing the complexities of their own views or even acknowledging the partial validity of opposite views when engaging in interpersonal persuasion. Hence it should not be too surprising for it to relate positively to the experience of being in the minority. Being in the minority probably helps prepare individuals to argue with disagreeing others, which is an ability needed when people persuade others to vote for a specific candidate.

On the whole, the concepts of position-taking and nonposition-taking activities seem to have the potential to clarify the contradictory findings in the existing literature. Of course, more case studies in different contexts and possibly a metaanalysis of existing research would be needed to further examine the explanatory power of the concepts.

It should be noted at this point that the cross-sectional design of the survey and hence the inability to convincingly establish causal relationships represents another limitation of this study. One may reinterpret Figures 1 and 2 as showing the impact of participation on discussion frequencies, but with protesters seeking out like-minded discussants and talk radio callers searching particularly for “disagreement.” Establishing the causal direction is beyond the resources available for this study’s design. Nevertheless, it can be stated that, regardless of the exact causality involved, Figures 1 and 2 do show that discussion, disagreement, and different types of political activities are structured in specific manners in the society. The pattern of findings would still have important implications on our understanding of the relationship between deliberation and participation.

As pointed out at the beginning parts of the article, this study is premised on the argument that empirical research about the influence of actually occurring political discussions, though necessarily being imperfectly deliberative, can nonetheless inform the theorization and practice of deliberative democracy (Mutz, 2008). One general theoretical implication of this study’s findings is that the tension between deliberation and participation is probably not as strong and serious as some researchers suggest. If deliberation adversely affects participation in some forms of activities, it is largely because those activities are not designed for deliberative democracy in the first place. Elections, for instance, are largely an institution of representative democracy, especially when electoral participation is structured along partisan lines. Protests, meanwhile, can be considered as a hallmark for more agonistic form of democracy (Mouffe, 2000). In this sense, the tension between deliberation and voting or protest participation is actually representative of the tension between different theoretical perspectives on the democratic system.

In fact, normative theorists of deliberative democracy have probably never expected deliberation to generate participation in, say, civil disobedience. The latter simply belongs to a different theoretical world. Rather, deliberation should not contradict with participation when we are concerned with forms of activities which can be regarded as the hallmark of deliberative democracy, such as, as this study shows, participation in public forums.

What proponents of deliberative democracy need to do, then, is to design and promote forms of participatory activities which are more deliberative. This, of course, does not mean that activities such as voting and protests should be abolished. Democracies in the real world are never going to fit perfectly with any ideal theoretical vision, be it deliberative, representative, agonistic, or other. But deliberative democrats can nonetheless strengthen the deliberative characteristic of the political process through promoting not only deliberation itself but also forms of participation or participatory institutions which are consistent with the concept of deliberation.

Footnotes

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the authorship and/or publication of this article.

Funding

The author disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The survey reported in this study was supported by a Strategic Research Grant offered to the author by the City University of Hong Kong.

Notes

Bio

Francis L. F. Lee is associate professor at the School of Journalism and Communication, Chinese University of Hong Kong. His research focuses mainly on political communication, journalism studies, and public opinion analysis. He is the lead author of Media, Social Mobilization and Mass Protests in Post-colonial Hong Kong (Routledge, 2010) and a co-editor of Media and Politics in Post-handover Hong Kong (Routledge, 2008)

References

Belanger

Eagles

(2007). Partisan cross-pressure and voter turnout: The influence of micro- and macro-environments. Social Science Quarterly, 88, 850-867.

Bennett

S. E.

Flickinger

R. S.

Rhine

S. L.

(2000). Political talk over here, over there, over time. British Journal of Political Science, 30(1), 99-119.

Chambers

(1996). Reasonable democracy. Ithaca, NY: Cornell University Press.

Cushion

Franklin

Court

(2006). Citizens, readers and local newspaper coverage of the 2005 UK general election. Javnost: The Public, 13(1), 41-60.

Dalton

(1996). Citizen politics. Chatham, NJ: Chatham House.

Delli Carpini

M. X.

Cook

F. L.

Jacobs

L. R.

(2004). Public deliberation, discursive participation, and citizen engagement: A review of the empirical literature. Annual Review of Political Science, 7, 315-344.

Eveland

W. P.

(1997). Interactions and nonlinearity in mass communication: Connecting theory and methodology. Journalism & Mass Communication Quarterly, 74, 400-416.

Eveland

W. P.

(2004). The effect of political discussion in producing informed citizens: The roles of information, motivation, and elaboration. Political Communication, 21, 177-193.

Eveland

W. P.

Jr. Hively

M. H.

(2009). Political discussion frequency, network size, and “heterogeneity” of discussion as predictors of political knowledge and participation. Journal of Communication, 59, 205-224.

10.

Eveland

W. P.

Thomson

(2006). Is it talking, thinking, or both? A lagged dependent variable model of discussion effects on political knowledge. Journal of Communication, 56, 523-542.

11.

Feldman

Price

(2008). Confusion or enlightenment? How exposure to disagreement moderates the effects of political discussion and media use on candidate knowledge. Communication Research, 35(1), 61-87.

12.

Fishkin

(1995). The voice of the people. New Haven, CT: Yale University Press.

13.

Guttman

Thompson

(1996). Democracy and disagreement. Cambridge, MA: Harvard University Press.

14.

Habermas

(1996). Between facts and norms. Cambridge, MA: MIT Press.

15.

Hardy

B. W.

Scheufele

D. A.

(2009). Presidential campaign dynamics and the ebb and flow of talk as a moderator: Media exposure, knowledge, and political discussion. Communication Theory, 19(1), 89-110.

16.

Hayes

Glynn

Huge

(2008). Cautions in the interpretation of coefficients and hypothesis tests from linear models with interactions. Paper presented at the annual meeting of the International Communication Association, Montreal, Canada, QC.

17.

Heaney

M. T.

Rojas

(2007). Partisans, nonpartisans, and the anti-war movement in the United States. American Politics Research, 35, 431-464.

18.

Huckfeldt

Johnson

P. E.

Sprague

(2004). Political disagreement. New York, NY: Cambridge University Press.

19.

Huckfeldt

Mendez

J. M.

Osborn

(2004). Disagreement, ambivalence, and engagement: The political consequences of heterogeneous networks. Political Psychology, 25(1), 65-95.

20.

Kim

Wyatt

R. O.

Katz

(1999). News, talk, opinion, participation: The part played by conversation in deliberative democracy. Political Communication, 16, 361-385.

21.

Kwak

Williams

A. E.

Wang

X. R.

Lee

(2005). Talking politics and engaging politics: An examination of the interactive relationships between structural features of political talk and discussion engagement. Communication Research, 32(1), 87-111.

22.

Lam

W. M.

(2004). Understanding the political culture of Hong Kong. New York, NY: M.E. Sharpe.

23.

Lee

F. L. F.

(2005). Ordinary political conversation and public opinion expression: Is existence of discord necessary? Journalism & Mass Communication Quarterly, 82, 891-909.

24.

Lee

F. L. F.

(2006). Collective efficacy, support for democracy, and political participation in Hong Kong. International Journal of Public Opinion Research, 18, 297-317.

25.

Lee

F. L. F.

(2007). Talk radio listening, opinion expression and political discussion in a democratizing society. Asian Journal of Communication, 17(1), 78-96.

26.

Lee

F. L. F.

(2009). The impact of political discussion in a democratizing society: The moderating role of disagreement and support for democracy. Communication Research, 36, 379-399.

27.

Leighley

J. E.

(1990). Social interaction and contextual influences on political participation. American Politics Quarterly, 18, 459-475.

28.

Leurdijk

(1997). Common sense versus political discourse: Debating racism and multicultural society in Dutch talk shows. European Journal of Communication, 12, 147-168.

29.

Lui

T. L.

Chui

S. W. K.

(Eds). (2000). The dynamics of social movements in Hong Kong. Hong Kong, People’s Republic of China: Hong Kong University Press.

30.

Chan

J. M.

(2006). Global connectivity and local politics: SARS, talk radio, and public opinion. In Davis

Siu

(Eds), SARS: Reception and interpretation in three Chinese cities (pp. 19-46). London, UK: Routledge.

31.

Mansbridge

(1999). Everyday talk in the deliberative system. In Macedo

(Ed.), Deliberative politics (pp. 211-240). New York, NY: Oxford University Press.

32.

McClurg

S. D.

(2003). Social networks and political participation: The role of social interaction in explaining political participation. Political Research Quarterly, 56, 449-464.

33.

McClurg

S. D.

(2006a). Political disagreement in context: The conditional effect of neighborhood context, disagreement and political talk on electoral participation. Political Behavior, 28, 349-366.

34.

McClurg

S. D.

(2006b). The electoral relevance of political talk: Examining disagreement and expertise effects in social networks on political participation. American Journal of Political Science, 50, 737-754.

35.

McGraw

K. M.

Bartels

(2005). Ambivalence toward American political institutions: Sources and consequences. In Craig

S. C.

Martinez

M. D.

(Eds.), Ambivalence and the structure of political opinion (pp. 105-126). New York, NY: McMillan.

36.

Meffert

M. F.

Guge

Lodge

(2004). Good, bad, and ambivalent: The consequences of multidimensional political attitudes. In Saris

W. E.

Sniderman

P. M.

(Eds.), Studies in public opinion (pp. 63-92). Princeton, NJ: Princeton University Press.

37.

Mouffe

(2000). The democratic paradox. New York, NY: Verso.

38.

Mutz

D. C.

(2002a). Cross-cutting social networks: Testing democratic theory in practice. American Political Science Review, 96(1), 111-126.

39.

Mutz

D. C.

(2002b). The consequences of cross-cutting networks for political participation. American Journal of Political Science, 46, 838-855.

40.

Mutz

D. C.

(2006). Hearing the other side. New York, NY: Cambridge University Press.

41.

Mutz

D. C.

(2008). Is deliberative democracy a falsifiable theory? Annual Review of Political Science, 11, 521-538.

42.

Nir

(2005). Ambivalent social networks and their consequences for participation. International Journal of Public Opinion Research, 17, 422-442.

43.

Noelle-Neumann

(1983). Spiral of silence. Chicago, IL: University of Chicago Press.

44.

Pan

Z. D.

Shen

L. J.

Paek

H. J.

Sun

(2006). Mobilizing political talk in a presidential campaign: An examination of campaign effects in a deliberative framework. Communication Research, 33, 315-345.

45.

Parsons

B. M.

(2010). Social networks and the affective impact of political disagreement. Political Behavior, 32, 181-204.

46.

Pattie

C. J.

Johnston

R. J.

(2008). It’s good to talk: Talk, disagreement and tolerance. British Journal of Political Science, 38, 677-698.

47.

Pattie

C. J.

Johnston

R. J.

(2009). Conversation, disagreement and political participation. Political Behavior, 31, 261-285.

48.

Scheufele

D. A.

(2002). Examining differential gains from mass media and their implications for participatory behavior. Communication Research, 29(1), 46-65.

49.

Scheufele

D. A.

Eveland

W. P.

(2001). Perceptions of “public opinion” and “public” opinion expression. International Journal of Public Opinion Research, 13(1), 25-44.

50.

Scheufele

D. A.

Hardy

B. W.

Brossard

Waismel-Manor

I. S.

Nisbet

(2006). Democracy based on difference: Examining the links between structural heterogeneity, heterogeneity of discussion networks, and democratic citizenship. Journal of Communication, 56, 728-753.

51.

Scheufele

D. A.

Moy

(2000). Twenty-five years of the spiral of silence: A conceptual review and an empirical outlook. International Journal of Public Opinion Research, 12(1), 3-28.

52.

Scheufele

D. A.

Nisbet

M. C.

Brossard

Nisbet

E. C.

(2004). Social structure and citizenship: Examining the impacts of social setting, network heterogeneity, and informational variables on political participation. Political Communication, 21, 315-338.

53.

Schudson

(1997). Why conversation is not the soul of democracy. Critical Studies in Mass Communication, 14, 297-309.

54.

Searing

D. D.

Solt

Conover

P. J.

Crewe

(2007). Public discussion in the deliberative system: Does it make better citizens? British Journal of Political Science, 37, 587-618.

55.

Thompson

(2008). Deliberative democratic theory and empirical political science. Annual Review of Political Science, 11, 497-520.

56.

Verba

Nie

(1972). Participation in America. New York, NY: Harper & Row.

57.

Wojcieszak

Baek

Y. M.

Delli Carpini

M. X.

(2010). Deliberative and participatory democracy? Ideological strength and the process leading from deliberation to political engagement. International Journal of Public Opinion Research, 22, 154-180.

58.

Wyatt

R. O.

Katz

Kim

(2000). Bridging the spheres: Political and personal conversation in public and private spaces. Journal of Communication, 50(1), 71-92.

59.

Young

I. M.

(2000). Inclusion and democracy. Oxford, UK: Oxford University Press.