“Ditto Heads”

Abstract

In three studies, we examined (a) whether conservatives possess a stronger desire to share reality than liberals and are therefore more likely to perceive consensus with politically like-minded others even for non-political judgments and, if so, (b) whether motivated perceptions of consensus would give conservatives an edge in progressing toward collective goals. In Study 1, participants estimated ingroup consensus on non-political judgments. Conservatives perceived more ingroup consensus than liberals, regardless of the amount of actual consensus. The desire to share reality mediated the relationship between ideology and perceived ingroup consensus. Study 2 replicated these results and demonstrated that perceiving ingroup consensus predicted a sense of collective efficacy in politics. In Study 3, experimental manipulations of affiliative motives eliminated ideological differences in the desire to share reality. A sense of collective efficacy predicted intentions to vote in a major election. Implications for the attainment of shared goals are discussed.

Keywords

perceived consensus political ideology group efficacy political behavior

Over the past two decades, the radio show of politically conservative commentator Rush Limbaugh has amassed a huge number of listeners, drawing over a million each day (Edwards, 2012). As the show began to attract a larger and more ideologically zealous crowd, followers frequently called in to express support for Limbaugh’s opinions. To speed up calls, listeners began by simply saying “ditto”—a colloquial way of endorsing statements that Limbaugh and earlier callers had already made. Over time, fans of the show have come to identify as “Ditto Heads” (Zasky, 2012).

Limbaugh’s “Ditto Heads” exemplify the phenomenon of like-minded individuals agreeing (or believing they agree) with one another and receiving social support for their convictions. The notion that individuals look to others to construct and validate their beliefs about the world has been a pervasive theme in social psychology (Allport, 1954; Festinger, 1954). Expanding on this theme, Hardin and Higgins (1996) proposed shared reality theory, which suggests that individuals are motivated to seek social verification for their beliefs. Achieving a sense of shared reality helps to satisfy epistemic and relational needs insofar as individuals come to think that their beliefs reflect objective reality and are shared by important others, such as friends and family members (Hardin & Conley, 2001).

Social psychologists have typically examined the motivation to share reality as a function of situational factors without accounting for chronic individual differences (e.g., Sinclair, Huntsinger, Skorinko, & Hardin, 2005). In the present research, we considered the possibility that there are meaningful individual differences in the desire to share reality and that these differences are politically significant. Specifically, we focused on the possibility that political conservatives, compared with liberals, are more motivated to perceive and maintain consensus among fellow ideologues. This possibility is consistent with the finding that conservatives possess stronger epistemic and relational needs than do liberals (Graham, Haidt, & Nosek, 2009; Jost, Ledgerwood, & Hardin, 2008).

We tested the hypothesis that ideological differences in the motivation to share reality would lead conservatives (more than liberals) to perceive consensus with like-minded others. In addition, we extended previous research on group motivation (e.g., Gibson, Randel, & Earley, 2000) by addressing the possibility that perceiving stronger ingroup consensus would give conservatives an edge in perceiving collective efficacy and striving to achieve group goals through a commitment to political action, including voting.

In three studies, we investigated whether there are relatively stable ideological differences in the motivation to share reality, such that conservatives would exhibit a stronger desire than liberals; whether ideological differences in shared reality motivation impact perceptions of ingroup consensus; and the political consequences, if any, of ideological differences in perceiving ingroup consensus. Our basic theoretical model is displayed in Figure 1.

Figure 1.

Theoretical model linking ideology to the desire to share reality, and linking both ideology and the desire to share reality to perceived ingroup consensus.

Ideological Differences in the Desire to Share Reality

Shared reality theory holds that individuals possess the desire to receive social verification for their beliefs and develop a shared understanding of the world (Echterhoff, Higgins, & Levine, 2009). Previous research has examined the cognitive and behavioral effects of temporarily increasing the desire to share reality with others (e.g., Sinclair et al., 2005) but has not tested whether there is dispositional variability in the strength of epistemic and relational motives to share reality. Regarding epistemic motives, conservatives evince stronger desires to attain certainty, order, stability, and closure than do liberals (e.g., Jost, Glaser, Kruglanski, & Sulloway, 2003). Regarding relational motives, conservatives place a higher value on conformity, loyalty, and group cohesion than liberals do (Feldman, 2003; Graham et al., 2009). Thus, we hypothesized that conservatives would possess a stronger desire than liberals to share reality with fellow ideologues (Figure 1, path a).

Developing Shared Reality Through Perceiving Ingroup Consensus

How do individuals achieve a sense of shared “social reality”? One way is to cultivate perceptions or beliefs that are similar to those held by valued others (Festinger, 1954; Hardin & Higgins, 1996; Sinclair et al., 2005). However, individuals do not often possess direct knowledge of how others think. Instead, they achieve a sense of shared reality by perceiving that others share their beliefs (Kruglanski, Pierro, Mannetti, & De Grada, 2006). This phenomenon of assuming similarity to others has been referred to as “projection” (Bramel, 1963) and the “false consensus effect” (Ross, Greene, & House, 1977).

According to shared reality theory, individuals seek to achieve a common perception of the world with close or valued others (Hardin & Higgins, 1996). Supporting this idea, individuals tend to perceive relationship partners (Kenny & Acitelli, 2001) and those they like (versus dislike; Bramel, 1963) as holding beliefs that are similar to their own. Ottati, Fishbein, and Middlestadt (1988) demonstrated that individuals who held favorable views toward a presidential candidate (either Ronald Reagan or Walter Mondale) overestimated how similar the candidate’s positions were to their own. More germane to the present research, individuals perceive ingroup (versus outgroup) members (Robbins & Krueger, 2005) as more likely to share their goals, values, and emotional experiences in general. Thus, we hypothesized that individuals with a stronger desire to share reality would be more likely to perceive consensus with fellow ideologues (Figure 1, path b). Importantly, shared reality theory suggests that individuals seek out social verification for “any experience—ranging from the immediate tactile sensation of a stone to the abstract understanding of a philosophical concept” (Hardin & Higgins, 1996, p. 29). Thus, we hypothesized that the desire to share reality would predict perceived consensus on virtually any type of judgment, including judgments that are unrelated to politics. Given that we predicted that conservatives would possess a stronger desire to share reality than liberals (path a), and that individuals with a stronger desire to share reality would perceive greater ingroup consensus (path b), it follows that conservatives would perceive greater ingroup consensus than liberals (path c).

Perceived Consensus: Actual or Imagined?

Conservatives might perceive more ingroup consensus than liberals, but do they actually possess more ingroup consensus? Because we predicted that conservatives would perceive greater ingroup consensus than liberals because of the motivation to develop a sense of shared reality, it is important to rule out the alternative possibility that conservatives perceive greater consensus simply because they possess accurate knowledge of ingroup consensus. Thus, we selected non-political domains in which we expected actual levels of consensus to vary considerably. For political judgments, we assumed that perceivers may have base-rate knowledge on which to base their perceptions. For non-political judgments, however, we theorized that participants would probably not possess base-rate knowledge concerning how similar their judgments would be to politically like-minded others. We therefore reasoned that, to the extent that conservatives’ perceptions are motivationally based, they would perceive greater consensus than liberals regardless of the actual degree of consensus they possess.

Political Consequences of Perceiving Ingroup Consensus

We additionally consider whether there are consequences of perceiving ingroup consensus, even on non-political judgments, for the achievement of political goals. We tested whether perceiving ingroup consensus would predict feelings of efficacy—that one’s political ingroup can effectively attain its goals (van Zomeren, Spears, Fischer, & Leach, 2004)—and whether this feeling predicted intentions to vote, a primary way in which groups strive to achieve political goals.

Goals of the Present Research

We investigated four general questions in the present research program. First, we assessed whether political ideology would be associated with individual differences in shared reality motivation (Question 1). Second, we explored whether ideological differences in shared reality motivation would affect perceptions of ingroup consensus (Question 2). Third, we investigated whether perceiving ingroup consensus would predict behavioral intentions in politics (Question 3). Finally, we examined when liberals and conservatives would and would not differ in the desire to share reality (Question 4). In summary, then, the present research seeks to illuminate the ways in which basic motivational differences between liberals and conservatives translate into politically meaningful outcomes.

Study 1

In Study 1, we addressed Questions 1 and 2. To begin, we investigated whether conservatives possess a stronger motivation to share reality than liberals. Given that there is no widely accepted individual-difference measure of the motivation to share reality, we sought to create a face-valid measure that would closely reflect Hardin and Higgins’s (1996) conceptualization. Previous work in this area, such as Feldman’s (2003) development of the social conformity-autonomy scale, has assessed relational motivation in specific contexts (e.g., politics, childrearing) through the endorsement of normative standards (e.g., whether individuals ought to adhere to societal standards), and in reference to other people in general. By contrast, we hew more closely to the theoretical foundations of shared reality and social comparison perspectives, both of which emphasize the psychological significance of seeking and finding common ground with similar or like-minded others regardless of the specific domain or context.

Testing Question 2, we investigated whether conservatives would perceive greater consensus with their ingroup than would liberals. To test whether conservatives perceive greater consensus with anyone, regardless of their political beliefs, we also examined perceptions of consensus with others in general.

To examine whether conservatives perceive greater ingroup consensus than liberals independent of their level of actual ingroup consensus, we selected two types of judgments on which we expected levels of actual consensus to vary: sexual orientation and birth month. We selected sexual orientation judgments because previous research has found that these judgments are often stereotype-based (i.e., phenotypically feminine faces are perceived as gay and masculine faces are perceived as straight), and conservatives are more likely than liberals to rely on these shared stereotypes when judging sexual orientation from faces (Stern, West, Jost, & Rule, 2013). We chose birth month judgments because we did not expect actual differences in ingroup consensus to emerge, insofar as liberals and conservatives are unlikely to possess pre-existing stereotypes about what individuals born in one month versus another look like. These judgments are also clearly devoid of political content. Thus, we expected little consensus with respect to birth month judgments but expected that conservatives, at least, would possess some degree of consensus for sexual orientation judgments. By assessing perceived consensus for both of these types of judgments, we were able to examine whether conservatives would perceive greater consensus than liberals independent of their actual level of consensus. For both types of judgments, we hypothesized that conservatives would perceive more consensus than liberals because of their stronger motivation to share reality.

Method

Stimuli

Participants viewed photos of 30 White undergraduate men chosen randomly from a database used in prior research (see Rule, Ambady, Adams, & Macrae, 2008). Targets posed facing forward in all images and possessed no jewelry, glasses, tattoos, or facial hair. All images were cropped at the target’s neck (hair and ears retained) and standardized in height.

Procedure

American participants (n = 107, 72 women; M_age = 34.66 years, range = 18-64) completed the study through Amazon’s Mechanical Turk website (see Buhrmester, Kwang, & Gosling, 2011, for a discussion of this platform as a research tool) using Qualtrics online survey software. After providing consent, participants read that they would be shown several male faces. Through random assignment, participants were told that they would either indicate each person’s sexual orientation (gay or straight; n = 54) or birth month (November or December; n = 53).

Face judgments

Participants viewed all faces individually in random order with two scales presented below each face. For sexual orientation judgments, participants used one scale to indicate how likely the target was to be gay, and the second to indicate how likely the target was to be straight. For birth month judgments, participants used the scales to indicate the likelihood that the target was born in November and December, respectively. All scales ranged from 1 (not at all likely) to 7 (very likely).

Perceived consensus

To measure perceived ingroup consensus, participants were asked, “What percent of participants who share your political beliefs made similar judgments as you did?” ranging from 0% to 100%. To determine whether their desire for consensus generalized to all others, participants were asked, “What percent of participants overall made similar judgments as you did?” To estimate perceived outgroup consensus, participants were asked, “What percent of participants who do not share your political beliefs made similar judgments to one another?” Participants received no demographic information concerning the other participants in the study.

Shared reality and political ideology

To measure the desire to share reality, participants were asked “How important is it that you see the world in a similar way as people who generally share your beliefs do?” from 1 (not at all important) to 7 (very important). To measure political ideology, participants were asked “Where on the following scale of political orientation would you place yourself?” from 1 (extremely liberal) to 9 (extremely conservative; M = 4.17, SD = 2.44). This single-item measure of ideology is commonly used and exhibits strong predictive validity (e.g., Graham et al., 2009; Jost, 2006). We counterbalanced whether participants were asked about their ideology at the beginning or end of all studies; order did not moderate the effects of ideology on the desire to share reality or perceptions of ingroup consensus in any studies and is not discussed further.

Results

Men (M = 4.79, SD = 2.37) were marginally more conservative than women (M = 3.86, SD = 2.45), t(104) = −1.85, p = .067, d = .36. Although participant gender did not interact with ideology in any of our analyses, we did include gender as a predictor in all regression analyses.¹ Effect sizes for predictors in all multiple regression models below are reported as semipartial correlations (r_sp; Aloe & Becker, 2012). This estimate of effect size provides a standardized relationship between the predictor variable and dependent variable while adjusting for all other predictors in the model. Unlike other standardized estimates in a regression output (e.g., standardized beta), this estimate can be squared to determine the percentage of variance in a dependent variable that a predictor variable explains.

Desire to share reality

As hypothesized, participants who reported being more conservative indicated a stronger desire to share reality with fellow ideologues, B = .18, SE = .06, t(103) = 2.90, p = .005, r_sp = .27.

Perceived ingroup consensus

Participants who reported being more conservative perceived more ingroup consensus, B = 1.83, SE = .82, t(101) = 2.24, p = .03, r_sp = .21 (Figure 2). There was neither a significant main effect of judgment type (coded 1 and −1 for sexual orientation and birth month, respectively),² B = 1.88, SE = 1.97, t(101) = .96, p = .34, r_sp = .09, nor an Ideology × Judgment type interaction, B = .18, SE = .81, t(101) = 0.22, p = .83, r_sp = .02.

Figure 2.

Perceived ingroup consensus in Study 1 as a function of judgment type and perceiver ideology (plotted at 1 SD below and above the ideology mean).

Next, we tested whether conservatives perceived greater ingroup consensus because they possessed a stronger desire to share reality compared to liberals. Judgment type was not included in the mediation analysis because it did not moderate any effects. As expected, participants who indicated a stronger desire to share reality perceived more ingroup consensus, B = 4.55, SE = 1.16, t(103) = 3.94, p < .001, r_sp = .36. We tested whether ideology predicted perceived ingroup consensus through the desire to share reality (Figure 3) using Model 4 (a model specification with one exogenous variable, one mediator, and one outcome variable) with 5,000 bootstraps in the PROCESS macro (see Hayes, 2013). The 95% confidence interval of the indirect effect did not contain zero [0.1959, 1.6706], suggesting significant mediation at α = .05. Thus, conservatives perceived more ingroup consensus than liberals partly because they possessed a stronger shared reality motive.

Figure 3.

Mediation model in Study 1 in which ideology predicted perceived ingroup consensus through the desire to share reality.

Perceived consensus with others in general

As expected, there was no main effect of ideology on perceived consensus with others in general, B = .84, SE = .92, t(101) = 0.92, p = .36, r_sp = .09; no main effect of judgment type, B = −.44, SE = 2.20, t(101) = −0.20, p = .84, r_sp = .02; and no Ideology × Judgment type interaction, B = .14, SE = .90, t(101) = 0.15, p = .88, r_sp = .02. Thus, conservatives’ inclination to perceive greater consensus did not significantly extend to others in general.

Perceived outgroup consensus

There was no main effect of judgment type on perceived outgroup consensus, B = .19, SE = 1.77, t(101) = 0.11, p = .92, r_sp = .01. There was a main effect of ideology, B = −1.88, SE = .73, t(101) = −2.56, p = .01, r_sp = .24, qualified by a significant Ideology × Judgment type interaction, B = −1.49, SE = .72, t(101) = −2.06, p = .04, r_sp = .20. Conservatives perceived less outgroup consensus than did liberals for sexual orientation judgments, B = −3.37, SE = 1.03, t(101) = −3.27, p < .001, r_sp = .31, but not for birth month judgments, B = −.39, SE = 1.03, t(101) = −.38, p = .71, r_sp = .04. Thus, conservatives believed that liberals disagreed in their judgments to a greater extent than liberals believed that conservatives disagreed in theirs, but only with respect to sexual orientation judgments.

Actual consensus

Analytic strategy

Participants’ judgments of targets as gay and straight were negatively correlated, r(1618) = −.76, p < .001.³ We created composite scores by reverse-coding straight judgments and averaging them with gay judgments for each participant (Ambady, Hallahan, & Conner, 1999).⁴ Participants’ judgments of birth month (November vs. December) were also negatively, but not strongly, correlated, r(1588) = −.23, p < .001. Thus, we analyzed the two ratings separately.⁵

We estimated participants’ actual consensus using Kenny’s (1994) Social Relations Model (SRM). The SRM is a componential analysis that decomposes each perceiver’s judgments into three sources of variance: target variance, which measures agreement among perceivers; perceiver variance, which measures individual differences in categorizing the faces; and relationship/error variance,⁶ which measures the remaining variance. We were most interested in target variance because it captures the degree of actual consensus in judgments.

The SRM estimates perceiver and target variances by analyzing dyadic perceptions among a group of individuals. As such, the model requires that individuals be separated into groups. Because we measured ideology continuously, we conducted a tertile split for these analyses. Individuals responding 1-3 were categorized as “liberal,” 4-6 as “moderate,” and 7-9 as “conservative.” We then estimated whether target variances for each of these three groups were significantly different from zero. Although we did not focus on moderates in analyses treating ideology as a continuous variable, we were able to estimate their similarity to liberals and conservatives when treating ideology as a categorical variable. For all of these analyses, we report Wald zs and p values. We do not present results of whether perceiver variances or relationship/error variances are different from zero in the main text because they are not of theoretical interest (see Note 6). Values of perceiver variances are presented in Tables 1, 2, and 6. Relative variances (i.e., absolute variance divided by total variance; Kenny, 1994) are listed in Table 1.

Table 1.

Relative Perceiver and Target Variances for Judgments in Study 1 as a Function of the Perceiver’s Ideological Group.

	Sexual orientation		Born in November		Born in December
Perceiver groups	Perceiver variance	Target variance	Perceiver variance	Target variance	Perceiver variance	Target variance
Liberals	.07*	.01	.05*	.00	.10*	.02
Moderates	.07*	.07*	.13**	.00	.18**	.01
Conservatives	.11^†	.10*	.15^†	.00	.15^†	.00
All perceivers	.08***	.05**	.09***	.00	.14***	.00

†

p < .10. *p < .05. **p < .01. ***p < .001.

Table. 2.

Relative Perceiver and Target Variances for Judgments in Study 2 as a Function of the Perceiver’s Ideological Group.

	Sexual orientation		Born in November		Born in December		Eats fruit		Eats vegetables
Perceiver group	Perceiver variance	Target variance	Perceiver variance	Target variance	Perceiver variance	Target variance	Perceiver variance	Target variance	Perceiver variance	Target variance
Liberals	.03	.00	.09*	.00	.08*	.02	.02	.00	.00	.02
Moderates	.12**	.10**	.15**	.00	.10**	.01	.01	.00	.01	.00
Conservatives	.19^†	.12*	.15^†	.00	.02	.00	.00	.00	.00	.02
All perceivers	.11***	.06**	.12***	.00	.08***	.00	.01	.01	.00	.00

†

p < .10. *p < .05. **p < .01. ***p < .001.

Target variance (actual consensus)

Target variance for sexual orientation judgments was significantly different from zero for conservatives, z = 2.29, SE = .12, p = .02, and moderates, z = 2.23, SE = .08, p = .03, but not liberals, z = 0.54, SE = .03, p = .59, indicating significant consensus only for moderates and conservatives (Table 1). As expected, there was no actual consensus in birth month judgments for any of the ideological groups (zs ≤ 1.09, ps ≥ .28).

Summary

As hypothesized, conservatives exhibited a stronger motivation to share reality and greater perceptions of ingroup consensus than did liberals. Judgment type did not moderate this effect, indicating that conservatives perceived more consensus than liberals for both sexual orientation and birth month judgments. Conservatives did not perceive significantly more consensus with participants overall or among non-like-minded individuals, suggesting that their stronger desire to share reality might most strongly impact perceived consensus with fellow conservatives.

Additional analyses revealed that conservatives perceived greater consensus than liberals both for judgments on which actual consensus was achieved (sexual orientation) and for judgments on which actual consensus was not achieved (birth month), suggesting that conservatives’ stronger desire to share reality might lead them to perceive greater consensus than liberals, independent of the amount of actual consensus they possess. In Study 2, we considered an additional type of social judgment and investigated whether perceiving consensus with like-minded others would be associated with perceiving one’s political party as efficacious.

Study 2

In Study 2, we aimed to replicate and extend the findings of Study 1 by including a new condition in which participants judged targets’ food preferences. We selected this type of judgment because (a) it is non-political, (b) we did not expect actual consensus to emerge given that there are no pervasive stereotypes about physical appearance and food preferences relevant to the present judgments, and (c) previous research has successfully assessed social perception processes by examining judgments of target individuals’ food preferences (Harris & Fiske, 2007). We expected the findings for food preferences to resemble those for birth month judgments.

In Study 2, we also addressed Question 3: Does perceiving ingroup consensus predict feelings of collective efficacy—the sense that one’s group possesses the capacity to produce a desired outcome? When a group is viewed as efficacious, its members are perceived as being able to work together to attain shared goals. Previous research suggests that perceiving consensus within one’s group should lead individuals to view it as more cohesive, confident, and capable of accomplishing shared goals (Bandura, 2000; Gibson et al., 2000). Here, we extended previous work by investigating whether perceptions of ingroup consensus in one domain would predict feelings of collective efficacy in a very different domain. We predicted that, insofar as conservatives perceive more ingroup consensus than liberals, they would also perceive greater collective efficacy in politics. In the American political system, individuals are most able to enact meaningful change by joining a political party and supporting the representatives and platforms of that party. Therefore, to assess participants’ perceptions of political efficacy, we focused on perceptions of their political parties.

Method

Participants, design, and stimuli

One-hundred fifty Americans (94 women; M_age = 34.03 years, range = 18-65) were recruited through Mechanical Turk for a study on the beliefs of individuals who actively belonged to a political party. Fifteen additional participants were excluded because they reported not belonging to a political party.

Stimuli and procedures were the same as in Study 1. Participants were randomly assigned to make judgments of sexual orientation (gay or straight; n = 51), birth month (November or December; n = 49), or food preferences (fruits or vegetables; n = 50). Participants who made food preference judgments used one scale to indicate the likelihood that the targets ate fruit and the other scale to indicate that they ate vegetables. All scales ranged from 1 (not at all likely) to 7 (very likely).

The desire to share reality and ideology were measured the same way as in Study 1 (M_ideology = 4.57, SD = 2.18). To measure perceived collective efficacy, we adapted six items from Chen, Gully, and Eden’s (2001) self-efficacy scale to tap expectations concerning their political party’s effectiveness in the upcoming election (α = .84). A sample item includes: “My political party will not be able to achieve most of the election goals (i.e., electing candidates) that it has set for itself” (reverse-coded). Participants responded to these items in the context of the 2012 electoral campaigns, using a 1 (completely disagree) to 7 (completely agree) scale.

Results

Men and women did not significantly differ in their self-reported ideology, t(148) = −1.00, p = .32, d = .16, and gender did not interact with ideology in any analyses. Nevertheless, we retained gender as a predictor in all analyses reported here (see Note 1).

Desire to share reality

Participants who reported being more conservative indicated a stronger desire to share reality, B = .19, SE = .06, t(147) = 3.25, p = .001, r_sp = .26.

Perceived ingroup consensus

Political ideology significantly predicted perceived ingroup consensus, collapsing across all three judgments types, B = 1.84, SE = .71, t(147) = 2.58, p = .01, r_sp = .21. Conservatives perceived more ingroup consensus than liberals (Figure 4).

Figure 4.

Perceived ingroup consensus in Study 2 as a function of judgment type and perceiver ideology (plotted at 1 SD below and above the ideology mean).

To examine whether judgment type moderated the relationship between ideology and perceived ingroup consensus, we created two orthogonal dummy-coded contrasts. We specified sexual orientation as the reference judgment because it was the only one for which we expected that participants would actually exhibit consensus (following Aiken & West, 1991). Predictors were political ideology (grand mean-centered), the birth month/sexual orientation contrast, the food preference/sexual orientation contrast, and the two-way interactions between ideology and each of the judgment contrasts. There were no main or interaction effects involving either of the contrast variables (ps > .47). Thus, conservatives again perceived greater ingroup consensus than did liberals, regardless of judgment type.

Perceived consensus with others in general

Collapsing across judgment type, ideology did not significantly predict ratings of perceived consensus with other participants in general, B = .81, SE = .70, t(147) = 1.16, p = .25, r_sp = .09, nor did it interact with judgment type to predict perceived consensus (ps > .30).

Perceived outgroup consensus

Collapsing across judgment type, ideology did not significantly predict perceived outgroup consensus, B = −.79, SE = .86, t(147) = −0.92, p = .36, r_sp = .08. However, we observed a marginally significant interaction between ideology and the sexual orientation/birth month contrast, B = 3.73, SE = 2.10, t(143) = 1.78, p = .078, r_sp = .14, and a significant interaction between ideology and the sexual orientation/food preference contrast, B = 4.69, SE = 2.06, t(143) = 2.27, p = .03, r_sp = .18. As in Study 1, conservatives perceived less outgroup consensus than liberals for sexual orientation judgments, B = −3.54, SE = 1.46, t(143) = −2.42, p = .02, r_sp = .20, but not for birth month judgments, B = .19, SE = 1.50, t(143) = 0.13, p = .90, r_sp = .01, or food preference judgments, B = 1.15, SE = 1.46, t(143) = 0.79, p = .43, r_sp = .06. Thus, conservatives again believed that liberals would disagree in their judgments to a greater extent than liberals believed that conservatives would, but only when it came to sexual orientation judgments.

Actual consensus

Analytic strategy

Participants’ judgments of targets as gay and straight were highly negatively correlated, r(1527) = −.82, p < .001; we therefore created a composite score using the same strategy as in Study 1. Judgments of how likely targets were to be born in November versus December, r(1468) = −.23, p = .001, and judgments of fruit and vegetable preferences, r(1484) = .08, p = .003, were not strongly correlated. Therefore, we analyzed these ratings separately (see Note 5). As in Study 1, we again used the SRM to decompose the judgment variance by target, perceiver, and relationship/error (Table 2).

Target variance (actual consensus)

For sexual orientation judgments, the target variance estimate significantly differed from zero for conservatives, z = 2.34, SE = .13, p = .02, and moderates, z = 2.89, SE = .08, p = .004, but not for liberals, z = .26, SE = .04, p = .80, indicating significant consensus in sexual orientation judgments only for moderates and conservatives. For birth month and food preference judgments, target variance estimates were not significantly different from zero for any of the ideological groups (zs ≤ .92, ps ≥ .35), replicating Study 1.

Path model testing the theoretical framework

Finally, we examined our hypothesized path model (Figure 5). Because judgment type did not affect perceived ingroup consensus, we did not include it in the model. We tested our hypothesized model using Mplus Version 7 (Muthén & Muthén, 1998-2012). We began by constructing a saturated model that included all possible paths (see Table 3 for zero-order correlations). Following the procedure recommended by Kline (2011), non-significant paths (.34 ≤ p ≥ .83) that were not predicted to be significant were trimmed from the final model, which exhibited good fit, χ²(3) = 2.18, p = .54, RMSEA < .000, comparative fit index (CFI) = 1.00. All predicted paths were significant and possessed coefficients of moderate size in the final model. We examined the indirect effect of political ideology on perceived ingroup consensus through the desire to share reality using the “indirect” command in Mplus with 5,000 bootstraps. The indirect effect yielded a 95% confidence interval that excluded zero [0.21, 1.35], suggesting significant mediation at α = .05. Thus, conservatives perceived more ingroup consensus in part because they possessed a stronger desire to share reality. Moreover, perceiving more consensus predicted confidence that one’s political group would perform efficaciously during the 2012 elections.

Figure 5.

Path model in Study 2.

Table 3.

Zero-Order Correlations Among Variables Presented in the Path Model in Study 2.

Measure	1	2	3
1. Ideology	—
2. Desire to share reality	.25**	—
3. Perceived ingroup consensus	.21*	.32***	—
4. Perceived ingroup efficacy	.14^†	.10	.21**

†

p < .10. *p < .05. **p < .01. ***p < .001.

Summary

In Study 2, we again observed that conservatives and moderates possessed actual consensus for sexual orientation judgments, whereas liberals did not. None of the ideological groups achieved consensus for birth month and food preference judgments. For all judgments, however, conservatives perceived more ingroup consensus than did liberals, suggesting that this ideological difference is independent of actual levels of group consensus. As in Study 1, conservatives did not perceive significantly greater consensus with others in general or among non-like-minded individuals, suggesting that their desire to share reality might most strongly impact perceived consensus with like-minded others.

As in Study 1, the desire to share reality mediated the relationship between ideology and perceptions of ingroup consensus. We also found that perceiving ingroup consensus predicted collective efficacy in the 2012 elections. This finding demonstrates a potential consequence of perceiving ingroup consensus that has implications for achieving collective political goals. Furthermore, it extends previous research on group efficacy by demonstrating that perceiving consensus (even on a non-political task) is associated with a sense of collective efficacy in another domain (politics). In Study 3, we sought more direct support for the theoretical model depicted in Figure 1 by experimentally manipulating the motivation to share reality and measuring its effects on behavioral intentions; namely, intentions to vote.

Study 3

In Study 3, we tested whether perceiving one’s political party as efficacious would predict the intention to engage in behavior that has direct implications for the success (or failure) of a political party being able to achieve its goals: voting in a major election. Whereas past research has linked feelings of personal efficacy to voting behavior (e.g., Finkel, 1985), Bandura (1997) proposed that, “[m]ost political participation is channeled through membership in various groups that unite shared interests into common goals” (p. 484), suggesting that feelings of group efficacy might likewise predict intentions to participate in politics. In addition, when individuals decide whether to vote, they often feel that their decision is diagnostic of what similar others will do (Acevedo & Krueger, 2004). As Quattrone and Tversky (1984) put it, “one may reason that if one votes, then one’s politically like-minded peers, who think and act like oneself, will also vote. Conversely, if one abstains, then one’s like-minded peers will also abstain” (p. 244). We hypothesized that perceiving one’s political party as more efficacious would be associated with stronger intentions to vote in a national election.

In Study 3, we addressed Question 4 by experimentally manipulating the desire to share reality by altering motives to affiliate (e.g., Sinclair et al., 2005). We drew from Banfield, Kay, Cutright, Wu, and Fitzsimons’s (2011) Person × Situation account of how individuals respond to temporary manipulations of motivation as a function of chronic levels of motivation. Specifically, Banfield and colleagues found that when participants’ motivation to defend the status quo was heightened, only participants who were chronically low in confidence in the sociopolitical system rejected organizations that promoted progressive system change. Participants who were chroncially high in system confidence were unaffected by the manipulation, presumably because it did not increase their (already strong) motivation to defend the status quo.

Based on Banfield et al.’s (2011) findings, we therefore hypothesized that liberals and conservatives would respond differently to situational manipulations that alter motives to affiliate. Studies 1 and 2 demonstrated that conservatives have a chronically stronger desire to share reality than do liberals. Thus, we expected that a manipulation that heightened affiliative motives would increase liberals’ (but not conservatives’) desire to share reality. Conversely, we expected that a manipulation that attenuated affiliative motives would significantly weaken conservatives’ (but not liberals’) desire to share reality.

Method

Participants, design, and stimuli

Three-hundred eleven Americans (210 women; M_age = 32.89 years, range = 18-70) were recruited online from Mechanical Turk. As in Study 2, we advertised the study as focusing on the beliefs of individuals who belonged to a political party. Twenty-nine additional participants were excluded from analyses because they reported not belonging to a political party.

Procedure

Affiliative motive manipulation

After providing consent, participants were randomly assigned to one of three affiliative motive conditions. In the attenuate motive condition (n = 101), participants read a passage intended to weaken their motivation to affiliate. Adapted from Sinclair et al. (2005), the passage asked participants to imagine returning from a weekend trip with friends feeling overwhelmed from the amount of social contact and then receiving a phone call from a friend asking to see a movie together. Participants wrote several sentences explaining how they would tell their friend that they needed time alone “to unwind and stop feeling socially overburdened.” In the induce motive condition (n = 105), participants read a similar passage but were asked to imagine returning from a weekend trip alone feeling socially isolated and then calling a close friend. They then wrote how they would explain to the friend that they needed some time with other people “to unwind and stop feeling socially distant.” In the no manipulation control condition (n = 105), participants proceeded to the judgment task without writing anything.

Face judgments

Procedures were identical to Study 2 with participants randomly assigned to judge either sexual orientation (gay or straight; n = 101), birth month (November or December; n = 106), or likelihood of eating certain foods (fruits or vegetables; n = 104). Stimuli were the same as in Studies 1 and 2.

Components of theoretical model

Except for the desire to share reality, measures were identical to Studies 1 and 2 (α_{perceived efficacy} = .85; M_ideology = 4.21, SD = 2.04). The desire to share reality with similar others was measured with three items using 1 to 7 scales (α = .75; for example, “I believe it is important that I see the world in a similar way as people who generally share my beliefs do”; 1 = completely disagree; 7 = completely agree). We adapted these items to measure participants’ desire to share reality with others in general (α = .73; for example, “I believe it is important that I see the world in a similar way as people in general do”).

Voting intentions

Participants expressed their intention to vote in response to the question, “How likely are you to vote in the 2012 elections?” from 1 (not at all likely) to 7 (very likely).

Results

Men and women did not significantly differ in their ideology, t(309) = −0.03, p = .97, d = .003, and gender did not interact with ideology in any analyses. Nevertheless, we retained gender as a predictor in all analyses reported here (see Note 1).

Desire to share reality with like-minded others

We tested whether inducing motives to affiliate would strengthen liberals’ (but not conservatives’) desire to share reality, and whether attenuating motives to affiliate would weaken conservatives’ (but not liberals’) desire to share reality. We created two orthogonal dummy-coded contrast variables to test for moderation by motive condition, specifying the control condition as the reference condition for both. One dummy-coded variable contrasted the attenuate motive condition with the control condition (Motive Contrast 1), and the other dummy-coded variable contrasted the induce motive condition with the control condition (Motive Contrast 2; Aiken & West, 1991).

Model predictors consisted of ideology (grand mean-centered), Motive Contrast 1, Motive Contrast 2, and the two-way interactions between ideology and each of the motive contrasts. The attenuate motive/control contrast was not significant, B = −.20, SE = .16, t(304) = −1.22, p = .22, r_sp = .07, but the induce motive/control contrast was, B = .35, SE = .16, t(304) = 2.14, p = .03, r_sp = .12. Participants assigned to the induce motive condition reported a stronger desire to share reality with similar others than did those in the control condition. Importantly, the hypothesized interaction effect emerged between ideology and the attenuate motive/control contrast, B = −.17, SE = .08, t(304) = −2.20, p = .03, r_sp = .12. In addition, the predicted interaction between ideology and the induce motive/control contrast was marginally significant, B = −.14, SE = .08, t(304) = −1.71, p = .089, r_sp = .10. To examine the specific nature of these interaction effects, we decompose each below (see simple effects analyses in Table 4).

Table 4.

Simple Effects Tests of the Desire to Share Reality With Similar Others in Study 3.

	B	SE	t	r _sp
Simple slopes of ideology
Control condition	0.16	0.06	2.80**	0.16
Attenuate motive condition	−0.01	0.05	−0.24	0.01
Induce motive condition	0.02	0.06	0.30	0.02
Motive contrast 1
Liberals	0.15	0.23	0.66	0.04
Conservatives	−0.55	0.23	−2.42*	0.13
Motive contrast 2
Liberals	0.63	0.23	2.73**	0.15
Conservatives	0.06	0.23	0.25	0.01

Note. Simple effects for conservatives and liberals tested at 1 SD above and below the ideology mean, respectively.

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

Effects of attenuating shared reality motivation

To unpack the interaction between ideology and Motive Contrast 1 (comparing the attenuate motive and control conditions), we separately examined the simple slopes of ideology in each condition (Figure 6). The main effect of ideology was significant in the control condition but, as predicted, not when affiliative motives were attenuated. Thus, conservatives assigned to the control condition again exhibited a stronger desire to share reality than liberals, as in Studies 1 and 2, but attenuating affiliative motives eliminated this difference.

Figure 6.

The desire to share reality with similar others plotted as a function of affiliative motive condition and perceiver ideology (1 SD below and above the ideology mean) in Study 3.

We next investigated whether attenuating affiliative motives would weaken conservatives’, but not liberals’, desire to share reality. We probed Motive Contrast 1 for conservatives (1 SD above the ideology mean) and liberals (1 SD below the ideology mean; Aiken & West, 1991). As predicted, conservatives in the attenuate motive condition reported a significantly weaker desire to share reality than did conservatives in the control condition. Liberals’ desire to share reality was unaffected by the manipulation.

Effects of inducing shared reality motivation

To unpack the interaction between ideology and Motive Contrast 2 (comparing the induce motive and control conditions), we examined the simple slope of ideology in the induce motive condition. Unlike the control condition (see above), the main effect of ideology was not significant. Thus, consistent with our hypothesis, inducing affiliative motives eliminated the ideological difference observed in the control condition.

To test whether inducing affiliative motives would strengthen the desire to share reality among liberals but not conservatives, we examined Motive Contrast 2 for conservatives (1 SD above the ideology mean) and liberals (1 SD below the ideology mean). As expected, liberals assigned to the induce motive condition reported a significantly stronger desire to share reality than did liberals assigned to the control condition. Conservatives’ desire to share reality was unaffected by the manipulation.

These results indicate that (a) ideological differences in the desire to share reality emerged in the control condition, consistent with Studies 1 and 2; (b) inducing or attenuating motives to affiliate eliminated this ideological difference; and (c) inducing motives to affiliate only affected liberals’ desire to share reality, and attenuating motives to affiliate only affected conservatives’ desire to share reality.⁷

Perceived ingroup consensus

We next examined whether altering motives to affiliate would affect perceptions of ingroup consensus. Model predictors consisted of ideology (grand mean-centered), Motive Contrast 1, Motive Contrast 2, and the two-way interactions between ideology and each motive contrast. The attenuate motive/control contrast was significant, B = −6.18, SE = 2.87, t(304) = −2.16, p = .03, r_sp = .12. Participants assigned to the attenuate motive condition perceived less ingroup consensus than did those in the control condition. However, neither the induce motive/control contrast, nor the two-way interactions between ideology and the motive contrasts were significant (ps > .15).

Collapsing across the three judgment types, ideology significantly predicted perceived ingroup consensus, B = 1.35, SE = .58, t(308) = 2.33, p = .02, r_sp = .13. As in Studies 1 and 2, conservatives perceived more ingroup consensus than did liberals. Ideology did not interact with judgment type (ps > .63).

Perceived consensus with others in general

Collapsing across judgment types, ideology did not significantly predict ratings of perceived consensus with others in general, B = .76, SE = .58, t(308) = 1.31, p = .19, r_sp = .07, nor did it interact with judgment type (ps > .44), replicating Studies 1 and 2.

Perceived outgroup consensus

Collapsing across judgment types, ideology did not significantly predict perceived outgroup consensus, B = −.57, SE = .57, t(308) = −1.00, p = .32, r_sp = .06. However, we observed a significant Ideology × Sexual orientation/Birth month contrast interaction, B = 3.31, SE = 1.31, t(304) = 2.52, p = .01, r_sp = .14, and a marginally significant Ideology × Sexual orientation/Food preference contrast interaction, B = 2.42, SE = 1.42, t(304) = 1.70, p = .09, r_sp = .10. As in Studies 1 and 2, conservatives perceived less outgroup consensus than liberals for sexual orientation judgments, B = −2.45, SE = .91, t(304) = −2.69, p = .008, r_sp = .15, but not for birth month, B = .86, SE = .94, t(304) = 0.91, p = .36, r_sp = .05, or food preference, B = −.03, SE = 1.09, t(304) = −0.03, p = .98, r_sp = .002, judgments. Thus, conservatives perceived that liberals would disagree in their judgments to a greater extent than liberals perceived that conservatives would, but only with respect to sexual orientation judgments.

Path model testing the theoretical framework

We next tested whether conservatives’ stronger desire to share reality would explain their greater perceptions of ingroup consensus (Figure 7). Given that the motive manipulations eliminated ideological differences in the desire to share reality, we expected that the desire to share reality would mediate the relationship between ideology and perceived ingroup consensus only in the control condition (i.e., we expected moderated mediation; Preacher, Rucker, & Hayes, 2007). Although ideology did not interact with the experimental conditions to predict perceptions of ingroup consensus, a direct interaction effect is not required for obtaining a meaningful moderated indirect effect (Shrout & Bolger, 2002).

Figure 7.

Path model in Study 3.

We first constructed a saturated model that included all possible paths. As previously mentioned, judgment type did not moderate any effects and was excluded from the model for the sake of parsimony. Because experimentally inducing and attenuating affiliative motives eliminated ideological differences in the desire to share reality, we combined the experimental conditions in the analysis (coded as no manipulation = 1; experimental conditions = −1; see Table 5 for zero-order correlations).

Table 5.

Zero-Order Correlations Among Variables Presented in the Path Model in Study 3.

Measure	1	2	3	4
1. Ideology	—
2. Desire to share reality	.09	—
3. Perceived ingroup consensus	.13*	.30***	—
4. Perceived ingroup efficacy	.14*	.12*	.27***	—
5. Intention to vote	.01	.02	.09	.28***

Note. Desire to share reality refers to items about sharing reality with similar others (as opposed to others in general).

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

After trimming non-significant paths (.11 ≤ p ≥ .88) that were predicted to be non-significant, the model fit the data well, χ²(9) = 7.40, p = .60, RMSEA < .000, CFI = 1.00. All predicted paths were significant and possessed coefficients of moderate size. We obtained indirect effects for each condition using Mplus specifications for Model 2 from Preacher et al. (2007) with 5,000 bootstraps. Model 2 provides simple indirect effects at each level of a moderator within the framework of a moderated mediation model where the moderator is specified to interact with the exogenous variable. As predicted, the confidence interval of the indirect effect of ideology predicting perceived ingroup consensus through the desire to share reality excluded zero [0.301, 1.510] in the control condition, suggesting significant mediation at α = .05. Thus, conservatives perceived more ingroup consensus in part because they possessed a stronger desire to share reality. The indirect effect in the experimental conditions contained zero [−0.464, 0.441], however, indicating that the desire to share reality no longer explained conservatives’ greater perceptions of ingroup consensus when ideological differences in the shared reality motive were eliminated. Additionally, perceiving greater ingroup consensus predicted increased confidence that one’s political party would be efficacious in the upcoming election, which predicted stronger voting intentions.

Actual consensus

Analytic strategy

Participants’ judgments of targets as gay and straight were again highly negatively correlated, r(3028) = −.80, p < .001, and so we created a composite score. Judgments of how likely targets were to be born in November versus December were again significantly but weakly correlated, r(3177) = −.22, p < .001, and judgments of fruit and vegetable preferences were not correlated, r(3115) = .002, p = .90, so we analyzed these measures separately. Using the SRM, we decomposed the variance in judgments into target, perceiver, and relationship/error variance (Table 6).

Table 6.

Relative Perceiver and Target Variances for Judgments in Study 3 as a Function of the Perceiver’s Ideological Group.

	Sexual orientation		Born in November		Born in December		Eats fruit		Eats vegetables
Perceiver group	Perceiver variance	Target variance	Perceiver variance	Target variance	Perceiver variance	Target variance	Perceiver variance	Target variance	Perceiver variance	Target variance
Liberals	.30***	.02^†	.15***	.00	.06**	.00	.00	.00	.01	.00
Moderates	.23***	.07***	.05**	.00	.03*	.00	.01	.00	.00	.00
Conservatives	.15*	.09*	.10*	.00	.06^†	.01	.00	.00	.00	.00
All perceivers	.23***	.06***	.10***	.00	.05***	.00	.00	.00	.00	.00

†

p < .10. *p < .05. **p < .01. ***p < .001.

Target variance (actual consensus)

For sexual orientation judgments, target variance was significantly different from zero for conservatives, z = 2.46, SE = .12, p = .01, and moderates, z = 3.22, SE = .06, p = .001, indicating significant consensus for these groups. For liberals, target variance was marginally different from zero, z = 1.75, SE = .02, p = .08, indicating a marginally significant degree of consensus. For birth month and food preference judgments, target variance was not significantly different from zero for any of the ideological groups (zs ≤ .69, ps ≥ .49), indicating a lack of consensus.

Summary

Study 3 replicated and extended Studies 1 and 2. We again observed that conservatives, but not liberals, exhibited a significant degree of consensus for sexual orientation judgments. For birth month and food preference judgments, however, neither group exhibited significant consensus. However, conservatives perceived more ingroup consensus than did liberals on all judgment types, suggesting that this ideological difference occurs independently of actual degrees of consensus. Conservatives did not perceive significantly greater consensus with others in general or among non-like-minded individuals, suggesting that their desire to share reality might most strongly impact consensus with like-minded others.

Study 3 examined more precisely when liberals and conservatives would differ in their desire to share reality. Results revealed that a manipulation attenuating affiliative motives weakened conservatives’ (but not liberals’) desire to share reality. Conversely, a manipulation enhancing affiliative motives strengthened liberals’ (but not conservatives’) desire to share reality. Thus, these experimental manipulations temporarily eliminated differences between liberals’ and conservatives’ desire to share reality. Furthermore, when the strength of liberals’ and conservatives’ motivation was equated, the desire to share reality no longer served as an explanation for why conservatives would perceive greater ingroup consensus.

Turning to consequences of perceiving consensus, we again found that individuals who perceived greater ingroup consensus perceived their political party as being more efficacious and capable of achieving collective goals. Furthermore, individuals who perceived their political party as more efficacious expressed stronger intentions to vote in the 2012 elections, demonstrating that perceiving consensus (even on non-political judgments) was associated with behavioral intentions that may facilitate the achievement of collective goals.

General Discussion

In the present research program, we demonstrated that conservatives are more likely than liberals to perceive consensus with fellow ideologues, and that this difference is partly due to differences between liberals and conservatives in the motivation to develop a shared sense of reality with like-minded others.

There are three reasons to conclude that ideological differences in perceptions of consensus are grounded in motivational differences. First, conservatives’ stronger perceptions of ingroup consensus were significantly mediated by the desire to share reality in all studies. Second, conservatives perceived greater ingroup consensus than liberals even when they did not actually exhibit greater judgmental consensus, revealing that conservatives’ estimates of consensus were independent of their actual levels of consensus. Third, Study 3 demonstrated that liberals and conservatives no longer differed in their desire to share reality when they received manipulations that either heightened or weakened affiliative motives.

We also examined the specificity of conservatives’ stronger perceptions of consensus—that is, whether they perceive greater consensus with everyone else or only fellow conservatives. Aggregating data across the three studies, we examined whether ideology predicted perceptions of ingroup consensus to a significantly greater extent than (a) perceptions of consensus with others in general, and (b) perceptions of outgroup consensus. We specified a model using generalized estimating equations (GEE) that accounted for the interdependence in participants’ perceptions of different types of consensus, specifying an exchangeable correlation matrix (Fitzmaurice, Laird, & Ware, 2012). Because there were three types of consensus (ingroup, general, and outgroup), we created two orthogonal dummy-coded contrasts with perceived ingroup consensus specified as the reference group (Aiken & West, 1991). The model included the main effect of ideology (grand mean-centered), the ingroup/general contrast, the ingroup/outgroup contrast, and the two-way interactions between ideology and each of the judgment contrasts. Gender was also included as a predictor. The model yielded a marginally significant Ideology × Ingroup/General-contrast interaction, B = −.79, SE = .41, z = −1.94, p = .052. Conservatives perceived more consensus than liberals in reference to both the ingroup, B = 1.58, SE = .38, z = 4.11, p < .001, and to others in general, B = .79, SE = .37, z = 2.12, p = .03, but ideology was a stronger predictor of perceptions of consensus with ingroup members than with others in general. The Ideology × Ingroup/Outgroup-contrast interaction was significant, B = −2.56, SE = .45, z = −5.67, p < .001. Although conservatives perceived more ingroup consensus than liberals (noted above), they perceived less outgroup consensus than liberals, B = −.98, SE = .40, z = −2.48, p = .01. These findings indicate that conservatives (in comparison to liberals) perceive greater consensus with fellow ideologues than with others in general or among outgroup members.

We have also demonstrated that perceiving consensus has implications for striving toward collective political goals; namely, the anticipation that one’s political party would exhibit more collective efficacy in the 2012 election (Studies 2-3). This sense of collective efficacy, in turn, predicted stronger intentions to vote in the election (Study 3). These findings suggest that perceiving consensus with respect to non-political judgments (e.g., guessing others’ birth month) has implications for outcomes that are politically meaningful.

Perceptions of Consensus and Group Efficacy: Facilitator of Group Achievement or Illusory Beliefs?

Does perceiving one’s group as efficacious facilitate the successful attainment of collective goals? Two theoretical perspectives speak to this issue. From one perspective, perceiving consensus and developing a sense of collective efficacy could mobilize group members and aid in overcoming obstacles to goal progress, especially in the incipient stages of forming a movement (Bandura, 1997; van Zomeren et al., 2004). Recent work suggests that conservatives may be motivated to perceive consensus, whereas liberals may be motivated to perceive their beliefs as relatively unique. Specifically, Stern, West, and Schmitt (2014) found that conservatives overestimated how similar their beliefs were to those that other conservatives espoused, whereas liberals underestimated how similar their beliefs were to those that other liberals held. Conservatives’ stronger perceptions of consensus within their ranks could help to rally their base, whereas liberals’ weaker perceptions of consensus could undermine the development of a truly cohesive movement. It may be telling that the conservative Tea Party movement crafted a highly unified set of goals early in its development and successfully formed a congressional caucus, whereas the progressive Occupy Wall Street movement resisted establishing a specific set of shared goals and, so far, has been largely unable to effect substantial change in American politics.

From another point of view, an exaggerated sense of consensus within one’s ranks could lead individuals to become overly confident that their group will achieve its goals and therefore undermine successful goal attainment. On the evening of the 2012 U.S. Presidential election, for example, many conservatives experienced shock and disbelief as Barack Obama recaptured the White House, suggesting that they had overestimated the likelihood that like-minded others would turn out to vote for Republican Mitt Romney. Furthermore, liberals’ reluctance to assume consensus for their opinions could work to their advantage by increasing their investment in organization to ensure that fellow liberals are “on the same page.” For instance, prior to the 2010 repeal of the U.S. military’s “Don’t ask, don’t tell” policy, liberals in the House of Representatives worried that they would not garner enough votes to undo the policy and worked extremely hard to broker consensus for their position among representatives. When the time came for actual consensus to be assessed, the policy was repealed seamlessly (Steinhauer, 2010).

In sum, perceiving ingroup consensus has implications for attaining collective goals. Perceiving ingroup consensus can be a beneficial way of mobilizing group members toward collective efforts, even if these perceptions are initially not grounded in reality (Gibson et al., 2000). However, if individuals continually base their understanding of groups’ capabilities on inaccurate perceptions of consensus, they are unlikely to fully recognize the hurdles that they must overcome to succeed.

Implications for Future Research

The present research elaborates on shared reality theory, demonstrating that political ingroup members can serve as sources of social verification—even when shared beliefs and perceptions are not directly related to the basis of group membership. Future research will be useful for establishing boundary conditions concerning precisely when individuals will (and will not) draw on ingroup members to verify their beliefs and perceptions.

In the present set of studies, we took an individual-differences approach to measuring and examining the consequences of the desire to share reality. Our approach is analogous to other approaches in which general perceptions of similarity to other people are used to predict perceptions of similarity on specific issues (e.g., Ames & Iyengar, 2005). Theoretically, individuals who possess a general motivation to share reality should desire consensus in nearly any domain and therefore be more likely to engage in behaviors that foster the development of consensus (e.g., social influence aimed at conformity). Therefore, the measure that we have created should be of interest to researchers examining consensus-building in a wide variety of domains. Nevertheless, it would be useful for future research to investigate these processes in the context of explicitly political conflicts (such as the abortion debate).

Our work also adds to research on consensus estimation, which has often emphasized cognitive factors (e.g., Krueger & Clement, 1997), by providing support for a complementary, more motivational account of why individuals perceive consensus with others. In the future, it would be useful to examine how perceptions of consensus derived through motivational and cognitive pathways might differentially relate to various outcomes (such as judgmental confidence or relationship commitment). Furthermore, given that there are more self-identified conservatives than liberals in the United States (Pew Research Center, 2012), future research would do well to determine whether the perceived size of one’s political ingroup might affect assumptions of ingroup consensus.

The present findings have direct implications for theory and research on group mobilization. It has been suggested that one reason individuals desire consensus is to develop empowerment or “collective self-efficacy” (i.e., “group locomotion”; see Festinger, 1950; Gibson et al., 2000). Our work extends this basic argument by highlighting the possibility that individuals might seek consensus for the purpose of satisfying a relational desire to share reality (versus a more explicit goal to develop a highly effective group), and yet perceptions of consensus, once established, may contribute to a sense of collective empowerment in different domains. Future research would do well to determine the extent to which feelings of group efficacy that develop in response to shared reality motivation effectively translate into successful cooperation and goal attainment. In any case, it is important to note that the research design for these studies was correlational. An interesting avenue for future research would be to systematically manipulate perceptions of consensus and to observe individuals’ feelings of collective efficacy and actual political behavior.

Conclusion

We have demonstrated that conservatives perceive greater ingroup consensus than liberals and that this is attributable to their stronger motivation to foster a shared sense of reality. Furthermore, these perceptions of consensus have implications for striving toward collective goals. As political movements of the left and right continue to play a prominent role in shaping public discourse, analyzing these movements in terms of their underlying psychological characteristics and motives may provide scientific insight to the likely causes of their successes and failures.

Footnotes

Declaration of Conflicting Interests

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

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by a National Science Foundation Graduate Research Fellowship to Chadly Stern and funding from New York University to Tessa V. West and John T. Jost.

Notes

References

Acevedo

Krueger

J. I.

(2004). Two egocentric sources of the decision to vote: The voter’s illusion and the belief in personal relevance. Political Psychology, 25, 115-134.

Aiken

West

(1991). Multiple regression: Testing and interpreting interactions. Newbury Park, CA: SAGE.

Allport

G. W.

(1954). The nature of prejudice. New York, NY: Holt.

Aloe

A. M.

Becker

B. J.

(2012). An effect size for regression predictors in meta-analysis. Journal of Educational and Behavioral Statistics, 37, 278-297.

Ambady

Hallahan

Conner

(1999). Accuracy of judgments of sexual orientation from thin slices of behavior. Journal of Personality and Social Psychology, 77, 538-547.

Ames

D. R.

Iyengar

S. S.

(2005). Appraising the unusual: Framing effects and moderators of uniqueness-seeking and social projection. Journal of Experimental Social Psychology, 4, 271-282.

Bandura

(1997). Self-efficacy: The exercise of control. New York, NY: W. H. Freeman.

Bandura

(2000). Exercise of human agency through collective efficacy. Current Directions in Psychological Science, 9, 75-78.

Banfield

J. C.

Kay

A. C.

Cutright

K. M.

E. C.

Fitzsimons

G. J.

(2011). A person by situation account of motivated system defense. Social Psychological & Personality Science, 2, 212-219.

10.

Bramel

(1963). Selection of a target for defensive projection. Journal of Abnormal and Social Psychology, 66, 318-324.

11.

Buhrmester

Kwang

Gosling

S. D.

(2011). Amazon’s Mechanical Turk: A new source of inexpensive, yet high-quality, data? Perspectives on Psychological Science, 6, 3-5.

12.

Chen

Gully

S. M.

Eden

(2001). Validation of a new General Self-Efficacy Scale. Organizational Research Methods, 4, 62-83.

13.

Echterhoff

Higgins

E. T.

Levine

J. M.

(2009). Shared reality: Experiencing commonality with others’ inner states about the world. Perspectives on Psychological Science, 4, 496-521.

14.

Edwards

(2012). Rush Limbaugh’s audience may be much smaller than you think. Retrieved from http://www.businessinsider.com/rush-limbaughs-audience-may-be-so-much-smaller-than-you-think-2012-3

15.

Feldman

(2003). Enforcing social conformity: A theory of authoritarianism. Political Psychology, 24, 41-74.

16.

Festinger

(1950). Informal social communication. Psychological Review, 57, 271-282.

17.

Festinger

(1954). A theory of social comparison processes. Human Relations, 7, 117-140.

18.

Finkel

S. E.

(1985). Reciprocal effects of participation and political efficacy: A panel analysis. American Journal of Political Science, 29, 891-913.

19.

Fitzmaurice

G. M.

Laird

N. M.

Ware

J. H.

(2012). Applied longitudinal analysis. Hoboken, NJ: John Wiley.

20.

Gibson

C. B.

Randel

A. E.

Earley

P. C.

(2000). Understanding group efficacy: An empirical test of multiple assessment methods. Group & Organization Management, 25, 67-97.

21.

Graham

Haidt

Nosek

B. A.

(2009). Liberals and conservatives rely on different sets of moral foundations. Journal of Personality and Social Psychology, 96, 1029-1046.

22.

Hardin

C. D.

Conley

T. D.

(2001). A relational approach to cognition: Shared experience and relationship affirmation in social cognition. In Moskowitz

G. B.

(Ed.), Cognitive social psychology: The Princeton symposium on the legacy and future of social cognition (pp. 3-17). Mahwah, NJ: Lawrence Erlbaum.

23.

Hardin

C. D.

Higgins

E. T.

(1996). Shared reality: How social verification makes the subjective objective. In Sorrentino

R. M.

Higgins

E. T.

(Eds.), Handbook of motivation and cognition: The interpersonal context (Vol. 3, pp. 28-84). New York, NY: Guilford Press.

24.

Harris

L. T.

Fiske

S. T.

(2007). Social groups that elicit disgust are differentially processed in mPFC. Social Cognitive and Affective Neuroscience, 2, 45-51.

25.

Hayes

A. F.

(2013). Introduction to mediation, moderation, and conditional process analysis. New York, NY: Guilford Press.

26.

Jost

J. T.

(2006). The end of the end of ideology. American Psychologist, 61, 651-670.

27.

Jost

J. T.

Glaser

Kruglanski

A. W.

Sulloway

F. J.

(2003). Political conservatism as motivated social cognition. Psychological Bulletin, 129, 339-375.

28.

Jost

J. T.

Ledgerwood

Hardin

C. D.

(2008). Shared reality, system justification, and the relational basis of ideological beliefs. Social & Personality Psychology Compass, 2, 171-186.

29.

Kenny

D. A.

(1994). Interpersonal perception: A social relations analysis. New York, NY: Guilford Press.

30.

Kenny

D. A.

Acitelli

L. K.

(2001). Accuracy and bias in the perception of the partner in a close relationship. Journal of Personality and Social Psychology, 80, 439-448.

31.

Kline

R. B.

(2011). Principles and practice of structural equation modeling. New York, NY: Guilford Press.

32.

Krueger

Clement

R. W.

(1997). Estimates of social consensus by majorities and minorities: The case for social projection. Personality and Social Psychology Review, 1, 299-313.

33.

Kruglanski

A. W.

Pierro

Mannetti

De Grada

(2006). Groups as epistemic providers: Need for closure and the unfolding of group-centrism. Psychological Review, 113, 84-100.

34.

Muthén

L. K.

Muthén

B. O.

(1998-2012). Mplus user’s guide (7th ed.). Los Angeles, CA: Author.

35.

Ottati

Fishbein

Middlestadt

S. E.

(1988). Determinants of voters’ beliefs about the candidates’ stands on the issues: The role of evaluative bias heuristics and the candidates’ expressed message. Journal of Personality and Social Psychology, 55, 517-529.

36.

Pew Research Center. (2012). Political ideology. Retrieved from http://www.pewresearch.org/data-trend/political-attitudes/political-ideology/

37.

Preacher

K. J.

Rucker

D. D.

Hayes

A. F.

(2007). Addressing moderated mediation hypotheses: Theory, methods, and prescriptions. Multivariate Behavioral Research, 42, 185-227.

38.

Quattrone

G. A.

Tversky

(1984). Causal versus diagnostic contingencies: On self-deception and on the voter’s illusion. Journal of Personality and Social Psychology, 46, 237-248.

39.

Robbins

J. M.

Krueger

J. I.

(2005). Social projection to ingroups and outgroups: A review and meta-analysis. Personality and Social Psychology Review, 9, 32-47.

40.

Ross

Greene

House

(1977). The “false consensus effect”: An egocentric bias in social perception and attribution processes. Journal of Experimental Social Psychology, 13, 279-301.

41.

Rule

N. O.

Ambady

Adams

R. B.

Jr. Macrae

C. N.

(2008). Accuracy and awareness in the perception and categorization of male sexual orientation. Journal of Personality and Social Psychology, 95, 1019-1028.

42.

Shrout

P. E.

Bolger

(2002). Mediation in experimental and nonexperimental studies: New procedures and recommendations. Psychological Methods, 7, 422-445.

43.

Sinclair

Huntsinger

Skorinko

Hardin

C. D.

(2005). Social tuning of the self: Consequences for the self-evaluations of stereotype targets. Journal of Personality and Social Psychology, 89, 160-175.

44.

Steinhauer

(2010, December 15). House votes to repeal “don’t ask, don’t tell.” The New York Times. Retrieved from http://www.nytimes.com/2010/12/16/us/politics/16military.html?_r=0

45.

Stern

West

T. V.

Jost

J. T.

Rule

N. O.

(2013). The politics of gaydar: Ideological differences in the use of gendered cues in categorizing sexual orientation. Journal of Personality and Social Psychology, 104, 520-541.

46.

Stern

West

T. V.

Schmitt

P. G.

(2014). The liberal illusion of uniqueness. Psychological Science, 25, 137-144.

47.

van Zomeren

Spears

Fischer

A. H.

Leach

C. W.

(2004). Put your money where your mouth is! Explaining collective action tendencies through group-based anger and group efficacy. Journal of Personality and Social Psychology, 87, 649-664.

48.

Zasky

(2012). Left turn: One man’s journey from dittohead to democrat. Retrieved from http://failuremag.com/feature/article/left_turn/

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB