Social Psychology

TheironicimpactofactivistsNegativestereotypesreducesocialchangeinfluence.pdf

Home >Psychology homework help >Social Psychology

European Journal of Social Psychology, Eur. J. Soc. Psychol. 43, 614–626 (2013) Published online 16 September 2013 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/ejsp.1983

Research article

The ironic impact of activists: Negative stereotypes reduce social change influence

NADIA Y. BASHIR1*, PENELOPE LOCKWOOD1, ALISON L. CHASTEEN1, DANIEL NADOLNY2 AND INDRA NOYES1 1 Department of Psychology, University of Toronto, Canada;

2 University of Waterloo, Canada

Abstract

Despite recognizing the need for social change in areas such as social equality and environmental protection, individuals often avoid supporting such change. Researchers have previously attempted to understand this resistance to social change by examining individuals’ perceptions of social issues and social change. We instead examined the possibility that individuals resist social change because they have negative stereotypes of activists, the agents of social change. Participants had negative stereotypes of activists (feminists and environmentalists), regardless of the domain of activism, viewing them as eccentric and militant. Furthermore, these stereotypes reduced participants’ willingness to affiliate with ‘typical’ activists and, ultimately, to adopt the behaviours that these activists promoted. These results indicate that stereotypes and person perception processes more generally play a key role in creating resistance to social change. Copyright © 2013 John Wiley & Sons, Ltd.

In 1964, the U.S. Congress enacted the Civil Rights Act, outlawing racial and gender discrimination (National Archives and Records Administration, 2011). More recently, in 2010, President Obama repealed the Don’t Ask, Don’t Tell policy, allowing openly gay Americans to serve in the military (The Library of Congress, 2011). These landmark events occurred only after activists spent many years actively challenging the status quo (House Committee on Armed Services, 1993; National Archives and Records Administra- tion, 2011), illustrating that long periods of slow progress typically precede social change. Such slow progress is at odds with research indicating that many individuals believe that it is important, socially desirable and moral to address social justice concerns (Beattie, 2010; Nelson et al., 2008). If individuals believe that social change is crucial and socially valued, they should generally be supportive of and responsive to the activists who advocate it. Yet although activists enthusiastically strive to address social justice concerns and are at times successful in promoting social change (e.g. Czopp, Monteith, & Mark, 2006), they often encounter substantial resistance from the public (Nelson et al., 2008; Superson & Cudd, 2002). Ironically, it may be this enthusiasm with which activists promote social change that undermines their impact: Rather than admiring their determination to address critical social issues, individuals may associate activists with negative stereotypes, viewing them as militant and eccentric. Accord- ingly, individuals may avoid affiliating with activists and disregard their pro-change initiatives. We examined this directly.

To date, researchers have attempted to understand resistance to social change by examining individuals’ perceptions of social issues, attitudes towards social change and personality traits.

*Correspondence to: Nadia Y. Bashir, Department of Psychology, University of E-mail: [email protected]

Researchers have assessed, for example, whether individuals avoid supporting social change because they deny or fail to perceive that a social issue or injustice exists (Gifford, 2011), perceive the issue to be personally irrelevant (Hodson & Esses, 2002) or believe that the status quo is acceptable (Morton, Postmes, Haslam, & Hornsey, 2009). In addition, researchers have examined whether individuals resist social change because they believe that it threatens positive aspects of the status quo (Kay & Friesen, 2011) or conflicts with their goals and beliefs (Feygina, Jost, & Goldsmith, 2010; Zárate, Shaw, Marquez, & Biagas, 2012). Finally, several studies have examined the personal characteristics that are associated with reduced support for social change, such as political conservatism and authoritari- anism (Agronick & Duncan, 1998; van Zomeren, Postmes, & Spears, 2008). Although this past research provides important insight into the theoretical basis for resistance to social change, one key element has been largely ignored: individuals’ perceptions of the people who strive to achieve this change, the activists themselves.

Indeed, even when individuals have perceptions of social issues and social change that are conducive to change (e.g. favourable perceptions of feminism), they are often still reluctant to identify with those who advocate this change (e.g. feminists; Aronson, 2003). Furthermore, individuals view activists in a variety of domains negatively: Feminists, for example, are typi- cally viewed unfavourably as aggressive, unconventional and unpleasant people (Berryman-Fink & Verderber, 1985; Twenge & Zucker, 1991). Similarly, portrayals of environmentalists and gay rights activists in government reports and sociological texts suggest that individuals view these activists as eccentric and militant (Brown, 2007; Federal Bureau of Investigation [FBI],

Toronto, 100 St. George Street, Toronto, ON M5S 3G3, Canada.

Received 22 September 2012, Accepted 20 August 2013

The ironic impact of activists 615

2001). Because activists, by definition, strive to effect change by publicly criticizing mainstream practices, they may be seen as hostile, unconventional and unpleasant. This tendency to associate activists with negative stereotypes may ultimately reduce individuals’ willingness to affiliate with activists and adopt the pro-change behaviours that activists espouse.

Specifically, because individuals strive to maintain a positive self-concept (Sedikides & Gregg, 2008) and consider their group memberships to be important components of their self-concepts, individuals typically desire membership in only those groups that they view positively (Tajfel & Turner, 1979). If individuals associate activists with negative stereotypes, therefore, they may avoid affiliating with activists who conform to these stereotypes (i.e. ‘typical’ activists), which may in turn reduce the likelihood that individuals will adopt behaviours that are characteristic of ‘typical’ activists. That is, because individuals have a strong need to belong and experience social acceptance (Baumeister & Leary, 1995), they may avoid engaging in behaviours that would character- ize them as individuals with whom it may seem unpleasant to affiliate. Given that individuals who are merely associated with stigmatized others can face prejudice and social rejection (Pryor, Reeder, & Monroe, 2012), individuals may fear that they too will be stigmatized and rejected by others if they affiliate with ‘typical’ activists and perform behaviours that are characteristic of such activists. Indeed, consistent with bal- ance theory (Heider, 1958), individuals may in part agree with social change ideologies but nevertheless avoid adopting pro-change behaviours because the ‘typical’ activists who advocate these behaviours seem dislikeable. By rejecting pro-change advocacy, individuals can distance themselves from individuals who are generally viewed negatively by soci- ety. In support of this possibility, evidence indicates that indi- viduals are less likely to adopt the opinions of stigmatized (e.g. Black or gay) versus nonstigmatized targets (Clark & Maass, 1988; White & Harkins, 1994). Although these studies do not show that negative stereotypes of stigmatized targets reduce individuals’ receptiveness to these targets, they are consistent with the possibility that individuals avoid adopting opinions espoused by targets whom they view negatively. Thus, individuals may at times resist social change, not neces- sarily because they have negative attitudes towards social issues or social change as previous research has indicated (e.g. Feygina et al., 2010; Hodson & Esses, 2002; van Zomeren et al., 2008) but rather because they have negative stereotypes of the agents of social change.

In five studies, therefore, we examined whether stereotypes of activists enhance resistance to social change by reducing individuals’ willingness to affiliate with activists and, ultimately, to adopt the pro-change behaviours that activists advocate. We first identified the stereotypes of two key activist groups and assessed how these stereotypes affect individuals’ willingness to affiliate with ‘typical’ activists (i.e. those who conform to activist stereotypes) and ‘atypical’ activists (i.e. those who do not conform to activist stereotypes; Studies 1 and 3–5). We then examined the extent to which activist stereotypes influence individuals’ motivation to adopt pro-change behaviours advocated by ‘typical’ versus ‘atypical’ activists (Study 2), because it affects their willingness to affiliate with these activists (Studies 3 and 4). We predicted that individuals

would be more likely to associate ‘typical’ activists with negative stereotypes, and consequently, they would avoid affiliating with and, ultimately, adopting the behaviours advocated by these activists.

PILOT STUDIES

We examined the influence of activist stereotypes on resistance to social change by focusing on two key activist groups: feminists and environmentalists. We chose to examine specific groups to show that, although there may be some differences in the specific traits that individuals associate with various activist groups, these stereotypes overlap considerably and have similar implications for resistance to social change. Because researchers have not previously examined the traits that individuals associate with environmentalists and because past research on feminist stereotypes may not reflect current perceptions of feminists, we first conducted a set of pilot studies to identify current stereotypes of these groups.

PILOT STUDIES A AND B

Method

Participants in Pilot Study A were 13 male and 26 female Americans recruited online via Amazon’s Mechanical Turk (Mage = 37.59 years, SD = 12.32) who received $0.20. Partici- pants in Pilot Study B were 49 male and 92 female undergrad- uate students (Mage = 19.44 years, SD = 2.02) who received course credit or $10.

Participants in Pilot Study A were randomly assigned to rate the extent to which 12 ‘militant/aggressive’ (e.g. ‘aggressive’ and ‘forceful’), 9 ‘eccentric/unconventional’ (e.g. ‘eccentric’ and ‘unusual’) and 12 ‘personable’ (‘friendly’ and ‘pleasant’; all Cronbach’s αs > .91) traits, which were selected on the basis of past research on feminist stereotypes (Berryman-Fink & Verderber, 1985; Twenge & Zucker, 1991), were characteris- tic of either a ‘typical’ feminist or a ‘typical’ American. Ratings were made along 7-point scales anchored at 1 (not at all characteristic of a ‘typical’ feminist/American) and 7 (very characteristic of a ‘typical’ feminist/American).

In Pilot Study B, participants assigned to one condition rated the extent to which 12 militant/aggressive (e.g. ‘militant’ and ‘abrasive’), 14 eccentric/unconventional (e.g. ‘eccentric’ and ‘odd-looking’) and 12 personable (e.g. ‘pleasant’ and ‘personable’; all αs > .81) traits were characteristic of a ‘typi- cal’ environmentalist. We note that these traits overlapped heavily with those used in aforementioned Pilot Study A but were selected on the basis of representations of environmental- ists in sociological texts and government reports (Brown, 2007; FBI, 2001). Participants assigned to a second condition rated a ‘typical’ university student, an individual whom student participants would view as a more mainstream member of society, on the same traits. Ratings were made along 7-point scales anchored at 1 (not at all characteristic of a ‘typical’ environmentalist/student) and 7 (very characteristic of a ‘typical’ environmentalist/student).

Eur. J. Soc. Psychol. 43, 614–626 (2013)

616 Nadia Y. Bashir et al.

Results

Pilot Study A

Independent t-tests revealed that participants viewed both militant and eccentric traits to be more characteristic of ‘typical’ feminists (Mmilitant = 5.36, SD = 1.26; Meccentric = 4.67, SD = 1.29) than of ‘typical’ Americans (Mmilitant = 4.05, SD = 1.23; Meccentric = 3.18, SD = 0.91), ts > 3.25, ps < .003, rs .47. Personable traits, in comparison, were viewed as less characteristic of ‘typical’ feminists (M = 3.55, SD = 1.29) than of ‘typical’ Americans (M = 4.60, SD = 0.69), t(37) = 3.31, p = .002, r = .48.

Pilot Study B

Two participants who were asked to rate a ‘typical’ student and indicated that there is no such thing as a ‘typical’ student were excluded. Thus, 49 male and 90 female participants were included in analyses. As was the case for feminists, partici- pants viewed ‘typical’ environmentalists (Meccentric = 5.06, SD = 0.60; Mmilitant = 3.59, SD = 0.79; Mpersonable = 4.09, SD = 0.64) as more eccentric, more militant and less person- able than ‘typical’ students (Meccentric = 3.92, SD = 0.52; Mmilitant = 3.29, SD = 0.57; Mpersonable = 4.54, SD = 0.56), ts > 2.25, ps < .04, rs > .19.1

Pilot Study C

Pilot Studies A and B indicate that individuals have negative stereotypes of two key activist groups: feminists and environ- mentalists. Given that the traits included in these pilot studies were selected from previous research and scholarly texts (e.g. Brown, 2007; Twenge & Zucker, 1991), we used these pilot studies as a basis for creating the stereotype measures used in Studies 1–3. We also, however, conducted an additional pilot study to verify that the traits identified in Pilot Studies A and B are not simply an artefact of the specific traits included in these initial pilot studies. We then used the traits identified in this additional pilot study to create the stereotype measure for Studies 4 and 5. In Pilot Study C, participants generated their own traits of ‘typical’ feminists and environmentalists.

Method

Participants were 228 Americans recruited online via Ama- zon’s Mechanical Turk (Mage = 33.75 years, SD = 11.71) who received $0.60. There were 92 male and 131 female participants. Five participants did not identify their biological sex. Participants generated 20 traits characteristic of a ‘typical’ feminist and 20 traits characteristic of a ‘typical’ environmentalist. The order of tasks was counterbalanced across participants.

1Participants may have viewed ‘typical’ environmentalists as less militant than personable because they were reluctant to endorse the extreme items (e.g. ‘criminal’) in the militant index. Participants were, nevertheless, more likely to associate militant traits with ‘typical’ environmentalists versus ‘typical’ students.

Results

As predicted, the majority of traits that participants listed as being characteristic of ‘typical’ feminists and environmental- ists were militant/aggressive and eccentric/unconventional traits. Table 1 contains the 30 most frequently listed traits for each group. Specifically, feminists were described in terms of militant/aggressive traits, such as ‘man hating’ and ‘force- ful’, and with eccentric/unconventional traits, such as ‘behaves like a man’ and ‘unhygienic’. Environmentalists were described in terms of militant/aggressive traits, such as ‘militant’ and ‘forceful’, and eccentric/unconventional traits, such as ‘eccentric’ and ‘tree-hugger’. Participants listed additional traits describing feminists or environmentalists that did not fall obviously into either of these categories (e.g. ‘animal lover’ and ‘Democrat’). Overall, however, the traits provided were overwhelmingly negative, with only a handful of more positive traits (e.g. ‘caring’ and ‘educated’) appearing on either list. Thus, it appears that individuals have negative perceptions of both feminists and environmentalists, viewing them primarily as aggressive militants and unconventional eccentrics rather than as pleasant and personable individuals. Notably, although we did not ask participants to describe either environmental or feminist ‘activists’ per se, they spon- taneously ascribed this trait to both groups: ‘activist’ was one of the top 10 most frequently listed traits for both groups.

STUDY 1

Maintaining a positive self-concept is a key goal for individuals (Sedikides & Gregg, 2008). Thus, if negative stereotypes char- acterize ‘typical’ activists as militant and eccentric, individuals may avoid affiliating with them. In Study 1, we examined this directly. Participants in one condition read about a ‘typical’ fem- inist, an individual who promoted women’s rights by organizing protests and challenging traditional representations of women. Participants in a second condition read about an ‘atypical’, personable feminist; this enabled us to examine whether partici- pants reacted negatively to all feminists or only to those who fit feminist stereotypes. Control participants read about an individ- ual whose stance on feminism was not described (i.e. an undefined target). All participants then rated the extent to which feminist stereotype traits were characteristic of the target, as well as their interest in affiliating with the target. We predicted that participants would be less interested in affiliating with the ‘typical’ feminist relative to the ‘atypical’ feminist and the undefined target, because they would be more likely to associate the ‘typical’ feminist with negative stereotypical traits.

We also used this study as an opportunity to rule out potential alternative explanations for our findings. Individuals who may be implicated in morally questionable behaviours dislike ‘moral rebels’, people who appear to condemn the behaviour by defending relevant social or moral values (Monin, Sawyer, & Marquez, 2008). Thus, individuals may express less desire in affiliating with ‘typical’ activists, not because they associate these targets with negative stereotypes but because they expect these activists to view them as immoral for failing to be similarly committed to promoting social change. Alternatively, given that

Eur. J. Soc. Psychol. 43, 614–626 (2013)

Table 1. Thirty most frequently mentioned traits describing ‘typical’ feminists and ‘typical’ environmentalists

Activist group Traits

‘Typical’ feminist Man hating (145), lesbian (133), unhygienic (95), angry (89), behaves like a man (87), unattractive (84), liberal (79), ambitious (73), loud (72), activist (72), mean (71), spinster (67), independent (65), forceful (65), assertive (64), whiny (63), abrasive (62), protester (57), competent (54), dresses like a man (53), self-righteous (51), bitter (49), overreactive (48), educated (47), strong-willed (45), strong (45), intolerant (44), irrational (42), annoying (42), bad dresser (42)

‘Typical’ environmentalist

Tree-hugger (151), vegetarian (124), hippie (124), liberal (111), unhygienic (91), militant (89), eccentric (85), activist (82), caring (81), protester (79), overreactive (68), unfashionable (63), self-righteous (61), educated (60), drug user (53), hairy (52), determined (52), stupid (51), intelligent (50), zealous (48), nontraditional (45), outdoorsy (43), forceful (43), animal lover (41), intolerant (40), helpful (40), Democrat (40), annoying (40), crazy (37), irrational (36)

Note: Values in parentheses indicate the number of participants who listed the trait.

The ironic impact of activists 617

similarity breeds liking (Byrne, 1971), individuals may express less desire to affiliate with ‘typical’ activists because these activists seem highly dissimilar to them. In Study 1, therefore, we examined not only participants’ tendency to attribute negative stereotypes to the target but also the extent to which they (i) believed that the target would view them as immoral and (ii) viewed themselves to be similar to the target.

2Stereotype traits items in Studies 1 and 2: radical, man hating, bitter, activist, aggressive, inflexible, argumentative, opinionated, outspoken, defensive, militant, preachy, hostile, forceful, nonconformist, confrontational, weird, strange, odd, ugly, eccentric, unconventional, nontraditional, unusual, approachable, fun, pleasant, friendly, popular, fashionable, social, cool and interesting (the last nine traits were reverse-scored).

Method

Participants

Participants were 17 male and 45 female undergraduate students (Mage =19.43years, SD= 4.99) who received course credit.

Procedure

Participants were randomly assigned to read a gender-neutral profile of a student who represented a ‘typical’ feminist, an ‘atypical’ feminist or an undefined target. Participants in the undefined target condition read about a target whose stance on feminism was not described (e.g. ‘When my weekends aren’t packed with schoolwork and volunteer activities, I usually spend the day watching TV or hanging out at coffee shops’). Participants in both the ‘typical’ and ‘atypical’ condi- tions read about a target who promoted feminism. However, whereas the ‘typical’ target used methods that were consistent with the stereotypes of feminists (e.g. ‘I also organize rallies outside corporate and political institutions in the community to pressure CEOs and politicians who don’t prioritize women’s rights issues into resigning’), the ‘atypical’ target promoted feminism in a nonstereotypical way, using nonabrasive and mainstream methods (e.g. ‘I’m involved in organizing social events at clubs and lounges to raise money for women’s rights organizations’). A separate group of participants rated the extent to which the ‘typical’ or ‘atypical’ feminist represented a ‘typical’ feminist on a scale anchored at 1 (not at all represen- tative of a ‘typical’ feminist) and 7 (very representative of a ‘typical’ feminist). The ‘typical’ feminist (M = 5.40, SD = 1.69) seemed more ‘typical’ of feminists than did the ‘atypical’ feminist (M = 4.38, SD = 1.48), t(39) = 2.84, p = .007, r = .41.

Next, participants rated the student on the 33 feminist stereotype traits identified in Pilot Study A (e.g. militant, eccentric and pleasant [reverse-scored]; α = .93) along 7-point scales anchored at 1 (not at all characteristic of the student)

and 7 (very characteristic of the student).2 Participants then rated seven items concerning their interest in affiliating with the student (e.g. ‘I can see myself being friends with this student’; adapted from Montoya & Horton, 2004; α = .89) along 7-point scales anchored at 1 (STRONGLY DISAGREE) and 7 (STRONGLY DISAGREE). Last, participants rated the extent to which they thought the student would view them as immoral (‘If he/she knew me, the student I read about would think that I am …’ [reverse-scored]; modified from Minson & Monin, 2012) and the extent to the student was similar to them (‘The student I read about is…’). These ratings were made along 7-point scales anchored at 1 (not at all moral/not at all similar to me) and 7 (very moral/very similar to me).

Analytic Strategy

We argue that individuals avoid affiliating with and adopting the behaviours advocated by ‘typical’ activists relative to ‘atypical’ activists and undefined targets because they are particularly likely to associate negative activist stereotypes with only those targets that are perceived to be ‘typical’ members of the group defined by these stereotypes (i.e. ‘typical’ activists). Although perceivers may attribute some of these stereotypical traits to the ‘atypical’ activists and undefined targets as well (because these traits are not necessarily unique to activists), we predicted that they would not do so to an extent that would elicit the same negative reactions as those elicited by the ‘typical’ activists. Given this specific hypothesis regarding individuals’ reactions to different types of activists, we conducted planned contrasts to compare ratings of ‘typical’ activists with those of the ‘atypical’ activists and undefined targets (Rosenthal & Rosnow, 1991). This procedure allowed us to minimize the familywise error rate of our statistical tests by reducing the number of pairwise comparisons conducted. We used this approach as our primary analytic strategy in all studies. Within a given study, degrees of freedom may differ slightly across analyses because of missing data.

Eur. J. Soc. Psychol. 43, 614–626 (2013)

Table 2. Mean ratings of moral reproach and perceived similarity in Study 1

‘Typical’ activist

‘Atypical’ activist

Undefined target

Moral reproach 4.27 (1.45) a 3.30 (1.30) b 2.95 (0.89) b

Perceived similarity 2.45 (1.26) a 3.10 (1.45) a, b 3.45 (1.28) b

Note: Standard deviations are in parentheses. Means within the same row with different lowercase letters are significantly different at p < .05.

618 Nadia Y. Bashir et al.

Results

Stereotypical Traits

An analysis of variance (ANOVA) revealed a significant effect of target on stereotypical traits, F(2, 59) = 34.14, p < .001, r = .46. The critical planned contrast comparing ratings of the ‘typical’ feminist (2) with those of the ‘atypical’ feminist (�1) and undefined target (�1) indicated that participants attributed more stereotypical feminist traits to the ‘typical’ feminist (M = 5.08, SD = 0.73) than to the ‘atypical’ feminist (M = 4.34, SD = 0.45) and undefined target (M = 3.60, SD = 0.50), t(59) = 7.19, p < .001, r = .68. Participants also attributed more stereotypical traits to the ‘atypical’ feminist versus undefined target, t(59) = 4.07, p < .001, r = .47.

Affiliation

An ANOVA revealed a significant target effect on affiliation, F(2, 59) = 4.19, p = .02, r = .35. Participants were less interested in affiliating with the ‘typical’ feminist (M = 3.82, SD= 1.42) than with the ‘atypical’ feminist (M= 4.29, SD=0.86) and undefined target (M=4.79, SD=0.82), t(59)= �2.49, p=.02, r=.31. Ratings of the ‘atypical’ feminist and undefined target did not differ significantly, t(59)= 1.48, p=.14, r= .19.3

Mediation Analysis

We argue that individuals avoid affiliating with ‘typical’ activists because they associate them with negative stereotypes rather than because they expect to be reproached by ‘typical’ activists or view themselves as dissimilar to ‘typical’ activists (see Table 2 for descriptive statistics for moral reproach and perceived similarity). We therefore tested a parallel multiple mediator model with 5000 bootstrapped resamples (Preacher & Hayes, 2008). Given our focus on comparing ‘typical’ activists with ‘atypical’ activists and undefined targets, we converted our target variable to a dichot- omous variable coded (0) ‘atypical’ feminist/undefined target and (1) ‘typical’ feminist. Results revealed that the 95% bias-corrected confidence interval (CI) [�1.81, �0.65] of the indirect effect of target on affiliation through stereotypical traits did not contain zero. The indirect effects through moral reproach (95% CI [�0.23, .23]) and perceived similarity (95% CI [�0.39, 0.00]), in comparison, were nonsignificant. Thus, stereotypical target perceptions mediated the effect of target on affiliation, whereas anticipated moral reproach and perceived similarity did not.4

Discussion

Participants expressed less desire to affiliate with the ‘typical’ feminist relative to the ‘atypical’ feminist and undefined target because they associated the ‘typical’ feminist with more

3We provide additional information regarding effect sizes for comparisons between individual conditions in a meta-analysis following Study 5. 4Participants in all studies rated the ‘typical’ activist as more militant, ps < .001, more eccentric, ps < .07, and less personable, ps < .03, than the other targets. Not surprisingly, therefore, we found the same pattern of medi- ation results when we reran these analyses and replaced the overall stereotypes index with an index of only the militant, eccentric or personable traits.

negative stereotypical traits. Furthermore, participants’ stereo- typical perceptions of the ‘typical’ feminist reduced their will- ingness to affiliate with this target over and above the influence of moral reproach and perceived similarity. Thus, the present findings provide evidence that stereotypes of activists have important implications for individuals’ reactions to activists.

STUDY 2

Study 1 shows that individuals avoid affiliating with activists because they associate them with negative stereotypes. This may have crucial implications for the extent to which activists have opportunities to transmit pro-change goals and values to others (Cohen, 2003; Walton, Cohen, Cwir, & Spencer, 2012). In Study 2, we directly examined the impact of negative activist stereotypes on how receptive individuals are to activists’ initiatives to promote social change.

Method

Participants

Participants were 90 undergraduate students who received course credit. One participant who provided the same response to all behavioural intentions items and whose score on this measure was more than three standard deviations below the mean was excluded. Accordingly, 46 male and 43 female participants (Mage = 18.63 years, SD = 1.98) were included in analyses.

Procedure

Participants were randomly assigned to read a bogus profile of a journalist who represented a ‘typical’ feminist, an ‘atypical’ feminist or an undefined target. These profiles were very similar to those used in Study 1 but described the target as a journalist rather than as a student to provide a rationale for subsequently presenting participants with an article written by the target. After reading the profile, participants read an article about the need for individuals to actively support women’s rights concerns:

women still face enormous obstacles: They are undervalued in the workplace, objectified in the media, and physically assaulted at alarmingly high rates. Furthermore, women living in less tolerant regions of the world still require substantial international help to achieve even very basic legal rights and to ensure their physical safety.

Eur. J. Soc. Psychol. 43, 614–626 (2013)

5To provide evidence that individuals’ stereotypes of activists influence their support for social change over and above the influence of their beliefs about social issues, we verified that stereotypical traits significantly mediated the effect of target on behavioural intentions in Study 2 (95% CI [�0.54, �0.06]), even when controlling for participants’ beliefs that gender inequality is still a serious problem.

The ironic impact of activists 619

All participants read the same article, but they were told that it had been written by the journalist whom they had just read about. Next, participants completed the stereotypical traits measure used in Study 1 (α = .85). They then rated 22 items concerning their intentions to perform pro-gender equality behaviours (‘I plan to get involved in pro-women’s rights initiatives at my school or in my community’; adapted from Bashir, Lockwood, Dolderman, Sarkissian, & Quick, 2011; α = .96) along 7-point scales anchored at 1 (strongly disagree) and 7 (strongly agree). Last, to verify that par- ticipants did in fact view the ‘typical’ feminist to be more ‘typical’ of feminists than the other targets, participants rated the extent to which the author of the article fit common stereotypes of feminists along a 7-point scale anchored at 1 (does not fit stereotypes of feminists) and 7 (fits stereotypes of feminists).

Results

Typicality Manipulation Check

An ANOVA on the single-item typicality rating revealed a significant effect of target, F(2, 86) = 8.66, p < .001, r = .41. Par- ticipants’ typicality ratings were higher for the ‘typical’ feminist (M = 6.03, SD = 0.84) than for the ‘atypical’ feminist (M = 5.54, SD = 1.14), t(86)= 1.74, p = .09, r = .18, and the undefined target (M = 4.87, SD = 1.28), t(86) = 4.15, p < .001, r = .41.

Stereotypical Traits

An ANOVA revealed a significant target effect on stereo- typical traits, F(2, 86) = 11.84, p < .001, r = .46. Participants attributed more stereotypical feminist traits to the ‘typical’ feminist (M = 4.90, SD = 0.45) than to the ‘atypical’ feminist (M = 4.57, SD = 0.58) and the undefined target (M = 4.28, SD = 0.46), t(86) = 4.27, p < .001, r = .42. Participants also attributed more feminist traits to the ‘atypical’ feminist versus undefined target, t(86) = 2.22, p = .03, r = .23.

Behavioural Intentions

An ANOVA revealed a marginal overall effect of target on behavioural intentions, F(2, 86) = 2.35, p = .10, r = .23. The critical planned contrast comparing the ‘typical’ feminist with the ‘atypical’ feminist and undefined target, however, was significant: Participants were less motivated to adopt pro- equality behaviours when these behaviours were advocated by the ‘typical’ feminist (M = 4.14, SD = 1.34) rather than by the ‘atypical’ feminist (M = 4.71, SD = 0.73) and undefined target (M = 4.55, SD = 0.91), t(86) = �2.10, p = .04, r = .22. Participants did not differ in their behavioural intentions when these behaviours were advocated by the ‘atypical’ feminist versus the undefined target, t(86) = .60, p = .55, r = .06.

Mediation Analysis

Using the same procedure as in Study 1, we found that stereotypical traits mediated the effect of target on behavioural intentions (95% bias-corrected CI [�0.59, �0.09]). Thus, participants were less likely to adopt pro-equality behaviours

when they were advocated by the ‘typical’ feminist because they attributed more negative stereotypical traits to this target.5

Discussion

This study revealed that participants were less motivated to adopt pro-equality behaviours when these behaviours were ostensibly advocated by the ‘typical’ feminist rather than when these same behaviours were advocated by the ‘atypical’ feminist or the undefined target, because participants were more likely to associate the ‘typical’ feminist with negative stereotypical traits. Thus, despite aggressively promoting social change, ‘typical’ activists may, ironically, undermine individuals’ motivation to engage in pro-change behaviours.

STUDY 3

Although Studies 1 and 2 provide converging evidence that negative stereotypes of activists undermine the pro-change influence of activists, it may be that the effects observed in these studies are specific to feminists and do not generalize to other activists. The pilot studies indicate, however, that individuals ascribe similarly negative traits to environmentalists, viewing members of both groups as militant/aggressive and eccentric/ unconventional. This suggests that negative stereotypes may indeed characterize activist groups in general and limit their success in promoting social change. In Study 3, we obtained additional support for this argument by replicating the effects observed in Studies 1 and 2 with environmentalist targets.

We also used this study as an opportunity to examine the relationships between stereotypes, affiliation and pro-change behavioural intentions within the same study. We have argued that individuals are reluctant to affiliate with activists because they associate them with negative stereotypes. This reluctance to affil- iate may in turn contribute to individuals’ rejection of the activists’ message. Indeed, because individuals who seem undesirable as interaction or relationship partners are rejected by others (Hebl, Williams, Sundermann, Kell, & Davies, 2012), individuals may wish to avoid engaging in behaviours that would characterize them in a similarly negative light. Thus, if individuals associate activists with negative stereotypes and therefore view them as peo- ple with whom it would be undesirable to affiliate, they may be less motivated to engage in the pro-change behaviours that are characteristic of activists. We examined this directly in Study 3.

Method

Participants

Participants were 140 Americans (Mage =34.10years, SD=12.23) recruited online via Amazon’s Mechanical Turk who received $0.75. There were 55 male and 74 female participants. Eleven participants did not identify their biological sex.

Eur. J. Soc. Psychol. 43, 614–626 (2013)

620 Nadia Y. Bashir et al.

Procedure

Similar to Study 2, participants in Study 3 were randomly assigned to read a bogus profile of a journalist who represented a ‘typical’ environmentalist (‘I hold rallies outside chemical research labs…to protest the production of harmful chemical substances’), an ‘atypical’ environmentalist (‘I’m involved in organizing social events at restaurants and lounges to raise money for grassroots-level environmental organizations’) or an undefined target (‘When my weekends aren’t packed with work and volunteer activities, I usually spend the day watching TV or hanging out at coffee shops’). As in Study 1, a separate group of participants rated the typicality of the targets described in the target profiles. Participants viewed the targets described in the ‘typical’ environmentalist profile (M = 6.15, SD= 0.93) to be more ‘typical’ of environmentalists than the targets described in the ‘atypical’ environmentalist (M = 5.12, SD = 1.57) and undefined target (M = 2.82, SD = 1.34) profiles, ts > 3.24, ps < .004, rs > .49.6

Next, participants read a fictitious pro-environmental article about climate change and the need for individuals to adopt sustainable lifestyles (e.g. ‘Scientific evidence suggests that within 10 years, hurricanes will become more frequent and intense than they are now, and as a result, they’ll devastate many regions around the world’). As in Study 2, all participants read the same article but were told that it had been written by the journalist whom they had read about. Participants then completed a 16-item version of the stereotypical traits measure used in Studies 1 and 2 (α =.90), which was created on the basis of the environmentalist traits identified in Pilot Study B.7 They also completed the affiliation measure used in Studies 1 and 2 (α = .97) and a 29-item measure assessing their pro-environmental behavioural intentions (e.g. ‘I plan to always recycle whatever materials I can’; α = .97). Participants rated their intentions along 7-point scales anchored at 1 (strongly disagree) and 7 (strongly agree).

Results

Stereotypical Traits

An ANOVA revealed a significant target effect on stereotypical traits, F(2, 137)=70.31, p < .001, r= .71. Participants attributed more stereotypical environmentalist traits to the ‘typical’ environ- mentalist (M =5.00, SD=0.96) than to the ‘atypical’ environmen- talist (M =3.73 SD= 0.64) and undefined target (M =3.18, SD= 0.54), t(137)=11.58, p< .001, r= .70. They also attributed more stereotypical traits to the ‘atypical’ environmentalist versus the undefined target, t(137)= 3.62, p< .001, r=.30.

6Participants may have reacted more favourably to the ‘atypical’ activists because they did not truly perceive these targets as environmentalists or feminists or because they perceived these targets as promoting subtle rather than considerable social change. A separate group of participants (N = 71) indicated, however, that both the ‘typical’ and ‘atypical’ target profiles in Studies 2–4 represented feminists or environmentalists (all ratings greater than 6 on a scale anchored at 1 [not at all a feminist/environmentalist] and 7 [very much a feminist/environmentalist]). Participants also rated the degree to which the targets advocated the need for substantial societal change. Ratings of the ‘typical’ and ‘atypical’ targets did not differ significantly, ps > .29. 7Study 3 stereotype traits items: radical, aggressive, forceful, annoying, cru- sader, activist, judgmental, eccentric, tofu-eating, scruffy, fun, pleasant, cool, friendly, approachable and social (the last six traits were reverse-scored).

Affiliation

An ANOVA revealed a significant effect of target on affiliation, F(2, 135)=34.38, p< .001, r=.58. As in Study 1, participants were less interested in affiliating with the ‘typical’ environmentalist (M=2.44, SD=1.62) than with the ‘atypical’ environmentalist (M = 4.53 SD = 1.31) and undefined target (M = 4.66, SD = 1.37), t(135) = �8.28, p < .001, r = .58. Their ratings of the ‘atypical’ environmentalist versus the undefined target did notdiffersignificantly, t(135) = .34,p = .73,r = .06.

Behavioural Intentions

An ANOVA revealed a marginal overall target effect on behavioural intentions, F(2, 130) = 2.73, p = .07, r = .20. Similar to the results obtained in Study 2, the critical planned contrast comparing the ‘typical’ environmentalist with the ‘atypical’ environmentalist and undefined target was significant: Participants were less motivated to adopt pro-environmental behaviours when these behaviours were advocated by the ‘typical’ environmentalist (M = 4.07, SD = 1.30) rather than by the ‘atypical’ environmentalist (M = 4.64, SD = 1.08) or undefined target (M = 4.59, SD = 1.39), t(130) = �2.30, p = .02, r = .20. Participants did not differ in their pro-environmental behavioural intentions when these behaviours were advocated by the ‘atypical’ environmentalist versus the undefined target, t(130) = .19, p = .85, r = .02.

Mediation Analysis

Using a serial multiple mediator model based on 5000 bootstrapped resamples (Hayes, 2013), we examined the extent to which stereotypical perceptions of the ‘typical’ environmentalist reduced interest in affiliating with this environmentalist, which ultimately reduced intentions to adopt the behaviours advocated by this target. The indirect effect of target on behavioural intentions through the two serial mediators, stereotypical traits and affiliation, was significant (95% bias-corrected CI [�0.63, �0.08]). Thus, because individuals associate ‘typical’ environmentalists with negative stereotypes, they are less interested in affili- ating with them and consequently less inclined to adopt the behaviours that these environmentalists advocate.

Discussion

This study provides converging evidence that individuals avoid adopting pro-change behaviours when these behav- iours are advocated by ‘typical’ activists rather than by ‘atypical’ activists or individuals who are not described as activists. Furthermore, the present findings demonstrate that individuals avoid adopting pro-change behaviours advo- cated by ‘typical’ activists because they associate these activists with negative stereotypes and therefore view them as people with whom it would be undesirable to affiliate.

Eur. J. Soc. Psychol. 43, 614–626 (2013)

The ironic impact of activists 621

STUDY 4

Studies 1–3 provide converging evidence that individuals avoid supporting social change in part because they associate activists with unfavourable stereotypes and therefore resist the efforts of these activists to promote change. In Study 4, we obtained additional evidence for this argument by replicating Study 3 using the more organically derived environmentalist stereotype traits identified in Pilot Study C.

Method

Participants

Participants were 238 Americans (Mage = 35.53 years, SD=12.84) recruited online via Amazon’s Mechanical Turk who received $0.65. There were 87 male and 145 female participants. Six participants did not identify their sex.

Procedure

Participants in Study 4 completed the same procedure used in Study 3 with the following exceptions. First, the stereotype measure was composed of the 30 most frequently listed environmentalist traits obtained in Pilot Study C (Table 1). To maintain consistency across studies, we also included the six personable traits used in Studies 1–3 in this measure (α = .90). Second, to provide evidence that participants did in fact view the ‘typical’ environmentalist to be more ‘typical’ of environmentalists than the other targets, at the end of the study, participants completed the same typicality manipula- tion check rating used in Study 2. Last, to verify that the iden- tity of the target had an impact on participants’ pro-change behavioural intentions over and above any effects that could be attributed solely to the article, participants who were randomly assigned to a second control group read the article and completed all measures without first reading a target profile (unidentified journalist condition).

Results and Discussion

Typicality Manipulation Check

An ANOVA on the single-item typicality rating revealed a significant target effect, F(3, 231) = 17.63, p < .001, r = .43. Participants’ ratings were significantly higher for the ‘typical’ environmentalist (M = 6.24, SD = 0.99) than for the ‘atypical’ environmentalist (M = 5.34, SD = 1.54), undefined target (M = 4.59, SD = 1.48) and unidentified journalist (M = 4.89, SD = 1.25), ts > 3.69, ps < .001, rs > .24.

Stereotypical Traits

An ANOVA revealed a significant target effect on stereotyp- ical traits, F(3, 232) = 38.40, p < .001, r = .58. Participants attributed more stereotypical environmentalist traits to the ‘typical’ environmentalist (M = 4.91, SD = 0.63) than to the ‘atypical’ environmentalist (M = 4.16, SD = 0.62), undefined target (M = 3.81, SD = 0.54) and unidentified journalist

(M = 4.10, SD = 0.56), t(232) = 10.16, p < .001, r = .55. They also attributed more stereotypical traits to the ‘atypical’ environmentalist versus the undefined target, t(232) = 3.23, p = .001, r = .21, and the unidentified target versus the undefined target, t(232) = 2.63, p = .009, r = .17. Ratings of the ‘atypical’ environmentalist and unidentified target did not differ significantly, t(232) = .58, p = .56, r = .04.

Affiliation

An ANOVA revealed a significant target effect on affiliation (α = .97), F(3, 234) = 20.44, p < .001, r = .46. Similar to Studies 1–3, participants were less interested in affiliating with the ‘typical’ environmentalist (M = 2.83, SD = 1.65) than with the ‘atypical’ environmentalist (M = 4.53, SD = 1.61), undefined target (M = 4.79, SD = 1.51) and unidentified jour- nalist (M = 4.35, SD = 1.24), t(234) = �7.66, p < .001, r = .45. Their ratings of the ‘atypical’ environmentalist, the undefined target and the unidentified journalist did not differ, ts < 1.57, ps > .11, rs < .10.

Behavioural Intentions

An ANOVA revealed a marginal overall effect of target on behav- ioural intentions (α =.97), F(3, 234)=2.29, p= .08, r= .17. As in Studies 2 and 3, the critical planned contrast was significant: Participants were less motivated to adopt pro-environmental behaviours when these behaviours were advocated by the ‘typi- cal’ environmentalist (M =4.41, SD= 1.13) rather than by the ‘atypical’ environmentalist (M =4.62, SD=1.26), undefined tar- get (M =4.75, SD=1.31) or unidentified journalist (M= 4.97, SD= 0.98), t(234) = �2.06, p = .04, r = .13. Participants’ behav- ioural intentions did not differ when these behaviours were advocated by the ‘atypical’ environmentalist, the undefined target or the unidentified journalist, ts < 1.68, ps > .11, rs < .11.

Mediation Analysis

As in Study 3, the indirect effect of target on behavioural intentions through the two serial mediators, stereotypical traits and affiliation, was significant (95% bias-corrected CI [�0.49, �0.23]).

Thus, even when participants rated a ‘typical’ environ- mentalist on the traits that most closely reflect individuals’ stereotypes of environmentalists, their tendency to attribute negative stereotypes to this target reduced their willingness to affiliate with this target and, ultimately, their motivation to adopt pro-change behaviours advocated by this target.

STUDY 5

In Studies 1–4, we manipulated target typicality using profiles that described the targets in a stereotype-consistent or stereotype-inconsistent manner, on the basis of the results of the pilot studies. Because activist stereotypes are largely negative, the profiles describing the ‘typical’ activists contained more negative information than did the profiles describing the other targets. Thus, to verify that participants’ reactions to the targets were influenced by the targets’ typicality rather than by

Eur. J. Soc. Psychol. 43, 614–626 (2013)

622 Nadia Y. Bashir et al.

only the valence of the profiles, we conducted an additional study in which participants simply read that the target was a ‘typical’ or ‘atypical’ environmentalist or feminist and did not receive any valenced target information.

Method

Participants

We recruited 194 Americans (Mage = 34.68 years, SD = 13.75) online via Amazon’s Mechanical Turk. They received $0.45. There were 73 male and 114 female participants. Seven participants did not identify their sex.

Procedure

Participants were randomly assigned to read about a ‘typical’ or ‘atypical’ environmentalist or feminist. Specifically, partic- ipants read that

A. M. Johnson is very committed to promoting [environ- mentalism/women’s rights], [and is a typical/but is an atypical] [environmentalist/feminist]. That is, Johnson is [very different from] what members of society expect [an environmentalist/a feminist] to be like.

Participants did not receive any specific or valenced target information. Participants in the environmentalist conditions then completed the stereotype trait measure used in Study 4 (α =.89). Participants in the feminist conditions completed the same measure except that the environmentalist traits were replaced with the top 30 feminist traits identified in Pilot Study C (Table 1; α =.94). Last, participants completed the affiliation measure (α =.94) and single- item typicality manipulation check used in the previous studies.

Results and Discussion

Typicality Manipulation Check

A 2 (manipulated typicality: ‘typical’, ‘atypical’) × 2 (activist domain: environmentalist, feminist) ANOVA on the single-item typicality rating revealed a main effect of manipulated typicality. Participants rated the ‘typical’ (M = 5.17, SD = 1.51) versus ‘atypical’ (M = 3.16, SD= 1.58) activists as more ‘typical’, F(1, 189) = 79.99, p < .001, r = .55. The main effect of domain and interaction were nonsignificant, Fs < 0.20, ps > .65, rs < .03.

Stereotypical Traits

A 2× 2 ANOVA revealed a main effect of manipulated typicality on stereotypical traits. Participants attributed more stereotypical feminist or environmentalist traits to the ‘typical’ (M = 4.51, SD = 0.67) versus ‘atypical’ (M = 3.85, SD = 0.75) activists, F(1, 190) = 41.47, p < .001, r = .42. The main effect of domain and interaction were nonsignificant, Fs< 0.15, ps> .70, rs< .03.

Affiliation

A 2 × 2 ANOVA also revealed a main effect of manipulated typicality on affiliation. Participants were less interested in

affiliating with the ‘typical’ (M = 3.91, SD = 1.51) versus ‘atypical’ (M = 4.32, SD = 1.36) activists, F(1, 189) = 4.28, p = .04, r = .15 Although participants were more willing to affiliate with environmentalists (M = 4.44, SD = 1.11) than feminists (M = 3.81, SD = 1.64), F(1, 189) = 10.19, p = .002, r = .23, activist domain did not moderate the manipulated typicality effect, F(1, 189) = 1.59, p = .21, r = .09.

Mediation Analysis

Consistent with Studies 1–4, participants were less willing to affiliate with the ‘typical’ versus ‘atypical’ targets because they attributed more negative stereotypical traits to these tar- gets (95% bias-corrected CI [�0.49, �0.22]).

Thus, even when perceivers do not receive any specific or valenced target information, they attribute more negative stereotypical traits to ‘typical’ versus ‘atypical’ activists and therefore react more negatively to them.

Meta-analysis of Effects

We have argued that individuals are less willing to affiliate with and adopt behaviours advocated by ‘typical’ activists relative to ‘atypical’ activists and undefined targets. To assess the overall evidence for this argument, we conducted a meta-analysis for the key comparisons involving these variables across all studies (Table 3). An unweighted means approach revealed an overall ‘typical’ versus ‘atypical’ activist effect on affiliation (r = .32, 95% CI [0.12, 0.52]) and behav- ioural intentions (r = .16, 95% CI [0.06, 0.26]). Similarly, there was an overall ‘typical’ activist versus undefined target effect on affiliation (r = .43, 95% CI [0.31, 0.55]) and behav- ioural intentions (r = .14, 95% CI [0.10, 0.18]). The overall ‘atypical’ activist versus undefined target effect, in contrast, was small for affiliation (r = .10, 95% CI [0.02, 0.19]) and did not differ significantly from zero on behavioural intentions (r = �.01, 95% CI [�0.07, 0.04]). Furthermore, when using a weighted means approach, the overall ‘atypical’ activist versus undefined target effect did not differ significantly from zero on affiliation (r = .08, 95% CI [�0.02, 0.17]) or behavioural intentions (r = .004, 95% CI [�0.09, 0.10]). For compari- sons of ‘typical’ activists with ‘atypical’ activists and ‘typical’ activists with undefined targets, the unweighted and weighted means approaches produced similar patterns of results. Thus, across studies, we found a consistent effect for the difference in responses to ‘typical’ and ‘atypical’ activists and also for the difference in responses to ‘typical’ activists and undefined targets. In contrast, individuals’ responses to ‘atypical’ activists and undefined targets did not differ consistently. The results of our meta-analysis thus provide additional evidence that individ- uals avoid affiliating with and adopting the behaviours advocated by ‘typical’ activists.

Testing Alternative Models

We argue that individuals avoid affiliating with ‘typical’ activists and adopting the pro-change behaviours that ‘typical’ activists advocate because they associate these activists with negative stereotypes. This possibility is consistent with

Eur. J. Soc. Psychol. 43, 614–626 (2013)

T ab le

3 .

S ig n ifi ca n ce

v al u es

an d ef fe ct

si ze s fo r p ai rw

is e co m p ar is o n s in

S tu d ie s 1 – 5

‘T y p ic al ’ v s. ‘a ty p ic al ’

ac ti v is t

‘T y p ic al ’ v s. u n d efi n ed

ta rg et

‘A ty p ic al ’ ac ti v is t

v s. u n d efi n ed

ta rg et

‘T y p ic al ’ ac ti v is t v s.

u n id en ti fi ed

ta rg et

‘A ty p ic al ’ ac ti v is t v s.

u n id en ti fi ed

ta rg et

U n d efi n ed

ta rg et

v s.

u n id en ti fi ed

ta rg et

p E ff ec t si ze

(r )

p E ff ec t si ze

(r )

p E ff ec t si ze

(r )

p E ff ec t si ze

(r )

p E ff ec t si ze

(r )

p E ff ec t si ze

(r )

S tu d y 1

T ra it s

< .0 0 1

.4 7

< .0 0 1

.7 3

< .0 0 1

.4 7

A ffi li at io n

.1 7

.1 8

.0 0 5

.3 5

.1 4

.1 9

S tu d y 2

T ra it s

.0 1

.2 6

< .0 0 1

.4 6

.0 3

.2 3

In te n ti o n s

.0 4

.2 2

.1 4

.1 6

.5 5

�. 0 6

S tu d y 3

T ra it s

< .0 0 1

.5 9

< .0 0 1

.7 0

< .0 0 1

.3 0

A ffi li at io n

< .0 0 1

.5 3

< .0 0 1

.5 2

.7 3

.0 6

In te n ti o n s

.0 3

.1 9

.0 7

.1 6

.8 5

�. 0 2

S tu d y 4

T ra it s

< .0 0 1

.4 2

< .0 0 1

.5 6

.0 0 1

.2 1

< .0 0 1

.4 4

.5 6

.0 4

.0 0 9

.1 7

A ffi li at io n

< .0 0 1

.3 7

< .0 0 1

.4 2

.3 7

.0 6

< .0 0 1

.3 4

.5 1

.0 4

.1 2

.1 0

In te n ti o n s

.3 5

.0 6

.1 3

.1 0

.5 5

.0 4

.0 1

.1 6

.1 2

.1 0

.3 1

�. 0 7

S tu d y 5

T ra it s

< .0 0 1

.4 2

A ffi li at io n

.0 4

.1 5

The ironic impact of activists 623

624 Nadia Y. Bashir et al.

research indicating that individuals draw on their impressions of targets (e.g. perceptions of targets’ traits) when forming evalua- tions of the targets and determining how they should react to the targets (Fiske & Neuberg, 1990; Macrae & Bodenhausen, 2000). It is also possible, however, that individuals form nega- tive perceptions of activists after they reject an opportunity to affiliate with ‘typical’ activists or adopt the behaviours that ‘typ- ical’ activists advocate, because they need to justify their initial negative reactions. This second explanation is not necessarily inconsistent with our argument. Indeed, it may be that individ- uals’ stereotypical perceptions of activists reduce their willingness to affiliate with ‘typical’ activists and adopt the pro-change behaviours that ‘typical’ activists advocate, which may ultimately reinforce individuals’ initial stereotypical perceptions. Nonetheless, we examined alternative mediation models by assessing models corresponding to all possible serial permutations of the mediators and dependent variable in Studies 1–5. This produced one alternative model in Studies 1, 2 and 5 and five alternative models in Studies 3 and 4. The indirect effects for the alternative model in Study 2, four of the alternative models in Study 3 and two of the alternative models in Study 4 were nonsignificant. Furthermore, although the remaining alterna- tive models did produce significant indirect effects, the size of the indirect effect in all alternative models was smaller than that of the original model in Study 1 (original model: r =.51; alternative model: r =.29), Study 2 (original model: r =.25; alternative model: r=.16), Study 3 (original model: r=.19; all alternative models: rs = .004–.12), Study 4 (original model: r= .33; all alternative models: rs =.05–.25) and Study 5 (original model: r=.34; alternative model: r= .14). Thus, although there is some evidence to suggest that individuals’ willingness to affiliate with activists and adopt the behaviours that they advocate may subse- quently influence their stereotypical perceptions of these activists, the evidence is more consistent with the possibility that individuals’ stereotypes of activists influence their desire to affiliate with and emulate activists. Indeed, the only consistent pattern of mediation across all studies was that in which stereo- typical perceptions served as a mediating variable rather than the dependent variable.

GENERAL DISCUSSION

The present research shows that individuals’ stereotypes of activists influence their support for social change. Previously, researchers have attempted to understand resistance to social change by examining individuals’ perceptions of social issues, attitudes towards social change and personality traits (e.g. Feygina et al., 2010; Hodson & Esses, 2002; van Zomeren et al., 2008). The present research, in comparison, assesses the extent to which individuals’ stereotypes of activists, the agents of social change, increase resistance to social change. Individ- uals believe that it is important and socially desirable to support social change (Beattie, 2010; Nelson et al., 2008), which suggests that they should view activists favourably and be receptive to their efforts. Unfortunately, however, the very nature of activism leads to negative stereotyping: By aggres- sively promoting change and advocating unconventional practices, activists become associated with hostile militancy

and unconventionality or eccentricity. Indeed, we show that the tendency to associate ‘typical’ activists with these nega- tive stereotypes mediates individuals’ unfavourable reactions to the activists. Specifically, individuals avoid affiliating with ‘typical’ activists (Studies 1 and 3–5) because they view them as militant/aggressive and eccentric/unconventional. Furthermore, this tendency to associate activists with negative stereotypes and perceive them as people with whom it would be unpleasant to affiliate reduces individuals’ motivation to adopt the pro-change behaviours that activists advocate (Studies 2–4). This research indicates, therefore, that stereotypes and person perception processes more generally influence individuals’ receptiveness to activists and their pro-change initiatives.

Our studies also reveal the nuances of these effects by showing that individuals do not avoid affiliating with and adopting the behaviours advocated by all activists. Indeed, although participants reacted negatively to the ‘typical’ activists, their willingness to affiliate with and adopt the behaviours advocated by ‘atypical’ activists and undefined targets did not differ. This suggests that it is not mere membership in an activist group but rather the degree to which an activist conforms to group stereotypes that influences perceivers’ reactions. Whereas individuals may avoid affiliating with and emulating activists who seem to conform to activist stereotypes, they may be more receptive to activists who defy these stereotypes by coming across as pleasant and approachable.

The present findings also contribute theoretically by illustrating effects that differ from those observed in research on message source typicality. On the basis of source typicality research (Ziegler & Diehl, 2011; Ziegler, Diehl, & Ruther, 2002), individuals should respond more favourably to messages containing strong arguments and less favourably to messages containing weak arguments when they are delivered by an ‘atypical’ activist rather than a ‘typical’ activist or undefined target. Indeed, ‘atypical’ activists possess ‘atypical’ combinations of personality traits (i.e. personable and environ- mentalist), whereas ‘typical’ activists and undefined targets do not. According to source typicality research, therefore, ‘atypical’ activists should disconfirm perceivers’ expectations and elicit greater information processing. In our studies, however, participants responded similarly to the message delivered by the ‘atypical’ activists and undefined targets.

Researchers have also examined message source typicality in terms of the fit between the position advocated in a message (e.g. pro-environmental stance) and the message source (e.g. director of an environmental group versus CEO of an oil com- pany; Eagly, Wood, & Chaiken, 1978). In our studies, both the ‘typical’ and ‘atypical’ activists were portrayed as individuals who advocate social change whereas the undefined targets were not. Indeed, participants in Study 3 viewed the ‘typical’ and ‘atypical’ environmentalists to be similarly environmen- tally friendly, t(96) = .06, p = .96, and significantly more envi- ronmentally friendly than the undefined target, ts > 5.51, ps < .001. Source typicality research would suggest, therefore, that individuals should respond similarly to messages delivered by ‘typical’ and ‘atypical’ activists and differently to messages delivered by these targets versus undefined targets. We show, however, that individuals differed in their pro-change behavioural intentions when these behaviours were advocated by ‘typical’ versus ‘atypical’ activists but not when these

Eur. J. Soc. Psychol. 43, 614–626 (2013)

The ironic impact of activists 625

behaviours were advocated by ‘atypical’ activists versus undefined targets. Thus, whereas our data are consistent with the argument that stereotypes influence individuals’ reactions to activists, they are inconsistent with alternative explanations based on source typicality research.

Although we examined the implications of activist stereotypes for social change by focusing on two key activist groups, we argue that the militant and eccentric traits that characterize these groups also characterize a variety of activist groups (e.g. gay rights, political democracy and Occupy Wall Street activists). Some activist groups (e.g. gun control advocates) maybe less likely to engage in overtly aggressive advocacy and may, therefore, be less likely to be perceived as militant. We note, however, that many of the stereotypical traits generated by partic- ipants in Pilot Study C describe activists as argumentative and confrontational but not necessarily violent (e.g. ‘forceful’, ‘self- righteous’, ‘assertive’ and ‘overreactive’). Because it is possible for individuals to be argumentative and confrontational even when advocating causes that explicitly condemn violence, individuals may associate militant and eccentric traits even with those activist groups that exemplify these traits less overtly than do feminists and environmentalists. In future research, it will be important to assess these possibilities directly.

We examined individuals’ reactions to activists without considering perceivers’ own identity as activists. Activist and nonactivist perceivers may, however, respond differently to activist targets. Indeed, because individuals generally view ingroup members positively (Sherif, Harvey, White, Hood, & Sherif 1961; Tajfel & Turner, 1979), activist perceivers may respond relatively favourably to ‘typical’ activists. On the other hand, because individuals have especially unfavourable impres- sions of group members who perform undesirable behaviours (Marques, Yzerbyt, & Leyens, 1988), activist perceivers who condemn the use of militant methods to promote social change (e.g. ‘atypical’ activists) may react particularly negatively to ‘typical’ activists. In future work, researchers may wish to examine whether perceiver identity moderates reactions to activists.

For many activists, the willingness to take a radical stand with- out regard for mainstream sensibilities is a point of pride. Indeed, environmental activist and author of ‘Tree Spiker’ Mike Roselle (as cited in Olafsson, 2009) defends his militant efforts to protect the environment, noting, ‘I don’t think there’s anything extreme about saying we have to stop pumping carbon into the air. If we’re extremists, so be it. The stakes are too high’ (para. 6). The present research suggests, however, that such seemingly zealous dedica- tion to a social cause may backfire and elicit unfavourable reac- tions from others. Indeed, individuals avoid affiliating with ‘typical’ activists and adopting the pro-change behaviours that these activists advocate because individuals associate them with negative stereotypes. Ironically and despite good intentions, there- fore, the very individuals who are most actively engaged in pro- moting social change may inadvertently alienate members of the public and reduce pro-change motivation.

ACKNOWLEDGEMENTS

This research was supported by a Social Sciences and Humanities Research Council (SSHRC) Canada Graduate Scholarship to the first author and a SSHRC grant to the second author.

REFERENCES

Agronick, G. S., & Duncan, L. E. (1998). Personality and social change: Individual differences, life path, and importance attributed to the women’s movement. Journal of Personality and Social Psychology, 74, 1545–1555. DOI: 10.1037/0022-3514.74.6.1545

Aronson, P. (2003). Feminists or “postfeminists?” Young women’s attitudes toward feminism and gender relations. Gender & Society, 17, 903–922. DOI: 10.1177/0891243203257145

Bashir, N. Y., Lockwood, P., Dolderman, D., Sarkissian, T., & Quick, L. K. (2011). Emphasizing jobs and trees: Increasing the impact of proenvironmental messages on migrants. Basic and Applied Social Psychology, 33, 255–265. DOI: 10.1080/01973533.2011.589319

Baumeister, R. F., & Leary, M. R. (1995). The need to belong: Desire for interpersonal attachments as a fundamental human motivation. Journal of Personality and Social Psychology, 117, 497–529. DOI: 10.1037/ 0033-2909.117.3.497

Beattie, G. (2010). Why aren’t we saving the planet? A psychologist’s perspective. London, UK: Routledge.

Berryman-Fink, C., & Verderber, K. S. (1985). Attributions of the term femi- nist: A factor analytic development of a measuring instrument. Psychology of Women Quarterly, 9, 51–64. DOI: 10.1111/j.1471-6402.1985.tb00860.x

Brown, G. (2007). Mutinous eruptions: Autonomous spaces of radical queer activism. Environment and Planning A, 39, 2685–2698. DOI: 10.1068/a38385

Byrne, D. (1971). The attraction paradigm. New York, NY: Academic Press. Clark, R. D., III, & Maass, A. (1988). Social categorization in minority

influence: The case of homosexuality. European Journal of Social Psychology, 18, 347–364. DOI: 10.1002/ejsp.2420180405

Cohen, G. L. (2003). Party over policy: The dominating impact of group influ- ence on political beliefs. Journal of Personality and Social Psychology, 85, 808–822. DOI: 10.1037/0022-3514.85.5.808

Czopp, A. M., Monteith, M. J., & Mark, A. Y. (2006). Standing up for a change: Reducing bias through interpersonal confrontation. Journal of Personality and Social Psychology, 90, 784–803. DOI: 10.1037/0022-3514.90.5.784

Eagly, A. H., Wood, W., & Chaiken, S. (1978). Causal inferences about communicators and their effect on opinion change. Journal of Personality and Social Psychology, 36, 424–435. DOI: 10.1037/0022-3514.36.4.424

Federal Bureau of Investigation. (2001). Terrorism 2000/2001. (Report no. 0308). Retrieved from http://www.fbi.gov/stats-services/publications/ terror/terrorism-2000-2001

Feygina, I., Jost, J. T., & Goldsmith, R. E. (2010). System justification, the denial of global warming, and the possibility of “system-sanctioned change”. Personality and Social Psychology Bulletin, 36, 326–338. DOI: 10.1177/0146167209351435

Fiske, S. T., & Neuberg, S. L. (1990). A continuum of impression formation, from category-based to individuating processes: Influences of information and motivation on attention and interpretation. In M. P. Zanna (Ed.), Ad- vances in experimental social psychology (Vol. 23, pp. 1–74). San Diego, CA: Academic Press.

Gifford, R. (2011). The dragons of inaction. Psychological barriers that limit climate change mitigation and adaptation. American Psychologist, 66, 290–302. DOI: 10.1037/a0023566

Hayes, A. F. (2013). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. New York, NY: The Guilford Press.

Hebl, M. R., Williams, M. J., Sundermann, J. M., Kell, H. J., & Davies, P. G. (2012). Selectively friending: Racial stereotypicality and social rejection. Journal of Experimental Social Psychology, 48, 1329–1335. DOI: 10.1016/j.jesp.2012.05.019

Heider, F. (1958). The psychology of interpersonal relations. Hillsdale, NJ: Lawrence Erlbaum Associates.

Hodson, G., & Esses, V. M. (2002). Distancing oneself from negative attributes and the personal/group discrimination discrepancy. Journal of Experimental Social Psychology, 38, 500–507. DOI: 10.1016/S0022-1031(02)00012-4

House Committee on Armed Services. (1993). Policy implications of lifting the ban on homosexuals in the military: Hearings before the Committee on Armed Ser- vices (HJLS.C. No. 103–18). Washington, DC: U.S. Government Printing Office.

Kay, A. C., & Friesen, J. (2011). On social stability and social change: Understanding when system justification does and does not occur. Current Directions in Psychological Science, 20, 360–364. DOI: 10.1177/ 0963721411422059

Macrae, C. N., & Bodenhausen, G. V. (2000). Social cognition: Thinking cat- egorically about others. Annual Review of Psychology, 51, 93–120. DOI: 10.1146/annurev.psych.51.1.93

Marques, J. M., Yzerbyt, V. Y., & Leyens, J.-P. (1988). The ‘black sheep effect’: Extremity of judgments towards ingroup members as a function of group identification. European Journal of Social Psychology, 18, 1–16. DOI: 10.1002/ejsp.2420180102

Eur. J. Soc. Psychol. 43, 614–626 (2013)

http://www.fbi.gov/stats-services/publications/terror/terrorism-2000-2001

626 Nadia Y. Bashir et al.

Minson, J. A., & Monin, B. (2012). Do-gooder derogation: Disparaging mor- ally motivated minorities to defuse anticipated reproach. Social Psychological and Personality Science, 3, 200–207. DOI: 10.1177/1948550611415695

Monin, B., Sawyer, P. J., & Marquez, M. J. (2008). The rejection of moral rebels: Resenting those who do the right thing. Journal of Personality and Social Psychology, 95, 76–93. DOI: 10.1037/0022-3514.95.1.76

Montoya, R. M., & Horton, R. S. (2004). On the importance of cognitive evaluation as a determinant of interpersonal attraction. Journal of Personal- ity and Social Psychology, 86, 696–712. DOI: 10.1037/0022-3514.86.5.696

Morton, T. A., Postmes, T., Haslam, S. A., & Hornsey, M. J. (2009). Theoriz- ing gender in the face of social change: Is there anything essential about es- sentialism? Journal of Personality and Social Psychology, 96, 653–664. DOI: 10.1037/a0012966

National Archives and Records Administration. (2011). Teaching with documents: The Civil Rights Act of 1964 and the Equal Employment Opportunity Commission. Retrieved from http://www.archives.gov/educa- tion/lessons/civil-rights-act/

Nelson, J. A., Liss, M., Erchull, M. J., Hurt, M. M., Ramsey, L. R., Turner, D. L., & Haines, M. E. (2008). Identity in action: Predictors of feminist self- identification and collective action. Sex Roles, 58, 721–728. DOI: 10.1007/s11199-007-9384-0

Olafsson, K. (2009, October 12). Environmental activist Mike Roselle. Time. Retrieved from www.time.com/time/arts/article/0,8599,1929701,00. html#ixzz15eA1Lykz

Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resampling strate- gies for assessing and comparing indirect effects in multiple mediator models. Behaviour Research Methods, 40, 879–891. DOI: 10.3758/ BRM.40.3.879

Pryor, J. B., Reeder, G. D., & Monroe, A. E. (2012). The infection of bad company: Stigma by association. Journal of Personality and Social Psychology, 102, 224–241. DOI: 10.1037/a0026270

Rosenthal, R., & Rosnow, R. L. (1991). Essentials of behavioral research: Methods and data analysis. New York, NY: McGraw-Hill.

Sedikides, C., & Gregg, A. P. (2008). Self-enhancement: Food for thought. Perspectives on Psychological Science, 3, 102–116. DOI: 10.1111/j.1745- 6916.2008.00068.x

Sherif, M., Harvey, O. J., White, B. J., Hood, W. R., & Sherif, C. W. (1961). Intergroup conflict and cooperation: The robbers cave experiment. Norman, OK: University of Oklahoma Book Exchange.

Superson, A. M., & Cudd, A. E. (2002). Theorizing backlash: Philosophical re- flections on the resistance to feminism. Lanham, MD: Rowman & Littlefield.

Tajfel, H., & Turner, J. C. (1979). An integrative theory of intergroup conflict. In W. G. Austin, & S. Worchel (Eds.), The social psychology of intergroup relations (pp. 33–47). Monterey, CA: Brooks/Cole Publishing Company.

The Library of Congress. (2011). Don’t Ask, Don’t Tell Repeal Act of 2010. Bill summary & status: H.R.2965. Retrieved from http://thomas.loc.gov/ cgi-bin/bdquery/z?d111:HR02965:@@@L&summ2=m&

Twenge, J. M., & Zucker, A. N. (1991). What is a feminist? Evaluations and stereotypes in closed- and open-ended responses. Psychology of Women Quarterly, 23, 591–605. DOI: 10.1111/j.1471-6402.1999.tb00383.x

Walton, G. M., Cohen, G. L., Cwir, D., & Spencer, S. J. (2012). Mere belonging: The power of social connections. Journal of Personality and Social Psychology, 102, 513–532. DOI: 10.1037/a0025731

White, P. H., & Harkins, S. G. (1994). Race of source effects in the elaboration likelihood model. Journal of Personality and Social Psychology, 67, 790–807. DOI: 10.1037//0022-3514.67.5.790

Zárate, M. A., Shaw, M., Marquez, J. A., & Biagas, D., Jr. (2012). Cultural inertia: The effects of cultural change on intergroup relations and the self- concept. Journal of Experimental Social Psychology, 48, 634–645. DOI: 10.1016/j.jesp.2011.12.014

Ziegler, R., & Diehl, M. (2011). Mood and multiple source characteristics: Mood congruency of source consensus status and source trustworthiness as determinants of message scrutiny. Personality and Social Psychology Bulletin, 37, 1016–1030. DOI: 10.1177/0146167211410438

Ziegler, R., Diehl, M., & Ruther, A. (2002). Multiple source characteristics and persuasion: Source inconsistency as a determinant of message scrutiny. Personality and Social Psychology Bulletin, 28, 496–508. DOI: 10.1177/ 0146167202287007

van Zomeren, M., Postmes, T., & Spears, R. (2008). Toward an integrative social identity model of collective action: A quantitative research synthesis of three socio-psychological perspectives. Psychological Bulletin, 134, 504–535. DOI: 10.1037/0033-2909.134.4.504

Eur. J. Soc. Psychol. 43, 614–626 (2013)

http://thomas.loc.gov/cgi-bin/bdquery/z?d111:HR02965:@@@L&summ2=m&