The effect of self-affirmation on sexual prejudice

Publication Year
2010

Type

Journal Article
Abstract

In three experiments, we explored the impact of a self-affirmation treatment on sexual prejudice (i.e., negative attitudes toward gays and lesbians). Studies 1 and 2 found that participants who were affirmed by valuing relationships with family and friends were significantly more prejudiced than participants who were affirmed by valuing other self-relevant characteristics. Relative to a non-affirmed control, the family/friends affirmation did not actually increase prejudice; however, other affirmations decreased bias. Study 3 replicated the finding that prejudice was higher among participants who affirmed to family/friends compared to those who affirmed to other values, and showed a mediator of the effect: the endorsement of traditional family values. That is, affirming to family/friends was associated with support for family values, which was positively associated with prejudice. These findings add to a growing body of evidence demonstrating the potential for self-affirmation to reduce bias, but establish that the type of value affirmed is an important consideration. Specifically, familial-based affirmations may undermine reduction of sexual prejudice because they remind individuals of values that many people see as being in conflict with expressing tolerant attitudes toward gays and lesbians.

Journal
Journal of Experimental Social Psychology
Volume
46
Pages
276–285
Type of Article
Journal Article
Full text

The following is an excerpt of the intervention methodology. For more information, please see the full text of the article on the publisher's website or through your institution's library.

Study 1

In Study 1, we experimentally manipulated whether participants were given the general opportunity to self-affirm or not. [...]

Participants Participants were 102 heterosexual, White college undergraduates. Fifty men and 52 women participated in the study, and the average age was 19.01 (SD = 1.20; range = 18–23). [...] In this and the subsequent studies, all participants were students from a large, Midwestern university in a state in which gay and lesbian individuals and couples do not receive explicit protections under the law and public opinion towards same-sex relationships was unfavorable at the time these studies were conducted.

Materials and procedure [...] Upon providing consent, participants were randomly assigned to receive (48) or not receive (54) a self-affirmation treatment, and then completed a questionnaire concerning social issues in which our dependent variables of interest were embedded.

Affirmation manipulation. At the beginning of the study, participants completed a standard self-affirmation manipulation. All participants were asked to rank 11 values in order of personal importance. Next, participants were asked to describe their most important value and write about why it is important to them personally (affirmation condition), or they were asked to describe their 9th-most important value and why it might be important to the average college student (non-affirmed condition).

Sexual prejudice. Ten items from Herek’s (1988) Attitudes toward Lesbians and Gay Men Scale were used to assess sexual prejudice [...], with five items assessing prejudice toward gay men and five items assessing prejudice toward lesbians. [...]

Religious fundamentalism. Ten items from Altemeyer and Hunsberger’s (1992) Religious Fundamentalism Scale [...] were given. [...]

Political conservatism. Five items were selected from the Conservatism–Liberalism Scale to assess individuals’ political orientations [...]

Demographics. Finally, participants completed a standard demographic questionnaire that included items about participant gender, race, age, and sexual orientation. [...]

Study 2

The goal of the second study was to see if we could replicate the findings from Study 1 when participants did not have the ability to choose the value to which they affirmed. Thus, participants were randomly assigned to affirm family/friends, to affirm another value of importance from a domain unrelated to anti-gay bias, or to participate in a non-affirmed control condition. [...]

Participants Participants were 188 heterosexual, White college undergraduates. One hundred eleven men and 77 women participated in the study, and the average age was 19.45 (SD = 1.18; range = 18–25). [...]

Materials and Procedure [...] Upon providing consent, participants were randomly assigned to one of three conditions: affirm family/friends (63), affirm sense of humor (63), or a no affirmation control (62). [...] Upon completing the affirmation manipulation, participants completed a questionnaire on social issues in which our dependent measures of interest were embedded.

Affirmation manipulation. The affirmation manipulation differed from that employed in Study 1 in that participants did not first engage in a value ranking task. In this study, participants were asked to write about why relationships with family/friends are important to them personally (familial affirmation condition), why their sense of humor is important to them personally (humor affirmation condition), or why a sense of creativity might be important to the average college student (non-affirmed condition). The non-affirmed condition was modeled after that used in Study 1, in which participants wrote about why a value of low personal relevance might be important to someone else. [...] After the manipulation, participants completed the following measures:

Sexual prejudice. We used the same subset of items from Herek’s (1988) Attitudes toward Lesbians and Gay Men Scale discussed in Study 1 [...].

Potential moderator variables. We used the same measures of religious fundamentalism [...] and political conservatism [...] as we did in Study 1. In addition, we administered the following new variables: sexual orientation group identification, as well as internal and external motivation to control prejudice. To assess sexual orientation group identification, participants completed a set of six items adapted from Schmitt, Branscombe, Kobrynowicz, and Owen’s (2002) measure of gender group identification [...]. The content of the items remained the same aside from the fact that the word ‘‘gender” was replaced with ‘‘sexual orientation.” [...] Internal and external motivation to control prejudice toward gays and lesbians were each assessed with five items adapted from Ratcliff, Lassiter, Markman, and Snyder (2006). [...]

Demographics. Finally, participants completed a demographic measure identical to that employed in the first study. [...]

Study 3

The goal of the third study was to consider a potential mediating mechanism of the effect of self-affirmation on sexual prejudice: endorsement of traditional ‘‘family values.” [...]

Participants Participants were 71 heterosexual, White college undergraduates. Thirty-two men and 39 women participated in the study, and the average age was 19.46 (SD = 1.14; range = 18–24). [...]

Materials and procedure The procedure was identical to that employed in Study 2, with the exception of the fact that participants were randomly assigned to either affirm family/friends (35) or affirm sense of humor (36) in the current study.

Affirmation manipulation. The affirmation manipulation was identical to that employed in Study 2, with the exception that a nonaffirmation control condition was not included. Thus, participants were simply asked to write about why relationships with family/ friends are important to them personally (familial affirmation condition) or why their sense of humor is important to them personally (humor affirmation condition). After the manipulation, participants completed the following measures: Sexual prejudice. We used the same subset of items from Herek’s (1988) Attitudes toward Lesbians and Gay Men Scale discussed in Study 1 [...].

Measures of traditional family values. Next, we administered three shortened measures of traditional ‘‘family values.” [...] They include three subscales: parent–child relationships [...] husband–wife relationships [...], and traditional male–female gender roles [...]. Each scale consisted of five items, and all items were rated on a 7-point scale ranging from 1 (strongly disagree) to 7 (strongly agree).

Potential moderator variables. We used the same measures of religious fundamentalism [...], political conservatism [...], internal and external motivation to control prejudice [...], and sexual orientation group identification [...] as we did in Study 2.

Demographics. Finally, participants completed a demographic measure identical to that employed in the previous two studies. [...]

Type of Prejudice/Bias
Country
Method