Effects of Outgroup Friendship Salience on Outgroup Attitudes and Moderating Conditions

Download data and study materials from OSF

Principal investigator:

Arthur Aron

State University of New York at Stony Brook

Email: Arthur.Aron@stonybrook.edu

Homepage: http://www.psychology.sunysb.edu/psychology/index.php?people/faculty/author_aron

Sample size: 443

Field period: 10/18/2006-11/01/2006


Intergroup prejudice is a major social problem and understanding its operation has been at the forefront of mainstream social psychological theorizing about intergroup relations, attitudes, and even basic social cognitive processes (e.g., categorization, social perception). The intergroup contact hypothesis, formulated by Allport (1954), and supported over the years (Pettigrew, 1998), is that under appropriate conditions contact with members of an outgroup can reduce prejudice towards that outgroup. Recently, studies have suggested that intergroup contact is most likely to reduce prejudice when it involves a close, intimate friendship with an outgroup member (Herek & Capitanio, 1996; Pettigrew, 1997). However, all published studies to date on this apparent friendship effect are correlational. The proposed study experimentally manipulates salience of an outgroup friendship, measures implicit and explicit attitudes towards the friend’s group, including explicit and implicit measures of the hypothesized mechanism of including other in the self, plus measures of a potential alternative moderator (admiration) and an additional potential moderator (self-esteem).


(1) Making an outgroup friendship salient improves attitudes towards that friend’s group.
(2) A key condition permitting this effect is the extent to which the outgroup friend is “included in the self.”
(3) All effects will remain after controlling for admiration of the outgroup friend.
(4) Self-esteem will moderate the moderation by including other in the self.

Experimental Manipulations

Three key conditions: (a) elaborate on the friendship with the minority group member, (b) elaborate on a friendship with a majority ingroup member with whom the participant feels similarly close as to his or her minority group friend, or (c) a no-elaboration control.


Dependent variables are measures of attitudes towards the ethnic group of their outgroup friend (along with measures of attitudes towards a different ethnic outgroup as a control).