Publication Date


Document Type


Degree Name

M.A. (Master of Arts)

Legacy Department

Department of Psychology


Social perception


The effects of both thinking in complex and multidimensional terms about others and of belonging to a positively-valued group (i.e., social identity) on the differential appraisals of desirable and undesirable in-group and out-group members were investigated. In the first of two sessions, 150 undergraduate students evaluated two targets; one was described in positive terms and the other in negative terms. Both were said to be either future partners of the participants in a problem-solving task planned for the second session (i.e., in-group) or members of a different group (i.e., out-group). To observe the joint effects of thought complexity and social identity, participants were either told to rate the targets by carefully considering, in a multidimensional manner, all the available information or to quickly form an impression using one piece of information. They were also told either that groups were composed of people with similar attitudes and beliefs, or assigned at random. A ruse required participants to evaluate the same targets again at the second session. For half of the participants, the target’s group membership was different (e.g., in-group targets became out-group targets) from session one. Targets were rated for favorability, desirability as a group member, and amount expected to contribute to the group. Essays written about each target were coded for cognitive complexity. Manipulations of thought complexity and social identity were unsuccessful in producing variations in, respectively, the degree to which participants considered targets in complex ways, and the degree to which one identified with the group. Therefore, to assess the effects of thought complexity and identity on judgments, quasi-median splits were conducted on the sum of the essays’ complexity scores and on a measure of the amount participants cared about the group’s performance. These splits formed the basis of the cognitive complexity and social identity variables, respectively. For each dependent variable, this yielded the factors of group membership, cognitive complexity, caring about the group, target valence, and time with repeated measures on the last two factors. When group membership was constant across sessions, out-group polarization emerged for ratings of favorability and desirability as a group member; the desirable target was evaluated more positively in an out-group than in an in-group, and the undesirable target was rated more negatively in an out-group than in an in-group. Greater thought complexity produced greater polarization. Thought complexity and group membership independently affected favorability ratings; when the effect of complexity was controlled, out-group polarization remained. For the target’s expected contribution to the group, participants who cared less about the group’s performance exhibited in-group polarization. Changes in the target’s group membership had no reliable effects on target evaluations of favorability, desirability as a group member, or expected contribution. Participants, especially those caring more about the group’s performance, anticipated that their group would outperform other groups. Implications for cognitive and social models of evaluation polarization are discussed. It was concluded that both cognitive and social (group membership) factors contribute to differential evaluation of in-group and out-group members.


Includes bibliographical references (pages [90]-94)


viii, 156 pages




Northern Illinois University

Rights Statement

In Copyright

Rights Statement 2

NIU theses are protected by copyright. They may be viewed from Huskie Commons for any purpose, but reproduction or distribution in any format is prohibited without the written permission of the authors.

Media Type