The effects of intergroup similarity and cooperative vs competitive orientation on intergroup discrimination Author Rupert Browna Publication Year 1984 Type Journal Article Abstract The prediction from several theories of interpersonal relations that intergroup similarity would lead to less differentiation and more attraction was contrasted to a hypothesis from Social Identity Theory which predicts the opposite outcome. In Expt 1, groups of schoolchildren expected to interact cooperatively with one another. The perceived characteristics of the out-group were varied in a 3 times 2 design. The independent variables were out-group Status (higher vs. same vs. lower relative to the in-group), and its prevailing Attitudes (similar to vs. different from those in the in-group). Evaluations of anticipated performance and liking clearly supported the similarity–attraction hypothesis. But on a Prisoner's Dilemma measure, more competition was observed against the similar group, as predicted by Social Identity Theory. In Expt 2, using a similar design, a third independent variable was incorporated: the nature of the Task (cooperative vs. competitive). Unexpectedly, this factor had few effects in isolation and its predicted interaction with similarity was not observed. But further analysis revealed that subjective goal orientation may be an important variable: similarity led to intergroup attraction only for non-competitive subjects; for competitive subjects the reverse was true. It was concluded that similarity may perform different functions in different contexts. Keywords attitude similarity, competition, cooperation, intergroup dynamics, status, interpersonal attraction Journal British Journal of Social Psychology Volume 23 Pages 21-33 Type of Article Journal Article DOI 10.1111/j.2044-8309.1984.tb00605.x 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. Experiment 1 Method Subjects. These were 102 children (43M, 59F) aged 12-13 years from two secondary schools in Bristol, England. Design. The experiment used a 3x2 complete factorial design with independent groups. [...] Procedure. [...] In each, children from each school were brought to the Department of Psychology at Bristol University and were taken to separate rooms without meeting the members of the other school. They were then randomly assigned to two-person teams, mostly of mixed sex. [...] The cooperative nature of the task was explained: a word matching game where each team had to match words with a team from the other school. Each team would have to work with the other to see how many words could be matched up in eight minutes. It was emphasized that the teams from the two schools could not compete since they must work together, and that a high level of general knowledge was important for success. The nature of the out-group. although not its actual identity, was then described to subjects by providing them with some information about both schools. This information was of two sorts: first about the presumed status of each school, and second about the kind of attitudes prevalent in them. [...] Subjects then answered a questionnaire which consisted of, among other things, evaluations of the two teams' expected performances on the task, and ratings of anticipated liking for the out-group. Finally. the experimenter asked subjects to distribute some money between the schools in return for helping with the experiment. using modified Prisoner's Dilemma (PD) matrices whose operation was first carefully explained. [...] Dependent Measures. The ratings were obtained by using bipolar scales represented by a 10 cm unmarked line (later converted into a 10-point scale. 0-9). [...] The two PD matrices, which always followed the ratings, are shown in Table I. Instead of the usual dichotomous choice between two alternatives. subjects were asked to show the strength of their choice on a rating scale. [...] Experiment 2 In order to test this supposition, a second experiment was designed in which the nature of the goal relationship linking the groups was varied. [...] Method Subjects. These were 166 children (82M, 84F) aged 12-13 years from two secondary schools in Bristol, England. [...] Design. The experiment used a 2 x 3 x 2 complete factorial design with independent groups. The factors were task (cooperative vs. competitive), status (higher vs. same vs. lower), and attitudes (similar vs. different). Procedure. There were three sessions. In each, children from each school were brought to the Department of Psychology and taken to separate rooms without meeting members of the other school. As in Expt 1. they were randomly assigned to two-person teams, nearly all of mixed sex. The purpose of the study and the nature of the task, which differed from session to session, were then outlined by the experimenter. In Session 1, both schools were told it was 'to see how well schools will work together on a general knowledge task', and the task described was a cooperative one. In Session 2, both schools were told it was 'to compare how well different schools will work on a general knowledge task', which was described as a competitive one: each team had to work against a team from the other school to see 'who gets the most' correct answers. In Session 3. one school received the cooperative instructions and the other the competitive. [...] As before, the perceived nature of the out-group was varied by randomly distributing different 'fact sheets' about the two schools. The information about children's own school stated that it was of 'average' standard in all subjects, and that nearly everyone at the school agreed that English, Maths, the ability to think, and General Knowledge were more important attributes than Latin, Music, Religious Knowledge, and Art. [...] According to experimental condition, the other school was described as being of very high, average. or very low standard in all subjects. and as holding two opinions about the relative importance of different attributes which were identical or nearly completely opposing to the in-group's attitudes. Subjects then answered a questionnaire which consisted of ratings of the two teams' expected performances on the task and anticipated liking for the out-group. Finally, subjects were asked to distribute some money between the schools in return for helping with the experiment, using modified PD matrices similar to those used in Expt I. [...] Dependent measures. Ratings were obtained from scales similar to those already described, in a randomized order for each subject. [...] Type of Prejudice/Bias Minimal/Artificial Country United Kingdom Method Lab Setting Elementary School (Grades 1-5) Google ScholarDOIBibTeX