|
|
|
|
Method The basic design of the study was the random assignment of White college students to small-group discussions in a 2x2 factorial design. The two independent variables are group racial composition and group opinion composition on a target social issue. The main outcome variable was the IC of students’ thinking about a target social issue. The experimental conditions were created through a research collaborator who acted as a participant in the discussion group. This collaborator was either Black or White and followed a pre-determined script that either agreed or disagreed with the opinions indicated by the White participants on a prior screening survey. Thirty-one collaborators were used in experiments; all were blind to the purposes of the study. The study was conducted at three selective research universities. Participants were recruited through campus flyers and e-mails. If they expressed interest, they were given a questionnaire to collect information on their race, background characteristics -including contact with racially diverse others1-, and their opinions on several social issues. Those who agreed with the most prevalent position of one of two target social issues (against child labor practices in developing countries or in favor of the death penalty) were asked to participate in a subsequent experimental session. Participants were blind to the purposes of the study and were debriefed subsequent to participation. Procedure A total of 357 White participants (135 men, 222 women; age M=20.0, SD=3.6) were assigned to same-sex experimental groups consisting of three participants and one research collaborator. In each experiment, a facilitator led the participants and the collaborator into the laboratory and sat them around a table. They were then given an issue prompt describing the same target social issue on which they had indicated their opinion on the screening survey. After reading the prompt and before any discussion took place, participants were asked to indicate their agreement or disagreement with the issue and to write a short essay describing their support for or opposition to the issue (the pre-discussion essay). They were given 15 minutes. Most of the participants (85 percent) indicated that they held the same position as they did on the screening survey. After completion of the first essay, participants were asked to discuss their opinions on the issue. The facilitator asked each participant to begin by orally stating his or her opinion. This was followed by an unstructured 15-minute discussion during which the collaborator followed a script written to express agreement or disagreement with the majority of the participants. Of the 357 participants, 108 were in groups in which the collaborator disagreed with everyone (i.e., extreme opinion minority condition) and 123 were in groups in which the collaborator agreed with everyone (i.e., extreme opinion majority condition). For 60 participants, the collaborator agreed with only 1 other group member and for the remaining 66 participants, the collaborator agreed with 2 of the other group members. Participants were then asked to write a second essay on the same topic (the post-discussion essay), for which they were given 15 minutes. After completion of the second essay, participants were given a second prompt asking them to indicate their agreement or disagreement with a different social issue (either child labor practices or the death penalty, the alternate of the first issue) and to write a short essay describing their support or opposition to this second issue. We call this the transfer essay because it tests whether any stimulation of complex thinking due to the group discussion on the first topic transfers to thinking on a second topic. Participants were given 15 minutes to complete their essay. Participants were subsequently asked to complete a questionnaire in which they rated each member of their group, including the collaborator, with regard to their contributions to the earlier discussion. Participants rated how much each group member made others think about the issue in different ways, introduced a novel perspective to the discussion, and was influential in the group. We averaged these three ratings of the collaborator (alpha of .90) to form an index of Perceived Novelty (scale range from 1 to 7). All essays were rated for IC (scale range from 1 to 7) by three independent judges who were blind to the purposes of the experiment. Procedures followed those established by Suedfeld et al. (1992). The interrater reliability was .70 for the pre-discussion measure, .62 for the post-discussion measure, and .66 for the transfer measure. Because the post-discussion measure is in essence a change measure and because it is acceptable that change scores have relatively low reliability (Overall & Woodward, 1975), the .62 value is methodologically acceptable. Moreover, because raters may have focused on different parts of the essay, interrater reliability may underestimate the reliability of the measure. Results Four different outcome variables were analyzed. First, we examined whether collaborator race and collaborator opinion had effects on perceived novelty. Second, we tested for the effect of collaborator race on IC in the pre-discussion measure (i.e., before the collaborator had the opportunity to express an opinion). Third, we tested for the effects of collaborator race and collaborator opinion on IC in the post-discussion measure. Fourth, we tested for the effects of collaborator race and opinion on IC in the transfer measure. All analyses were multilevel regression analyses that allowed initially for group-level effects. When group effects were not present, they were dropped from the model. We also tested for main effects of university site, issue, age, gender, and contact with racially diverse others, as well as interactions of these variables with collaborator race. Only statistically significant results are reported. Perceived Novelty There were statistically significant main effects for collaborator race and collaborator opinion on perceived novelty (for Race, t(108)=2.05, p=.042, d=.29; for Opinion, t(108 = -6.39, p < .001, d= -1.07). That is, participants judged the collaborator's contribution to the discussion as more novel when the collaborator was Black, even though the White collaborator followed the same script in the group discussion (least squares mean of 5.56 for the Black collaborator and 5.27 for the White collaborator). In addition, participants who were in opinion minority groups rated the collaborator higher for perceived novelty (M=5.95) than participants who were in opinion majority groups (M=4.88). The interaction of these factors was not statistically significant, but an examination of the means suggested that in groups in which the collaborator agreed with everyone else in the group, the Black collaborator was seen as more novel than the White collaborator. Pre-discussion Integrative Complexity The IC of the pre-discussion essays was analyzed for effects of collaborator race and participant background characteristics. The effect of collaborator opinion was not considered because the participants were unaware of the positions of the other participants or the collaborator at this point in the experiment. A marginally significant main effect of collaborator race was indicated (for the White collaborator M=1.83 and for the Black collaborator M=1.94; t(352)=1.70, p=.09, d=.18). No significant interactions of race with university site, issue, age, gender, and contact with racially diverse others were detected. Post-discussion Integrative Complexity We found a main effect for collaborator opinion [t(351)= -3.91, p < .001, d = -.51], such that participants in groups in which the collaborator held a minority opinion showed higher IC (M=1.88) than those in groups in which the collaborator agreed with the three members of the group (M=1.63). There was no effect of collaborator race, nor did it interact with other variables. There was, however, a significant main effect for diversity of racial contact such that participants reporting higher levels of diverse racial contact showed higher levels of IC [t(351)=2.47, p=.014, r=.13]. Transfer There were no main effects of collaborator race or collaborator opinion. Similar to the post-discussion analysis, there was a positive effect on IC for contact with racially diverse others [t(352) =2.66, p=0.008, r=.14]. The analysis of interaction effects indicated one significant interaction of collaborator race by issue [t(352)= -1.98, p=.049]. For those now writing on the topic of child labor, those who had a Black collaborator in the group had higher IC scores (M=1.91) than those with a White collaborator (M=1.52; d=.73); however, for those now writing on the topic of the death penalty the race difference was much smaller (M=1.71 of Black collaborator and M=1.68 for White collaborator; d=.06). |
|
Send mail to
wperez@stanford.edu with
questions or comments about this web site.
|
