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.

Method

Participants College students (N — 576; 288 women and 288 men) enrolled in a general psychology course participated in partial fulfillment of a research (readings—participation option) course requirement. During an earlier pretesting session, half of these students indicated that they identified with the Democratic Party, and half indicated that they identified with the Republican Party. Each experimental session was composed of two 3-person groups (three Democrats and three Republicans) that were homogeneous with regard to gender. Sixteen pairs of 3-person groups were randomly assigned to each condition within the 3 (interaction: full, partial, none) X 2 (common fate: yes, no) design.

Procedure

Subgroup formation. In each session, two 3-person groups (Democrats and Republicans) were assigned to two ostensibly different experiments [...]. In general, the procedures were modeled after those of Gaertner et al. (1989, 1990). At the beginning of the session, each group was informed that the 3 participants were selected to participate in the session on the basis of their common political party identification as Democrats or Republicans, which they indicated earlier in the semester. Tape-recorded instructions stated that the purpose of the study was to explore ways in which people of different political persuasions make decisions across a variety of issues and that we would be recording their group's interaction as they attempted to reach consensus to the winter survival problem. This problem is engaging and requires participants to imagine that their plane has crash-landed in the woods of northern Minnesota in mid-January and to rank-order 10 items salvaged from the plane [...] in terms of their importance for the group's survival. To further increase the salience of each participant's membership in this 3-person group, the members were asked to create a name for themselves [...]. Also, the members of each 3-person group were asked to wear the same color T-shirts (either blue or maroon) [...]. Following the 3-person group discussion, participants were informed of their subsequent contact with another group that would take place soon in a larger room within the laboratory. Participants were also told that the political party affiliation of the other group was different than their own. [...]

Manipulation of intergroup interaction and common fate. The members of each group sat at different trapezoidal tables [...] in different colored chairs [...]. Although no mention was made of the participants' common university affiliation, one of two experimenters indicated that the groups had different political affiliations as Democrats and Republicans and mentioned their earlier derived group names during an announcement ostensibly prepared for the video record. Before group discussions began, the participants were informed about the contingencies relating to the manipulation of common fate. In this phase of the experiment, the groups considered a different problem that was designed to relate to the differentiation in their political ideologies. The problem was named the budget deficit problem, and the participants were asked to prioritize a list of spending programs to cut and taxes to increase in order to most effectively reduce the United States's budget deficit. Specifically, participants were asked to consider six spending programs [...] and to rank-order these programs in terms of which should receive the largest to smallest cut. [...] Participants were led to believe that a single best solution had been developed by a bipartisan team of economic experts and that this solution would represent the standard against which the relative effectiveness of solutions generated in our experiment would be judged. [...] In the full interaction condition, the two 3-person groups [...] reached a single consensus solution to the problem within a standard 8 min time limit. In the partial and no-interaction conditions, each 3-person group solved the problem separately without conversing with members of the other group. [...] In the partial interaction condition, however, after each group reached its separate solution to the budget deficit problem and then deposited its entry into a lottery box, members of each group reported their solution to the other group by reading the rank it assigned each item. [...] In the common fate with full interaction condition, after the groups interacted and reached a single consensus to the budget deficit problem, each person signed his or her name to the solution form which was then placed into a single envelope before it was deposited into a sealed box labeled Lottery Box. [...] In the no-common-fate-with-full-interaction condition, after the groups interacted and reached a consensus solution, the members of each 3-person group were given a 1-min opportunity to discuss and modify that 6-person consensus solution. Then, the members of each 3-person group separately recorded the group's final solution and the names of its three members on a special form that was placed into its own separate envelope, and then both envelopes were deposited into the lottery box. Even if members of each group entered identical solutions, the fate of each group would be independent of the other because these solutions were sealed in separate envelopes. [...] In the no-common-fate with partial- and no-interaction conditions, after each group reached its separate solution to the budget deficit problem without having interacted with the other group, the members of each 3-person group signed their names to the solution, placed it into an envelope, and then both envelopes were deposited into the lottery box. Subsequently, in the partial interaction condition, members of each group shared their solutions with one another. In the partial and no-interaction conditions, the members of each group did not work together, nor would they necessarily share the fruits of their labors with members of the other group.

Measures Following the experimental manipulations, the participants were given a postexperimental questionnaire containing items to check the efficacy of the experimental manipulations, to examine several of the processes proposed to mediate the relationship between the elements of intergroup cooperation and bias, and to measure intergroup bias. As a check of the common fate and interaction manipulations, participants were asked to indicate the extent (on a scale ranging from 1 = not at all to 7 = very much) to which they agreed with the following statements: "If 3 of us win the monetary prize, the other 3 people will win the prize also"; "During the session, the two groups were communicating"; and "It felt as though both groups were cooperating with each other." Also, participants rated each of the five other participants in terms of how much each person revealed about himself or herself. Intergroup bias was measured with evaluative rating scales in which each participant rated (on a scale from 1 to 7) each of the others (not including themselves) in terms of how much they liked each of the others, as well as each person's cooperativeness, trustworthiness, and valuableness. Separate in-group and out-group ratings were constructed for each measure and then averaged for each 6-person group. [...] Finally, as a less reactive indicator of the effects of common fate on the evaluations of in-group and out-group members, observer ratings of one participant's facial expressions while in-group and out-group members were contributing their ideas in the full interaction conditions were undertaken. [...]