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

Subjects Research subjects in the experiment were 56 white males with limited military service, aged 18-21 [...]. Subjects selected from this pool of volunteers were from small towns in the southern United States. The intent of selecting subjects by geographical background was to obtain individuals with preexisting, antiblack attitudes. Public opinion surveys have shown that antiblack attitudes are more prevalent in the rural South than elsewhere in the country. [...]

Experimental Task The cooperative task forming the context in which the experiment was conducted was a "management training activity." It involved the operation of a railroad business with lines serving 10 cities and carrying a variety of items in 500 cars of six different types. The business received orders to ship specified materials from one city to another. Profits depended upon having the appropriate types of cars in the proper city to fill each order quickly. The three-person staff filled orders, redistributed cars (at a loss if empty), kept financial records, etc. [...] There were three positions in the business. The executive officer received the requests for shipment, consulted with other staff members, made the final decisions, and "phoned in" the information about how each order was to be executed. The shipping officer suggested the type of car and route to be used and kept the financial records. The equipment officer kept records regarding the distribution of cars, noted the type of available cars at the city of origin, and recommended the redeployment of cars to points where they might be needed. The research subject always occupied the latter position. The two other positions were filled by experimental confederates.

Independent Variables Success-failure of the group, degree of participation in decision making, and race of the group's executive officer were systematically varied in a 2X2X2 factorial design with a random assignment of subjects to conditions. The success-failure variable was introduced as follows: 1. The subjects were to!d at the beginning of the experiment that they would receive a $5 reward for exceeding the performance of a stated comparison group. 2. The comparison groups were said to be military officers (high status) for the success condition and basic enlisted personnel (low status) for the failure condition. 3. Success or failure was reported to the work group at the end of each workday by plotting on a display board the team's progress in relation to the appropriate comparison group. [...] The subject's level of participation in decision making was controlled by the behavior of the two trained confederates. [...] In the low-participation condition the insights underlying these changes were "discovered" and decided upon totally by the confederate occupying the executive role; the executive explained the reasons for his decisions to his co-workers. By contrast, in the high-participation condition the confederates guided the discussion in such a way that the subject was likely to discover the principles of efficient operation. [...] Race was varied as follows: For half the subjects the executive officer was black; for the other half he was white. When the executive officer was black, the shipping officer was always white, and vice versa. [...]

Subjective Experience Variables Two rating scales were employed to assess the subjects' perception of the success-failure and the high-participation-low-participation variables. A three-item measure of perceived success and a fouritem measure of perceived participation in decision making were developed for this purpose. [...] To provide a check on the adequacy of the experimental manipulations, every session was tape recorded and later evaluated by a trained observer who judged the quality of the confederates' efforts. When these were judged to be inadequate the subject was discarded and replaced [...].

Criterion Variables Each subject's postexperimental attraction ratings for both the executive officer and shipping officer (i.e., the two confederates) were obtained in private with assurances that his group would not assemble again and that the ratings were confidential. [...] Attraction for each confederate was measured by a 21-item scale. Six of these items asked for ratings of respect and liking. [...] Nine of the items were adjective ratings from the evaluative factor of the semantic differential. The adjective ratings also utilized 7-point continua [...]. The last six items were intended to specify real-life choices for the subjects, for example, how much they would like to work with each confederate on a military training program, and how much they would like to spend off-duty time with each. [...] In order to minimize the likelihood that the predicted attraction outcomes might be influenced by confederate expectations, the two confederates were misled about the nature of our hypotheses. They were told that we expected subjects to like their groupmates more when the group failed and when their personal participation was minimal. [...]