Compositions in Black and White: Determinants of interaction in interracial dyads.

The initial, unstructured interactions of 40 interracial (black-white) dyads were studied in a design in which three factors were systematically varied. These factors were (a) disposition of white dyad members to either seek out or shun (i.e., approach vs. avoid) interaction with blacks, (b) race (black vs. white) of experimenter, and (c) gender composition (male-male vs. female-female) of dyads. The results revealed that within dyads, white dyad members displayed more interactional involvement (i.e., more talking, smiling, and gazing) than their black partners but at same time appeared to experience interactions as more stressful and uncomfortable. Additional findings qualified these results by revealing that (a) whites predisposed to avoid interaction with blacks looked and smiled at their partners less than those predisposed to initiate interaction; (b) both black and white members of these avoidance dyads reported heightened feelings of anxiety and concern about their interactions; but (c) moderating influences of whites' approach-avoidance dispositions on interaction behavior were essentially limited to conditions in which experimenter was black and white subject was therefore a solo minority. The author suggests that blackwhite partner effects may be attributable to differing amounts of crossrace contact typically experienced by blacks and whites. He interprets black-white experimenter effects in terms of Taylor's (1981) hypothesis that stereotypes and related dispositions are activated in social contexts in which group membership is made salient. Stephan (in press) concluded a recent review of literature on intergroup relations by noting that the ecological validity of much of work in this area is low and it remains to be determined when the phenomena of interest operate in everyday social interactions. This conclusion may apply with even greater force to more specific literature on black-white relations in U.S. and Canada. Not only are there relatively few studies in this literature in which actual social interactions of whites and blacks are examined, but there appear to be none in which a range of behavioral and self-report measures is used