Sociability: Traits, settings, and interactions.

Abstract

Sociability was examined in relation to the trait, setting, and interaction controversy. Sociability, measured behaviorally as verbal participation in an unstructured conversation, was expected to be greater for participants who were affiliative and exhibitionistic but not defensive (traits); for sociopetal seating arrangements and for greater environment-elicited pleasure, arousal, and dominance (settings); and for interactions among and between these trait and setting measures. Sociability did, as hypothesized, increase with the traits of affiliativeness and exhibitionism, the setting variables of pleasure and arousal, one Person X Person interaction, and one Setting X Setting interaction. However, seating arrangements and 1 2 Person X Setting interactions were notably unrelated to sociability; assertions that Person X Setting interactions account for large portions of variance in behavior are unsupported. Instead, the results suggest that more attention should be paid to the predictive utility of interactions among person measures and among setting measures. By now it is clear that sociability, defined here as verbal participation, is influenced by a wide range of variables, from the subtle microkinesic cues people display in entering a meeting room (Gary, 1976) to the enduring interpersonal relationships making u p a group's personality, or syntality (Cattell, 1948). Person-based measures have been the most frequently studied correlates; two decades ago Mann (1959) could already review 17 studies of the relationship between individual difference measures and verbal activity rate. On the other hand, interest i n the relationship of physical setting variables to verbal interaction began a s a n afterthought (Steinzor, 1950) but increased with the general upsurge of interest i n the physical environment and behavior (e.g., Osmond, 1957; Sommer, 1969). Very few studies o f sociability have included both person and environment measures in the same investigation o r investigated statistical interactions between them. The assistance of Patsy Mclnnes and the Social Sciences and Humanities Research Council (Grant 41077-0509) is gratefully acknowledged.