GENDER AND FORMAL AUTHORITY

This paper specifies three theoretical approaches that identify conditions under which gender differences in leader-subordinate interaction should occur: gender-role socialization, structural position, and expectation states. From these perspectives are derived predictions of leaders' and subordinates' task and socioemotional behavior in formal task groups. These predictions then are tested in a laboratory experiment consisting of a simulated organization with formal positions. The structural approaches-structural position and expectation states-were supported more strongly than the socialization approach, an indication that formal position and associated power resources have a stronger effect than gender on leader and subordinate behavior. Several unexpected findings, however, suggest that the leader-subordinate interaction in mixed-sex groups is qualitatively different from that in same-sex groups. Mixed-sex groups may contain some social tension that is absent in same-sex groups.