Abstract
The Green and Mitchell (1979) model of leadership response is based on the attribution theories of Weiner et al. (1972) and Kelley (1972). The present study extended the model to include measures of locus of control (LOC) and situational control. It was hypothesized that supervisors with an external LOC, or those who used more institutional control, would be less sensitive to subordinate performance and would evaluate it more negatively than would supervisors with an internal LOC, or those who used less institutional control. Further, supervisors were expected to prefer to use consensus or base-rate information when there was incompatibility between the situation and their personal beliefs and to prefer personal or individuating information when there was compatibility. To test these hypotheses, 288 Australian undergradutes were allocated to 18 groups in a 3 × 3 × 2 design. They responded to 16 descriptions of subordinate performance that varied in outcome, consistency, distinctiveness, and consensus. Results generally supported the hypotheses, providing further support for the Green and Mitchell formulation and, in particular, for the inclusion of variables of personal and situational control.