Doubts about the conceptual and empirical status of context‐free and firm‐specific control expectancies: A reply to Boone and de Brabander

Boone and De Brabander (1993) contend mat Hodgkinson's (1992) strategic locus of control scale will not lead to interesting research results and that researchers should continue to adopt the well known Rotter (1966) I‐E scale. Central to their argument is the assertion that responses to domain‐specific control expectancy scales, such as the strategic locus of control scale, largely reflect actors' perceptions of their current circumstances, whereas responses to the I‐E scale are a function of stable personality differences. In this reply the literature on the locus of control construct is briefly reviewed, in order to show that the accumulated empirical research evidence does not support the notion of generalized control expectancies as a simple unidimensional personality trait, but points overwhelmingly to the conclusion that control expectancies are more appropriately construed as a multidimensional, domain‐specific, cognitive variable shaped by the combined effects of disposition, prior learning experiences, reinforcement histories and current circumstances. The rationale for the development of the strategic locus of control scale is further explained, in order to clarify a number of other misconceptions.