ITEM CONTENT VERSUS ITEM WORDING - DISENTANGLING ROLE-CONFLICT AND ROLE AMBIGUITY

Rizzo, House, and Lirtzman's (1970) Role Ambiguity Scale and Role Conflict Scales assess ambiguity with 6 negatively worded items and conflict with 8 positively worded items, respectively. This methodological confound between item wording and content precludes unambiguous interpretation. In the present study, confirmatory factor analysis of these 2 scales and Beehr, Walsh, and Taber's (1976) Role Overload Scale (which has positively and negatively worded items) was used to disentangle this confound. Across 2 independent samples (N = 767 and N = 363), a 3-factor model consistent with conceptual definitions of role ambiguity, conflict, and overload fit the data better than models with (a) one general role-stress factor, (b) a general role-stress and a method (item wording) factor, or (c) two method (positive and negative wording) factors. These results support the construct validity of Rizzo et al.'s (1970) scales; the consistency of the results across 2 independent samples suggests their generalizability.