Is it “trustworthy”? A multiple-levels-of-analysis reexamination of an ohio state leadership study, with implications for future research

A moderator subgroup study, originally published by Schriesheim and Murphy (1976), is reexamined using the within-group agreement coefficient (rwg; James, Demaree & Wolf, 1984), moderated raw-score multiple linear regression, and moderated within- and between-entity analysis (WABA; Dansereau, Alutto & Yammarino, 1984; Schriesheim, forthcoming). The results show serious discrepancies in results between the subgroup and regression approaches, the need to exercise care in interpreting the rwg and raw-score regression results, and the necessity of conducting WABA and supplemental within-entity WABA moderation analyses to gain meaningful insight into the phenomena being investigated. Implications and future directions for leadership theory and research are considered.