A MULTI-LEVEL ANALYSIS OF THE EFFECTS OF SURFACE- AND DEEP-LEVEL DIVERSITY, IDENTIFICATION, AND PAST PERFORMANCE ON SUPERVISORY MENTORING

Abstract

Researchers have provided inconsistent results regarding the effects of diversity on mentoring. The current study examines the effect of surface-level and underlying diversity on mentoring. I hypothesize that diversity negatively impacts upon identification (an emotional process by which one empathizes with, emulates, and feels similar to another person and/or feels that another person is a part of one's idealized self-image) and highlight the importance of identification in mentoring. As theorists suggest that protégé past performance predicts mentoring, I also investigate the effect of this factor on mentoring.

Using hierarchical linear modeling, I examine whether supervisors provide different levels of mentoring to subordinates based on similarity and past performance. Results suggest that underlying diversity does not negatively impact upon identification. However, gender diversity negatively impacts upon both subordinate and supervisor identification, and age diversity negatively impacts supervisor identification. Limited evidence suggests that supervisors provide increased mentoring to high performing subordinates.