Date of Award

Winter 2013

Document Type


Degree Name

Doctor of Philosophy (PhD)


Psychology and Behavioral Sciences

First Advisor

Tilman Sheets


The process of using multiple sources or raters (i.e., self, supervisor, peers, subordinates, and others) in the assessment of managerial performance has been used pervasively in organizations with the primary goal of motivating behavioral change through feedback (Bracken, Timmrick, & Church, 2001). Multi-source or 360-degree feedback programs are especially suited to help measure behaviors related to performance and assess outcomes, such as leadership, interpersonal relationships, coaching, and communication (London & Smither, 1995). Typically, 360-degree feedback dimensions are measured by meta-categories of behavior called competencies.Bartram (2005) stated that these competencies could be defined as the search for characteristics that separate the best workers from the rest, usually related to characteristics, such as personality traits, that span across all jobs. However, DeNisi and Kluger (2000) stated that problems arise when managerial feedback is related to components of the ideal self (e.g., traits or individual differences) rather than feedback related to performance.

Spencer and Spencer (1993) proposed two kinds of competencies: skill-based competencies and trait-based competencies. Due to the negative outcomes associated with feedback disrupting the ideal self (i.e., decreased self-esteem, self-efficacy, and productivity), the present study sought to test whether 360-degree feedback competencies are related to personality traits of a person. Moreover, the present research tested Schmidt and Hunter's (1996) claim that interpreting the 360-degree feedback ratings as inter-rater correlations serving as reliability coefficients between ratings could represent a method of assessing the construct validity of 360-degree feedback ratings. Using confirmatory factor analysis (CFA), the present research modeled 360-degree feedback competencies by averaging across rater types (with and without self-ratings) and hierarchically across feedback items. Confirmatory models were then transformed into structural models in which personality characteristics of the Big Five were hypothesized to globally predict trait-based competencies, while not predicting skill-based competencies.

The present study indicates that hierarchical confirmatory models of the 360-degree feedback competencies have the most clear fit indices and validity coefficients. Mixed results were found for the hypothesis of personality characteristics of the Big Five predicting trait-based competencies, while the hypothesis regarding skill-based competencies was not supported. Detailed findings and implications of the research are discussed.