Exploring the predictive validity of personality for job performance across occupations using a person-centered approach
The goal of this study is to shed light upon the complex and long-debated relationship between personality and job performance from a new angle. Using a person-centered approach to examine personality, this study is the first to examine the criterion-related validity of personality profiles in predicting job performance in a corporate sample while accounting for occupational membership. More specifically, using an archival dataset from a Fortune 100 company, the current study involves hypotheses and research questions related to the existence and distribution of personality profiles across occupations, incremental validity of personality profiles in predicting performance, differential predictive validity for personality profiles across occupations, and the distribution of personality profiles among top performers within occupations. Four organization-based personality profiles were identified: adaptable, rigid, confident, and nervous. Occupation-based personality profiles were also identified for the occupations of sales, accounting and finance, manufacturing engineering, and research and development. The identified occupation-based personality profiles included some of the organization-based profiles as well as some distinctive profiles. Testing for the criterion-related validity of personality profiles showed somewhat lower validities in comparison with personality traits. Examination of the incremental validity of personality profiles above and beyond personality traits showed limited evidence of incremental validity for organization-based profiles and mixed evidence for occupation-based profiles, with a few cases of notable incremental validity for occupation-based profiles in predicting specific job performance dimensions. In addition, an exploration of the distribution of personality profiles among the top 10% of performers indicated that the confident profile was most common and the rigid profile was least common. Theoretical and practical implications are discussed along with potential future research directions.