Clarifying the Functional Form of the Personality-Performance Relationship Using More Appropriate Measurement

University Of Georgia Research Foundation Inc, Athens GA

Investigators

Abstract

The association between desirable personality traits and job performance has long been assumed to be positive such that higher levels of these traits lead to better performance. This assumption has driven both theory and practice in the areas of employee selection, promotion, and team assembly. However, more recent theories have suggested that at extreme levels these desirable traits can result in sub-optimal performance, consistent with other work linking extreme levels of adaptive traits (e.g., conscientiousness) to maladaptive traits (e.g., obsessive compulsion). Despite convincing theoretical arguments, studies testing for such relations between personality traits and job performance have been mixed. These mixed results represent a serious challenge for both our understanding of behavior in the workplace and can have deleterious effects on the quality of decisions made by organizations. This research clarifies this relationship through two overlooked means. First, the research capitalizes on recent advances in the measurement of personality by using statistical algorithms that have been shown to more accurately distinguish persons with moderately high levels of personality traits from those with extreme levels of those traits. By utilizing these algorithms to both develop personality measures and index persons? levels of personality traits, the idea that extreme trait levels can result in sub-optimal human performance can be more accurately tested. Second, the research integrates this methodological advance with theoretical advances in the person-situation interaction by considering the degree to which the relevance of a given personality trait to the desired performance behaviors of the particular job in which they are working. For example, although extreme extraversion may be adaptive in jobs with a high social orientation (e.g., sales), it may result in maladaptive behaviors in jobs that are lower in social orientation (e.g., engineer). Finally, this research extends empirical work regarding the influence of extreme levels of desirable traits to the team context, an area that has been greatly under-studied and has also shown inconsistent empirical results. This research has a large impact on both the scientific study and practical application of the link between personality and behavior. Long-standing assumptions that seemingly-desirable traits always contribute to increased performance have largely informed our understanding of personality in the workplace and our use of personality tests for organizational decision-making. This research greatly increases the accuracy of scientist and practitioner communities? activities surrounding the areas of personality and workplace behavior, employee selection, team assembly, and personality measurement. This research utilizes two basic data sources. First, Procter & Gamble has supplied the research team with a large dataset including performance metrics, job titles, and responses to personality surveys that were designed to conform to ideal point models, a statistical test-scoring algorithm that results in more accurate distinction between extreme- and moderately-high levels of personality traits. This research utilizes these data to test for curvilinear relations between various personality traits and dimensions of job performance. In particular, this research tests the hypothesis that when the relevance of a particular personality trait for the persons? occupation is high, extreme levels of that trait are beneficial, resulting in a linear relationship. However, when the relevance of that particular trait to performance is only moderately high, then a moderate level of the trait is optimal for performance, whereas extreme levels of the trait result in sub-optimal performance, resulting in curvilinear relationship. When trait relevance is low, the personality trait is expected to be irrelevant to performance, resulting in a null relationship. This research integrates perspectives on curvilinear personality-performance relationships with those on the person-situation interaction, such as trait activation theory. Second, this research tests this hypothesis in large samples of adult workers using more faceted measures of personality to replicate and extend the test of this hypothesis in more representative samples. These data collections also serve the purpose of the development and free dissemination of measures developed and scored according to the more accurate ideal point scaling approach, and the development of a software program that allows researchers to easily implement the appropriate scoring for these measures in small samples, which currently requires large sample sizes and high quantitative expertise. Finally, this research capitalizes on collaboration between personality researchers and experts on team assembly to acquire and utilize lab and field samples of work teams to extend the research on curvilinear personality-performance relations to the team composition context. This research represents a major development in uncovering the true functional form of the personality-performance relationship and specifying its dependence on context. By integrating recent advances in methodological and theoretical perspectives on personality and leveraging this integration across multiple contexts, the research represents an important step forward in our theoretical and practical understanding of the operation of personality traits in organizational settings and provides researchers and practitioners with the tools to continue to generate innovations in this area of scientific inquiry.