Demands,Resources, Well-Being and Strain: Meta-Analyzing Moderator Effects of Workforce Racial Composition

Racial diversity in the U.S. workforce is increasing, and many organizations are more racially diverse than ever before. Racial minority employees experience the ecosystem of work demands and resources differently to white employees; including perceiving higher demands, lower control and support, greater stress and psychological strain, and less well-being. However, there remains little research on how relationships between these work characteristics and well-being and strain, and interrelationships differ across more racially diverse versus homogeneous (e.g., predominantly white) workplaces. This limits understanding of optimal job redesign practices as workplaces continue to become more racially diverse. Through the lens of the job demands, control, support (JDC[S]), and the job demands-resources (JD-R) frameworks, we build on previous meta-analyses by examining workplace racial composition as a moderator of demand-resource relationships with well-being and strain, and interrelationships, in 63 studies of U.S. workers (N_{(Individuals)} = 93,974). Our findings show several moderation effects. For example, as the proportion of racial minority employees increases, the positive control and well-being relationship increases, and the positive relationship between supervisor support and well-being decreases. Further, as the proportion of African-American employees (versus white and all other racial subgroups) increases, the positive control and supervisor support relationship decreases, as does the positive coworker support and well-being relationship. Our results offer new insights into the role of workplace racial composition on how work is experienced, and job design recommendations in a time of increasing workplace racial diversity.

     

Using a Computational Model to Understand Possible Sources of Skews in Distributions of Job Performance

The typical assumption that performance is distributed normally has come under question in recent years (e.g., O'Boyle & Aguinis, 2012). This paper uses a dynamic, computational model of performance‐as‐results to examine possible sources of such distributions. That is, building off the classic model of job performance (Campbell & Pritchard, 1976), components of a dynamic model are examined in 4 separate experiments using Monte Carlo simulations. The experiments indicate that positively skewed distributions can arise from pure luck, multiplicative combinations of factors where 1 of those factors has a zero origin, Matthew effects associated with learning, and feedback effects of performance on resource allocation policies by external agents. The results are discussed in terms of explanations for positively skewed performance distributions and the use and expansion of the computational model for examining dynamic performance more generally.