Revisiting the nature and strength of the personality-job performance relations: New insights from interpretable machine learning.

Prior research on the relations between the five-factor model (FFM) of personality traits and job performance has suggested mixed findings: Some studies pointed to linear relations, while other studies revealed nonlinear relations. This study addresses these gaps using machine learning (ML) methods that can model complex relations between the FFM traits and job performance in a more generalizable way, particularly interpretable ML techniques that can more effectively reveal the nature (linear, curvilinear, interactive) and strength (feature/relative importance) of the personality-job performance relations. Overall, the results based on a sample of 1,190 employees suggest that nonlinear ML methods perform slightly yet consistently better than linear regression methods in modeling the relation of job performance with FFM facets, but not with factors. On the factor level, conscientiousness exhibits a noticeable curvilinear relation with job performance, and it also interacts with other FFM factors to predict job performance. Conscientiousness displays the strongest feature importance across job types, followed by agreeableness. On the facet level, most FFM facets show limited evidence for curvilinear and interactive (with other facets) relations with job performance. While several conscientiousness facets (order, deliberation, self-discipline) display the strongest feature importance in predicting job performance, some agreeableness (straightforwardness, altruism) and extraversion (positive emotionality) facets also emerge as important features for different sales job types (corporate vs. individual sales). We discuss the implications of these findings for research and practice. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Adverse impact reduction and job performance optimization via pareto-optimal weighting: A shrinkage formula and regularization technique using machine learning.

In personnel selection practice, one useful technique for reducing adverse impact and enhancing diversity is the Pareto-optimal weighting approach of De Corte et al. (2007). This approach produces a series of hiring solutions that characterize a diversity-job performance trade-off and can lead to more optimal selection outcomes (sometimes doubling the number of job offers for minority applicants without changing the job performance outcomes of personnel selection). Despite these advantages, recent research has identified a potential problem with the Pareto-weighting technique-Pareto solutions suffer from shrinkage upon cross-validation. To address the problem of shrinkage in the Pareto trade-off curve (i.e., diversity shrinkage and validity shrinkage), we offer two contributions. First, a shrinkage approximation formula is introduced (similar to a formula for adjusted R², but applicable to the entire Pareto trade-off curve). Second, we derive a novel technique for the regularization of Pareto-optimal predictor weights (borrowed from the field of machine learning), which is designed to produce predictor weights that are less vulnerable to shrinkage (similar to ridge regression and adapted from the elastic net technique). Both approaches-the proposed Pareto shrinkage formula and the proposed regularization technique-are then evaluated using Monte Carlo simulation. Recommendations are provided for approximating potential diversity-performance trade-off curves in personnel selection, while accounting for shrinkage. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Design and methods of the Longitudinal Eating Disorders Assessment Project research consortium for veterans.

INTRODUCTION: Military service members must maintain a certain body mass index and body fat percentage. Due to weight-loss pressures, some service members may resort to unhealthy behaviors that place them at risk for the development of an eating disorder (ED). OBJECTIVES: To understand the scope and impact of EDs in military service members and veterans, we formed the Longitudinal Eating Disorders Assessment Project (LEAP) Consortium. LEAP aims to develop novel screening, assessment, classification, and treatment tools for veterans and military members with a focus on EDs and internalizing psychopathology. METHODS: We recruited two independent nationally representative samples of post-9/11 veterans who were separated from service within the past year. Study 1 was a four-wave longitudinal survey and Study 2 was a mixed-methods study that included surveys, structured-clinical interviews, and qualitative interviews. RESULTS: Recruitment samples were representative of the full population of recently separated veterans. Sample weights were created to adjust for sources of non-response bias to the baseline survey. Attrition was low relative to past studies of this population, with only (younger) age predicting attrition at 1-week follow-up. CONCLUSIONS: We expect that the LEAP Consortium data will contribute to improved information about EDs in veterans, a serious and understudied problem.

Family Relationships Under Work From Home: Exploring the Role of Adaptive Processes.

Work-from-home (WFH) influences both work and life, and further impacts family relationships. The current study explored the impacts of WFH on family relationships during the COVID-19 pandemic and identified effective adaptive processes for maintaining family relationships under WFH. Using the Vulnerability-Stress-Adaptation (VSA) model, the study examined the roles of adaptive processes (spending time with family members and balancing work and life) and demographic differences (gender, age, marital status, and education level) in the relation between WFH and family relationships. Path analysis results based on an online survey (N = 150) suggested that, overall, WFH improved family relationships through proper adaptive processes. WFH had a positive relation to time spent with family members, and this relation was especially salient for workers with lower education levels. While there was no statistically significant overall relation between WFH and work-life balance, older workers tended to engage in increased work-life balance during WFH. Both adaptive processes were positively related to family relationship quality. The findings advance the understanding of family relationships and WFH and provide practical recommendations to enhance family relationships under WFH.

Developmental structure of personality and interests: A four-wave, 8-year longitudinal study.

Personality traits and vocational interests capture different aspects of human individuality that intersect in certain ways. In this longitudinal study, we examined developmental relations between the Big 5 traits and RIASEC vocational interests over 4 timepoints from late adolescence to young adulthood (age 16-24) in a sample of Icelandic youth (N = 485) well-representative of the total student population. Results showed that interests and personality traits were similarly stable over time, but showed different patterns of mean-level change. There was evidence of personality maturation but a lack of cumulative changes in interest levels. For the most part, gender differences in developmental trends were minimal. In addition, latent growth curve analyses revealed broad and specific correlated changes between personality and interests. Changes in general factors of personality and interests were moderately related (r = .32), but stronger correlated changes were found among specific personality-interest pairs that share situational content. Overall, results reveal how interests and personality are related across different types of continuity and change. While there was little correspondence between group-level changes, substantial correlated change occurred at the individual level. This means that when a person's personality changes, their interests tend to change in predictable ways (and vice versa). Integrative theories that link different aspects of psychological functioning can benefit by incorporating these findings. (PsycInfo Database Record (c) 2020 APA, all rights reserved).

Do differences between individuals who are healthy weight or overweight on self-report measures of disinhibited eating and restrained eating reflect reality or item "bias"?

In light of increasing rates of overweight and obesity worldwide, there is a critical need for accurate self-report measures of disinhibited and restrained eating behaviors across the weight spectrum. Item response theory was used to determine whether differences in disinhibited and restrained eating between healthy weight and overweight or obese individuals were due to item bias (i.e., differential item functioning). Study 1 participants were healthy weight (n = 510) or overweight or obese (n = 304) adults recruited from the community. Study 2 participants were healthy weight (n = 778) or overweight or obese (n = 320) college students. Study 1 participants completed the Eating Disorder Examination-Questionnaire (EDE-Q), Eating Disorder Inventory-3, Dutch Eating Behaviors Questionnaire, Restraint Scale, and Three-Factor Eating Questionnaire. Study 2 participants completed the Eating Pathology Symptoms Inventory (EPSI). Items on the Restraint Scale demonstrated the most evidence for bias (60% of items), whereas the majority of other scales demonstrated low to moderate levels of item bias (17-38% of items). However, EDE-Q Restraint and EPSI Binge Eating, Cognitive Restraint, Excessive Exercise, Muscle Building, and Negative Attitudes Toward Obesity scales did not show any evidence of differential item functioning among weight groups. Participants with the same level of disordered eating responded differently to certain eating disorder self-report items due to weight-bias, rather than true between-groups differences. Nevertheless, EDE-Q Restraint, EPSI Cognitive Restraint, and EPSI Binge Eating did not exhibit any evidence of bias and are ideal for assessing restrained and disinhibited eating across the weight spectrum in both research and clinical settings. (PsycInfo Database Record (c) 2020 APA, all rights reserved).

Diversity shrinkage: Cross-validating pareto-optimal weights to enhance diversity via hiring practices.

To reduce adverse impact potential and improve diversity outcomes from personnel selection, one promising technique is De Corte, Lievens, and Sackett's (2007) Pareto-optimal weighting strategy. De Corte et al.'s strategy has been demonstrated on (a) a composite of cognitive and noncognitive (e.g., personality) tests (De Corte, Lievens, & Sackett, 2008) and (b) a composite of specific cognitive ability subtests (Wee, Newman, & Joseph, 2014). Both studies illustrated how Pareto-weighting (in contrast to unit weighting) could lead to substantial improvement in diversity outcomes (i.e., diversity improvement), sometimes more than doubling the number of job offers for minority applicants. The current work addresses a key limitation of the technique-the possibility of shrinkage, especially diversity shrinkage, in the Pareto-optimal solutions. Using Monte Carlo simulations, sample size and predictor combinations were varied and cross-validated Pareto-optimal solutions were obtained. Although diversity shrinkage was sizable for a composite of cognitive and noncognitive predictors when sample size was at or below 500, diversity shrinkage was typically negligible for a composite of specific cognitive subtest predictors when sample size was at least 100. Diversity shrinkage was larger when the Pareto-optimal solution suggested substantial diversity improvement. When sample size was at least 100, cross-validated Pareto-optimal weights typically outperformed unit weights-suggesting that diversity improvement is often possible, despite diversity shrinkage. Implications for Pareto-optimal weighting, adverse impact, sample size of validation studies, and optimizing the diversity-job performance tradeoff are discussed. (PsycINFO Database Record