The validity and incremental validity of knowledge tests, low-fidelity simulations, and high-fidelity simulations for predicting job performance in advanced-level high-stakes selection

In high-stakes selection among candidates with considerable domain-specific knowledge and experience, investigations of whether high-fidelity simulations (assessment centers; ACs) have incremental validity over low-fidelity simulations (situational judgment tests; SJTs) are lacking. Therefore, this article integrates research on the validity of knowledge tests, low-fidelity simulations, and high-fidelity simulations in advanced-level high-stakes settings. A model and hypotheses of how these 3 predictors work in combination to predict job performance were developed. In a sample of 196 applicants, all 3 predictors were significantly related to job performance. Both the SJT and the AC had incremental validity over the knowledge test. Moreover, the AC had incremental validity over the SJT. Model tests showed that the SJT fully mediated the effects of declarative knowledge on job performance, whereas the AC partially mediated the effects of the SJT.     

A colorful walk, but is it on the mental number line? Reply to Cohen Kadosh, Tzelgov, and Henik

Cohen Kadosh, Tzelgov, and Henik [Cohen Kadosh, R., Tzelgov, J., and Henik, A. (2008). A synesthetic walk on the number line: The size effect. Cognition, 106, 548-557] present a new paradigm to probe properties of the mental number line. They describe two experiments which they argue to be inconsistent with the exact small number model proposed by Verguts, Fias, and Stevens [Verguts, T., Fias, W., Stevens, M. (2005). A model of exact small-number representation. Psychonomic Bulletin and Review, 12, 66-80]. We discuss the data, assumptions, and conclusions of Cohen Kadosh et al.'s paper in relation to existing models of numerical cognition.     

Maladaptive personality traits and psychopathology in childhood and adolescence: the moderating effect of parenting

This study examines child and adolescent psychopathology from a maladaptive trait perspective, incorporating both parental and child ratings of parenting as a moderator of the personality-psychopathology association. Hierarchical moderated regression analyses were conducted on a combined sample of referred and nonreferred children and adolescents (N=862, parental ratings of parenting and N=396, child ratings of parenting). The results indicated positive main effects of maladaptive traits on externalizing and internalizing problems, and positive main effects of parental negative control on externalizing problems. Significant interactions were found for Disagreeableness and Emotional Instability with parental Negative Control and for Disagreeableness x Positive Parenting in explaining externalizing problems. The discussion focuses on the contribution of these findings to a better understanding of the trait-psychopathology relationship at young age.     

Validity and Adverse Impact Potential of Predictor Composite Formation

Previous research on the validity and adverse impact (AI) of predictor composite formation focused on the merits of regression‐based or ad hoc composites. We argue for a broader focus. Ad hoc chosen composites are usually not Pareto‐optimal, whereas the regression‐based composite represents only one element from the total set of Pareto‐optimal composites and can, therefore, provide only limited information on the potential for validity and AI reduction of forming predictor composites when both validity and AI are of concern. In that case, other Pareto‐optimal composites may provide a better benchmark to decide on the merits of the predictor composite formation. We summarize a method to determine the set of Pareto‐optimal composites and apply the method to a representative collection of selection predictors. The application shows that the assessment of the AI and validity of predictor composite formation can differ substantially from the one arrived at when considering only regression‐based composites.     

A longitudinal and multi-method examination of interest–occupation congruence within and across time

Remarkably little research has addressed change and stability in person–environment fit across time. The objective of the current study was to address this gap in the literature by investigating Holland interest–occupation congruence across time for a sample of college alumni (N = 167) that were tracked during the first third of their professional career. Congruence was examined in all its complexity, including a repeated assessment of both occupations and interests, the use of objective (O*NET) and subjective (self-report) environment assessment methods, and adopting sophisticated congruence calculation methods (i.e., Euclidean distance and profile correlation). This resulted in a total of 12 interest–occupation comparisons within and across time that could be related to general and career specific well-being. The results first indicated moderate levels of stability in interests and occupations across the 15-year time interval, yet also with room for change. Congruence analyses indicated significant interest–occupation fit at the beginning of the career and 15 years later, with the magnitude of congruence slightly varying depending on which occupation assessment and which congruence method was used. Profile correlation congruence did not change over time in absolute terms and evidenced relative stability. Euclidean distance indices had little relative stability over time but did manifest some absolute changes in levels of congruence. Finally, job change moderated the association between interest–occupation congruence and life satisfaction in such a way that higher levels of satisfaction were reported only when little job change was present.     

Organizational career growth and subsequent voice behavior: The role of affective commitment and gender

Although research has established a connection between career growth and turnover intentions, there continues to be a need to study how employee career growth contributes positively to organizations. In the present research, we studied in particular how employees' organizational career growth is related to voice behavior. Employing theories of social exchange, organization-based self-esteem, and psychological attachment, we developed six hypotheses pertinent to this relationship, including the mediating role of affective organizational commitment and the moderating effect of gender. We tested our hypotheses using data from 328 employees in Mainland China over three time periods. As we hypothesized, we found positive relationships between the three dimensions of organizational career growth and subsequent voice behavior. Our results also verified that these relationships are partially mediated by affective organizational commitment and partially moderated by gender.     

A meta‐analysis of psychopathy‐, antisocial PD‐ and FFM associations

This research meta‐analytically summarizes the relationships of the Five‐Factor Model (FFM) with psychopathy and Antisocial Personality Disorder (APD). Effect sizes of the associations between psychopathy, APD and the FFM were compiled from 26 independent samples (N = 6913) for psychopathy and 57 independent samples (N = 16 424) for APD. The results revealed predominantly points of similarity and some differences in the FFM associations of both disorders. Symptoms of psychopathy and APD were negatively associated with Conscientiousness and Agreeableness facets and positively with scores on Angry–Hostility (N2), Impulsiveness (N5), Excitement Seeking (E5) and negatively with Warmth (E1). Only psychopathy had a small negative association with Anxiety (N1) and was characterized by stronger negative associations with Agreeableness and Straightforwardness (A2), Compliance (A4) and Modesty (A5) compared to APD. The moderator analyses showed that sample type, use of the NEO‐PI‐R and APD instrument moderated the APD FFM associations, while psychopathy instrument and age group were moderators in the psychopathy MA. Implications of this research for the assessment of APD and psychopathy relying on dimensional models of personality pathology are discussed. Copyright © 2009 John Wiley & Sons, Ltd.     

Temperament and Risk for Depressive Symptoms in Adolescence: Mediation by Rumination and Moderation by Effortful Control

The present study examined the relations between temperament, ruminative response style and depressive symptoms both cross-sectionally and prospectively (1 year follow-up) in a community sample of 304 seventh- through tenth-graders. First, higher levels of negative affectivity (NA), lower levels of positive affectivity (PA) and lower levels of effortful control (EC) were found to be associated with higher levels of depressive symptoms. Second, the association between NA and PA on the one hand and depressive symptoms on the other was significantly moderated by level of EC (low PA and high NA are associated with depressive symptoms only if EC is low) and these relations were moderated by sex in the cross-sectional data. In the prospective data, T1 depressive symptoms and PA predicted T2 depressive symptoms; with EC approaching significance. Third, rumination also predicted T1 as well as T2 depressive symptoms. Finally, support was found for a model of moderated mediation: higher levels of NA were associated with higher levels of ruminative response style, which was in turn related to more depressive symptoms but only in individuals with low EC and this was true for the cross-sectional as well as the prospective data albeit with noteworthy differences in pattern. These findings confirm and extend previous findings on the associations between temperament, response styles and depression in adolescence and, as such, add to the growing body of research providing support for the applicability of cognitive vulnerability theories to depression in younger populations.

Creating Alternate In‐Basket Forms Through Cloning: Some preliminary results

Research on constructing alternate forms of assessment center exercises is very scarce. This study examines the effectiveness of a cloning procedure (incident isomorphic approach) for developing alternate forms of a computerized in‐basket. In this approach, original and alternate items are essentially similar (they are based on the same critical incident), while being superficially different (they are situated in a different context). Results showed there was no significant difference between the overall in‐basket score across the alternate forms. In addition, these overall scores correlated .66, with projected estimates for the full in‐basket approaching .80. Implications and limitations of the use of cloning in designing alternate assessment center exercises are discussed.