Differential prediction and the use of multiple predictors: the omitted variables problem

Moderated regression is widely used to examine differential prediction by race or gender. When using multiple predictors in a selection system, guidance as to whether differential prediction analysis should be conducted on each predictor individually, or on the set of predictors in combination, is lacking. Analyzing predictors individually creates the possibility of an omitted variable problem. Army Project A data were used to examine differential prediction by race with the use of personality measures for 79 predictor-criterion combinations. Traditional analysis indicated predictive bias by intercept in 45 instances and by slope in 7 instances; the inclusion of an Armed Services Vocational Aptitude Battery general factor as an additional predictor changed the conclusion in 32 cases for the intercept and in 3 cases for the slope.     

METAANALYSES OF VALIDITY STUDIES PUBLISHED BETWEEN 1964 AND 1982 AND THE INVESTIGATION OF STUDY CHARACTERISTICS

Review and metaanalyses of published validation studies for the years 1964-1982 of Journal of Applied Psychology and Personnel Psychology were undertaken to examine the effect of (1) research design; (2) criterion used; (3) type of selection instrument used; (4) occupational group studies; and (5) predictor-criterion combination on the level of observed validity coefficients. Results indicate that concurrent validation designs produce validity coefficients roughly equivalent to those obtained in predictive validation designs and that both of these designs produce higher validity coefficients than does a predictive design which includes use of the selection instrument. Of the criteria examined, performance rating criteria generally produced lower validity coefficients than did the use of other more "objective" criteria. In comparing the validities of various types of predictors, it was found cognitive ability tests were not superior to other predictors such as assessment centers, work samples, and supervisory/peer evaluations as has been found in previous metaanalytic work. Personality measures were clearly less valid. Compared to previous validity generalization work, much unexplained variance in validity coefficients remained after corrections for differences in sample size. Finally, the studies reviewed were deficient for our purposes with respect to the data reported. Selection ratios, standard deviations, reliabilities, predictor and criterion intercorrelations were rarely and inconsistently reported. There are also many predictor-criterion relationships for which very few validation efforts have been undertaken.     

Complex mediating effects of rumination facets between personality traits and depressive symptoms

This study investigates whether facets of rumination statistically mediate the relationships between Big Five personality traits and depressive symptoms. Self-reported personality traits and rumination were investigated as predictors of depressive symptoms in a cross-sectional sample of 3043 participants aged 18–60 years (68.8% female). Multiple regression analysis investigated which personality traits and rumination facets best explained variance in depressive symptoms. Structural equation modelling was used to determine whether facets of rumination mediated the relationships between personality traits and depressive symptoms. Multiple regression analysis found that variance in depressive symptoms was best explained by the personality traits neuroticism, extroversion, conscientiousness; and both facets of rumination, brooding and reflection. Structural equation modelling added that the effects of neuroticism, extroversion, conscientiousness and openness on depressive symptoms were statistically mediated by brooding; the effects of neuroticism, extroversion and openness to depressive symptoms were statistically mediated by reflection. Rumination facets statistically mediated the effects of various personality traits on depressive symptoms. These results provide insights into which individuals may be best suited to treatments for depression targeting rumination.     

VALIDITY GENERALIZATION: A REVIEW AND CRITIQUE OF THE CORRELATION MODEL

This paper presents a review and critique of the validity generalization research based on the correlation model. Emphasis was placed on integrating the applied validity generalization studies as well as the computer simulation studies which have tested the accuracy of the various validity generalization procedures in estimating the mean and variance of true validity coefficients. In general, this review indicated that the validity generalization procedures are fairly accurate in estimating the mean and variance of true validities, that the preponderance (e.g., 90%) of corrected validity coefficients for a single predictor-criterion relationship, based on a series of cumulative studies, when placed in distributional form are positive, and that a substantial proportion of observed validity coefficient variation can be attributed to statistical artifacts with sampling error accounting for the majority of the artifactual variance. Moreover, potential problems with validity generalization/meta analysis based on criterion-related validity coefficients were discussed. It was concluded that validity general-ization/meta-analysis is an important development in summarizing cumulative research, however, caution should be exercised in utilizing and interpreting the findings of such analyses. Directions for future research aimed at establishing trait-performance relationships were discussed.     

Degradation of validity over time: a test and extension of Ackerman's model

The dynamic nature of the predictor-criterion relationship has long been a concern in psychology, especially with regard to the deterioration of validity over time. The authors examine P. L. Ackerman's ( 1987, 1988) hypothesized relationships between different types of predictors and criteria over time using data from previous longitudinal studies. Expert ratings categorized predictors and criteria according to P. L. Ackerman's model. Regression results support the predicted negative curvilinear relationship between cognitive ability and consistent and inconsistent task performance but do not support the predicted relationships between perceptual speed ability and psychomotor ability and consistent and inconsistent task performance. Deterioration of validity was more ubiquitous than has been suggested previously, and the pervasive form of deterioration was cubic with a negative trend. Findings are discussed in the context of catastrophe-chaos models.     

Sex-role attitudes: A comparison of sex-race groups

This study investigated the differences in the three dimensions of sex-role attitudes among four sex-race groups and also the ability of sex and race jointly to condition the effects of background characteristics on these attitudes. In a sample of 409 college students significant differences were found on all three dimensions of sex-role attitudes between the two sex groups, but not between the two racial groups, implying that gender is the most significant variable in eliciting differences in sex-role attitudes. Stepwise multiple regression analysis did not find a single common predictor to explain the variance in the wife role, the mother role, and the father role for the four sex-race group models. Some predictors were found in common in the four models explaining the variance in three dimensions of sex-role attitudes when the effects of sex and race were examined separately. However, the interaction of race and sex in conditioning the effect of background variables on the sex-role attitudes was quite different for each of the four sex-race subsamples.     

Gender differences in the psychosocial correlates of suicidal ideation among adolescents.

Gender differences in the psychosocial correlates of suicidal ideation were studied. A sample of 613 high school students (ages 14-19) completed measures of suicidal ideation, depression, hopelessness, life stress, loneliness, alcohol and drug use, and reasons for living. The results of a discriminant function analysis indicated that males reported higher loneliness and substance abuse scores than females whereas females reported greater suicidal ideation, depression, and reasons for living. The results of multiple regression analyses found that, although the same four variables, depression, hopelessness, substance abuse, and few reasons for living emerged as significant predictors of suicidal ideation in both samples, the predictive equation accounted for more of the variance in ideation scores in females (57%) than in males (46%). In a final analysis a discriminant function analysis of the subscales of the reasons for living inventory revealed that females have a greater fear of death and injury whereas males have a greater fear of social disapproval over having suicidal thoughts. This may account for the greater rate of suicide completing among males. Fear of social disapproval, more anger and impulsivity, and less help-seeking behavior among males are offered as potential variables to explain the observed gender differences.     

Personality and Cognitive Ability as Predictors of Typical and Maximum Managerial Performance

Despite long-established distinctions between typical and maximum performance variables on both the predictor and criterion side, little previous research has directly addressed the extent to which these distinctions translate into differential predictor-criterion relationships. Using a sample of candidates for managerial positions, we examined relations of predictors conceptually linked to typical (i.e., broad, narrow, and compound personality traits) and maximum (i.e., broad and narrow cognitive abilities) performance with corresponding criterion measures (N = 84–873). Supervisory ratings of managerial performance served as the typical performance criterion, whereas maximum performance was assessed via an assessment center. Confirmatory factor analyses supported the distinction between typical and maximum performance. Results also confirmed our hypothesis that cognitive abilities are more strongly correlated with maximum performance than with typical performance and largely supported the expectation of the opposite pattern with personality traits as predictors.     

The Relation Between Anger Coping Strategies,Anger Mood and Somatic Complaints in Children and Adolescents

Attempts to explain the experience of somatic complaints among children and adolescents suggest that they may in part result from the influence of particular strategies for coping with anger on the longevity of negative emotions. To explore these relationships British (n = 393) and Dutch (n = 299) children completed a modified version of the Behavioral Anger Response Questionnaire (BARQ), and two additional questionnaires assessing anger mood and somatic complaints. A hierarchical regression analysis showed that for both the UK and Dutch samples two coping styles, Social support-seeking and Rumination, made a significant contribution to somatic complaints, over and above the variance explained by anger mood. A tendency to repeatedly think or talk about an angering event as a way of coping seems to underlie the observed negative health effects. In addition, tentative support is given for a broader range of strategies to cope with anger than just the traditionally studied anger-out and anger-in styles.     

执行功能子成分与学业成就的关系

目的:探讨执行功能三个子成分(抑制功能、转换功能和信息刷新功能)与学业成就的关系。方法:采用认知任务对85名大学生的执行功能子成分进行测量,并进行执行功能子成分的Stroop效应、转换任务效应及信息刷新任务效应对学业成就的逐步回归分析。结果:抑制子功能的Stroop效应及信息刷新子功能的任务效应对学业成就具有负向预测作用,两者共同解释学业成就变异的15.1%。其中,抑制子功能的Stroop效应预测作用最大,可解释学业成就变异的8.5%。结论:执行功能抑制子功能的Stroop效应及信息刷新子功能的任务效应可作为学业成就的预测因子。     

Failure Avoidance Motivation in a Goal-Setting Situation

Scholars have consistently identified contextual performance or organizational citizenship behavior as a core component of job performance. The current literature on this issue has been dominated by a single-level approach, typically conducted at the individual level of analysis. This study adopts a multilevel approach to simultaneously examine main effects of and cross-level interactions among individual- and group-level predictors of interpersonal helping behavior. Results from a large-scale longitudinal data set show that at the individual level, helping behavior was predicted by perceived organizational support (POS), fairness, and affective commitment. At the group level, helping behavior was predicted by trust among group members. Trust among members also significantly moderated the individual-level relationships between POS and helping behavior and between fairness and helping. These crosslevel moderations indicated that the group- and individual-level predictors were complementary (instead of mutually reinforcing) in predicting interpersonal helping behavior. This finding indicates that various antecedents of interpersonal helping are characterized by distinct dynamics at the individual and group levels of analysis.     

Age-of-acquisition ratings for 30,000 English words

We present age-of-acquisition (AoA) ratings for 30,121 English content words (nouns, verbs, and adjectives). For data collection, this megastudy used the Web-based crowdsourcing technology offered by the Amazon Mechanical Turk. Our data indicate that the ratings collected in this way are as valid and reliable as those collected in laboratory conditions (the correlation between our ratings and those collected in the lab from U.S. students reached .93 for a subsample of 2,500 monosyllabic words). We also show that our AoA ratings explain a substantial percentage of the variance in the lexical-decision data of the English Lexicon Project, over and above the effects of log frequency, word length, and similarity to other words. This is true not only for the lemmas used in our rating study, but also for their inflected forms. We further discuss the relationships of AoA with other predictors of word recognition and illustrate the utility of AoA ratings for research on vocabulary growth.     

Client,Treatment, and Therapist Variables Related To Outcome in Brief,Systems-Oriented Family Therapy

Client, therapist, and treatment characteristics were examined with respect to how much of the variance they could account for in a variety of outcome measures. Multiple regression analyses were used to examine relationships among each of the client, therapist, and treatment characteristics studied and the various outcome measures. For the 219 families that were treated with brief family therapy, only a relatively low amount of variance in any of the outcome measures could be accounted for. The amount of explained variance, however, varied considerably from one outcome measure to the other. When the more homogeneous groups of clients (single parents or adolescent identified patients) were considered, the amount of explained variance generally showed an increase. Different sets of client, therapist, and treatment variables accounted for the variance in these outcome measures across client groups, demonstrating both the complexities of the relationships and the relative independence of various outcome measures. New variables are suggested for future research.     

Self-monitoring and sex as moderator variables in the statistical explanation of self-reported marijuana and alcohol use

Self-monitoring and sex are examined as likely moderators of some of the known effective predictors of self-reported use of marijuana and alcohol. The theory of self-monitoring implies that in comparisons between low and high self-monitors, dispositional predictors account for more of the variance in reported use among low self-monitors and environmental predictors account for more of the variance among high self-monitors. With the two classes of predictors established via principal-components analysis and with subgroups defined by scores on Snyder's Self-Monitoring Scale, data from 807 college students showed that self-monitoring does not affect the action of either dispositional or environmental predictors. Thus, the self-monitoring construct promises explanatory leverage that in these tests, the scale is not able to deliver. When subgroups are defined by sex, analyses of reported alcohol use yield small but significant moderator effects: Women's use is more predictable from environmental variables, whereas men's use is more predictable from dispositional variables. These effects are consistent with the conclusions of recent meta-analyses of results from studies of sex differences in conformity to group pressures.     

Consumer Masculinity Behavior Among College Students and Its Relationship to Self Esteem and Media Significance

The purpose of the study was to investigate male college students in the southeastern United States to identify factors associated with consumer masculinity behavior. The variables of self-esteem and media significance were tested as possible predictors of consumer masculinity behavior. A total of 219 surveys were used as the sample for this study. Pearson correlation analysis tested relationships among the variables, and linear regression was used to further test the nature of the relationships identified. Results indicated that media significance and appearance-related self-esteem were significant predictors of consumer masculinity behavior. Limitations and implications are discussed.     

Predictors of organizational commitment: The study of part-time army reservists

A predictive model built upon a multiple role conflict/felt responsibility conceptual framework and containing six previously untested predictors and nine previously tested organizational-related predictors of organizational commitment was investigated. The sample consisted of two battalions of U.S. Army Reserve members. Stepwise multiple regression analysis utilizing a double cross-validation design on each battalion was the data analytic technique. Increased job satisfaction and stronger intent to stay consistently entered each equation as the first and second predictors, respectively. Group cohesion also appeared in more than one equation. The variance explained across the four cross-validated samples averaged .46 and the results were very stable. The importance of the investigation of commitment for this part-time, voluntary organization was discussed, as were future research directions regarding organizational commitment.     

效度概化：预测效度元分析的30年成果述评

效度概化是通过元分析技术对普遍化的预测效度的估计。元分析是研究效度概化问题的技术,是对具有“预测因子-效标”特征的相关数据进行定量综合的方法。效度概化促进了预测因子与效标之间关系的理论研究和应用研究,是应用心理学领域近30年（1977~2007）来最重要的进展之一。30年来的效度概化研究表明,认知能力测验、知识和技能测验、人格测验、结构化面试和评价中心技术等的预测效度具有鲁棒性、对应性和联合增值性     

An Empirical Comparison of Commitment Scales

In a previous review of the literature, Morrow (1983) identified a redundancy or overlap among the various measures of commitment. The present study employed a factor analytic design to examine the intercorrelations among various job, organizational, and value commitment scales. In addition, possible consequences of the various scales were analyzed. Results from 216 police officers identified a general factor as well as unique components. The commitment scales were able to explain between 20% and 40% of the variance of several self-report measures. Explanations and implications of the results were discussed.     

Self-Efficacy,Perceptions of Success,and Intrinsic Motivation for Exercise1

A number of investigators have attempted to identify the determinants of continued participation in exercise regimens. Within a social cognitive framework, the present study examined the relationships among self-efficacy cognitions, perceptions of success, and multidimensional intrinsic motivation in the exercise domain. Multivariate analyses of variance indicated that highly efficacious individuals were more intrinsically motivated toward aerobic dance than were their less efficacious counterparts. However, multiple regression analyses revealed both perceptions of success and self-efficacy to account for significant variance in intrinsic motivation at both the composite and dimensional levels. Moreover, perceptions of success were considerably stronger predictors than were efficacy cognitions. The results are discussed in terms of the need for multiple measures of perceived success, the employment of longitudinal designs to determine the direction of causality among these variables, and possible implications for adherence issues.     

Workplace discrimination: A meta‐analytic extension,critique, and future research agenda

Despite a large and growing literature on workplace discrimination, there has been a myopic focus on the direct relationships between discrimination and a common set of outcomes. The aim of this meta‐analytic review was both to challenge and advance current understanding of workplace discrimination and its associations with outcomes by identifying the pathways through which discrimination affects outcomes, examining boundary conditions to explain when discrimination is most harmful for employees, and exploring a potential third variable explanation for discrimination–outcome relationships. Mediation tests indicated that workplace discrimination is associated with employee outcomes through both job stress and justice. Moderator analyses showed that discrimination appears to be most detrimental when it is observed rather than personally experienced, interpersonal rather than formal, and measured broadly rather than specifically. We also found that discrimination–outcome relationships differ across work and nonwork contexts and as a function of the social identity targeted by discrimination. Discrimination generally explained meaningful incremental variance in outcomes after controlling for the effects of negative affectivity, but the relationships between discrimination and health were substantially decreased. We conclude by offering a constructive critique of the empirical discrimination literature and by detailing an agenda for future research.