Special Section on Theoretical Models and Conceptual Analyses: Second Installment.

The Journal of Applied Psychology's call for theoretical models and conceptual analyses brought a terrific response. The first set of articles accepted in response to the call appeared in the December 2004 issue. This installment contains the second set of articles. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Further exploration of typical and maximum performance criteria: Definitional issues, prediction, and White-Black differences.

Addresses differences between typical and maximum performance criteria with a threefold purpose. First, the growing recognition that it is important to distinguish between typical and maximum performance criteria is examined. Second, clarification of definitional and measurement issues are addressed. Third, results of an empirical study are offered to contribute to the new knowledge base regarding typical and maximum performance criteria. The study investigates differences in validities of cognitive ability predictor measures against supermarket cashier typical and maximum performance criteria reflecting speed and accuracy. White–Black mean score differences on typical and maximum performance criteria are also examined. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Relations between measures of typical and maximum job performance.

In this study we proposed that understanding of interrelations among criterion measures can be improved by differentiating between measures of typical and maximum job performance. We offered defining characteristics of typical and maximum performance measures and requirements for a fair examination of relations between the two. Typical and maximum performance measures of the speed and accuracy with which supermarket cashiers process items were obtained from two large samples (Ns?=?635 and 735) of cashiers from 12 supermarket chains. Relatively low correlations between typical and maximum performance measures were found. We discuss the implications of these findings for the use of maximum performance measures as criteria in the validation of selection systems and for the understanding of the impact of ability and motivation on job performance. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Identification of moderator variables by discriminant analysis in a multipredictable group validation model.

Used a multivariate discriminant analysis procedure to identify moderator variables in a multipredictable group validation (mpgv) model. Results from mpgv were compared to those of the univariate simple algebraic and absolute difference techniques. 17 variables, determined for 418 undergraduates (assigned equally to an experimental and a cross-validation group), were investigated as potential moderators of the relationship between the predictor and the criterion (gpa). The 3 techniques revealed moderators different from each other. In addition, number and types of moderators varied according to the composition of the prediction groups in mpgv. Results are discussed with respect to problems with moderator variable techniques. (27 ref.) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Comparison of "joint moderators" in three prediction techniques.

Examined the existence of 2 potential "joint moderator" variables (anxiety and study habits), and compared their operation according to 3 different prediction techniques: (a) subgroup analysis, (b) differential prediction of predictability, and (c) moderated regression. Results with 393 undergraduates indicate that anxiety moderated the relationship between GPA and American College Test scores using subgroup analysis and differential predictability techniques. A comparison of moderated regression with linear regression indicated that study habits operated as an independent predictor. None of the techniques identified joint moderators in this situation. Results are discussed in the light of problems and assumptions of each technique. The need for systematic identification of moderators from a large mass of data is indicated. (17 ref.) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Validity and fairness of physical ability tests for predicting performance in craft jobs.

Problems relating to performance, accidents, and turnover in outdoor telephone craft jobs stimulated 2 experiments aimed at developing and validating a physical test battery. Based on job analysis results, a battery of 9 measures was administered to a sample of 128 Ss (83 males and 45 females) in Exp I. A 2-test battery (dynamic arm strength and reaction time), valid for predicting job task performance and turnover, was selected. Regression equations for males and females were not significantly different. Exp II included a sample of 210 Ss (132 males and 78 females). A 3-test battery consisting of a body density measure, a balance test, and a static strength test was selected based on relationships with training performance. No significant differences were found in the regression equations for males compared to females. The Exp II battery was also significantly related to field performance, training completion, and accidents and was valid for the Exp I criteria. (18 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Behavioral expectations: Development of parallel forms and analysis of scale assumptions.

Describes the development of parallel forms of behavioral expectation scales (BES) as well as a check on the adequacy of examples contributed as documentation for BES numerical evaluations. Data generated by O. Harari and S. Zedeck (see record 1974-11883-001) were used to develop 2 forms of BES, A and B. Results from 95 undergraduates indicate that the forms had, in general, equivalent dimension means and variances as well as the same degree of correlation with other variables (overall evaluation and satisfaction with the instructor). An independent sample of 56 students assessed 39 contributed examples provided to support actual evaluations. Results indicate high agreement between the original value and the independent assessments. The flexibility and increased use of the BES data are discussed. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Development of behaviorally anchored rating scales as a function of organizational level.

2 levels in a large medical center-8 head nurses (supervisory level) and 9 registered nurses (subordinate level)-independently participated in the development of behaviorally anchored rating scales. Although the 2 organizational levels identified similar behavioral dimensions, the behavioral incidents defining the dimensions were valued differently. Alternative explanations for the obtained differences are offered, and a discussion of behavioral expectation scale methodology is presented. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

The scalability of behavioral expectation scales as a function of developmental criteria.

Conducted a scalogram analysis of the items on behavioral expectation scales (BES) to test the assumption that such scales are unidimensional and cumulative. The effects of varying developmental criteria on the scalability of BES were also examined. BES items developed for faculty evaluation were randomly ordered on a checklist and presented to 48 undergraduates. The reproducibility coefficients of the original scales ranged from .65 to .81. Analyses indicated that higher coefficients resulted from a free ordering of items or elimination of items on the basis of applying a more stringent standard deviation criterion. Reallocation percentage agreement did not affect reproducibility. Scalogram analysis is suggested as an addition to BES developmental procedures. (15 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)