VOLUNTARY AFFIRMATIVE ACTION AND PREFERENTIAL TREATMENT: LEGAL AND RESEARCH IMPLICATIONS

This paper provides a review of recent Supreme Court decisions stemming from the implementation of voluntary AAPs that include preferential treatment as a remedy; the evidence and proof required to establish and defend against reverse discrimination claims is identified. Overall, the courts sanctioned the use of preferential treatment if it occured as part of a well-conceived AAP. However, differences were found in the manner in which the courts adjudicated Title VII and constitutional claims of preferential treatment. The discussion examines research on the potential effects of voluntary AAPs on important organizational and societal outcome variables. We conclude that the impact of AAPs is likely to be far-reaching, exerting an effect possibly well beyond the limits considered in court decisions.     

ON THE NEED FOR QUALITY CONTROL IN VALIDATION RESEARCH

The aptitude tests used to help make personnel decisions about military recruits were validated against hands-on tests of job performance in two Marine Corps occupational specialties, radio repairers and automotive mechanics. The tests were administered by Marine Corps noncommissioned officers. Marine Corps units provided the test administrators, testing facilities, and examinees. Data collected under such conditions are filled with errors that reduce the accuracy of validity coefficients. This paper shows how validity coefficients can be made more accurate by exercising quality control during the statistical analysis.     

ISSUES CONCERNING THE USE OF REGRESSION ANALYSIS IN SALARY DISCRIMINATION CASES

Twenty-nine salary discrimination court cases were examined to determine ways an organization can refute a regression analysis that leads to the inference of discrimination in compensation. It was found to be equally effective for the organization to introduce its own regression, other statistics, or no statistics. The plaintiff won all of the cases when discrimination was proven in promotion. The defendant won all cases when discrimination was not proven in promotion. Defendants also won all cases when plaintiffs inappropriately treated jobs as fungible or when they failed to include important variables. Market variables may be included in regressions if defendants can show they were applied consistently to determine salaries. The implications of the courts' acceptance of some regressions for comparable worth is presented, and a list of recommendations is made for organizations that may be faced with results from regression analysis in court.     

SOME NEW FRONTIERS IN PERSONNEL SELECTION RESEARCH

Current research developments in several areas of personnel selection methodology are reviewed. The areas reviewed represent attempts to develop alternatives to traditional paper-and-pencil methods of test development. These areas include computer-interactive tests in the employment situation, new ways of using life-history information, the use of physical tests in selecting individuals for physically demanding jobs, and the development of job-related medical standards for job assignment.     

A SURVEY OF MANAGEMENT TRAINING AND EDUCATION PRACTICES IN U.S. COMPANIES

A survey was systematically developed to assess issues related to management training and education, and it was sent to 1,000 randomly selected U.S. companies with at least 1,000 employees. Based on a 61% response rate, findings were obtained regarding issues such as needs assessment, management training and development approaches, reasons for selecting particular programs, characteristics of participants, how decisions are made regarding who will participate, preparation and follow-up of participants, evaluation of management training programs, future management training trends, and needed training content. Significant findings by company size and industry type also are reported.     

THE RELATIONSHIP BETWEEN PAY-FOR-PERFORMANCE PERCEPTIONS AND PAY SATISFACTION

A study was conducted with 104 hospital employees to assess the relationship between pay-for-performance perceptions and pay satisfaction. Unlike previous research examining this relationship, a multi-item measure of pay-for-performance perceptions and a multidimensional measure of pay satisfaction were used. As hypothesized, the results indicated a positive relationship between pay-for-performance perceptions and pay-raise satisfaction, pay-level satisfaction, and overall pay satisfaction even after the effects of salary level, salary increases, performance ratings, job tenure, job satisfaction, and promotions were controlled. The importance of gathering perceptual data on characteristics of the pay system believed to have differential relationships with the subdimensions of pay satisfaction are discussed.     

Heart-rate response to water-immersion isolation as a function of spatial orientation

The present study investigates the effect of spatial orientation of the body on heart rate during conditions of water-immersion sensory isolation. Two groups of three men and three women each were exposed to one of two isolation conditions. In one condition the subject was suspended in a buoyant but vertical attitude; in a second condition the subject was suspended in a buoyant but horizontal attitude. Heart-rate data were obtained continuously during each session. There was no significant difference in mean heart rate between the two conditions. The data were interpreted as providing support for the position that studies comparing recumbent subjects in air isolation with underwater subjects in a vertical suspension attitude are valid, at least for some measures, and are not an artifact of spatial bodily orientation. The possible importance of neutral buoyancy in accounting for the nonsignificant effect of body attitude on heart rate is discussed.     

Aggregate item response analysis

A stochastic postulate is given for the multiple-item, successive-intervals scaling of populations. The logistic equivalent of this postulate provides an aggregate item response model in which a unidimensional submodel may be nested. This reduction provides a subtractive conjoint measurement of several items and stimuli on the same latent scale. Generalized-least-squares methods are used to estimate and test the multiple-item model, and its unidimensional reduction, on aggregate survey responses. The entire procedure is illustrated with an analysis of semantic-differential attitude data. This analysis exhibits an item selection procedure that is applicable to various social constructs.The authors dedicate this paper to the memory and contributions of Clyde Coombs.The programming and data analyses for the present paper were carried out by José Ventura of the Department of Industrial and Systems Engineering, and Jerry Meiten of the Department of Statistics, University of Florida.The study was also supported by the College of Business Administration, University of Florida, and the Faculty of Social Sciences, Hebrew University of Jerusalem.     

Test theory without an answer key

A general model is presented for homogeneous, dichotomous items when the answer key is not known a priori. The model is structurally related to the two-class latent structure model with the roles of respondents and items interchanged. For very small sets of respondents, iterative maximum likelihood estimates of the parameters can be obtained by existing methods. For other situations, new estimation methods are developed and assessed with Monte Carlo data. The answer key can be accurately reconstructed with relatively small sets of respondents. The model is useful when a researcher wants to study objectively the knowledge possessed by members of a culturally coherent group that the researcher is not a member of.This research was supported by NSF Grant No. SES-8320173 to the authors. We gratefully acknowledge comments and suggestions from John Boyd, Tarow Indow, and Kathy Maher as well as the editor and several anonymous referees.     

Sample size determinations for the two rater kappa statistic

This paper gives a method for determining a sample size that will achieve a prespecified bound on confidence interval width for the interrater agreement measure,. The same results can be used when a prespecified power is desired for testing hypotheses about the value of kappa. An example from the literature is used to illustrate the methods proposed here.