Degree Type

Dissertation

Date of Award

1986

Degree Name

Doctor of Philosophy

Department

Psychology

Abstract

An important issue in personnel research concerns the assessment of the efficacy of such human resources functions as selection, training, and performance appraisal. Traditionally used measures such as correlation coefficients and subjective reactions have been heavily criticized for their inadequacy in accurately describing the value of a given human resources function. The regression-based general utility function from decision theory, proposed as an alternative, represents a major improvement, as it is intended to allow assessment of virtually any personnel function in monetary terms. It is also intended to account for such factors as the number of employees involved, the costs of the personnel function, the validity of the personnel function, and the variability of job performance. Two important variables of the utility function--SD(,y), the standard deviation of performance, and (phi)/p, the selection ratio--were examined in terms of their sensitivity to departures from underlying mathematical assumptions. Because job performance is assumed to be normally distributed, the effects of progressive truncation of a standard normal distribution were assessed. It was found that this term is relatively robust, unless the truncation was extreme. The robustness of the (phi)/p term was assessed by replacing the normal distribution of predictor performance with a reflected gamma distribution. It was found that this term is more affected by a departure from normality. The combined effects of these two terms were assessed by simultaneously examining the effects of varying selection standards and varying job performance on the expected per-person increase in standardized job performance. It was found that such effects were nonsymmetrical and depended on base rates of job performance. Implications for the implementation of the function and directions for further research are discussed.

DOI

https://doi.org/10.31274/rtd-180813-6811

Publisher

Digital Repository @ Iowa State University, http://lib.dr.iastate.edu/

Copyright Owner

Sue Margaret Anderson

Language

en

Proquest ID

AAI8703683

File Format

application/pdf

File Size

103 pages

Share

COinS