Honesty Testing for Personnel Selection: A Quantitative Review
MICHAEL A. MCDANIEL AND JOHN W. JONES
There has been an increasing interest in the use of paper-and-pencil measures designed to identify dishonest job applicants ( Sackett & Harris, 1984). The purpose of this chapter is to summarize the validity evidence for a published dishonesty measure, the Dishonesty subscale of the Personnel Selection Inventory (PSI) (London House, 1980). Terris ( 1985) has provided a narrative review of the validity of this scale. The research examined here employed meta-analysis methods ( Hunter, Schmidt & Jackson, 1982) to summarize the validity evidence and to test the hypothesis that the validity covaries with a severity-of-consequences moderator. Specifically, we hypothesize that respondents who anticipate adverse consequences following accurate reporting of theft-related attitudes will be less likely than others to respond accurately and that the correlation between the predictor and the criterion will be lower.
Validity studies involving the Dishonesty scale were obtained from its publisher. The senior research psychologist of the test publisher stated that the authors were given a complete set of all validity studies on the scale (personal communication, W. Terris, 1986). Thus, reporting bias ( Schmidt, Hunter, Pearlman, & Hirsch, 1985) does not appear to be a problem in the research.
Hirsh and McDaniel ( 1986) have recommended that meta-analytic investigators explicitly state the reasons some research is excluded from the analysis