Dependence of the Relative Productivity Gains of Two Personnel Selection Tests on the Applicant Pool Size

Abstract

(1980), Gross and Su (1975), Petersen (1976), Huynh (1976, 1977), Petersen and Novick (1976), and Novick and Petersen (1976) have proposed &dquo;threshold utility&dquo; models, in which utility is viewed as a two-valued function of criterion performance. Lindley (1978, 1979) have described what they consider to be &dquo;more realistic forms that utility functions might take in selected ... applications&dquo; (p. 181). Their model defined utility for members of a group in terms of a normal ogive whose parameters are specified by decision makers (administrators) for that group. They outlined advantages of their normal-distribution-based utility function over the &dquo;threshold utility&dquo; functions.The Brogden-Cronbach-Gleser Model Another model in which the utility function is associated with a normal distribution function was developed by Cronbach and Gleser ( 1965) from an earlier model of Brodgen ( 1946, 1949). This model allows estimation of the utility, in dollars, of personnel selection procedures. Recently the BrogdenCronbach-Gleser model has received much attention (see Bobko,