Section Onemanagers for a specific job whose productivity is higher than
those hired without the test. The calculations of utility might
involve several variables. For example, validity of the selection
test would be a critical variable, in that it provides an indication
of the predictive ability of the test. Additionally, the increased
production, its contribution to profitability, and the standard
deviation of the contribution, would be variables in the
calculations. Finally, other variables might be included in the
analysis, such as the cost of testing enough applicants to obtain
a sufficient number having scores above the cut-off point.^15
Brian Becker and Mark Huselid’s study in a national
retailing company provides another example of an application
of utility theory. Becker and Huselid’s analysis explained return
on sales for each store on the basis of the performance
appraisals of the store supervisors. Their statistical analysis also
controlled for differences in the supervisors’ educational levels
and their commitment to the company. Their study
demonstrated that better estimates of the standard deviation of
the performance appraisal variable could be obtained through a
model based on the use of accounting data (return on sales)
rather than the more commonly used subjective approaches.
This study helps to enhance the legitimacy of utility theory for
applications in real business environments.^16