On the Use of a Cumulative Distribution as a Utility Function in Educational or Employment Selection.
Abstract
Formal decision theory can make important contributions to educational or employment decision-making, provided one can quantify the utilities of different possible outcome such as test scores, grade-point averages or other common outcome variables. Utility is usually a monotonic increasing function of true ability or performance score. A cumulative probability function is then very convenient for describing one's utilities. Moreover, calculations of expected utility of a decision is greatly simplified when the utility and the probability function have the same functional form, e.g. both normal. A least-squares procedure for fitting a utility function is described and applied to truncated normal and beta distribution functions. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Feb 01, 1981
- Accession Number
- ADA097465
Entities
People
- James J. Chen
- Melvin R. Novick
Organizations
- University of Iowa