UTIL 2.1 is a free downloadable Excel workbook with an interactive display representing an implementation of Berger and Tsujimoto’s (1983) classification program.
Berger, D. E., & Tsujimoto, R. N. (1983). UTIL: A BASIC program for determining optimal cutting scores adjusted for base rates and classification utility. Educational and Psychological Measurement, 43, 173‑175.
Citation: Berger, Dale E. (2018). Maximizing Utility in Group Classification with UTIL2.1. Retrieved from https://wise.cgu.edu
Input Parameters
UTIL allows users to input the following parameter values based upon either observed or theoretically expected values:
- Population means for two groups
- Population standard deviations for both groups
- Population base rates for the two groups
- Utility values of each of the four possible outcomes of classification decisions:
-
- Classification of a Group 1 case into Group 1 (correct decision)
- Classification of a Group 1 case into Group 2 (incorrect decision)
- Classification of a Group 2 case into Group 1 (incorrect decision)
- Classification of a Group 2 case into Group 2 (correct decision)
Classification Statistics
- The interactive program displays the two population distributions and the distributions weighted by base rates and utility of the four possible decision outcomes.
- The program also displays the cut scores and decision rule for maximizing expected utility. The decision rule states whether there are zero, one, or two cut points, as well as how the classification groups are defined by the cut point boundaries.
- Statistical results include the predicted number and proportion of correct classifications for cases from Group 1 and Group 2, and the relative utility of classification based on the computed cut points.
Important Assumptions
- The populations of test scores are assumed to be normally distributed. Transformations may be useful to obtain normal distributions.
- Accuracy of classification depends on accuracy of the parameters provided by the user.
Questions or comments? Contact: Dale Berger