Abstract
Educational tests are used extensively for predicting future performance of individuals. Multiple regression is the most commonly used approach for this purpose, but occasionally it is desirable to use multiple cutoff scores. This paper provides Taylor-Russell type tables for determining the theoretical proportion of successful students in a selected subpopulation as a function of: the proportion successful in the unselected population; various intercorrelations; and the percentage cut on each of the two variables.
Keywords
Get full access to this article
View all access options for this article.
References
1.
2.
3.
4.