Key points are not available for this paper at this time.
The recent significant work of Schmidt and Hunter and their colleagues, and of Callender and Osbura, has resulted in several procedures for testing whether validity is generahzable Monte Carlo studies based on these procedures have thus far yielded good support, however, the degree of accuracy of these procedures in estimating the appropriate population parameters still has room for improvement In this article, two new procedures, which are based on less restrictive assumptions than those used by the previous investigators, are proposed for studying validity generalization Several Monte Carlo studies were earned out to test the accuracy of the new procedures in estimating population true validity mean and vanance and to compare these estimates with those of the other currently available procedures In general, the results indicated that one of the new procedures provided slightly more accurate estimates of the mean and vanance of population true validity than the procedures of Schmidt, Hunter, and their colleagues or of Callender and Osburn From a practical point of view, however, the estimates from the vanous procedures are quite comparable
Raju et al. (Sat,) studied this question.