An Empirical Investigation of Matrix Sampling Involving Multiple Item Samples in a Two-Factor Analysis of Variance Design
James Archie Newell · University of North Texas Digital Library (University of North Texas) · 1971
The primary purposes of this study were: (1) to study empirically differences that might occur among item-samples and (2) to compare empirically the effect of test item samples on matrix sampling estimates of the mean and variance of a population of test scores, and (3) to study empirically an analysis of variance design through multiple matrix sampling.