The science achievement scores of a large sample of 13-year-olds were regressed to test a psychological theory of educational productivity, and to explore the usefulness of the National Assessment of Educational Progress data for secondary analysis for policy purposes. (Author/GK)

Eight of the hypotheses supported by the Harvard data were again supported in the Montreal sample, though only two of the replications reached statistical significance. (Authors)

See EJ 030 570. (CK)

The hypothesis that the correlation of measures on the same characteristic at two points in time will approach unity when the relevant, intervening environment is added to the prediction equation is tested on a sample of high school physics students. (Author/AG)

Similarities between regression analysis and analysis of variance are noted and it is argued that the former has advantages over the latter. It is also argued that canonical correlation analysis is more suitable than factor analysis in certain cases. The argument is illustrated with four recent pieces of educational research. (DG)

Cut-and-try techniques that point to appropriate transformations of variables and to the selection of sets of variables for an equation that may improve understanding of a social process are illustrated. (MV)