A simplex process model for describing differences between cross-lagged correlations.

Presents a model that is based on the use of the diagonal method of factoring to describe a simplex process and that explains differences between cross-lagged correlations. When the model is applied to the data of R. Atkin et al (see record 1978-21951-001), which showed that Listening tapped most directly the causes of later intellectual development, quite good fits are obtained for both 2- and 4-yr lags; 4 groups were studied on 4 occasions, at Grades 5, 7, 9, and 11. The best fit, however, is based on an estimate of the effects of a 3-yr lag between changes in the rank order of individual differences on the Listening test and the same changes on the intellectual composite. It is shown that accurate knowledge of reliabilities and specificities of the measures is necessary for interpreting cross-lagged differences but that stationarity per se is not an essential assumption. (9 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)