ON TESTING FOR STOCHASTIC DISSOCIATIONS

Methods for examining stochastic relationships have been proposed as p owerful ways to dissociate different underlying psychological processe s, but a number of problems have undermined conclusions based upon the se methods. These testing methods and criticisms are reviewed, and the statistical methods for the testing of stochastic dependence are exam ined using computer simulation. With each of the methods examined, the re were difficulties dealing with certain situations that are likely t o occur in experiments examining dependence between successive tests. Examination also showed that the sample sizes of some previous studies were insufficient for findings of moderate amounts of dependence, cal ling some conclusions of stochastic independence into question. The re sults of the studies presented here suggest that testing for statistic al dependence is a statistically perilous technique, but they also sug gest several ways in which dedicated users of this form of testing can strengthen its application.