UvA

Reports an error in the original article by Jeroen G. W. Raaijmakers, Joseph M. C. Schrijnemakers, Frans Gremmen (Journal of Memory and Language, 1999[Oct], Vol 41[3], 416-426). Due to a programming error, the degrees of freedom for the error term of the minF' statistics are incorrect. The corrected form is given in the current article. (The following abstract of this article originally appeared in record 1999-01034-006). Although H. H. Clark's (see record 1974-05129-001) critique of statistical procedures in language and memory studies (the "language-as-fixed-effect fallacy") has had a profound effect on the way such analyses have been carried out in the past 20 yrs, it seems that the exact nature of the problem and the proposed solution have not been understood very well. In this article, the authors review Clark's solution and show that the F-sub-1 x F-sub-2 criterion, although widely used, leads to positive bias. Next, the authors discuss alternatives to the minF' approach and consider the effects of commonly used variations in the exact nature of the design (such as matching of items and counterbalancing of lists) that affect the outcome of the analysis. The authors attempt to show that it is necessary to take the details of the experimental design into account before deciding on the particular ANOVA to be preformed.