A limitation of the prediction analysis of 2 × 2 contingency tables is pointed out. Specifically, when competing predictions are being compared, a direct comparison of selected cell (or marginal) frequencies gives' virtually the same information as that provided by prediction analysis. This limitation does not apply in the more general case of R × C cross-classifications.
References
1.
FromanT.HubertL. J.Application of prediction analysis to developmental priority. Psychological Bulletin, 1980, 87, 136–146.
2.
HildebrandD. K.LaingJ. D.RosenthalH.Prediction analysis of cross classifications. New York: Wiley, 1977.
