In pretest-posttest designs, residual scores created in the analysis of covariance commonly are preferred to difference scores because the latter suffer from reliability problems. This paper develops the reliability of residuals adjusted for a single covariate for comparison to the reliability of difference scores. Comparisons reveal that little is gained by analysis of covariance, which also suffers from reliability problems as the pretest and posttest reliabilities decrease and as their intercorrelation increases.
Get full access to this article
View all access options for this article.
References
1.
Cronbach, L. J. and Furby, L. (1970). Measuring "change."Psychological Bulletin, 71, 68-80.
2.
O'Connor, E. F. (1974). Extending classical test theory to the measurement of change. Review of Educational Research, 42, 73-92.
3.
Thorndike, E. L. (1924). The influence of chance imperfections of measures upon the relation of initial score to gain or loss. Journal of Experimental Psychology, 7, 225-232.
4.
Webster, H. and Bereiter, C. (1963). The reliability of changes measured by mental test scores. In C. W. Harris (Ed.), Problems in measuring change (pp. 39-59). Madison, WI: University of Wisconsin Press.
