Difsim, a BASIC and FORTRAN program for computing and determining the statistical significance of difference and similarity indexes computed on several independent or dependent samples of ratings, is described. The procedures employed in the program are exact probability analogs of the randomized groups and repeated measures analyses of variance for small samples.
Get full access to this article
View all access options for this article.
References
1.
Aiken, L. R. and Aiken, T. A. (1986). Difference tests for distributions of ratings. Educational And Psychological Measurement, 46, 871-881.
