ERIC Number: ED393940
Record Type: RIE
Publication Date: 1995-Nov-9
Reference Count: N/A
Unequal Cell Size ANOVA and the Meaning of the Interaction.
Rakow, Ernest A.
Analysis of variance (ANOVA) is a frequently used statistical procedure in education and the social sciences. Very often the use of ANOVA involves situations with unequal cell sizes. When confronted with data to analyze from an unbalanced design, the researcher should select very carefully from the method or option in the statistical package being used for estimation of the sums of squares. As is illustrated, conflicting results may occur depending on the options chosen in the analysis. One approach for doing ANOVA is the use of unweighted means analysis. Interaction effects in such analysis are explored. Authors who have most clearly explained the alternatives for analyzing designs with unequal cell sizes have done so by emphasizing comparisons of models through the use of sums of squared errors for competing models, and they have also tended to use regression methodology to explicate the various models for nonorthogonal designs. Some of their points are reviewed through examples using the Statistical Analysis System and the Statistical Package for the Social Sciences. (Contains 8 tables and 13 references.) (SLD)
Publication Type: Reports - Evaluative; Speeches/Meeting Papers
Education Level: N/A
Authoring Institution: N/A
Identifiers: Statistical Analysis System; Statistical Package for the Social Sciences; Unequal Cell Size; Weighting (Statistical)
Note: Paper presented at the Annual Meeting of the Mid-South Educational Research Association (24th, Biloxi, MS, November 1995).