NotesFAQContact Us
Search Tips
ERIC Number: ED074085
Record Type: RIE
Publication Date: 1972
Pages: 28
Abstractor: N/A
Reference Count: 0
Unequal Cell Frequencies in Analysis of Variance: A Review and Extension of Methodology for Multiple Missing Observations.
Proger, Barton B.; And Others
Many researchers assume that unequal cell frequencies in analysis of variance (ANOVA) designs result from poor planning. However, there are several valid reasons why one might have to analyze an unequal-n data matrix. The present study reviewed four categories of methods for treating unequal-n matrices by ANOVA: (a) unaltered data (least-squares solution and unweighted means solution); (b) data substitution (grand mean method, cell mean method, Winer method, Snedecor-Cochran method); (c) data deletion, and (d) data clustering (unreplicated cell mean method, unreplicated random data clustering method, replicated random data clustering method). The methods were compared empirically and theoretical problems with each were discussed. (Author)
Publication Type: N/A
Education Level: N/A
Audience: N/A
Language: N/A
Sponsor: Research and Information Services for Education, King of Prussia, PA.; Montgomery County Intermediate Unit 23, Blue Bell, PA.; Pennsylvania Resources and Information Center for Special Education (PRISE), King of Prussia, PA.
Authoring Institution: N/A