NotesFAQContact Us
Search Tips
ERIC Number: ED314483
Record Type: Non-Journal
Publication Date: 1989-Mar
Pages: 31
Abstractor: N/A
Reference Count: N/A
Commonality Analysis with Multivariate Data Sets.
Daniel, Larry G.
Commonality analysis may be used as an adjunct to general linear methods as a means of determining the degree of predictive ability shared by two or more independent variables. For each independent variable, commonality analysis indicates how much of the variance of the dependent variable is unique to the predictor and how much of the predictor's explanatory power is common to or also available from one or more of the other predictor variables. Commonality analysis is particularly useful in social science research involving multivariate data sets with at least one predictor at the interval level of scale, since, unlike many analyses of variance techniques, it does not require that all the independent variables be converted to the nominal level of scale. A small, hypothetical data set is presented to illustrate the value of commonality analysis and to demonstrate its usefulness in interpreting results from educational experiments using both univariate and multivariate methods. The statistical examples provided serve as models to researchers of ways of implementing commonality analysis as an adjunct to various univariate and multivariate statistical methods. Ten data tables are provided. (TJH)
Publication Type: Reports - Research; Speeches/Meeting Papers
Education Level: N/A
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: N/A