ERIC Number: ED194568
Record Type: RIE
Publication Date: 1980-Sep
Reference Count: 0
Using Composite Variables in Multivariate Analysis: Why Weight?.
Gabriel, Roy M.
Multivariate analysis techniques employ various methods of optimally weighting variables to form composites. The interpretation of these composites in subsequent analyses is rarely straightforward, fraught with difficulties based upon the statistical properites of the estimated weights. This paper synthesizes interpretive guidelines for four commonly used multivariate techniques: multiple linear regression, canonical correlation, multivariate analysis of variance (MANOVA) and linear discriminant analysis. Three methods of weighting variables are discussed and illustrated on simulated data. The complexity of the correlation structure of the entire set is shown to be the crucial determinant in selecting the weighting method of choice. Guidelines for the use of the various weighting schemes on all four multivariate techniques are presented. (Author/MH)
Publication Type: Speeches/Meeting Papers; Reports - Research
Education Level: N/A
Authoring Institution: N/A
Identifiers: Data Interpretation; Linear Discriminant Function; Multiple Linear Regression; Weighted Variables
Note: Paper presented at the Annual Meeting of the American Psychological Association (88th, Montreal, Quebec, Canada, September 1-5, 1980).