Canonical Correlation Programs: a Comparative Analysis of Performance

Abstract

This study was designed to compare and contrast generally available computer programs which provide canonical correlation analysis. The same set of data was analyzed by each of four canonical correlation programs. The four programs compared came from the Biomedical Computer Programs (BMD), Miami Multivariate Statistical Programs (Miami), Statistical Analysis System (SAS), and Statistical Package for the Social Sciences (SPSS). Following the analyses detailed comparisons were tabulated using the following criteria: (1) Run Characteristics, (2) Limitations, and (3) Analysis Features and Options. The newer programs (i.e., SPSS CANCORR and SAS CANCORR) tended to have more features and greater versatility. However, each program had certain uniquenesses which might make it more nearly appropriate for a given analysis requirement.

Get full access to this article

View all access options for this article.

References

Clyde, D. J. , Cramer, E. M. , and Sherin, R. J. Multivariate statistical programs. Coral Gables, Florida : Biometric Laboratory, University of Miami, 1966.

Cooley, W. W. and Lohnes, P. R. Multivariate data analysis. New York: Wiley, 1971.

Dixon, W. J. Biomedical computer programs (3rd ed.). Los Angeles : University of California Press, 1973.

Hotelling, H. Relations between two sets of variates. Biometrika, 1936, 28, 321-377.

Kerlinger, F. N. , and Pedhazur, E. J. Multiple regression in behaviorial research. New York: Holt, Rinehart and Winston, 1973.

Morrison, D. F. Multivariate statistical methods . New York: McGraw-Hill, 1967.

Nie, N. H. , Hull, C. H. , Jenkins, J. G. , Steinbrenner, K. , and Bent, D. H. Statistical package for the social sciences (2nd ed.). New York: McGraw-Hill Book Company, 1975.

Service, J. A user's guide to the statistical analysis system. Raleigh, North Carolina: Student Supply Stores, North Carolina State University, 1972.