Range restrictions for product-moment correlation matrices
- 241 Downloads
It is well-known that for a trivariate distribution if two correlations are fixed the remaining one is constrained. Indeed, if one correlation is fixed, then the remaining two are constrained. Both results are extended to the case of a multivariate distribution. The results are applied to some special patterned matrices.
Key wordsconstrained correlations multiple correlation partial correlation patterned correlation matrices
Unable to display preview. Download preview PDF.
- Glass, G. V. & Collins, J. R. Geometric proof of the restriction of possible value ofr xy whenr xz andr yz are fixed.Educational and Psychological Measurement, 1970,30, 37–39.Google Scholar
- Gleser, L. J. On bounds for the average correlation between subtest scores in ipsatively scored tests.Educational and Psychological Measurement, 1972,32, 759–765.Google Scholar
- Hubert, L. J. A note on the restriction of range for Pearson product-moment correlation coefficients,Educational and Psychological Measurement, 1972,32, 767–770.Google Scholar
- Leung, C.-K. & Lam, K. A note on the geometric representation of the correlation coefficients.The American Statistician, 1975,29, 128–130.Google Scholar
- McCornack, R. L. A criticism of studies comparing item-weighting methods.The Journal of Applied Psychology, 1956,40, 343–344.Google Scholar
- Priest, H. F. Range of correlation coefficients.Psychological Reports, 1968,22, 168–170.Google Scholar
- Radcliffe, J. A. Some properties of ipsative score matrices and their relevance for some current interest tests.Australian Journal of Psychology, 1963,15, 1–11.Google Scholar
- Stanley, J. C. & Wang, M. D. Restrictions on the possible values ofr 12 givenr 13 andr 23.Educational and Psychological Measurement, 1969,29, 579–581.Google Scholar