Abstract
The internal correlation, a measure of dependency in a set of variables, is discussed and generalized. This coefficient is an upper bound to the product moment correlations, multiple correlations, and canonical correlations that can be defined in a set of variables. Applications of the internal correlation coefficient and its generalizations are given for a number of data-analytic situations. Where appropriate, we discuss tests of significance. We illustrate the internal correlation and expand the concept to a series of additional indices: local internal, up-internal, and down-internal correlations. Uses of these indices are illustrated in several areas: multicollinearity, ridge regression, factor analysis, principal components analysis, and test reliability.
Get full access to this article
View all access options for this article.