Showing papers on "Intraclass correlation published in 1975"

Measuring agreement between two judges on the presence or absence of a trait.

Abstract: At least a dozen indexes have been proposed for measuring agreement between two judges on a categorical scale. Using the binary (positive-negative) case as a model, this paper presents and critically evaluates some of these proposed measures. The importance of correcting for chance-expected agreement is emphasized, and identities with intraclass correlation coefficients are pointed out.

A review of statistical methods in the analysis of data arising from observer reliability studies (Part II)

Abstract: Summary This paper reviews research situations in medicine, epidemiology and psychiatry, in psychological measurement and testing, and in sample surveys in which the observer(rater or interviewer) can be an important source of measurement error. Moreover, most of the statistical literature in observer variability is surveyed with attention given to a notational unification of the various models proposed. In the continuous data case, the usual analysis of variance (ANOVA) components of variance models are presented with an emphasis on the intraclass correlation coefficient as a measure of reliability. Other modified ANOVA models, response error models in sample surveys, and related multivariate extensions are also discussed. For the categorical data case, special attention is given to measures of agreement and tests of hypotheses when the data consist of dichotomous responses. In addition, similarities between the dichotomous and continous cases are illustrated in terms of intraclass correlation coefficients. Finally, measures of agreement, such as kappa and weighted-kappa, are discussed in the context of nominal and ordinal data. A proposed unifying framework for the categorical data case is given in the form of concluding remarks.

On the correlation between correlations

Abstract: Specific tests for the equality of two dependent interclass correlations, estimated from a multivariate sample, have been examined in some detail by Dunn & Clark (1969, 1971). In this paper I wish to consider the dependencies among correlation coefficients in more generality, whether interclass or intraclass, and hence to suggest appropriate large-sample tests of hypotheses concerning them. This problem is of particular interest to geneticists, who may be concerned with the analysis of correlations between pairs of relatives. Rao, Morton & Yee (1974) have recently presented a general linear model for such an analysis, in which an attempt is made to apportion the total variability of a trait to environmental and genetic causes when independent estimates of the various correlations are available. In practice, however, correlations such as parent-offspring and full-sib are often estimated from a single set of nuclear families, parents and children, and so the estimates are not independent. After deriving asymptotic expressions for the correlations between each of the five different pairs of correlation coefficients possible, I indicate how they can be used to test hypotheses and to obtain pooled correlation estimates from families of different sizes.

On Estimation and Hypothesis Testing Problems for Correlation Coefficients.

Abstract: A selection of statistical problems commonly encountered in psychological or psychiatric research concerning correlation coefficients are re-evaluated in the light of recently developed simplifications in the forms of the distribution theory of the intraclass correlation coefficient (exact theory), of the product-moment correlation coefficient and of the Spearman rank correlation coefficient (approximate).

Testing the equality of covariance matrices under intraclass correlation models

Abstract: The modified likelihood ratio test for the equality of covariance matrices under intraclass correlation models is obtained and its asymptotic distributions are derived This test is compared with the test derived by using Roy's union-intersection principle, by a Monte Carlo study It is found that in general the modified likelihood ratio test has a larger power When the covariance matrices are such that one has small eigenvalues, one has large eigenvalues and the eigenvalues of the rest are in the middle, the two tests have about the same power