A note on the unbiased estimation of the intraclass correlation

The intraclass correlation,, is a parameter featured in much psychological research. Two commonly used estimators of, the maximum likelihood and least squares estimators, are known to be negatively biased. Olkin and Pratt (1958) derived the minimum variance unbiased estimator of the intraclass correlation, but use of this estimator has apparently been impeded by the lack of a closed form solution. This note briefly reviews the unbiased estimator and gives a FORTRAN 77 subroutine to calculate it.The first author was supported by an All-University Fellowship from the University of Southern California.     

The robustness of the usual correction for restriction in range due to explicit selection

Corrections for restriction in range due to explicit selection assume the linearity of regression and homoscedastic array variances. This paper develops a theoretical framework in which the effects of some common forms of violation of these assumptions on the estimation of the unrestricted correlation can be investigated. Simple expressions are derived for both the restricted and corrected correlations in terms of the target (unrestricted) correlation in these situations.The author is grateful to D. Holt, C. J. Skinner and T. M. F. Smith (all University of Southampton) for their helpful comments. Research was initially supported by grant No. HR7152 from the Economic and Social Research Council.     

Range restrictions for product-moment correlation matrices

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.     

Canonical/redundancy factoring analysis

The interrelationships between two sets of measurements made on the same subjects can be studied by canonical correlation. Originally developed by Hotelling [1936], the canonical correlation is the maximum correlation betweenlinear functions (canonical factors) of the two sets of variables. An alternative statistic to investigate the interrelationships between two sets of variables is the redundancy measure, developed by Stewart and Love [1968]. Van Den Wollenberg [1977] has developed a method of extracting factors which maximize redundancy, as opposed to canonical correlation.A component method is presented which maximizes user specified convex combinations of canonical correlation and the two nonsymmetric redundancy measures presented by Stewart and Love. Monte Carlo work comparing canonical correlation analysis, redundancy analysis, and various canonical/redundancy factoring analyses on the Van Den Wollenberg data is presented. An empirical example is also provided.Wayne S. DeSarbo is a Member of Technical Staff at Bell Laboratories in the Mathematics and Statistics Research Group at Murray Hill, N.J. I wish to express my appreciation to J. Kettenring, J. Kruskal, C. Mallows, and R. Gnanadesikan for their valuable technical assistance and/or for comments on an earlier draft of this paper. I also wish to thank the editor and reviewers of this paper for their insightful remarks.     

Extension of Feldt's approach to testing homogeneity of coefficients of reliability

Limitations and extensions of Feldt's approach to testing the equality of Cronbach's alpha coefficients in independent and matched samples are discussed. In particular, this approach is used to test equality of intraclass correlation coefficients.This work was supported by the National Institute of Mental Health Specialized Research Center Grant # MH-30854.     

The comparison of interdependent correlations between optimal linear composites

A general procedure is provided for comparing correlation coefficients between optimal linear composites. The procedure allows computationally efficient significance tests on independent or dependent multiple correlations, partial correlations, and canonical correlations, with or without the assumption of multivariate normality. Evidence from some Monte Carlo studies on the effectiveness of the methods is also provided.This research was supported in part by an operating grant (#67-4640) to the first author from the National Sciences and Engineering Research Council of Canada. The authors would also like to acknowledge the helpful comments and encouragement of Alexander Shapiro, Stanley Nash, and Ingram Olkin.     

‘Attentional Bias’ in correlation judgments? Smedslund (1963) revisited

Smedslund (1963) reports one of the first studies that investigated human judgments of correlation. Smedslund's conclusion, that people reason about correlation mostly from a consideration of the number of times two variables co-occur, has become textbook wisdom (e.g. Baron's "attentional bias", 1994). Yet the data reported in Smedslund's paper fall short of endorsing such a conclusion. After reviewing the original paper, we present the method with which we replicated Smedslund's main experiment. In Experiment 1 subjects were presented with symptom-disease correlation data through a simulated medical diagnosis task. Subjects clearly discriminated between data sets which shared an equal number of symptom-disease co-occurrences but which otherwise showed different levels of correlation. Subjects' diagnoses showed a propensity to predict the disease in the presence of the symptom, and symptom-disease co-occurrences were overestimated in two of the five data sets presented to the subjects. Experiment 2 used a novel abstract scenario with a symmetric predictor variable. Judgments again indicated good discrimination, and biases in prediction responses and case recall were eliminated. In both experiments, judgments of zero correlation were a function of the outcome base rate. We evaluate and contrast the extent to which Cheng's (1997) model of causal induction and two associative models, Rescorla and Wagner (1972) and Pearce (1987) can anticipate the observed pattern in the mean judgments.     

On the mean and variance of the tetrachoric correlation coefficient

Estimates of the mean and standard deviation of the tetrachoric correlation are compared with their expected values in several 2 × 2 tables. Significant bias in the mean is found when the minimum cell frequency is less than 5. Three formulas for the standard deviation are compared and guidelines given for their use.This research was performed when the first author was on leave at the University of California at Los Angeles and was supported in part by NIH Special Research Resources Grant RR-3. The second author was also supported by NIH Fellowship 5 F22 GM00328-02.     

Maximum likelihood estimation of the polychoric correlation coefficient

The polychoric correlation is discussed as a generalization of the tetrachoric correlation coefficient to more than two classes. Two estimation methods are discussed: Maximum likelihood estimation, and what may be called two-step maximum likelihood estimation. For the latter method, the thresholds are estimated in the first step. For both methods, asymptotic covariance matrices for estimates are derived, and the methods are illustrated and compared with artificial and real data.This paper was read at the 1978 European Meeting on Psychometrics and Mathematical Psychology in Uppsala, Sweden, June 1978.Research reported in this paper has been supported by the Bank of Sweden Tercentenary Foundation under project Structural Equation Models in the Social Sciences, project director Karl G. Jöreskog.