The Effect of Skew on the Magnitude of Product-Moment Correlations

Abstract The effect of transforming actual skewed variables toward symmetry on the intercorrelations among the transformed variables was examined. The results showed that average correlations increased in all 20 cases studied, even though the transformations to correct skew produced slightly greater nonlinear relationships in some cases. Because the F and t tests for the significance of r are directly proportional to the size of r, correcting skew increases power and at the same time better meets underlying assumptions of those tests. The implications of the findings as well as potential difficulties associated with correction of skew in correlational research are discussed.

[1]  M. R. Novick,et al.  Statistical Theories of Mental Test Scores. , 1971 .

[2]  R. Fowler,et al.  Power and Robustness in Product-Moment Correlation , 1987 .

[3]  W. Dunlap,et al.  Data transformation, power, and skew: A rejoinder to Games. , 1983 .

[4]  R. Gooding,et al.  A Meta-Analytic Review of the Relationship between Size and Performance: The Productivity and Efficiency of Organizations and Their Subunits. , 1985 .

[5]  Lee J. Cronbach,et al.  Psychological tests and personnel decisions , 1958 .

[6]  W. Dixon,et al.  BMDP statistical software , 1983 .

[7]  N. Raju,et al.  An empirical Monte Carlo test of the accuracy of the correlation, covariance, and regression slope models for assessing validity generalization. , 1989 .

[8]  M. J. Allen.  Introduction to Measurement Theory , 1979 .

[9]  D. Cox,et al.  An Analysis of Transformations , 1964 .

[10]  G. A. Ferguson,et al.  The factorial interpretation of test difficulty , 1941 .

[11]  S. Halperin.  Spurious correlations—causes and cures , 1986, Psychoneuroendocrinology.

[12]  J. Carroll.  The nature of the data, or how to choose a correlation coefficient , 1961 .

[13]  William P. Dunlap,et al.  A computer program for determining optimal data transformations minimizing skew , 1974 .

[14]  J. Aitchison,et al.  The Lognormal Distribution. , 1958 .

[15]  M. Burke,et al.  A New Approach for Utility Analysis , 1990 .

[16]  H. E. Brogden,et al.  On the interpretation of the correlation coefficient as a measure of predictive efficiency. , 1946, Journal of educational psychology.

[17]  John B. Cullen,et al.  Blau's Theory of Structural Differentiation Revisited: A Theory of Structural Change or Scale? , 1986 .

[18]  Masoud Yasai-Ardekani,et al.  Effects of Environmental Scarcity and Munificence on the Relationship of Context to Organizational Structure , 1989 .

[19]  C. Glisson,et al.  Productivity and efficiency in human service organizations as related to structure, size, and age. , 1980, Academy of Management journal. Academy of Management.

[20]  J. M. Cortina,et al.  What Is Coefficient Alpha? An Examination of Theory and Applications , 1993 .

[21]  Jacob Cohen,et al.  Applied multiple regression/correlation analysis for the behavioral sciences , 1979 .

[22]  Vincent H. Swoyer,et al.  Some Empirical Results Concerning the Power of Bartlett’s Test of the Significance of a Correlation Matrix , 1967 .

[23]  William P. Dunlap,et al.  Power of the F test with skewed data: Should one transform or not? , 1982 .

[24]  J. Kimberly.  Organizational size and the structuralist perspective: a review, critique, and proposal , 1976 .

[25]  P. Games.  Curvilinear transformations of the dependent variable. , 1983 .

[26]  T. Greer,et al.  Skew reduces test-retest reliability , 1994 .

[27]  T. Micceri.  The unicorn, the normal curve, and other improbable creatures. , 1989 .

[28]  R. Blair,et al.  A more realistic look at the robustness and Type II error properties of the t test to departures from population normality. , 1992 .

[29]  Jacob Cohen.  Statistical Power Analysis for the Behavioral Sciences , 1969, The SAGE Encyclopedia of Research Design.

[30]  H. O. Lancaster.  Some properties of the bivariate normal distribution considered in the form of a contingency table , 1957 .