Sage Journals: Discover world-class research

Abstract

It is shown that Goodman and Kruskal's Tau-b, a PRE coefficient of association between two nominal variables, is a more general form of the correlation ratio. Deriving from this, it is shown that similar correlation ratios can be derived for relationships between independent ordinal and interval variables and a dependent nominal variable. It is suggested that this makes it possible to express the relationship between any variable at any level of measurement-nominal, ordinal, or interval-in terms of linear equations and correlation ratios, and that this provides a useful and conceptionally parsimonious technique for practitioners. It also opens the way to greater flexibility in multivariate analysis using variables at different levels of measurement.

Get full access to this article

View all access options for this article.

References

Andrews, F.M. and R.C. Messenger (1973) Multivariate Nominal Scale Analysis. Ann Arbor, Mich.: Survey Research Center.

Blalock, H.M., Jr. (1972) Social Statistics New York: McGraw-Hill.

Costner, H.L. (1965) "Criteria for measures of association." Amer. Soc. Rev. 30 (June): 341-353.

Daniels, H.E. (1944) "The relation between measures of association in the universe of sample permutations." Biometrika 33 (August): 129-135.

Fennessey, J. (1968) "The general linear model: a new perspective on some familiar topics." Amer. J. of Sociology 74 (July): 1-27.

Goldberger, A.S. (1964) Econometric Theory. New York : John Wiley.

Goodman, L.A. and W.H. Kruskal (1972) "Measures of association for cross-classifications IV: simplification of asymptotic variances." J. of the Amer. Statistical Assn. 67 (June): 415-421.

--- ( 1963) "Measures of association for cross-classifications III: approximate sampling theory." J. of the Amer. Statistical Assn. 58: 310-364.

--- ( 1954) "Measures of association for cross-classifications." J. of the Amer. Statistical Assn. 49: 732-764.

10.

Hawkes, R.K. (1971) "The multivariate analysis of ordinal measures ." Amer. J. of Sociology 76 (March): 908-926.

11.

Kendall, M.G. (1955) Rank Correlation Methods. New York: Hafner.

12.

Kim, J.-O. (1971) "Predictive measures of ordinal association." Amer. J. of Sociology 76 (March): 891-907.

13.

Kruskal, W.H. and W.A. Wallis (1952) "Use of ranks in one-criterion variance analysis ." J. of the Amer. Statistical Assn. 47: 583-621.

14.

Labovitz, S. (1967) "Some observations on measurement and statistics ." Social Forces 47 (December): 151-160.

15.

Scheffe, H. (1959) The Analysis of Variance. New York: John Wiley.

16.

Senders, V.L. (1958) Measurement and Statistics. New York: Oxford Univ. Press.

17.

Somers, R.H. (1968) "On the measurement of association." Amer. Soc. Rev. 33 (April): 291-292.

18.

--- (1962a) "A new asymmetric measure of association for ordinal variables." Amer. Soc. Rev. 27 (December): 799-811.

19.

——— (1962b) "A similarity between Goodman and Kruskal's Tau and Kendall's Tau, with a partial interpretation of the latter." J. of the Amer. Statistical Assn. 57 (December): 804-812.

20.

Southwood, K.E. (1973) "On the interpretation of measures of ordinal association ." (unpublished)

21.

Suits, D.B. (1958) "Use of dummy variables in regression equations ." J. of the Amer. Statistical Assn. 53 (December): 548-551.

22.

Van De Geer, J.P. (1971) Introduction to Multivariate Analysis for the Social Sciences. San Francisco: W. H. Freeman .

23.

Wilson, T.P. (1968) "A proportional-reduction-in-error interpretation for Kendall's Tau-b." Social Forces 47 (March): 340-342.

24.

Zar, J.H. (1972) "Significance testing of the Spearman rank correlation coefficient." J. of the Amer. Statistical Assn. 67 (September): 578-580.

Goodman and Kruskal's Tau-b as Correlation Ratio

Abstract

Get full access to this article

References