Abstract
Chapter 10 is the second of two chapters describing connections, equivalencies, and relationships relating to measures of ordinal association. First, Kendall’s \(\tau _{b}\) measure of ordinal association and Stuart’s \(\tau _{c}\) measure of ordinal association are described and the connection linking Kendall’s \(\tau _{b}\), Stuart’s \(\tau _{c}\), and permutation test statistic \(\delta \) is established. An example illustrates the connections among \(\tau _{a}\), \(\tau _{b}\), \(\tau _{c}\) and \(\delta \). Second, Goodman and Kruskal’s \(\gamma \) measure of ordinal association is presented and the connection linking Goodman and Kruskal’s \(\gamma \) and permutation test statistic \(\delta \) is detailed. Third, Somers’ \(d_{yx}\) and \(d_{xy}\) measures of ordinal association are described and the connections linking Somers’ \(d_{yx}\) and \(d_{xy}\) and permutation test statistic \(\delta \) are established. Fourth, Wilson’s e measure of ordinal association is described and the connection linking Wilson’s e and test statistic \(\delta \) is delineated. Fifth, the connection linking Kendall’s S measure of ordinal association and Mann and Whitney’s U two-sample rank-sum statistic is described. Finally, Cohen’s weighted kappa measure of inter-rater agreement is presented and the connection linking \(\kappa _{w}\) and permutation measure \(\Re \) is described.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
For a more comprehensive description of ordinal measures of association, see The Measurement of Association by Berry, Johnston, and Mielke [14].
- 2.
It is common in statistics to denote cross-classified categorical variables as a and b instead of x and y.
- 3.
It should be noted that all Monte Carlo permutation analyses based on L random arrangements of the data in this chapter were initiated with a common seed.
- 4.
Although Type I error is not relevant to permutation statistical methods, it can be noted that the asymptotic probability value of \(P = 0.0657\) is greater than the traditional level of significance of \(\alpha = 0.05\) and the Monte Carlo probability value of \(P = 0.0292\) is less than \(\alpha = 0.05\).
- 5.
It is customary in statistics to reserve the Greek letter \(\gamma \) for the population parameter and to use the Latin letter G for the sample statistic, even though Goodman and Kruskal originally used \(\gamma \) to denote the sample statistic. In this book the authors use Goodman and Kruskal’s \(\gamma \) notation.
- 6.
- 7.
It should be noted that Kendall did not refer to concordant (C) and discordant (D) pairs. The C and D notation was introduced some years later by Leo Goodman and William Kruskal [76]. Kendall denoted “pairs in their natural order” as p, corresponding to C, and “pairs in their inverse order” as q, corresponding to D.
- 8.
When converting tied raw scores to rank scores, each rank is converted to the average of the assigned ranks. For example, in Table 10.19 there are four raw scores of 20, corresponding to ranks, 1, 2, 3, and 4. Each of the four raw scores is assigned the average of the ranks, i.e., \((1+2+3+4)/4 = 10/4 = 2.5\).
- 9.
- 10.
Permutation test statistic \(\delta \) is multivariate and can easily accommodate multiple response measurements [150, p. 145].
References
Berry, K.J., Johnston, J.E., Mielke, P.W.: The Measurement of Association: A Permutation Statistical Approach. Springer, Cham, CH (2018)
Blalock, H.M.: A double standard in measuring degree of association. Am. Sociol. Rev. 28, 988–989 (1963)
Burr, E.J.: The distribution of Kendall’s score S for a pair of tied rankings. Biometrika 47, 151–171 (1960)
Cicchetti, D.V., Allison, A.: A new procedure for assessing reliability of scoring EEG sleep recordings. Am. J. EEG Technol. 11, 101–109 (1971)
Cohen, J.: A coefficient of agreement for nominal scales. Educ. Psychol. Meas. 20, 37–46 (1960)
Cohen, J.: Weighted kappa: Nominal scale agreement with provision for scaled disagreement or partial credit. Psychol. Bull. 70, 213–220 (1968)
Ezekiel, M.J.B.: Methods of Correlation Analysis. Wiley, New York (1930)
Goodman, L.A., Kruskal, W.H.: Measures of association for cross classifications. J. Am. Stat. Assoc. 49, 732–764 (1954)
Hays, W.L.: Statistics. Holt, Rinehart and Winston, New York (1963)
Hemelrijk, J.: Note on Wilcoxon’s two-sample test when ties are present. Ann. Math. Stat. 23, 133–135 (1952)
Jonckheere, A.R.: A distribution-free k-sample test against ordered alternatives. Biometrika 41, 133–145 (1954)
Kelley, T.L.: An unbiased correlation ratio measure. Proc. Natl. Acad. Sci. 21, 554–559 (1935)
Kendall, M.G.: The treatment of ties in ranking problems. Biometrika 33, 239–251 (1945)
Kendall, M.G.: The Advanced Theory of Statistics, Vol. II. Griffin, London (1946)
Kendall, M.G.: Rank Correlation Methods. Griffin, London (1948)
Kendall, M.G., Babington Smith, B.: The problem of m rankings. Ann. Math. Stat. 10, 275–287 (1939)
Kruskal, W.H.: Historical notes on the Wilcoxon unpaired two-sample test. J. Am. Stat. Assoc. 52, 356–360 (1957)
Leach, C.: Introduction to Statistics: A Nonparametric Approach for the Social Sciences. Wiley, New York (1979)
Mann, H.B.: Nonparametric tests against trend. Econometrica 13, 245–259 (1945)
Mann, H.B., Whitney, D.R.: On a test of whether one of two random variables is stochastically larger than the other. Ann. Math. Stat. 18, 50–60 (1947)
Mielke, P.W., Berry, K.J.: Permutation Methods: A Distance Function Approach, 2nd edn. Springer, New York (2007)
Moses, L.E.: Statistical theory and research design. Annu. Rev. Psychol. 7, 233–258 (1956)
Randles, R.H., Wolfe, D.A.: Introduction to the Theory of Nonparametric Statistics. Wiley, New York (1979)
Robinson, W.S.: The statistical measurement of agreement. Am. Sociol. Rev. 22, 17–25 (1957)
Scott, W.A.: Reliability of content analysis: The case of nominal scale coding. Public Opin. Quart. 19, 321–325 (1955)
Somers, R.H.: A new asymmetric measure of association for ordinal variables. Am. Sociol. Rev. 27, 799–811 (1962)
Spearman, C.E.: ‘Footrule’ for measuring correlation. Brit. J. Psychol. 2, 89–108 (1906)
Stuart, A.: The estimation and comparison of strengths of association in contingency tables. Biometrika 40, 105–110 (1953)
Vanbelle, S., Albert, A.: A note on the linearly weighted kappa coefficient for ordinal scales. Stat. Methodol. 6, 157–163 (2008)
Whitfield, J.W.: Rank correlation between two variables, one of which is ranked, the other dichotomous. Biometrika 34, 292–296 (1947)
Wilcoxon, F.: Individual comparisons by ranking methods. Biometrics Bull. 1, 80–83 (1945)
Wilson, T.P.: Measures of association for bivariate ordinal hypotheses. In: Blalock, H.M. (ed.) Measurement in the Social Sciences: Theories and Strategies, pp. 327–342. Aldine, Chicago (1974)
Author information
Authors and Affiliations
Preview of Chap. 11
Preview of Chap. 11
Chapter 11 describes connections, equivalencies, and relationships relating to nominal-level (categorical) variables. First, Pearson’s chi-squared test of independence is described and illustrated with an example analysis. Second, Pearson’s \(\phi ^{2}\), Pearson’s C, Tschuprov’s\(T^{2}\), and Cramér’s\(V^{2}\) measures of nominal association are described and the connections linking the measures to Pearson’s \(\chi ^{2}\) are established. Third, the connections linking Goodman and Kruskal’s\(t_{a}\) and \(t_{b}\) measures with \(\chi ^{2}\) and permutation test statistic \(\delta \) are established. Fourth, McNemar’s\(Q_{\text{M}}\) and Cochran’s\(Q_{\text{C}}\) measures of change are described and the connections linking \(Q_{\text{M}}\), \(Q_{\text{C}}\) and permutation test statistic \(\delta \) are presented. Fifth, Cohen’s unweighted kappa (\(\kappa \)) and weighted kappa (\(\kappa _{w}\)) are described and the connections linking \(\kappa \), \(\kappa _{w}\), and Mielke and Berry’s permutation-based measure of chance-corrected agreement, \(\Re \), are established. Finally, the connections linking Pearson’s\(\chi ^{2}\) test of independence and Pearson’s \(r_{xy}^{2}\) product-moment correlation coefficient are described and illustrated for \(r \times c\) contingency tables.
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this chapter
Cite this chapter
Berry, K.J., Johnston, J.E. (2023). Measures of Ordinal Association II. In: Statistical Methods: Connections, Equivalencies, and Relationships. Springer, Cham. https://doi.org/10.1007/978-3-031-41896-9_10
Download citation
DOI: https://doi.org/10.1007/978-3-031-41896-9_10
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-41895-2
Online ISBN: 978-3-031-41896-9
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)