The Analysis of k x 2 and other Two Way Tables

  • Lothar Sachs
Part of the Springer Series in Statistics book series (SSS)

Abstract

The information content of frequencies is small. Nevertheless, analysis of fourfold tables (cf., Section 4.6), the simplest two way tables, offers a number of possibilities. We can test these 2 by 2 tables for independence, homogeneity, correlation, and symmetry. These and other tests are discussed in this chapter for tables of size 3 by 2 or greater. Especially important is Section 6.2.1.

Keywords

Manifold Assure 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

[8:6] Chapter 6

  1. Altham, Patricia, M. E.: The measurement of association of rows and columns for an r - s contingency table. J. Roy. Statist. Soc. B 32 (1970), 63–73MATHGoogle Scholar
  2. Armitage, P.: Tests for linear trends in proportions and frequencies. Biometrics 11 (1955), 375–386CrossRefGoogle Scholar
  3. Bartholomew, D. J.: A test of homogeneity for ordered alternatives. I and II. Biometrika 46 (1959), 36–48 and 328–335 [see also 63 (1976), 177–183, 647–654; J. Roy. Statist. Soc. B 23 (1961), 239–281. J. Amer. Statist. Assoc. 67 (1972), 55–63, 75 (1980), 454–459 as well as Biometrics 30 (1974), 589–597]MathSciNetMATHGoogle Scholar
  4. Benedetti, J. K., and Brown, M. B.: Strategies for the selection of log linear models. Biometrics 34 (1978), 680–686MATHCrossRefGoogle Scholar
  5. Bennett, B. M.: Tests for marginal symmetry in contingency tables. Metrika 19 (1972), 23–26MathSciNetMATHCrossRefGoogle Scholar
  6. Bennett, B. M., and Hsu, P.: Sampling studies on a test against trend in binomial data. Metrika 5 (1962), 96–104MATHCrossRefGoogle Scholar
  7. (1).
    Bennett, B. M., and E. Nakamura: Tables for testing significance in a 2 × 3 contingency table. Technometrics 5 (1963), 501–511.MathSciNetMATHGoogle Scholar
  8. (2).
    Bennett, B. M., and E. Nakamura: The power function of the exact test for the 2 × 3 contingency table. Technometrics 6 (1964), 439–458MathSciNetMATHGoogle Scholar
  9. Bennett, B. M., and Kaneshiro, C.: Small sample distribution and power of the binomial index of dispersion and log likelihood ratio tests. Biometrical Journal 20 (1978) 485–493MATHCrossRefGoogle Scholar
  10. Berg, Dorothy, Leyton, M., and Maloney, C. J.: Exact contingency table calculations. Ninth Conf. Design Exper. in Army Research Developments and Testing (1965), (N.I.H., Bethesda Md.)Google Scholar
  11. Bhapkar, V. P.: On the analysis of contingency tables with a quantitative response. Biometrics 24 (1968), 329–338MathSciNetCrossRefGoogle Scholar
  12. (1).
    Bhapkar, V. P., and Koch, G. G.: Hypotheses of “no interaction” in multidimensional contingency tables. Technometrics 10 (1968), 107–123.MathSciNetGoogle Scholar
  13. (2).
    Bhapkar, V. P., and Koch, G. G.: On the hypotheses of “no interaction” in contingency tables. Biometrics 24 (1968), 567–594CrossRefGoogle Scholar
  14. Bishop, Yvonne, M. M.: Full contingency tables, logits, and split contingency tables. Biometrics 25 (1969), 383–399 (see also 119–128)CrossRefGoogle Scholar
  15. Bishop, Yvonne, M. M., Fienberg, S. E., and Holland, P. W.: Discrete Multivariate Analysis. Theory and Practice. (MIT Press: pp. 557) Cambridge, Mass. 1975 [cf. Biometrical Journal 22 (1980), 159–167, 779–793, and 795–789]Google Scholar
  16. Bowker, A. H.: A test for symmetry in contingency tables. J. Amer. Statist. Assoc. 43 (1948), 572–574 [see also Biometrics 27 (1971), 1074–1078]MATHGoogle Scholar
  17. Bresnahan, J. L., and Shapiro, M. M.: A general equation and technique for the exact partitioning of chi-square contingency tables. Psychol. Bull. 66 (1966), 252–262CrossRefGoogle Scholar
  18. Castellan, N. J., Jr.: On the partitioning of contingency tables. Psychol. Bull. 64 (1965), 330–338CrossRefGoogle Scholar
  19. Caussinus, H.: Contribution à l’analyse statistique des tableaux de corrélation. Ann. Fac. Sci. Univ. Toulouse. Math., 4. Ser., 29 (1965), 77–183MathSciNetCrossRefGoogle Scholar
  20. (1).
    Cochran, W. G.: Some methods of strengthening the common χ 2 tests. Biometrics 10 (1954), 417–451.MathSciNetMATHCrossRefGoogle Scholar
  21. (2).
    Cochran, W. G.: Analyse des classifications d’ordre. Revue Statistique Appliquée 14, 2 (1966), 5–17Google Scholar
  22. Dunn, O. J.: Multiple comparisons using rank sums. Technometrics 6 (1964), 241–252Google Scholar
  23. Eberhard, K.: FM - Ein Maß für die Qualität einer Vorhersage aufgrund einer mehrklassigen Variablen in einer k · 2-Felder-Tafel. Z. exp. angew. Psychol. 17 (1970), 592–599Google Scholar
  24. Enke, H.: On the analysis of incomplete two-dimensional contingency tables. Biometrical Journal 19 (1977), 561–573 [and 20 (1978), 229–242, 22 (1980), 779–793]MathSciNetCrossRefGoogle Scholar
  25. Everitt, B. S.: The Analysis of Contingency Tables. (Chapman and Hall; pp. 128) London 1977Google Scholar
  26. Fairfield Smith, H: On comparing tables. The Philippine Statistician 6 (1957), 71–81Google Scholar
  27. Fienberg, S. E.: The Analysis of Cross-Classified Categorical Data. (MIT Press; pp. 151) Cambridge, Mass. 1978Google Scholar
  28. Gabriel, K. R.: Simultaneous test procedures for multiple comparisons on categorical data. J. Amer. Statist. Assoc. 61 (1966), 1080–1096MathSciNetGoogle Scholar
  29. Gart, J. J.: Alternative analyses of contingency tables. J. Roy. Statist. Soc. B 28 (1966), 164–179 [see also Biometrika 59 (1972), 309–316 and 65 (1978), 669–672]MathSciNetMATHGoogle Scholar
  30. Gokhale, D. V., and Kullback, S.: The Information in Contingency Tables. M. Dekker, New York, 1978MATHGoogle Scholar
  31. (1).
    Goodman, L. A.: On methods for comparing contingency tables. J. Roy. Statist. Soc., Ser. A 126 (1963), 94–108.MathSciNetCrossRefGoogle Scholar
  32. (2).
    Goodman, L. A.: Simple methods for analyzing three-factor interaction in contingency tables. J. Amer. Statist. Assoc. 59 (1964), 319–352.MathSciNetMATHGoogle Scholar
  33. (3).
    Goodman, L. A.: On partitioning χ 2 and detecting partial association in three-way contingency tables. J. Roy. Statist. Soc. B 31 (1969), 486–498.MATHGoogle Scholar
  34. (4).
    Goodman, L. A.: The multivariate analysis of qualitative data: interactions among multiple classifications. J. Amer. Statist. Assoc. 65 (1970), 226–256.Google Scholar
  35. (5).
    Goodman, L. A.: The analysis of multidimensional contingency tables. Stepwise procedures and direct estimation methods for building models for multiple classifications. Technometrics 13 (1971), 33–61 [see also J. Amer. Statist. Assoc. 68 (1973), 165–175 and 74 (1979), 537–552 as well as Biometrika 68 (1981), 347–355]MATHGoogle Scholar
  36. Goodman, L. A., and Kruskal, W. H.: Measures of association for cross classifications, IV. Simplification of asymptotic variances. J. Amer. Statist. Assoc. 67 (1972), 415–421 [see also 49 (1954), 732–764; 52 (1957), 578; 54 (1959), 123–163; 58 (1963), 310–364; 74 (1979), 537–552 as well as A. Agresti 71 (1976), 49–55]MATHGoogle Scholar
  37. Grizzle, J. E., Starmer, C. F., and Koch, G. G.: Analysis of categorical data by linear models. Biometrics 25 (1969), 489–504 [see also 26 (1970), 860; 28 (1972), 137–156, J. Amer. Statist. Assoc. 67 (1972), 55–63 and Method. Inform. Med. 12 (1973), 123–128 as well as Int. Stat. Rev. 48 (1980), 249–265]MathSciNetMATHCrossRefGoogle Scholar
  38. Hamdan, M. A.: Optimum choice of classes for contingency tables. J. Amer. Statist. Assoc. 63 (1968), 291–297 [see also Psychometrika 36 (1971), 253–259]MathSciNetCrossRefGoogle Scholar
  39. Hartwig, F.: Statistical significance of the lambda coefficients. Behavioral Science 18 (1973), 307–310CrossRefGoogle Scholar
  40. Hettmansperger, T. P., and McKean, J. W.: On testing for significant change in c x c tables. Commun. Statist. 2 (1973), 551–560MATHCrossRefGoogle Scholar
  41. Ireland, C. T., Ku, H. H., and Kullback, S.: Symmetry and marginal homogeneity of an r r contingency table. J. Amer. Statist. Assoc. 64 (1969), 1323–1341MathSciNetGoogle Scholar
  42. Ireland, C. T., and Kullback, S.: Minimum discrimination information estimation. Biometrics 24 (1968), 707–713 [see also 27 (1971), 175–182 and Biometrika 55 (1968), 179–188]CrossRefGoogle Scholar
  43. (1).
    Kastenbaum, M. A.: A note on the additive partitioning of chi-square in contingency tables. Biometrics 16 (1960), 416–422.CrossRefGoogle Scholar
  44. (2).
    Kastenbaum, M. A.: Analysis of categorical data: some well-known analogues and some new concepts. Commun. Statist. 3 (1974), 401–417MathSciNetMATHCrossRefGoogle Scholar
  45. Killion, R. A., and Zahn, D. A.: A bibliography of contingency table literature: 1900 to 1974. International Statistical Review 44 (1976), 71–112MathSciNetGoogle Scholar
  46. Kincaid, W. M.: The combination of 2 × m contingency tables. Biometrics 18 (1962), 224–228MathSciNetCrossRefGoogle Scholar
  47. Kramer, C. Y.: A First Course in Methods of Multivariate Analysis. Virginia Polytechnic Institute and State University, Blacksburg 1972Google Scholar
  48. Ku, H. H.: A note on contingency tables involving zero frequencies and the 21 test. Technometrics 5 (1963), 398–400MATHGoogle Scholar
  49. Ku, H. H., and Kullback, S.: Interaction in multidimensional contingency tables: an information theoretic approach. J. Res. Nat. Bur. Stds. 72B (1968), 159–199MathSciNetGoogle Scholar
  50. Ku, H. H., Varner, R. N., and Kullback, S.: On the analysis of multidimensional contingency tables. J. Amer. Statist. Assoc. 66 (1971), 55–64 [see also 70 (1975), 503–523 and 624–625]MathSciNetMATHGoogle Scholar
  51. Kullback, S.: Information Theory and Statistics. (Wiley, pp. 395) New York 1959MATHGoogle Scholar
  52. J. Adam, H. Enke and G. Enderlein: Biometrische Zeitschr. 14 (1972), 305–323, 15 (1973), 53–78, 17 (1975), 513–523, 22 (1980), 779–793MATHGoogle Scholar
  53. (1).
    Kullback, S., Kupperman, M., and Ku, H. H.: An application of information theory to the analysis of contingency tables, with a table of 2n ln n, n = 1(1)10, 000. J. Res. Nat. Bur. Stds. B 66 (1962), 217–243.MathSciNetMATHGoogle Scholar
  54. (2).
    Kullback, S., Kupperman, M., and Ku, H. H.: Tests for contingency tables and Markov chains. Technometrics 4 (1962), 573–608MathSciNetMATHGoogle Scholar
  55. Kullback, S., and Leibler, R. A.: On information and sufficiency. Ann. Math. Statist. 22 (1951), 79–86MathSciNetMATHCrossRefGoogle Scholar
  56. Lancaster, H. O.: The Chi-Squared Distribution. (Wiley, pp. 356), New York 1969MATHGoogle Scholar
  57. Lewontin, R. C., and Felsenstein, J.: The robustness of homogeneity tests in 2 × n tables. Biometrics 21 (1965), 19–33MathSciNetCrossRefGoogle Scholar
  58. (1).
    Mantel, N.: Chi-square tests with one degree of freedom; extensions of the Mantel-Haenszel procedure. J. Amer. Statist. Assoc. 58 (1963), 690–700 [see also Biometrics 29 (1973), 479–486].MathSciNetMATHGoogle Scholar
  59. (2).
    Mantel, N.: Tests and limits for the common odds ratio of several 2 × 2 contingency tables: methods in analogy with the Mantel-Haenszel procedure. Journal of Statistical Planning and Inference 1 (1977), 179–189MATHCrossRefGoogle Scholar
  60. Mantel, N., and Haenszel, W.: Statistical aspects of the analysis of data from retrospective studies of disease. J. Natl. Cancer Institute 22 (1959), 719–748 [see also W. J. Youden: Cancer 3 (1950), 32–35 and Biometrika 66 (1979), 181–183, 419–427; Biometrics 35 (1979), 385–391, 36 (1980), 355–356, 381–399]Google Scholar
  61. Martini, P. : Methodenlehre der theraupeutisch-klinischen Forschung. Springer-Verlag, Berlin-Göttingen-Heidelberg, 1953Google Scholar
  62. Maxwell, A. E.: Analysing Qualitative Data. London 1961MATHGoogle Scholar
  63. Meng, R. C., and Chapman, D. G.: The power of chi-square tests for contingency tables. J. Amer. Statist. Assoc. 61 (1966), 965–975 [see also Amer. Statist. 31 (1977), 83–85]MathSciNetMATHGoogle Scholar
  64. Mosteller, F.: Association and estimation in contingency tables. J. Amer. Statist. Assoc. 63 (1968), 1–2MathSciNetCrossRefGoogle Scholar
  65. L. A. Goodman, 74 (1979), 537–552, J. R. Landis and G. G. Koch, Statistica Neerlandica 29 (1975), 101–123, 151–161MathSciNetMATHCrossRefGoogle Scholar
  66. L. A. Goodman, Biometrika 66 (1979), 413–418CrossRefGoogle Scholar
  67. J. Wahrendorf, Biometrika 67 (1980), 15–21MathSciNetMATHCrossRefGoogle Scholar
  68. Nass, C. A. G.: The χ 2 test for small expectations in contingency tables with special reference to accidents and absenteeism. Biometrika 46 (1959), 365–385MATHGoogle Scholar
  69. Nelder, J. A.: Log linear models for contingency tables. Applied Statistics 23 (1974), 323–329 [cf. 25 (1976), 37–46 and The Statistician 25 (1976), 51–58]MathSciNetCrossRefGoogle Scholar
  70. Odoroff, C. L.: A comparison of minimum logit chi-square estimation and maximum likelihood estimation in 2 × 2 × 2 and 3×2×2 contingency tables: tests for interaction. J. Amer. Statist. Assoc. 65 (1970), 1617–1631 [see also 69 (1974), 164–168]MATHGoogle Scholar
  71. Pawlik, K.: Der maximale Kontingenzkoeffizient im Falle nichtquadratischer Kontingenztafeln. Metrika 2 (1959), 150–166MathSciNetMATHCrossRefGoogle Scholar
  72. Plackett, R. L.: Analysis of Categorical Data. 2nd rev. ed. (Griffin; pp. 215) London 1981MATHGoogle Scholar
  73. Ryan, T.: Significance tests for multiple comparison of proportions, variances and other statistics. Psychological Bull. 57 (1960), 318–328CrossRefGoogle Scholar
  74. Sachs, L.: Statistische Methoden. 6th ed. (Springer; pp. 133) New York, Berlin, Heidelberg, Tokyo 1984Google Scholar
  75. Shaffer, J. P.: Defining and testing hypotheses in multidimensional contingency tables. Psychol. Bull. 79 (1973), 127–141MathSciNetCrossRefGoogle Scholar
  76. (1).
    Upton, G. J. G.: The Analysis of Cross Tabulated Data. (Wiley; pp. 148) New York 1978.Google Scholar
  77. (2).
    Upton, G. J. G.: A comparison of alternative tests for the 2 × 2 comparative trial. J. Roy. Statist. Society A 145 (1982), 86–105MathSciNetCrossRefGoogle Scholar
  78. Wisniewski, T. K. M.: Power of tests of homogeneity of a binomial series. Journal of the American Statistical Association 67 (1972), 680–683MATHGoogle Scholar
  79. Woolf, B.: The log likelihood ratio test (the G-Test). Methods and tables for tests of heterogeneity in contingency tables. Ann. Human Genetics 21 (1957), 397–409CrossRefGoogle Scholar
  80. Yates, F.: The analysis of contingency tables with groupings based on quantitative characters. Biometrika 35 (1948), 176–181 [see also 39 (1952), 274–289]MATHGoogle Scholar

Copyright information

© Springer-Verlag New York Inc. 1984

Authors and Affiliations

  • Lothar Sachs
    • 1
  1. 1.Abteilung Medizinische Statistik und Dokumentation im Klinikumder Universität KielKiel 1Federal Republic of Germany

Personalised recommendations