Download PDF
Methods Inf Med 1990; 29(02): 104-112
DOI: 10.1055/s-0038-1634776
Statistical Analysis
Schattauer GmbH

Fitting and Interpreting Loglinear Interactions in Cross-Classifications from Health Policy and Medicine

M. L. Brown
1   Department of Mathematics, Simmons College, Boston Mass, USA
› Author AffiliationsThe author is grateful to Paul Densen, Professor of Community Health and Medical Care Emeritus, Harvard University, for comments on a draft of this paper. The author also wishes to thank John Rowe, M. D., Professor of Medicine, Harvard University, and Carol Greenfield and Lawrence Kirsch, both of the Harvard School of Public Health, for their contributions to the earlier study cited in the paper.
Further Information

Publication History

Publication Date:
06 February 2018 (online)

    Permissions and Reprints

Abstract

A nontechnical exposition is presented of current statistical techniques for the analysis of multidimensional tables of counted data. Performing an original analysis of a data set of interest to researchers in health policy and medicine, the paper considers what kinds of questions an analysis by loglinear modeling can address, and what kinds of answers it can obtain and how they may be sought. Unlike most previous expository accounts seeking to provide introductions to this field, this paper does not require a background from the reader in either regression or the analysis of variance. By a thoroughgoing use of odds ratios and higher-order odds ratios, it nevertheless provides a technically accurate account of the key concepts of higher-order interactions among variables, and of models being hierarchical.

Statistically more advanced readers are provided with a means of effectively expositing their loglinear modeling methods and conclusions to nonstatisticians; a number of footnotes are directed toward such readers.

Key-Words

Categorical Data - Contingency Tables - Cross-Classifications - Health Policy - Interaction Terms - Loglinear Models - Odds Ratios - Stepwise Selection
 

  • REFERENCES

  • 1 Agresti A. Analysis of Ordinal Categorical Data. New York: John Wiley and Sons; 1984
    Google Scholar
  • 2 Bishop Y, Fienberg S, Holland P. Discrete Multivariate Analysis. Cambridge: M. I.T. Press; 1975
    Google Scholar
  • 3 Fienberg S. The Analysis of Cross-classified Categorical Data. 2nd ed. Cambridge: M.I.T. Press; 1980
    Google Scholar
  • 4 Fleiss JL. Statistical Methods for Rates and Proportions. 2nd ed. New York: Wiley-Interscience; 1981
    Google Scholar
  • 5 Gokhale DV, Kullback S. The Information in Contingency Tables. New York: Marcel-Dekker; 1978
    Google Scholar
  • 6 Haberman SJ. Analysis of Qualitative Data. Vol. 1: Introductory Topics. New York: Academic; 1978
    Google Scholar
  • 7 Haberman SJ. Analysis of Qualitative Data. Vol. 2: New Developments. New York: Academic; 1978
    Google Scholar
  • 8 Plackett RL. The Analysis of Categorical Data. 2nd ed. London: Griffin; 1981
    Google Scholar
  • 9 Forthofer RN, Lehnen RG. Public Program Analysis, A New Categorical Data Approach. Belmont, CA: Lifetime Learning Publications; 1981
    Google Scholar
  • 10 Gilbert GN. Modeling Society. London: Allen & Unwin; 1981
    Google Scholar
  • 11 Hout M. Mobility Tables. Sage University Series on Quantitative Applications in the Social Sciences. Beverly Hills: Sage Publications, Inc; 1983
    Google Scholar
  • 12 Knoke D, Burke P. Log-Linear Models. Sage University Series on Quantitative Applications in the Social Sciences. Beverly Hills: Sage Publications, Inc; 1980
    Google Scholar
  • 13 Kawasaki S. An analysis of multivariate ordinal polytomous data using the loglinear probability model. J Math Sociol 1982; 09: 15-31.
    CrossrefPubMedGoogle Scholar
  • 14 Sullivan JL. Preface to [15].
    PubMedGoogle Scholar
  • 15 Mosteller F. Association and estimation in contingency tables. J Amer Statist Ass 1968; 63: 1-28.
    PubMedGoogle Scholar
  • 16 Jewell NP. On the bias of commonly used measures of association for 2 × 2 tables. Biometrics 1986; 42: 351-58.
    CrossrefPubMedGoogle Scholar
  • 17 Goodman LA. Some multiplicative models for the analysis of cross-classified data. Proc. 6th Berkeley Symp Math Statist Probability, I. Berkeley: University of California Press; 1972: 649-96.
    Google Scholar
  • 18 Goodman LA. The analysis of cross-classified data having ordered and/or unordered categories: association models, correlation models, and asymmetry models for contingency tables with or without missing entries. Ann Statist 1985; 13: 10-69.
    CrossrefPubMedGoogle Scholar
  • 19 Davis LJ. Exact tests for 2 × 2 contingency tables. Amer Statistician 1986; 40: 139-41.
    PubMedGoogle Scholar
  • 20 Dale J. Asymptotic normality of goodness of fit statistics for sparse product mul-tinominals. J Roy Statist Soc B 1986; 48: 48-59.
    PubMedGoogle Scholar
  • 21 Haberman SJ. A warning on the use of chisquared statistics with frequency tables with small expected cell counts. J Amer Statist Ass 1988; 83: 555-60.
    CrossrefPubMedGoogle Scholar
  • 22 Benedetti JK, Brown MB. Strategies for the selection of loglinear models. Biometrics 1978; 34: 680-86.
    CrossrefPubMedGoogle Scholar
  • 23 Goodman L. The analysis of multidimensional contingency tables: stepwise procedures and direct estimation methods for building models for multiple classifications. Technometrics 1971; 13: 33-61.
    CrossrefPubMedGoogle Scholar