In recent years, the loglinear model has been proposed and used for analysing frequency data in multidimensional contingency tables. The primary focus of the literature has been on model building and only secondarily on hypothesis testing and estimation. This paper extends Kennedy's (1988) description by presenting post hoc procedures for statistically evaluating treatment effects, contrasts, and confidence intervals. It illustrates methods for main effect and interaction contrasts and pays special attention to odds ratios and their interval estimates. Procedures are described for treating the variables as interdependent and for the case where there are independent and dependent variables—both ordered and unordered.
AgrestiA. (1984). Analysis of Ordinal Categorical Data.New York: Wiley.
2.
AitkinM. (1980). A note on the selection of log-linear models.Biometrics, 36, 173–8.
3.
AltmeyerB.K., LockeB.J., GriffinJ.C., RickettsR.W., WilliamsD.E., MasonM., & StarkM.T. (1987). Treatment strategies for self-injurious behavior in a large service-delivery network.American Journal of Mental Deficiency, 91, 333–40.
4.
BrownM.B. (1981). BMDP Statistical Software.Berkeley, CA: Univeristy of California Press.
5.
CochranW.G. (1954). Some methods for strengthening the common chi-square tests.Biometrics, 10, 417–51.
6.
DunnO.J. (1961). Multiple comparisons among means.Journal of the American Statistical Association, 56, 52–64.
7.
EverittB.S. (1977). The Analysis of Contingency Tables.London: Chapman & Hall.
8.
FienbergS.E. (1980). The Analysis of Cross-classified Categorical Data (2nd ed.). Cambridge, MA: MIT Press.
9.
FienbergS.E. (1985). Comments on and reactions to Freedman, statistics and the scientific method. In MasonW.H., & FienbergS.E. (Eds.), Cohort Analysis in Social Science Research: Beyond the Identification Problem (pp. 371–83). New York: Springer-Verlag.
10.
FleissJ.L. (1981). Statistical Methods for Rates and Proportions (2nd ed.). New York: Wiley.
11.
FreedmanD.A. (1985). Statistics and the scientific methods. In MasonW.H., & FienbergS.E. (Eds.), Cohort Analysis in Social Science Research: Beyond the Identification Problem (pp. 343–66). New York: Springer-Verlag.
12.
FreedmanD.A. (1987). As others see us: A case study in path analysis.Journal of Educational Statistics, 12, 101–28.
13.
GoodmanL.A. (1970). The multivariate analysis of qualitative data: Interactions among multiple classifications.Journal of American Statistical Association, 65, 226–56.
KennedyJ.J. (1988). Applying log-linear models in educational research.Australian Journal of Education, 32, 3–24.
16.
MarascuiloL. A., & BuskP.L. (1987). Loglinear model: A way to study main effects and interactions for multidimensional contingency tables with categorical data.Journal of Counseling Psychology, 34, 443–55.
17.
MarascuiloL.A., & LevinJ.R. (1970). Appropriate post hoc comparisons for interaction and nested hypotheses in analysis of variance designs: The elimination of Type IV errors.American Educational Research Journal, 7, 397–421.
18.
MarascuiloL.A., & McSweeneyM. (1977). Nonparametric and Distribution-free Methods for the Social Sciences.Monterey, CA: Brooks/Cole.
19.
MarascuiloL.A., & SerlinR.C. (1988). Statistical Methods for the Social and Behavioral Sciences.New York: W.H. Freeman.
