A COMPUTER-PROGRAM FOR THE ANALYSIS OF OVER-DISPERSED COUNTS AND PROPORTIONS

Over-dispersed binary and count data occur frequently in many fields o f application. Examples include occurrence of cavities in one or more teeth, and development of tumors in one or more animals of a litter. M ethods of statistical analyses that ignore correlation between observa tions underestimate the standard errors. Consequently, coverage propor tions of confidence intervals and significance levels of tests are dis torted. To implement methods for the analysis of correlated binary or count data requires a level of sophistication for data analysis such t hat one can specify a model for over-dispersion and the correlation be tween observations. To analyze the over-dispersed binary or count data , one could postulate a specific statistical model and use maximum lik elihood methods for the estimation of parameters. However, it may be p referable to employ an approach that does not rely on modeling because the true model is hard to know with confidence. Rao and Scott (J.N.K. Rao and A.J. Scott, Biometrics 48 (1992) 577-585)y and Scott and Rao (A.J. Scott and J.N.K. Rao, submitted for publication, 1995) proposed simple methods for analyzing correlated binary and count data exhibiti ng over-dispersion relative to a binomial and homogeneous Poisson mode l. This paper presents the SAS program to implement their methods to a nalyze over-dispersed binary and count data. To demonstrate the implem entation and the usefulness of their methods, we present an applicatio n involving sensitivity of a monoclonal antibody and the number of mam mary tumors developing in rats. (C) 1997 Elsevier Science Ireland Ltd.