Pooling Issues and Methods in Regression Analysis: Some Further Reflections

Recently, Brobst and Gates [3] provided comments on our article in which we discussed pooling issues and methods [1]. The main point Brobst and Gates appear to be making is that we should have used Wallace's test to determine whether pooling would be appropriate using weaker criteria than those inherent in the conventional homogeneity test. We will briefly comment on this aspect of our study, and discuss some other issues that we believe deserve further emphasis.