The purpose of our article(Fornell and Yi 1992 [this issue]) was to identify the assumptions implicitly made under two-step approaches and to evaluate the extent to which they can be met. In total, four sets of assumptions were discussed. In a response to our article, Anderson and Gerbing(1992[this issue]) dismiss all of them as a result of misunderstandings and incomprehension on our part. Although we do admit to a very different understanding of basic statistical principles, we do not believe that Anderson and Gerbing have made a convincing case for the merits of the two-step approach.