Rejoinder to Dykstra and Kousser

Abstract

In our article “Alchemist’s Gold”, we tried to make three main points. First, a large number of models of individual behavior can fit observed ecological patterns, and these models entail different assumptions about the nature of the individual relationship within and across the ecological units. This topic has been most impressively developed by Gudmund Iversen (1981). Second, there is nothing in the ecological data or the various types of ecological analysis that informs the researcher about how to choose among the many dissimilar models. All the models are compatible with the ecological data, and we have no way of knowing whether a particular inference about individual behavior is quite accurate or extremely misleading. Because of the inherent ambiguity over the unobserved individual behavior within ecological units, we made a third point of recommending techniques for reducing the amount of uncertainty surrounding estimates.