Constructing Causal Networks
Jerald T. Hage, Barbara Foley Meeker · 2025
Finding causal networks requires a very different strategy from what has been discussed in Chapter 2 for testing whether or not there is an association between two state variables and between one causal network and some dependent variable. When testing , the researcher is interested in finding quantitative data collected under the most rigorous of temporal designs, preferably multiple time point quasi-experimental designs ( Campbell and Stanley, 1963 ). In contrast, when constructing causal networks, frequently the theorist should be interested in finding qualitative data collected with very different kinds of research designs from those that are typical in the field of the social sciences today and especially sociology. The researcher and the theorist, the tester and the constructer, are concerned with different kinds of data. The researcher is concerned with what is measurable and observable; the theorist interested in causality has to worry about what may be unmeasurable and nonobservable. The irony is that these unmeasurables suggest how the interventionist can change the behaviors of individuals or organizations.