An Interrupted Time-Series Test of the Distributive Theory of Military Policy-Making

IN RECENT years several researchers have reported negative findings regarding the hypothesis that constituencies represented on the standing military committees in Congress benefit disproportionately from the distribution of military procurement expenditures.l Their findings suggest that whether or not a constituency is represented on the congressional military committees makes no difference in the amount of military procurement awards it receives. Faced with negative findings, political scientists have two options. They can either accept the findings and revise their substantive theory, or they can question the validity of the findings and devise other tests. Both approaches are consistent with the goal of producing scientific generalizations about political processes. However, when the negative findings are based on fairly crude tests and the theory that is called into question is more plausible than the available alternatives, we think the second approach should take precedence over the first. Thus the present paper concerns an alternative test of the hypothesis that constituencies represented on the military committees benefit from the distribution of military procurement expenditures.