Integrating Voting Theory and Roll Call Analysis: A Framework

Scholars of legislative studies typically use ideal point estimatesfrom scaling procedures to test theories of legislative politics.We contend that theory and methods may be better integratedby directly incorporating maintained and to be tested hypothesesin the statistical model used to estimate legislator preferences.In this view of theory and estimation, formal modeling (1) providesauxiliary assumptions that serve as constraints in the estimationprocess, and (2) generates testable predictions. The estimationand hypothesis testing procedure uses roll call data to evaluatethe validity of theoretically derived to be tested hypothesesin a world where maintained hypotheses are presumed true. Wearticulate the approach using the language of statistical inference(both frequentist and Bayesian). The approach is demonstratedin analyses of the well-studied Powell amendment to the federalaid-to-education bill in the 84th House and the Compromise of1790 in the 1st House.