Date of Thesis



This thesis explores the relationship between coalition structure between Congress and the President and its effect on legislative output. Situated in macro-political study, my research expands on my prior work in this area of study where I found a significant relationship between unified government and high political polarization (an effect I named the Interaction Effect) and the number of pieces of significant legislation passed using David Mayhew's process of analyzing significant legislation from Divided We Govern. Thus, my central question for this study is does government structure and increased levels of political polarization have an effect on legislative productivity when productivity is quantified using different measures? To answer this question, I utilized three different measures of legislative productivity to test the effect of the Interaction Effect. First, I used David Mayhew's measure of significant legislation to extend my prior research. Second, I obtained and customized Sarah Binder's gridlock score to incorporate the legislative agenda into my analysis of productivity. Finally, I incorporated Clinton and Lapinski's legislative accomplishment measure to quantify legislative productivity over a large span of congressional history. After tying the three datasets together using three categories of independent variables (unified/divided government, polarization, and public mood), I ran multivariate linear regressions in each author's dataset. I found that high polarization, conditional on divided government, does not affect significant legislative output as defined by Mayhew or Clinton and Lapinski. However, I found that high polarization under divided government does affect Sarah Binder's gridlock score: gridlock significantly increases under divided government and high polarization. As for the Interaction Effect, I found that high political polarization, conditional on unified government, had a significant relationship with all three measures of legislative productivity. However, my results for Mayhew and Clinton and Lapinski were impacted by the presence of Stimson's public mood variable. Thus, my conclusions on these results are limited. Overall, the results show that high polarization, conditional on divided government, does not have an effect on legislative productivity when measured absolutely. However, high polarization, conditional on divided government, does have a significant impact on the instance of legislative gridlock.


legislative productivity, divided government, unified government, interaction effect, political polarization, significant legislation, macro-politics

Access Type

Honors Thesis (Bucknell Access Only)

Degree Type

Bachelor of Arts


Political Science

First Advisor

Scott R. Meinke