Using Bayesian Scaling to Examine Gender in a Polarizing Congress

• In Event: Poster Session: Women and Politics Research

Abstract

The conventional wisdom is that women in Congress are more liberal than their male co-partisans. This paper adopts new methodological approaches to re-evaluate this expectation. I expect that women vote more liberal than their male colleagues, especially on women’s issues, but that this trend has diminished over time due to polarization. Using more than 40,000 roll call votes across 40 years in the House and 25 years in the Senate, this project uses Bayesian item response theory to scale legislators on 21 policy topics. I find that gender differences vary across chamber, party, issue, and time. For House Republicans, there has been a decline in gender differences since Gingrich’s reign, and polarization has made today’s House Republican women indistinguishable from Republican men, even on women’s issues; Republican women in the Senate follow a similar trend. For House Democrats, following the 1992 “Year of the Woman,” I find women were increasingly more liberal than their men co-partisans, and today’s House women remain more liberal than Democratic men. However, for Senate Democrats, there are sparsely any gender differences across time. Across both parties and chambers, gender differences are most consistent for legislation on women’s issues, civil rights, and health. This project provides new insights into how women legislators differ in voting across chamber, party, issue, and time. Ultimately, this reveals what influence gender identity can still have on the legislative process in today’s polarized Congress.

Author

• Sarina Rhinehart