Doctoral Dissertation Research: Partisanship, Gender, and Policy Frames

University Of Oklahoma Norman Campus, Norman OK

Investigators

Abstract

Despite gains made over the past three decades, women remain the most underrepresented group in Congress relative to the general population. The purpose of this project is to better understand how women in Congress represent women generally. To this end, the project examines the relative influence of partisanship and gender on the content of congressional communications about women's issues and voters' responses to those messages. To accomplish this task, the project analyzes the content of more than 40,000 congressional newsletters about women's issues and uses the findings to generate a survey experiment to test how men and women in the electorate interpret messages about women's issues differently based on the gender of the messenger. This project contributes to an important, ongoing debate about representation --do women substantively represent women, are all members purely partisan, or is it something in between? And, how are voter opinions about women's issues affected by the gender of the person who presents them? This research will push forward our understanding of the character of women's representation in Congress, and will broaden the knowledge held by both voters and members of Congress concerning the quality of descriptive and substantive representation of women on women's issues. The purpose of this project is to evaluate the relative influence of partisanship and gender on the content of congressional communications about women's issues and voters' responses to those messages. To accomplish this task, the project uses structural topic modelling to identify how gender and partisanship influence communication about women's issues in congressional newsletters. It utilizes these findings in a survey experiment that randomizes the gender of the representative delivering the content to test how men and women in the electorate interpret messages about women's issues differently based on the gender of the messenger. This project presents several distinct contributions to the field of American politics generally, and scholarship on representation and political communication in particular. Methodologically, it utilizes an innovative methodological tool --structural topic modelling--to analyze a large corpus. It compares the content of these texts between categories available in the metadata, including gender and partisanship. The project then employs these findings in a survey experiment to assess the effect of the differing content on the electorate. This project will be the first scholarship, as of this date, to use this method to identify distinct patterns of communication between male and female members of legislatures and to assess the impact of these differences on the electorate. This contribution, and future research that utilizes a similar methodological approach, will allow us to better understand the character of women's representation and will facilitate a broader understanding of the content and impact of political messages generally. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.