Date of Award

Spring 5-15-2021

Author's School

Graduate School of Arts and Sciences

Author's Department

Political Science

Degree Name

Doctor of Philosophy (PhD)

Degree Type



This dissertation is composed of three papers on interpersonal political communication. In the first, entitled "Measuring Agenda Setting in Interactive Political Communication," I seek to measure the agenda setting that occurs within political debates, deliberations, and discussions. In these settings, actors must negotiate who is speaking and what is being discussed, thus actors have the opportunity to influence the agenda as it develops in real time. I measure this form of agenda setting drawing on topic modeling research from computer science. Validation exercises show that the model accurately identifies topic shifts and infers coherent topics, and three empirical applications validate the agenda-setting measure within different political settings. The second paper, entitled "The Consequences of Interparty Conversation on Outparty Affect and Stereotypes," evaluates whether interpersonal communication in the form of conversation can improve the heightened polarization that characterizes the American political climate today. I conduct two experiments where I manipulate whether a pair of opposing party members converse with each other or not and whether they discuss a political or non-political topic. To conduct these experiments, I designed a software called "Chatter" by which participants can have real time, written conversations online. I find that both political and non-political conversation can improve how partisans feel about opposing party members. The third paper of the dissertation, entitled "Improving Balance in Experiments with Social Interaction with Blocked Cluster Designs," discusses the benefits of blocked cluster experimental designs. However, the literature lacks guidance on how to implement this design when researchers bring participants together for interpersonal communication, like in the dissertation's second paper. Therefore, I propose an algorithm that constructs a blocked cluster design in this setting, allowing researchers often constrained to small sample sizes to ensure balance and improve their experiment’s efficiency and power. I demonstrate the benefits of the algorithm and experimental design with simulated data, data from a published experiment, and original data from the dissertation.


English (en)

Chair and Committee

Jacob M. Montgomery

Committee Members

Betsy Sinclair, Steve Smith, Taylor Carlson, Samara Klar,