Stanford Sporadic Psychometrics Seminar

Stanford Sporadic Psychometrics Seminar (SSPS) is a virtual seminar run 3-4 times per year focused on current research in psychometrics. The seminar series is organized by Prof. Ben Domingue. Future communication about the SSPS seminar will be handled via infrequent communications from this listserv. If you would like information about future seminars, please subscribe to the mailing list.

Upcoming Seminar

Measuring Issue Specific Ideal Points from Roll Call Votes -- Sooahn Shin (Harvard)

Ideal points are widely used to measure the ideology and policy preferences of political actors, ranging from voters and legislators to sovereign states. Yet, an outstanding challenge is to estimate ideal points specific to a single issue area. The conventional approach resorts to subsetting voting data, which results in the loss of valuable information and makes comparisons across issue areas ambiguous. To address this, I introduce IssueIRT, a hierarchical Item Response Theory (IRT) model that uses roll-call votes and user-supplied issue labels to estimate issue-specific axes, each running from left to right positions. This approach first estimates multidimensional ideal points using all available voting data, which are then projected onto issue-specific axes to generate single-dimensional, issue-specific ideal points. I demonstrate that IssueIRT effectively captures issue-specific voting behaviors through simulations and applications studies. Specifically, I show cross-party, regional division over monetary policy in the US House of Representatives during the depression of the 1890s. Finally, I show that polarization in Congress has increased overall from 1979 to 2023, though the degree and pattern of change vary across 33 separate issues. The R package issueirt is available for implementing IssueIRT. Full draft available here: https://sooahnshin.com/issueirt.pdf.

Time: Nov 5 @11 AM Pacific Standard Time

Zoom Link: https://stanford.zoom.us/j/92138618572?pwd=V7PSYN9aLzBH4upyG0ubKoLbYH8R1c.1. Password: 706364

Add this event to your calendar via this link and feel free to share this!