The Princeton R Group will be hosting its final event of the semester with a set of lightning talks! These are informal talks where people across policy and the social sciences can come showcase an interesting application of R for their work. Speakers will present for 10-15 minutes and take questions at the end of their talk. The goal is to spark additional ideas for how R can be used to further social science and policy research.

Speaker #1: Zac Lim (SOC)

Zac is a visiting 5th year PhD candidate in Sociology from Harvard. He works on using text data to study behavior. Specifically, he works with a large dataset of around 1M articles in the field of education to understand how researchers in the field navigate methodological disputes. He is currently working with Professor Brandon Stewart in Sociology on research involving large-scale text data.

Quantitative sociological semantics and conceptual ethics: Concepts are tools that we use to organize our experiences, communicate our intentions, and coordinate action. Just as how archaeologists study tools to tell us about people living in different times, I study concepts as they are used in specific communities as a window into their social milieu using various natural language processing tools and techniques. As an initial case study, I examine the use of philosophical language in the field of education, especially in the context of methodological debates.

Speaker #2: Muhammet Emre Coskun (SPIA)

Emre is a Research Specialist at the Niehaus Center for Globalization at SPIA. He is a quantitative social scientist and data scientist. He received his PhD in Public Policy from Georgia State and Georgia Tech (joint). He uses R for most parts of his research, along with Stata. He is currently working on a research project exploring socio-economic factors underlying the rise of populist right political parties in Europe.

Running 100 imputations on HPC Clusters: I will demonstrate how to run multiple imputation (MI) analyses in R using parallel computing on HPC clusters. Multiple imputation is used to obtain less biased estimations compared to single imputations by accounting for uncertainty in imputed values of previously missing data. The process can take a long time depending on the data size. I will demonstrate how to run MI in parallel on HPC clusters, reducing time significantly.

Speaker #3: Jamie Caldwell (HMEI)

Jamie is an Associate Research Scholar at the High Meadows Environmental Institute. She is a quantitative ecologist who studies infectious disease dynamics in wildlife and people. Her research investigates climate and ecological drivers of disease transmission and uses those relationships to predict outbreaks in the future. She was previously a postdoctoral fellow at Stanford University where her research focused on climate-driven patterns of mosquito-borne diseases in South America and Sub-Saharan Africa. She received her Ph.D. in 2017 from the University of Hawai‘i at Mānoa (UHM) where she studied infectious disease dynamics in coral reefs.

Creating effective Shiny apps: Online dashboards are a great way to visualize and communicate research with a variety of audiences. Shiny apps, developed in R, are one such resource that allows you to build online applications, and, they are very easy to create! In this lightening talk, I'll give a brief overview of how you can get started developing Shiny apps and walk through an example of an application I built to predict disease risk in coral reefs (https://coraldisease.com/).

*Lunch will be provided!*  Please RSVP below by Tuesday November 28!

Please contact tigeRs co-organizers Kim Kreiss ([email protected]), DDSS Grad Fellow, or Angela Li ([email protected]), 4th-year Social Policy and Sociology PhD, with any questions.