Back to the program

Topic-Contributed Paper Session

Statistical Innovation with Text as Data in Causal Inference Contexts: Emerging Trends and Future Directions

Mon, Aug 5, 2:00 PM - 3:50 PM

Luke MiratrixChair

Section on Text Analysis

4 Presentations

ChatGPT vs. Linear Regression: Assessing the Efficacy and Accuracy of Large Language Models for Automated Essay Scoring

2:05 PM - 2:25 PM

Youngwon Kim (Harvard Graduate School of Education)

New Experimental Designs with Text as Treatment: Can a Large Language Model be Too Large?

2:25 PM - 2:45 PM

Alexander Volfovsky (Duke University)

Text as a High-Dimensional Causal Confounding Variable: An Empirical Framework for Evaluation

2:45 PM - 3:05 PM

Katherine Keith (Williams College)

LLM annotations in the Social Sciences: A General Framework of Using Predicted Variables in Statistical Analyses

3:05 PM - 3:25 PM

Naoki Egami (Princeton University)

Discussant

Reagan Mozer (Bentley University)

↑