Causal Inference

Estimating causal effects from observational and interventional data

I work on causal inference in the social network and content moderation context. My work evaluates the impact of policy interventions on the spread of unreliable content and misleading information online. Here are the projects I am pursuing in this direction:

  1. Identifying the Causal Effects of Twitter’s interventions on Trump’s tweets slides
  2. Identifying the Causal Effect of the Reduction of Feedback on the Sharing of Low-Quality News Online

Causal Effects of Interventions on Misinformation

I am interested in examining the impact of interventions taken by social networks to limit the spread of misinformation. In recent work with Jim Bisbee at NYU CSMAP, I extend the analysis of Sanderson et. al, 2021 to analyse the causal effects of warning labels and tweet removal on Twitter, Facebook, Instagram, and Reddit. The results were presented at the Stanford Internet Observatory’s first Trust and Safety Research Conference, 2022.

Causal effects are typically challenging to estimate in a network context due to the confounding between the cause and effect so I’m excited to be able to move the envelope on approaches to identify novel interventions and use synthetic counterfactuals to estimate the average treatment effect in different settings.