Skip to content

Experimentation

What interventional mechanisms do we need to enable citizen science (especially citizen social science!)?

Backstory!

A few years ago now, Eytan Bakshy designed a domain-specific language for online field experiments, PlanOut. I had the great fortune to intern with him at Facebook, where we investigated possible applications of static analysis to the experimental design space. While we started out focusing on generating contrasts and inferring probability of treatment assignment, we found there were errors and threats to validity of the experiments that only existed at the intersection of programs and experiments. At various points in this project, I worked with UMass folks Emery Berger and Eliot Moss on the PL side of things, and had a wonderful resource in causal inference in David Jensen. I presented this work at NEPLS in 2016 and OOPSLA 2019. It was recognized by my professional organizations via a SIGPLAN research highlight in 2020 and a CACM Research Highlight in 2021 and a grant proposal based on this work was accepted for funding by the NSF as part of the Formal Methods in the Field program. After moving to Northeastern, Chris Martens joined the project as co-PI and has been instrumental in the formal methods aspects of this work.

Current work

I am currently working with several collaborators on research problems related to programmatically-defined experiments, including but not limited to: robustness checks for p-hacking, type systems for interventions, integration of experiments with passive/observational data collection mechanisms and subsequent data fusion, and inference over proxy variables in systems that do not allow for intervention.

How to get involved

I are interested in developing collaborations with empiricists/domain experts across a variety of computational backgrounds and am specifically looking to apply the tools we have been developing to a working scientist's workflow. I am especially interested in working with faculty and industrial collaborators and can typically fund Northeastern undergraduate coop students. This line of work is particularly central to my research and I am happy to explore other collaboration opportunities.

I am also interested in chatting with anyone who is teaching a research methods course to discuss how this work might be useful to you.

Please reach out for a meeting with me if you'd like to have a chat.