As part of the Center for Social & Behavioral Science Methods Series, CSBS partnered with the National Center for Supercomputing Applications to host a full-day workshop led by Dr. Jeff Levy, Assistant Instructional Professor at the University of Chicago Harris School of Public Policy, focused on how machine learning can be applied in social and behavioral science.
Using real-world data, Dr. Levy introduced core machine learning concepts including supervised learning, model evaluation, and interpreting results. Participants engaged in interactive, hands-on exercises using tools like Python and R.
View videos from the Workshop:
Access the dataset and code here.
Dr. Jeff Levy is an Assistant Instructional Professor at the University of Chicago Harris School of Public Policy and has extensive experience in machine learning and data programming for social and behavioral science.
What can attendees expect to learn in the workshop?
What are some of the core similarities and differences between the causal inference statistics commonly relied upon by social scientists, and the models used in machine learning? What are the foundational concepts and terminology one needs to get started in machine learning? And finally, how do we begin using ML tools alongside the ones we already know?
What are some exciting examples of machine learning applications in social and behavioral science?
“Predicting poverty and wealth from mobile phone metadata” – Blumenstock, Cadamuro, and On, 2015
Using data from cell phone usage and the observed wealth of a subset of users, the authors are able to predict wealth across all of Rwanda, generating estimates that match well with other sources of this information. This is exciting, since the areas we most often wish to study and plan interventions for are also often data-poor, but cell phone usage is ubiquitous and automatically collected.
“A generalizable and accessible approach to machine learning with global satellite imagery” – Rolf et. al, 2021
An efficient method for parsing satellite imagery into many desirable global measures, including forest cover, pollution, land use, poverty, and population density.
Tell us a fun fact about yourself.
As a grad student, I had a rule: at every conference, I had to ask at least one question. But at an IMF conference, just as I stood up to ask mine, Paul Krugman stepped in line at the microphone in front of me… and I completely chickened out!