Access the full transcript for this episode

“We're kind of in an early phase among most social scientists, trying to figure out what's new here, what's different, and how to integrate it with our standard social science methodological concerns, which I don't think we should abandon. Thinking about the relationship between theory, concept and measurement. For example, that's one of the things that social scientists bring to the table in data science projects: thinking about questions of representativeness, generalizability, and questions of causal inference.”

Welcome to the season 8 premiere! In this episode, we sit down with David J. Harding, a professor in the sociology department at UC Berkeley. David shares his unique academic journey in sociology and data science, emphasizing the integration of social science methodologies with data science tools. He discusses his work on poverty, inequality, and incarceration, and the challenges of using administrative data in research. The conversation delves into future directions for his research on adolescents and urban communities, the importance of bridging social science and data science education, and strategies for creating inclusive classroom environments.

“A standard complaint about running and estimating models in the social sciences is that we make a lot of assumptions, and then don't have the ability to test them. Then right along comes the kind of more machine learning type workflow, which allows us to learn what the model should look like from a portion of the data, and then test it and validate it on another portion of the data. I think social scientists should be building that sort of workflow into our normal work process all the time.”

This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit datascienceeducation.substack.com