Access the full transcript for this episode

“So what we do in Data Feminism is try to synthesize a whole lot of feminist ways of thinking about the world, that have to do with questions of bias and oppression, that have to do with questions of sort of unequal power, and who gets to make choices about how to design systems — with these sort of really broad social questions, we try to apply them to data science as both a field and as a practice.”

Join us as we engage in a conversation with Lauren F. Klein, Associate Professor at Emory University and Director of the Digital Humanities Lab. Klein shares her unique journey from a background in comparative literature to pioneering the field of digital humanities, where she bridges the gap between computational methods and humanistic inquiry. We delve into her innovative projects, including her influential "Data Feminism" book and the "Data by Design" project, exploring how these works challenge traditional data science perspectives and emphasize the importance of context, history, and ethics in data visualization.

“The point that I'm trying to make in this project is that if we take this historicized, almost literary and critical, humanistic lens to this history, we can see how the people who were designing data visualizations were either asking very similar questions to the kinds of questions about responsible data visualization that we're asking today, or they weren't. And because of that, we can see how their visualizations — far from being some sort of neutral representation of data — in fact, represented a certain policy sort of unreflective politics that I think we also need to be able to identify again, so that we don't reproduce that in the present.”

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.”

Access the full transcript for this episode

“You can be hoodwinked with data in the same way that you can be hoodwinked by a car salesman. And so the idea of [Calling B******t] was to step away from all the details of the black box: that's the statistical procedures, the algorithms, etc. (Not to say that we don't pay attention to what we do.) But the idea is to really pay attention to the input data that's coming in—to think about things like selection bias—to think about where that data is coming from.”

Join us in our Season 7 finale as we host Jevin West, an associate professor at the University of Washington and a co-founder of the Center for an Informed Public. Dive into a deep discussion about the intersection of data science and misinformation, the challenges of big data, and the ethical considerations that come with it. Jevin shares his experiences from the early days of data science programs, his insights on combating misinformation through education, and the evolution of his course and book, "Calling B******t." Whether you're a data science professional or a student, listen in to explore how data science education can empower us to make informed decisions and foster a more truthful society.

“One of the most important skills that we're going to want to enhance more and more is humaneness…things like being able to ask questions, to sort of work through logic to really tease out things, like correlation versus causation. Machines don't tend to do so well [with those things]—they don't have access to the physical world. That's one of their weaknesses. So you want to lean into your strategic advantages as humans…maintain that humaneness by doing things that machines can't do.”