Sarah Kelley is a co-program area director of data science and senior data scientist whose research focuses on applying data science techniques— especially natural language processing, computer vision, and machine learning—to answer questions related to social science and education.
She has worked on a diverse set of social science problems, from using social media data to explore public conversations around child abuse in Haiti, to using big data and geospatial statistics to understand drivers of the opioid crisis, to using natural language processing to gain insight into political polarization online. Sarah is particularly excited about the potential of data science methods to augment traditional research methods. She seeks to make her work accessible to general audiences through data visualization. She also hopes to support quantitative researchers by using data science methods to create and integrate rich data sets, providing data access through application programming interfaces, and linking data sets using machine learning methods.
Previously, Sarah was a data scientist at the American Institutes for Research, where she led projects focused on applying computer vision algorithms in education contexts, developing large-scale data processing pipelines, and using machine learning techniques to improve predictive modeling. She holds a bachelor’s degree in sociology from Yale University and a master’s degree in data science from the University of California, Berkeley.