Geiger studies the relationships between science, technology, and society — not only how science and technology have substantial impacts on society, but also how they are social institutions in themselves. Much of his work focuses on machine learning, particularly in how user generated platforms like Twitter and Wikipedia are moderated. He has examined how values and biases are embedded in these technologies and how communities make decisions about how to use or not use them. He is a methodological pluralist and specializes in mixed methods, such as combining the rich and thick descriptions of cultural context that come from qualitative methods with large-scale quantitative and computational methods from Natural Language Processing. Geiger also studies the development of data science itself as an academic and professional field.