New data sources and analytic techniques have enabled educational researchers to ask new questions and work to address enduring problems. Yet, there are challenges to those learning and applying these methods. In this chapter, we provide an overview of a nascent area of both scholarship and teaching, educational data science. We define educational data science as the combination of capabilities related to quantitative methods in educational research, computer science and programming capabilities, and teaching, learning, and educational systems. We demonstrate that there are two distinct—but complementary—perspectives on educational data science in terms of being both in education (as a research methodology) and for education (as a teaching and learning content). We describe both of these areas in light of foundational and recent research. Lastly, we highlight three future directions for educational data science, emphasizing the synergies between these two perspectives concerning designing tools that can be used by both learners and professionals, foregrounding representation, inclusivity, and access as first-order concerns for those involved in the growing community, and using data science methodologies to study teaching and learning about data science. We highlight the potential for the growth of educational data science within learning design and technology as situated with the broader data science domain and in education more broadly.