The integration of data, machine learning, and artificial intelligence (AI) in education has ushered in a new era of data-driven decision-making, particularly in efforts to enhance equity and inclusion within educational institutions. However, despite the significant increase in using data for these purposes, there remains limited empirical research exploring the effectiveness and potential unintended consequences of such practices. This dissertation aims to bridge this gap by presenting a series of studies that provide empirical evidence on the backlash to using data for Equity, Diversity, and Inclusion (EDI). Moreover, it offers recommendations on effectively using data for EDI while minimizing future backlash. The dissertation explores several research questions, including how learning analytics dashboard research is being used to improve justice, equity, diversity, and inclusion (JEDI) in higher education, as well as the potential maintenance or exacerbation of inequitable outcomes in this context. It also delves into scalable measures for equality and inclusion in courses and examines how educators might use equality and inclusion metrics to make decisions regarding EDI. Additionally, the dissertation investigates the impact of explaining complex algorithms on educators' attitudes and intent to use AI-powered tools, shedding light on the differences in educator perceptions of tools using simple versus complex algorithms. By addressing these questions, the dissertation aims to contribute to the understanding of using data for EDI and to provide practical insights for educators and educational institutions. Through these studies, the dissertation seeks to provide valuable insights into the challenges and opportunities associated with using data for EDI in education. By examining the implications of data-driven decision-making for equity and inclusion, it aims to offer actionable recommendations for educators and educational organizations to navigate this complex landscape effectively.