Since the murder of George Floyd on May 25, 2020, and the racial justice protests that followed, many institutions, including the academy, pledged their support for policies and practices that combat on-going racial injustice. Social justice and anti-racism initiatives abound on college campuses, including programming, hosting speakers, and proposing required ‘diversity’ classes for all students. For all this rhetoric, college and university administrators have remained silent when it comes to diversity, equity, and inclusion practices as they relate to research. And yet, extant research documents the ways in which racial and gender biases have consistently shaped every level of research from the development of the research question, to the diversity (or not) of the sample, the availability of funding, and the probability of publishing. In this paper we focus on one aspect of the research process: the assembling (or not) of diverse research teams. We explore the benefits that diversity in research teams brings to the integrity of the data as well as the obstacles to both assembling a diverse research team and managing it successfully. Specifically, this paper focuses on the myriad ways in which diversity in research teams is treated as a set of boxes to check, rather than an epistemology that underscores positionality and power. We present a series of case examples that highlight the ways in which diversity, equity, and inclusion are successfully and unsuccessfully achieved in research teams, both in terms of outcomes and experiences. These case examples focus specifically on power relations along all forms of diversity, including race and gender as well as rank. The case examples also serve to unpack the ways in which research teams can rely on positionality as a tool for addressing power at three distinct levels: in conducting social science research generally, between the researcher and the “researched,” and among the research team itself.