This paper considers what it could mean to say that epistemologies and methodologies are racially just or unjust. It has been argued that this has nothing to do with whether an individual researcher is racist: he or she could be anti-racist but still use an epistemology that is racially biased. To explore this issue, some pioneering research by Allison Davis, an influential African-American scholar in the 1940s and 1950s, is examined. This research dealt with the effects of social class cultures on academic achievement in the United States. While Davis was well-known at the time, his work in this field is now largely forgotten. Two senses of 'racially-just' are examined in this paper, relating: to the intrinsic character or origin of epistemologies and methodologies; or to the implications for and likely consequences of their use. Davis's investigations employed methods and assumptions that are rejected by many educational researchers today. I will address the question of whether these can be criticised as racially unjust; and, if so, on what basis. This evaluation will lead to a consideration of the broader question of the terms in which it is appropriate to assess educational research, and in particular the epistemologies and methodologies on which it depends.