Beyond Black: Biracial Identity in America (review)

In recent years there has been a flood of articles and books about multiracials. Unfortunately, nearly all these studies are based on small convenience samples. One consequence of this dearth of generalizable findings is that when the federal government recently revised its guidelines for the collection and analysis of racial data, it found little clear direction from the literature on multiracials.

In Beyond Black:Biracial Identity in America, Kelly Ann Rockquemore and David L. Brunsma identify this gap in the literature and seek to fill it. The foundation for their study is a methodologically diverse research design that collects data from 191 young adults with one white parent and one black parent.

After discussing how black biracials have historically been classified in the U.S.,Rockquemore and Brunsma review 31 empirical studies of black/white biracials. They conclude that although each of these works presents interesting findings, they all fail to provide a coherent understanding of biracial identity, primarily because they rely on small convenience samples.

Beyond Black offers a different approach. First, Rockquemore and Brunsma conducted in-depth interviews with 14 undergraduates from a private Catholic university. Themes from these interviews and the literature on biracial identity were used to construct a survey instrument. In phase 2 of the project, survey data was collected from 177 undergraduates who attended one of two selected schools. Last, in-depth interviews were conducted with 25 of the Phase 2 survey respondents.

In three chapters, Rockquemore and Brunsma use these data to assess biracial identity profiles, the effect of social networks and parental socialization on biracial identity, and the relationship between complexion and identity. The biracial identity profiles are perhaps the book's most important contribution. Rockquemore and Brunsma argue that much previous work has wrongly assumed that all black/white biracials possess or should possess the same racial identity. By contrast, their data suggest that black/white biracials think of themselves as biracials (border identity), always white or always black (singular identity), sometimes white and sometimes black (protean identity), or raceless (transcendent identity).

Despite these contributions, Beyond Black disappoints on both substantive and methodological grounds. Substantively, the book fails to deliver on its promise of informing the federal debate about racial data. The concluding chapter revisits existing arguments about the continuing and compelling need for racial data and uses its ownfindings to argue that biracials cannot be collapsed into any single racial category. Yet the book makes no specific recommendations about how data on multiracials should be collected or coded.

Methodologically, Beyond Black disappoints because it fails to deliver on its promise of providing generalizable results. First, it is not always clear how the authors get from their survey data to their findings. There is no discussion of how some key variables are constructed (e.g., four identity types), and very few survey results are presented. It is understandable that the authors would not want to fill a book aimed at a general audience with complex statistics, but surely even the general reader would appreciate seeing an occasional table or graph to support the book's conclusions.

Further, Beyond Black disappoints methodologically because its data are almost certainly less generalizable than the authors claim. The survey sample was drawn from official school lists of all students who identified themselves as black or some other race on their application, or who failed to provide a race. Valid survey data was obtained from 177 of these 4,532 students. Rockquemore and Brunsma conclude that their relatively small sample size could be due to (1) there being a small number of black/white biracials at the two schools, (2) a large number of eligibles opting out of the survey, or (3) a nontrivial number of students with one white and one black parent identifying as white on their application. Unfortunately, the authors have no sound basis for arguing that their sample is a random draw from...