9538.1	Assessment
Research Question (Clear, unambiguous)	"In the United States, does race affect choice of college major/type of degree conferred?"  Okay, this is a reasonable question, although it will require considerable specification to become meaningful.
Relevant Literature	The hodgepodge of things mentioned do not really represent what we mean by the relevant literature.  This should point to the scholarly work that informs us about the research question.
Causal Interpretations (clear, all parts defined, mechanisms, controls, plausible)	By invoking "perception", the first causal interpretation becomes obscure and the second one seems irrelevant (obviously, other things influence choice of major, but simply choosing one does not make much sense).
Data (variables, sample, comparison)	The NCES data will allow you to look at the distribution of graduates by major and "race" (crudely coded) across American colleges.   It will not, however, get you far in explaining people's choices of majors, as that would require individual level data.  (The two additional bullet points you supply make no apparent sense.)
Research Value	To the degree that the research contributes new findings, which depends on past research, it would seem inherently valuable.  The reasons specified for the project's value seem rather weak, however.
Overall	The general goal of the project seems good.  Its prospective value depends on what past research has already accomplished and the availability of data to pursue the project's intended goals.