Breaking down data by race is one of the most common forms of cross-tabulation researchers look for, and, unfortunately, also one of the most complicated. Most government statistical agencies in the US use the racial categories defined by the Census Bureau, but those categories have changed dramatically over the years. For example, early censuses recognized only two races (white and black), and race designations were made by census-takers who gathered the statistics. In contrast, the 2000 Census had six main race categories (American Indian or Alaska Native, Asian, Black or African American, Native Hawaiian or Other Pacific Islander, White, and Other), determined race solely by self-identification (a "white person" was anyone who checked the box labelled "White"), and was the first census to allow people to check multiple race boxes.
Another factor which many people find confusing is that of Hispanic origin. With a few historical exceptions, the Census Bureau has not recognized a racial category for Latinos. Instead, in most recent decades there is a separate yes/no question regarding Hispanic origin. So anyone who is "of Hispanic origin" is also counted under one of the main racial categories.
It's always best to double-check the subject definitions used by the Census Bureau in each separate census publication. Those definitions are usually available in the back of the book (for print products) or in separate technical documentation files (for digital products).