The two sets of tables above - one comparing National Research Council occupational data to the NIH workforce and the other comparing 1990 Census occupational data to the NIH workforce - illustrate how the choice of baseline data can shift the underrepresentation index (URI) for general biological scientists at NIH. Note the difference between the two data sets in the URI for Hispanics. However, it must be emphasized that neither of these "pools" adequately represents the "populations" from which NIH recruits its scientists. For example, for black and Hispanic scientists, NIH often needs to look at M.D.s who might be interested in research. These data could also be misleading because they represent only working scientists as opposed to recruitable researchers.