In a recent article in this journal, Naglieri and Ford (2003) claimed that Black and Hispanic students are as likely to earn high scores on the Naglieri Nonverbal Ability Test (NNAT; Naglieri, 1997a) as White students. However, the sample that Naglieri and Ford used was not representative of the U.S. school population as a whole and was quite unrepresentative of ethnic subgroups within that population. For example, only 5.6% of the children in their sample were from urban school districts, and both Black and Hispanic students were relatively more likely to be from high socioeconomic status (SES) families than were White families. Further, the mean and standard deviation in the sample differed significantly from the NNAT fall test norms, which were based on the same data. Finally, White-Black and White-Hispanic differences reported by Naglieri and Ford were smaller than those reported by Naglieri and Ronning (2000a) in a previous analysis of the same data set in which students were first matched on several demographic variables. For these reasons, I argue that the authors' claims are not supported by the data they present. Putting the research to use: Do nonverbal ability tests really identify equal proportions of majority and minority students? Although recent evidence presented in this journal suggests that the NNAT achieves this goal, I argue that the data do not support the claims that were made. Those who have depended on these claims to implement selection systems need to look carefully at the data they have collected. When doing so, users should look not only at the number of minority students identified, but also at the number of students who have subsequently succeeded in attaining academic excellence. The larger issue of the proper role of nonverbal, figural reasoning tests in the selection of academically gifted students needs much closer scrutiny than it has received.