According to recent reports, the Educational Testing Service (ETS), which administers the Scholastic Aptitude Test (SAT), is seeking to change the SAT so that it will "give colleges a tool for bringing social equity into the admissions process." The proposed change computes a handicap for the SAT to identify "strivers," or students who score better than would have been predicted given a set of 14 background characteristics.

Unfortunately, this change is flawed, both conceptually and methodologically.

Conceptually, the SAT is supposed to provide information to admissions departments about the ability of students to succeed in college. It attempts to predict the grade point averages (GPAs) of students at the end of their first year in college. When trying to predict students' performance in college, it is not clear how it would be helpful to "adjust" for their background characteristics. We want to predict how well students will do given who they are, not how they would presumably be if their lives were different.

Attempting to control for students' backgrounds would only be useful if we wanted to diagnose why some students tend to be less prepared for college than others. This would give us ideas about problems that should be addressed or successes that should be emulated. But the purpose of the SAT is to guide the decisions of admissions departments, not correct inequities in the quality of K-12 education.

The proposed change to the SAT is also methodologically flawed because it assumes that the difference between the scores that we would expect students to receive given their backgrounds and the scores they actually receive is the result of their "striving."

But this is nonsense. The difference between what we expect students to score and what they actually score is simply the error in our prediction. There is no reason to assume that all or most of that error is caused by the gumption of individual students in overcoming their circumstances. The error could also be caused by failure to control for all of the background advantages or disadvantages that students possess or the mismeasurement of the background characteristics for which we do control.

For example, some students might score better than expected because they had the advantage of having great teachers. But because the proposed SAT change does not take teacher quality into account, test scorers would award "striver" points to a student who had great teachers, while another student who managed to receive the same SAT score despite lousy teachers might not be scored as a striver. Assuming that a better-than-expected score is the sole result of students' "striving" fails to recognize that there might be other factors that explain the students' scoresfactors unrelated to the student's ambition, such as having great teachers.

There is a simple way to test whether "striving" is something which college admissions departments should even be interested in. If "strivers" really are likely to do better than expected, we should be able to detect this in their freshman GPAs. The freshman GPA of striversthose who did better than expected on the SATshould be significantly higher than that of non-strivers with the same SAT. If strivers do not outperform non-strivers on their freshman GPAs, then college admissions departments should be uninterested in knowing who is a striver. It is unclear if ETS has ever performed this comparison, but it should before it considers changing the SAT.

Testing officials at ETS are probably aware of the flaws in this proposed change, so why are they interested in promoting it? One explanation might be that they are feeling political pressure to allow colleges to consider race in admissions without appearing to do so. Of the 14 background factors used to compute the handicap for students, race is likely to have one of the largest single influences on that adjustment.

Unfortunately, the "striver" SAT score could send a strongly patronizing and arguably racist message to test-takers. It is as if ETS would be saying to poor, black students, "You did very well for a person like you." This is a terrible and insulting way to promote greater social equity in admissions.

For the time being, it appears that this proposal is unlikely to be implemented. The College Board, the organization that contracts with ETS to produce the SAT, has rejected the proposed change. But the idea is sure to be proposed again in the future because it is attracting strong support within the education community.

The SAT's usefulness as an admissions tool would be seriously compromised by the proposed "strivers" score. Educators and parents who care about academic excellence should strive to see that it stays out of the test.