Standards Without Substance: Resisting Minimum Test Scores for College Admission

In the wake of declining enrollment, University of Tennessee Chancellor, Loren Crabtree, has suggested abandoning an idea floated by the school’s former President, which, if implemented would require applicants to receive an ACT score of at least 26 (out of 36) in order to gain admission. While rethinking such a score cutoff makes sense, the best reason for junking the plans of ex-President John Shumaker is not simply to fill classrooms. Rather, UT should scuttle the idea because such a minimum score would be arbitrary, unrelated to the academic quality of the school’s enrollment, and would especially hurt applicants of color and low-income applicants who were otherwise qualified, but whose scores disproportionately fall below the minimum.

Even the company that writes the ACT says that setting a minimum score for enrollment is a flawed practice. Indeed, the test makers insist their exam should never be used to deny admission to college. One reason the test-makers oppose minimum cutoffs is because of the margin of error known to exist on the ACT. Score swings of 2-3 points can occur purely by chance due to inherent inconsistencies in standardized testing, and thus, score gaps of 4-6 points between any two students should be considered entirely random, and not indicative of different levels of ability. Setting a minimum score for admission to college ignores this basic truth, and would result in the exclusion of persons whose academic talents were utterly indistinguishable from many of those admitted.

Supporters of raising the minimum score say doing so will help UT attract the “best and brightest,” yet ACT scores have little to do with picking the best students. According to all available evidence, high school grades are better predictors of success in college than the ACT. Furthermore, even the testing company admits that their exam, at best, can only partially predict freshman grades, while having nothing to do with overall college GPA or graduation rates.

Studies have also found that the ACT is especially bad at estimating the academic abilities of people of color, who tend to score 3-5 points lower on average than whites even when their grades would indicate they are equally capable. In fact, roughly 75 percent of black student performance in their first year is totally unrelated to their ACT scores. Furthermore, the ACT underpredicts college graduation rates for students of color, indicating that the test is incapable of picking and choosing who can and cannot succeed. If a minimum score of 26 becomes the policy at UT, the result would be the exclusion of many persons of color who would have done well as students.

A primary reason for the racial score gaps mentioned above is the role of economic background in determining test scores. Those who attend affluent high schools, who are disproportionately white of course, receive higher level preparation and often outperform others on tests, even when they aren’t any better as students. According to 2002 data, the average score for poor students is below 18, and even students from families with annual income of $50-$60,000 only average a score of 21. At every income level scores improve, indicating that results on the ACT are more related to zip code than actual ability. ACT scores are also susceptible to coaching, which is why the test company sells preparation software for $500 to those seeking to boost their scores. Needless to say, only the most affluent families can afford this expense, giving the wealthy yet another advantage over those from middle-class, working-class and poor families.

Ultimately, if UT is serious about wanting the best and brightest students (and not merely the richest and whitest), they will move away from any plan to raise the minimum ACT needed for admission. Instead of placing even more emphasis on these tests than is currently the case, the school should develop holistic methods of evaluating applicants, which could assess the many talents that cannot be gauged by a single exam. To go down the road proposed by the former President would sacrifice the interests of both educational equity and true academic excellence. Furthermore, it would foreclose opportunities for thousands of capable students whose families pay taxes to support the very institution to which they would be denied access.


Leave a Reply