Using test scores in a vacuum for admissions purposes is poor practice. In my experience as a faculty member, department chair and graduate dean, I have seen firsthand how important it is to consider all of the information in a candidate’s application. Some of my very best graduate students had “nontraditional” career paths, and it is only by considering factors such as research experience, grades and work experience, that you are able to fully evaluate a student’s likelihood for success in graduate education.
Lately, many graduate (and undergraduate) schools have been adopting “test-optional” policies in an effort to increase the diversity of their applicant and registered student pools, arguing that standardized tests are poor predictors of academic success and biased against women and underrepresented groups.
While there are many good educational reasons for supporting efforts to increase the diversity of students pursuing graduate education, it is important to consider both the intended and unintended consequences of changes in admissions practices — sometimes what seems like a good idea may result in negative outcomes.
For example, evidence shows that programs that have dropped or made standardized tests optional do not see improved diversity in their applicant pools. Research from the University of Georgia showed that undergraduate schools that adopted a test-optional policy concerning the SAT® test have not experienced the expected increase in applicant diversity. In contrast, what these schools have seen increase is their selectivity because more students apply and those that report scores typically report higher scores.
On the other hand, proper use of test scores in the admissions process can help institutions ensure a diverse and talented pool of applicants and enrolled students because standardized test scores are the only measure common across all applicants in the admissions process.
It is useful to recall why these tests were created in the first place. Prior to the development of standardized admissions tests, students from privileged family backgrounds who attended selective and expensive private schools had a much better chance of getting admitted to college than talented but less affluent students who attended a public high school. When making admissions decisions, selective institutions consider applicants’ undergraduate GPA, the prestige of the undergraduate school attended, transcripts, interviews and personal relationships. These factors are subjective and vary from institution to institution.
Standardized admissions tests, however, level the playing field. Dropping test scores from admissions processes can potentially negatively impact underrepresented groups from non-selective institutions and actually increase bias in the admissions process. Data show that underrepresented groups frequently start at nonselective schools immediately putting them at a disadvantage. Standardized tests, however, offer programs a fair and impartial way to compare and evaluate students from different backgrounds, particularly international candidates where other admissions documents may not be well understood or are unfamiliar.
There is an appropriate role for test scores, undergraduate grades and noncognitive factors such as “grit,” in the graduate school admissions process. Putting too much emphasis on a test, or any other single predictor, is poor educational practice. It’s critical that admissions officers balance the limitations of any single measure of knowledge, skills or abilities as they alone cannot completely represent the potential of any person.
In fact, Educational Testing Service (ETS) guidelines recommend that multiple sources of information always be used when making admissions decisions, particularly when assessing the abilities of educationally disadvantaged applicants. Studies show that the combination of grades and test scores is a more effective predictor of a student’s readiness than either one alone.
Test scores typically correlate with students’ socioeconomic status, race, and gender. This, however, is not a fault in a test; rather it reflects the reality that more educational resources are available to students from wealthier families, and these real educational advantages are reflected in test scores. There are a number of factors that contribute to observed differences in scores, such as variation in course-taking patterns, interests, knowledge and skills, or educational, economic and social systems in which everyone does not receive equal opportunity. Society creates many of these differences in scores, and the test reflects them. Reflecting real differences is not bias. Economic disadvantages and inferior school systems are not fixed by making a test that is blind to the differences caused by these inequalities.
While it is impossible for a standardized exam to predict the numerous life events and economic forces that may make succeeding in graduate school more difficult for many Americans, the loss of the tests seriously harms the ability to fairly and impartially identify students who have the potential for academic success.
Note: SAT is a registered trademark of the College Entrance Examination Board.