|Posted On February 19, 2014|
JUDY WOODRUFF: Itís one of those times of the year when high school juniors aiming for college are getting ready to take the SAT or the ACT, but a large new study is challenging the value of these well-known standardized tests.
Researchers looked at 33 public and private colleges and universities where itís optional for applicants to submit their test scores. In all, the study examined the records of 123,000 students from more than 20 states. It found that test scores didnít correlate with how well a student did in college based on grades and graduation rates.
The paper has raised a variety of questions from several corners.
And we turn to its lead author, William Hiss. Heís the former dean of admissions at Bates College in Maine, which is, we should note, a test-optional school.
Bill Hiss, thank you very much for joining us.
First of all, why did you undertake this study?
WILLIAM HISS, Former Dean of Admissions, Bates College: I have been looking at this issue for over 30 years.
Iím trained as an ethicist, so I wasnít trying to find the perfect formula to admit students when I was dean of admissions. Rather, I was trying to say, how do we understand human intelligence?† How do we understand promise?
And, originally, we began looking at whether standardized testing helped us to choose talented students or whether it artificially truncated the pools of people who would succeed if we would give them a chance.
JUDY WOODRUFF: And so you were ó you had an idea in your mind of what the answer would be before you undertook this?
WILLIAM HISS: Well, we have done at five-year intervals for 30 years studies of the Bates students who were submitters and non-submitters.
JUDY WOODRUFF: Meaning ó submitters meaning students who did or didnít submit their test scores.
WILLIAM HISS: Correct. Submit their testing, correct.
And we wanted to do a study that would be a national study. I was frankly surprised to find how closely the national studyís results lined up with the long history of studies we have done at Bates.
JUDY WOODRUFF: And so tell us, what were the main findings?
WILLIAM HISS: The main finding is that if students have strong high school records, good grades in high school, their odds of doing welcome in college are very good even with a quite wide range of testing.
Of the 123,000 students in our study, we had approximately 30 percent who had not submitted their tests or not had them used for admissions decisions. The difference in cumulative GPAs in college turned out to be five-one-hundredths of a GPA point. The difference in graduation rates turned out to be six-tenths of 1 percent.
By any definition, by any statistical standard, those are trivial differences.
JUDY WOODRUFF: So, itís commonly been accepted, though, I think, for many years that these tests help in determining whether students should be accepted at college.
What do you think these results should say to colleges that are today insisting that students send in, submit their test results?
WILLIAM HISS: I think that colleges will probably be doing some hopefully analysis of their own predictors college by college. This is a national study, so it looks across institutional types.
Of the 123,000 students in our study, 71,000 were at public universities. Another 12,000 were at minority-serving institutions. We had 20 private colleges and universities in our study. So thereís a very wide variety. But I hope institutions will ask themselves the question, would we open up our applicant pools more, would we see larger numbers of students who would succeed if in one way or another we deemphasized the testing?
And that essentially is what Bates has found. Over the years, our applicant pool went from 2,200 to 5,200 for about the same size class. Do you get a better class with two-and-a-half times as many applicants?† Of course. Of course you do, and on all the scales, not just academic success, but contributions to the college community in a variety of ways.
JUDY WOODRUFF: What do you say to representatives of the College Board?† This is the body that administers the SAT, that they say multiple surveys they have done over the years, tests they have done, evaluations, show that this test is an accurate predictor of college success.
And they also say that ó Iím quoting now ó ďHigh-quality research shows neither the SAT nor high school grade-point average should be used alone when making college admission decisions.Ē
WILLIAM HISS: Well, I would tend to agree with the second point.
And I would say students should try to show us what they have brought to some high level of success. It may be a viola or debate or soccer goalie or community service. But I agree we should look at more than a single instrument.
I, frankly, was surprised to find how reliable the high school GPA was. And I thought about it a lot. And I myself think that we are looking at two four-year-long demonstrations, high school and college, of self-discipline, curiosity, intellectual drive, if you will, the ability to get your homework done on time, get your papers written.
And so I think there are reasons that the high school GPA turns out to be a very solid predictor. And ours is by far not the first study that has turned up the same results.
JUDY WOODRUFF: And just quickly, finally, pin down for us what harm you see in insisting that students take these tests and submit them.
WILLIAM HISS: Well, the biggest harm is to the country. You truncate the applicant pools of people who would succeed.
And the people who are non-submitters in higher percentages are just who we need to get through, first generation to college, minority students, students with learning disabilities like dyslexia, somewhat more women than men, Pell Grant recipients who are low-income students. These are the people we have to get through college.
So the danger is, you truncate your pool of the people you really need to have in your college.
JUDY WOODRUFF: Bill Hiss, we thank you for talking with us.
WILLIAM HISS: Thank you.