Two weeks ago I wrote a post entitled, “The College Board’s Terrible, No Good, Very Bad Week,” borrowed respectfully and lovingly from a book title by the children’s author Judith Viorst. I wrote about the class action lawsuit filed against the College Board in response to its on-line AP Exams this spring and the decision by the University of California to end use of the SAT and ACT over the next five years. The UC system is the College Board’s biggest customer.
For the past two weeks the nation’s consciousness has been appropriately focused on the protests of police violence in the wake of the tragic death of George Floyd, and the depth and breadth of the outrage felt by citizens from all backgrounds promises transformational change. The past two weeks have also been potentially transformational for the admission testing industry.
· Marten Roorda has left his position of CEO of ACT, and there have been open calls for the removal of his counterpart at the College Board, David Coleman.
· The College Board’s attempt to open registration on May 28 for test dates in August, September, and October became another public relations fiasco, as students attempting to register after having been prevented from testing this spring found phone lines jammed for hours and some test centers filling up in the first few hours.
· The College Board issued a press release acknowledging challenges in providing universal access to the SAT for high school students in the class of 2021, announcing that it will not offer a version of the SAT that can be taken on-line at home, and asking colleges to be flexible with students in recognition of the difficulties most will have in completing testing.
· The number of colleges and universities going test-optional for at least the coming year continues to grow, with at least four Ivies and the University of Virginia joining the list.
I applaud the College Board for acknowledging its difficulties after initially maintaining that the SAT registration process was, like its claim about AP exam completion and submission, a huge success. I do wonder why the College Board always tends to communicate with its members through press releases.
The statement reveals that the biggest threat facing the admission testing industry is supply chain disruption. During the pandemic that same issue has affected products ranging from toilet paper to meat. Now add admission tests to the list.
The supply chain for both the SAT and ACT relies on local test centers, most of them high schools. Now the social distancing requirements in many states mean that test centers won’t be able to test as many students. Will students be squeezed out of testing, or be forced to travel 100 miles to find an open test center? Will school districts not allow high schools to open up to strangers on a Saturday? And will the colleges that require admission tests step up and offer themselves as test centers? At a time when there is a lot of discussion about testing and its future, the biggest threat to the SAT and ACT might be the supply chain. It’s hard to make money when customers can’t buy your product.
My last post reported on the University of California’s decision to give up requiring the SAT and ACT, and loyal reader Ethan Lewis pointed out that I had misreported part of the plan put forth by UC President Janet Napolitano. I somehow missed that the plan to end the requirement for test scores will apply only to California residents. Out-of-state applicants will still have to submit scores from the ACT, SAT, or the new test that the university is talking about creating. I apologize for my oversight on that point and appreciate Ethan bringing it to my attention.
That raises an interesting question. Is it legitimate for a college or university to require testing for some students but make it optional for others? I have asked that question before about universities that are test-optional, except for engineering students or scholarship candidates.
I suppose that depends on your justification for being test-optional. Are you test-optional for philosophical reasons, because you question the validity of the test? Are you skeptical of how much value test scores add to the validity of admission decisions? Are you not selective enough for test scores to make a difference in admission decisions? Or are you test-optional as a tool for profile enhancement, allowing you to admit students you want on your campus but whose test scores might hurt your data for ranking purposes?
In the case of the University of California, the argument might be that it serves California residents and knows the high school curriculum offered at California high schools. Requiring out-of-state students to submit test scores provides an additional data point to evaluate the credentials of applicants from schools that the UC admissions offices may not know as well.
I wonder whether there are other iterations of the concept of requiring admission tests for only some applicants. We know that test scores have context. If you and I have the same SAT score but I come from a wealthy family and a good school and have spent a sizable amount on test prep, while you don’t have any of those advantages, then the scores don’t mean the same thing.
Given that we now know that SAT and ACT scores have a high correlation with family income, and that the value test scores add to predictability is minimal, should we require test scores only from applicants who come from families above a certain income, with students from lower socioeconomic groups able to receive a decision based only on their high school transcript? That’s the test-optional I’d like to see.
In the last post I argued that it is odd that after being test-optional for the next two years the University of California will allow students to submit the SAT and ACT for scholarship consideration but not for admission. Why use test scores for a purpose for which they are not intended, scholarship consideration, when you are not going to use them for admission, the purpose for which the test was designed?
That leads to one other issue. The purpose of the SAT is predicting freshman-year GPA. That’s certainly helpful, but shouldn’t the college admissions process be intended to predict a much broader definition of success, say likelihood of graduation or success beyond graduation? If testing is going to be part of our admissions calculus, shouldn’t we develop tests that measure the qualities that lead to success?
Do we measure what we value, or do we value what we can easily measure?