Should the SAT be mandatory and free of charge?

Updated Aug. 22nd at 2:50am by Jason Pontius

This idea isn’t ours. But the point of this blog is to present examine seemingly simple, even simplistic ideas that have the potential to affect real change, and this is an ideal example. In this New York Times article Susan Dynarski (@dynarski), a professor at the University of Michigan, argues that making college admissions tests mandatory would put more low-income students on colleges’ radar:

There is widespread concern about over-testing in schools. Yet we need all students to take the right tests if low-income and minority children are to have a good shot at a quality college education.

The two standard college admission tests — the SAT and the ACT — could be administered universally and free of charge to students. That would reduce the administrative barriers to applying to college, help identify talented disadvantaged children, and increase the likelihood that they will attend a college that matches their skills.

On its face, it seems like a no-brainer. Standardized college admissions test scores are reported to colleges and universities; that’s why high school juniors and seniors’ mailboxes (both real and virtual) become stuffed with college marketing materials. Low-income students take these tests at lower rates, because of roadblocks both financial (costs for the SAT can range to $100 or more) and practical (the paperwork and deadlines associated with testing are harder to manage for students juggling responsibilities and pressures outside the classroom).  If all students were required to take the SAT, all their scores would be reported to colleges, and admissions offices— many of which are actively trying to recruit low-income and/or first-generation college students— could locate those prospects. 


Dynarski links to a number of case studies and research papers reporting on experiments with universal testing, often with strikingly effective results. From one example, in an academic article by Sarena Goodman:

Between 2000 and 2010, five U.S. states adopted mandates requiring high school juniors to take a college entrance exam. In the two earliest-adopting states, nearly half of all students were induced into testing, and 40% to 45% of them earned scores high enough to qualify for selective colleges. […] I conclude that a large number of high-ability students appear to dramatically underestimate their candidacy for selective colleges.

Case studies are always subjective— we only see the cases that Dynarski’s chosen to present. But there’s some powerful stuff here.


The immediately discordant element to this proposal is the fact that the SAT and ACT are widely regarded as imperfect instruments for measuring academic potential. It seems quite inarguable that standardized testing is socioeconomically biased; wealthier students are able to afford tutors and test preparation courses. Many selective colleges have made these tests optional, based on strong evidence that test scores do not equal academic merit. That being the case, it seems odd to make the SAT the basis of an effort to expand educational opportunity to low-income students.

Dynarski acknowledges this:

Many people worry that college admissions tests are biased against low-income and nonwhite students. But disadvantaged students who do not take the tests are out of the running for selective colleges. While we may wish for a better approach, these tests are a gateway to selective schools.

What is important to her proposal in fact has very little to do with the nature of the test itself. It rather simply takes advantage of an existing fact— that test scores are reported to colleges— and repurposes that channel to serve the purpose of putting more prospective students on the radar screen.

Admissions offices simply have no way to find a student who doesn’t take the SAT or ACT. Those students’ college options will be limited to the places they learn about firsthand— often a community college or two-year degree. If all high school students were reported to colleges, then admissions offices could make their own decisions about how, and to whom, to reach out.

Making it happen

The College Board isn’t likely to make the SAT free; although it’s officially a nonprofit, it is highly profitable, and many articles question whether it ought to be allowed to function as a nonprofit at all. So until the SAT is replaced with something better, the goal of free standardized testing is likely to be achieved through fee waivers, grants, and perhaps private philanthropy.

And of course it’s much more likely to happen on a state, regional, or municipal level.  The cost to the state of Maine for implementing mandatory testing (including test fees and payments to proctors and educational support teams, and a range of other support services) was estimated at $1 million per year— not pocket change, but well within the scope of a state budgetary allocation. An experimental test at the county or municipal level would be considerably less expensive.

The fact is, making the SAT free is probably an easier task than making it mandatory— because the latter means working with school districts, local governments, and a variety of regional stakeholders. The Maine project involved a great deal of support beyond the test itself:

First, the state transformed all public high schools into SAT testing centers. Second, school districts were provided with funding for school busses to transport students from their homes to school. Students unable to take the exam on the official administration date were provided with an official makeup date the following month. An unofficial administration was also provided by the state of Maine in which students were granted accommodations not officially sanctioned by the CB, and therefore not reportable for college admissions purposes. Finally, students were registered for the exam in bulk.

Other schoolwide adjustments were also undertaken in Maine in response to the SAT mandate. Students were offered online courses for both the PSAT and the SAT to increase student preparation for the exam. Breakfast and lunch were offered to students on the exam day. As a reward for student participation, a “comp” day was provided during the following week for SAT test-takers.

That’s a lot of infrastructure. For an experiment in universal testing to have its best chance at success, it will require partnerships among many different kinds of stakeholders.

Edit post