It funny when everyone talks about the age-old question of ‘Which is more important, SAT/ACT scores or GPA?’
I have heard so many presentations of Statistical Studies, by Admission Professionals in various organizations, often identifying very compelling ‘significance test’ to validate their studies, that identify the SAT/ACT as the best identifier of ‘aptitude’ but the GPA as the absolute best indication of success overall in college and, most importantly, completion-capability of a prospective student. SAT/ACT is a snapshot. Yes, the standardized tests supposedly are a more accurate and un-bias identifiers of ‘aptitude’ but again a single snapshot. A GPA, while possibly skewed a bit, based on the HS identifying that GPA, is over a four-year period and often that GPA identifies the level of effort expended by the Student, and that has been proven to be the best indicator of success in both college and life. (re Forbes, MIT’s study in 2017, even NACAC’s admissions in 2018)
So why then are so many significant Merit Aid awards linked primarily to SAT/ACT?
Probably because it’s standardized across the nation and isn’t skewed by individual schools districts, etc…
Of course, that doesn’t account for groups of poor students without access to tutors, etc… but it’s probably better, in the eyes of colleges, than nothing.
Agree, objective is important BUT if Colleges identify that Graduation Rates (in 4 or 6 year) are one of the most important statistics to THEM… why wouldn’t they reward what gives that to them… and that has been the effort identified by a GPA. Not my view but that of almost every single ‘qualified’ study I have seen presented.
To me, if say ‘Life Expectancy’ is best predicted by eating apples… bring on the orchards!
It will be interesting to see where we are going on this issue, as over 1,000 4 year colleges/universities do not require a standardized tests for admission, not sure if any of those require the tests to get a merit award though. Awarding merit does become more difficult without a standardized test, as so many US high schools have grade inflation, and standardized tests have historically allowed AOs to glean more info about a given applicant’s potential (among other things). But that’s not perfect as we all know the access issues with standardized tests.
Merit awards have been declining both in amount and by number of schools offering them, and generally this decline is offset in an increase in need-based aid (like Tulane’s stated goals to shift this balance over the next few years). Maybe we will see a model where there is only merit at schools lower in the rankings, as a way to compete on price. OTOH maybe we will see very few merit programs, aside from the schools that have a few marquee type awards (will those still require test scores?), and the lower ranked schools will simply reduce their prices (we are seeing some that now, but not too much yet). Should be interesting as college affordability, along with the desire to increase access, will continue to major issues.
Because colleges generally aren’t giving merit money to kids who are on the borderline of being able to graduate but those who would be in the top percentiles at that college, and they probably assume those kids would not have trouble graduating so they are awarding merit money for their own institutional purposes.
GPAs are not only generally inflated in many HS, there is no consistency between schools regarding what level of work constitutes an A. When students “fail” out of our local magnet HS (meaning they get Cs at the magnet school), they generally end up at the local general HS where they receive straight As. Did the student suddenly develop better study habits or their IQ increase? Nope, what constitutes C level work at the magnet HS is easy A level work at the general HS. So how do you compare GPAs when there is no consistency? You really can’t with any level of accuracy so that’s where a standardized test comes in.
Many colleges and their admissions staffs have figured out how to sort thru HS grade inflation, relative rigor and such. Bowdoin, as one example has been test optional for 50 years…they straight out say they don’t need test scores to know which students are more likely to succeed there.
Many of the 1,000+ schools that have gone test optional are smaller to medium size in terms of number of students…so it may pose difficulties for larger schools to sort thru these issues on such a large scale, but it can be done, especially with the incredible amounts of data available as well as predictive analytic tools.
I do not believe for one second that any college has enough data to accurately convert 30,000+ individual HS level of difficulty to one comparable number. No way. Do they sort of have a feel for which schools are “harder”? Sure. But that’s much different than knowing for school #12,736 a 3.7 = a 3.8 at school #27,002.
I’m not a supporter of the test optional model for that reason and many others.
SAT is correlated ~0.8 with intelligence, which is the single biggest factor in predicting life outcomes. If you have to bet on one single metric that is commonly available, a standardized score is the best thing we’ve got. Sure, you could add other metrics to get a better prediction for any individual, but state schools don’t have the time to go crazy trying to figure out HS rigor, course selection, etc. as regards thousand of candidates.
It is absolutely possible that the state schools, and schools offering merit dollars, will continue to require standardized tests.
As an example of a larger test optional school… GWU gets 30K apps…that admissions staff has a process that doesn’t need standardized test scores to make decisions…and the administration and trustees were comfortable enough with that process to make the change to test optional a few years ago.
Regardless, it will be interesting to see how things shake out over time, this is but one factor in an industry that is rapidly changing.
Why should they care so much about college performance? We are only talking about the top students here - those under consideration for academic scholarships. We’re not trying to figure out whether an 1100 SAT student can or can’t complete a major and affect the 4-6 year graduation numbers.
What scholarship winners do after college is going to be a whole lot more important than whether they got a 3.5 or a 3.9 in college, especially for raising the profile of the state institution awarding the scholarship. Standards for graduation are relatively low anyway; it is not like graduation is a big hurdle for the vast majority of the candidates for academic scholarships.
It will be interesting to see studies on the Redesigned SAT. Not even the college performance validity study on the new test has been published - maybe this summer - and presumably a study correlating IQ is a ways off yet.
Over the past school year, College Board decided to give significantly fewer hard questions (these can be counted via the difficulty designation on the SAS), which led to some interesting scoring scales where a single question can sometimes cost 20-30 points at the top end - this has occurred in math and writing. I imagine that score differences at the top end of the scale are accordingly less-predictive on these demonstrably easier tests than on older versions of the SAT and possibly even on Redesigned SAT test forms given prior to June 2018. (My guess is they may have gotten themselves into a standardization problem, with poor-quality equating, but that’s a separate issue altogether.)
This is not necessarily true. Some schools’ merit scholarships kick in for ACT scores in the mid-20’s, further some use weighted GPAs in their merit matrices. In my kid’s highly competitive high school, with class sizes typically in the 900-1,000 range and and average ACT of 28, students in the bottom 50% of the class can qualify for significant merit at some schools. Smart kids to be sure, but top students? Not really.
A big part of the reason schools (including test optional schools) will still pay for a high standardized test score is because potential students use them to judge schools and they go into rankings. Which in turn, attracts more applicants. When people and subsequently ranking systems stop caring what the average SAT/ACT range for a school is, then schools will stop rewarding it.
Musak, why would I stop caring about average SAT/ACT range for a school might be? I don’t care about weather, access to beaches/surfing, frats or sororities, whether there is sushi in the dining hall, or how many games the basketball team won last year. What if I ONLY care about the academics? And what if I know- based on my HS experience- that sitting through an introductory econ class with kids who are 500-ish math SAT scorers is fundamentally different from sitting in an introductory econ class with kids who are 750-ish scorers? And that a discussion about “what was Tolstoy’s view of the aristocracy” when dissecting the ballroom scene in War and Peace is night and day different depending on the critical reading/verbal skills of the kids in the class?
Of course that’s how prospective students judge schools. These are academic institutions after all-- the rankings are a by-product, not the goal.
On CC it’s fine if your Olympic caliber athlete kid doesn’t want to join the neighborhood “fooling around” team in his sport- everyone understands that in order to get better, an athlete needs to compete with other athletes playing at a comparable or higher level. But the 780 SAT kid is supposed to ignore the academic metrics of his/her fellow students??? And not supposed to “judge” a college by how “thick” it is with high scoring kids?
@blossom - I was saying that purely theoretically. It makes sense to me why colleges still use the metric and I don’t anticipate a time in the near future that no one will care. I have a senior with a top 1% score and having a quirky and engaged academic peer group is definitely part of our selection process. Though I think many kids can have great experiences at schools when they are higher than the 75%.
That said, I also find value in exposure to diversity on a campus and that single score isn’t telling for every single student. It’s one piece of data collected on a single day. Standardized test scores are tightly coupled with parental education level and family income so I think it’s good to take it with a grain of salt. I think there are other more important measures in the mix for assessing school fit. It’s good that some schools recognize they can still get valuable applicants for a campus community without a test score.
Yes, many kids can have great experiences at schools when they are well above the 75% range. But that doesn’t invalidate the use of the scores as an important metric- it just means that if you are studying musical theater or majoring in cello performance, someone’s math ability isn’t going to impact your own day-to-day academic experience all that much. Or if you’re an actuary major, where the departmental scores are typically 100 or 150 points above the university average it means that the SAT scores of the recreation management majors or the kids studying travel and tourism aren’t that relevant for you.
Actually as an art major at Harvard it made a huge difference to me that the mathy kids were brilliant. I was pretty good at math, and I had a great time hanging out with them. I learned a lot about programming and watched a friend build a computer. (He also came back from a summer at Bell Labs, telling us about this cool new device called a mouse.) I didn’t go to art school because I wanted a real college experience, and I wasn’t sure what someone who loved art, history and math about equally should do. I ended up in architecture, which I might not have done at a school with great artists, but not so much of the other stuff.