kitchen table math, the sequel: speaking of SAT scores...

Wednesday, May 20, 2009

speaking of SAT scores...

Colleges Acknowledge SAT and ACT Score Cut-Offs in Admissions

This is news only because journalists continually write articles quoting college admissions officers saying SAT scores don't matter, or are going out of style, or are used holistically, or are "optional," or god-knows-what.

Trust nothing a college admissions officer tells a newspaper.

Some of the comments are great.
I was “waitlisted” at two of my top schools and when I called to follow-up, I was told point blank it was for my SAT scores. Sorry, but I have a hard time buying the holistic bit. Particularly given the number of applications some of these schools receive. They need to have an initial filter; test scores give it to them.

On a related note, I was also shown a list of applicants, while applying to grad school, and they were sorted by GRE score. Again, the person who showed me the list said, point blank, we sort first by scores.


When my first child and oldest daughter was readying to choose a college, I asked the admissions department of my alma mater whether she should apply. I was told, first, before any other topic was raised or question was asked, that the college board cutoff was such and such–about fifty points higher than my daughter’s score.

She had gone to a pricey progressive private high school, where her grades were not bad, though not straight As. She did well in English, but also in biology and math–her only really weak course was chemistry.

Her guidance counselor, a Harvard graduate, was unable to interest the admissions departments of any major ivy league school in considering my daughter.

I asked the college admission department if a legacy mattered at all. “Somewhat,” I was told. Eventually, my third child was admitted, but he made the board score cut.

This practice has at least one side effect that has not been well understood. It is actually part of the decline of the humanities in American colleges and Universities, because the boards emphasize quantifiable ability–obviously so in mathematical skills, but still true in the “verbal” part of the exam: what else could such an exam do but quantify? Even the writing exam is graded, not on quality of the writing, but on quatifiable features, e.g. whether it has topic sentences at the beginnings of paragraphs (a custom universal in America but nowhere else, and only due to college entrance exams, including AP English), whether it has a standard three-paragraph form, whether it has an intro and a conclusion, and so on. THe content can be invented; can be virtually nonsense, but the template of the essay form to which all high school English writing is now taught can be measured, counted, and scored. Likewise the multiple choice questions which test for information processing ability only. There is no way to build a standardized test for humanitas.

— droll

Our child is a high school senior. He scored very well on his ACT but his cumulative GPA at the time of applications was just below 2.5. He was, nonetheless, offered scholarships and entrance into honors programs at different schools. You don’t need to convince me that these test scores are really where it’s at.


Having retired from the college counseling field, I recall when it became non-PC to use a single test score as a cutoff for admissions. There was some discussion of this being racist and not being the purpose for which tests were standardized (and thus leaving schools open to lawsuits). So colleges were coached to say they viewed test scores as “only one factor in a holistic assessment.” It my observation that some college admission teams actually employ holistic assessment and some only pretend to do so. Some admissions staffers value test scores because they value tests but don’t know an instrument’s limitations; others are racists. It’s hard to discern who is who isn’t it?


The holistic approach apparently does not apply to the NJ state schools. My sister’s twin daughters did not do well on the SATs, but they each have excellent gradepoints, extra-curriculars, outside jobs, sports, and stellar course grades for the college courses they have taken in their junior and senior years of high school. They and my sister were each told flat out that it was meet the cut-off score or we don’t look any further. Luckily, the non-state schools that they each applied to accepted them(2 acceptances each). I wonder if this policy is linked in any way to the budgetary issues in NJ government?

Jim McNerney

I have had years of experience at graduate school admissions for my department. First, the TOEFL had an institution-imposed minimum for admission, and experience told us that it was too low. Also, there is doubt about testing conditions in some countries.

Second, we came to understand that there are only three scores on the General portion of the GRE: high, middle, and low. Enough students take the GRE more than once, so we were able to see the variances in scores not attributable to additional education.

Third, subject-area GRE scores are useful. Since we did not require them (but were encouraged), when good scores showed up we regarded applicants as confident in their disciplinary knowledge.

Fourth, general and major GPAs are unreliable because they depend on the standards in place where they were earned. From less-highly-regarded institutions, applicants may have perfect GPAs, stellar letters of recommendation, and miserable GRE scores. When admitted, their failure rates were high, unless they dug in and made up for what they had not learned. OK GPAs from top institutions, or from students who had lots of jobs to get through college, were treated as just fine, and that’s mostly how they turned out.

Leading Edge Boomer

preparation for College Admission Exams


Niels Henrik Abel said...

This "news" really shouldn't come as a surprise to anyone. GPAs probably say more about a school's grading policy (think "Lake Woebegone effect") than they do about a student's ability. Standardized tests, on the other hand, are impartial in their assessment: either Johnny knows his stuff (in which case he'll do fine on the SAT/ACT), or he doesn't (in which case no amount of wheedling, cajoling, or BSing will raise that score). I hear the same type of thing from students who supposedly studied the material and did all the homework, yet their quiz and test scores are in the cellar.

Given a choice between Suzy (straight As in four years of accelerated high school math, yet gets a 425 on the math portion of the SAT) and Johnny (lackluster B/Cs in four years of the same math classes, yet scores 700+ on the same math portion of the SAT), who would you wager is more likely to be successful in college and complete a degree?

I'm not an admissions officer, but my money's on Johnny.

Luke said...

Tests are useful for finding out how well students can take tests. Which, if that's how you decide who to graduate, it makes perfect sense to base your admissions off the tests.

However, depending on your major, your professors, and the goal of the school, test scores may prove less than useful. But, yes, schools have to start somewhere.


ChemProf said...

Test scores are also often used as a measure of IQ, essentially (whether they should be or not is a topic for another day). I know admissions often look at low grades and high test scores as "a bright flake, will either get it together in college or not, worth a try."

There are limits, though. I know we've had students in our postbac premed program (for students who decide they want to go to med school after they've graduated) who didn't get admitted to med school despite high MCAT scores because of low undergraduate GPAs -- the med schools weren't willing to have the student screw up their average GPA.

Catherine Johnson said...

the med schools weren't willing to have the student screw up their average GPAMeaning med schools rate themselves partly on undergrad GPA of admitted students?

ChemProf said...

"Meaning med schools rate themselves partly on undergrad GPA of admitted students?"

So they tell me, and rankings like US News definitely do use average undergraduate GPA of the entering class. Schools care about these rankings more than they should, but being in the top ten (or twenty or whatever) is helpful in recruiting and in fundraising.

School rankings have all kinds of odd effects. I remember one college I worked at had three sections of General Chemistry, with an average enrollment of ~55. The administration wanted us to limit two sections to 49 and make the third section bigger, because US News ranked colleges based on the fraction of classes with 50 or more students.