Since I work at a small college, my sample size is small. But that also
gives me an advantage, which I will explain in a minute.
Some of our scholarships are based upon test scores. Because of this,
high-school students often take the SAT or ACT multiple times in the hopes
of improving enough to get a scholarship or to move into a higher
scholarship category. At my college it seems the average number of times a
student takes the SAT or ACT is two, but this is not a "bell curve." About
one third take it once; one third take it twice; one third take it three
times. Therefore I see a fair number of double and triple scores.
These double and triple scores are almost always identical. This would seem
to indicate it measures something. Some students can consistently score a
30 while others consistently score 21. What is this telling us?
In general, students scoring above 25 will succeed in science; but the 25s
will have to work harder than the 30s. In general, students scoring below
25 will drop out of science. It is rare for someone below 22 to complete a
science major. Therefore, when we see an entering science freshman class
with a whole slug of 19-23 ACT scores, we know we are in trouble. If we see
a slug of 25-30 then we start smiling.
The problem is applying the trend to any individual. I clearly cannot
predict or expect a particular 22-ACT student to fail, nor can I expect a
particular 30-ACT student to succeed. This is where the other stuff comes
in, and this is where I have an advantage being at a small college.
If the 22-ACT student is highly motivated and really wants to learn, he will
be fine. If the 30-ACT student doesn't like her roommate, has problems at
home, can't find friends, has poor sleep habits... she is in trouble. I can
predict with good accuracy whether the 22-ACT student will succeed or fail
once I know her better, which happens during the first semester here. The
30-ACT student can tolerate some depression or poor sleep habits better than
the 22-ACT student, but he can't afford to shut-down and do little or
nothing. The energetic, curious, hard-working 22-ACT student will be a joy
to work with, but will struggle, especially with abstract concepts.
The ACT score in conjunction with the high-school rank often works well.
The rank gives some indication of how eager the student is to excel, whereas
the ACT gives an indication of IQ and basic preparation. Letters of
recommendation are also helpful in discerning the student's motivation and
overall stability, but we have to trust high school teachers and guidance
counselors to be honest and frank (and many are not).
We have an admissions and scholarship committee which is supposed to look at
all these types of things for all students who apply and meet our basic
admissions criteria. That's a lot of work, but we are small and can do it.
I don't know how the big places do this.
I find the SAT and ACT scores useful as I try to figure out why a student
does what she does, and how I might be able to help. Used alone these
scores are much less useful, but not worthless. I suppose we ought treat
these scores about the same as we treat teaching evaluations. We once had a
good dean who said, of course evaluations, "We take these seriously, but not
too damn seriously."
Michael D. Edmiston, Ph.D. Phone/voice-mail: 419-358-3270
Professor of Chemistry & Physics FAX: 419-358-3323
Chairman, Science Department E-Mail edmiston@bluffton.edu
Bluffton College
280 West College Avenue
Bluffton, OH 45817