SAT/ACT

Are the SAT/ACT reliable representations of how freshman will perform in their first year of college?


I'm going to go ahead and start out by saying no. Multiple-choice question tests in general are unreliable. As my 7th grade Latin teacher was so fond of saying, as we prepared for the National Latin Exam, "A monkey could ace this thing." And it's true. Usually you are given 3-4 answers to choose from, so for each question you have about a 25% chance of getting it right if you just guess random answers. The probability that you will ace the test if there are 40 questions and you guess randomly for each one is very small, yet quite possible.
Further, the SAT is not a simple multiple choice test, it does some funky thing where you lose more points for answering wrong than you would for leaving the question blank. The result is that a family with excess money can spend thousands of dollars on tutors who teach their kid all of the little tricks for getting a few extra points, and in general, how to work the system. Thus someone with below average intelligence can easily get a much higher score on the SAT than a kid with above average intelligence, simply because he has been taught how the system works. This particular phenomenon is personal for me, since one such rich family stole one of my favorite high school teachers and paid him ridiculous sums of money to tutor their kids about how to work the SAT. We never got him back.

I suppose I have only answered a part of the question at this point, that is, no, the SAT/ACT are not reliable representations of the intelligence/reasoning ability of those who take it. People who get high scores might be dumb as a stone, while those with lower scores might be quite intelligent. As for whether the tests are reliable at predicting the academic performance of first year college students, again, I say no. I have at least two friends at Ithaca whose SAT scores were higher than mine, and yet their college GPA is far lower than mine. I watch them struggle in classes where I have very little trouble. Thus our relative SAT scores are unable to predict how we perform in our first year of college.

I do acknowledge that my data pool consists of only three people out of the many thousands of college students in the world. However, it is only an example of my point, and that is that any multiple-choice test, particularly one with easily manipulated factors, such as the SAT, is unreliable in determining how a student will fare in their first year of college. Also, the SAT only attempts to predict pure academic ability. It does not take into account the psychological effects of being effectively stripped of your family and friends and given a place to sleep in completely unfamiliar surroundings. It does not take into account the need to balance a social life with an academic life. It does not take into account the need to find a job, maintain it, and balance the time that takes with your social and academic life. Thus a kid with below average reasoning abilities might end up being a fantastic organizer, and thus get better scores on college tests because they get a good night's sleep every night, and they have general peace of mind and reduced stress.

In conclusion, I say once again that SAT scores are unreliable representations of how a student will do in their first year of college.