Standardized Testing: Can it predict college success?

Candian schools focus on students grades, while American schools focus on students standardized test scores. So which system prevails? Are grades or standardized testing a better predictor of college success?

There are hundreds of posts online all stating different perspectives on this discussion, I will try to synthesize these perspectives as well as analyze the data myself. However, before we can do this we must define what college success means. As it is very hard to compare qualitative measures of success such as happiness, we will measure quantitative measures of success. We will mainly focus on college GPA and graduation rate.

One study, conducted by the college board, that measured the correlation between SAT scores and college GPA found a positive correlation between these variables. This does not show causation, however, and many critics to standardized test scores say that this relationship is caused by the lurking variable of family income and value of education that affects both of these variables.

2016 College SAT Score by Income Graph (c) Joshua Lorincz, CC0 Creative Commons

This is a graph of family income vs. SAT score, created with data from the college board. It is clearly visible that there is a strong correlation between SAT score and Family income. Families with more income are more likely to be able pay for SAT/ACT prep-classes. Children of these families are also able to afford tutoring and not work in college giving them a major leg up on other students. It is clear that there are flaws behind Standardized test taking but are grades a better alternative?

The problem that many find with this, however, is that grades contain the same flaws involving wealth disparities due to tutoring. They also involve other biases such as some schools purposely giving higher grades to help their students get into better colleges. Despite this numerous studies such as this one have shown that grades are a better predictor of college success than Standardized test scores.  This study only sampled students at the University of Anchorage, however, it shows that the relationship between these predictors of success and actual success are not trivial. Since the release of this study, numerous other papers have come out finding results on bother sides of the argument.

The uncertainty of the study and complexity of the problem, in my opinion, shows that neither measure is a sound predictor of success in college. Both measures should be used in tandem to make admissions decisions, and should still be taken with a grain of salt as they both clearly contain bias.

Speak Your Mind