New SAT I Study Reveals Exam’s Limitations

Status: 
Archived
Subject: 
University Testing

A new study sponsored by the University of California Office of the President (UCOP) adds to the mounting evidence that the SAT I is a poor predictor of success in college.

 

The four-year study of more than 80,000 students enrolled in the University of California system looked at the relationship between high school GPA (HSGPA), SAT I scores, SAT II scores, and first-year undergraduate grades (FGPA, which is what the SAT series is designed to predict). The weakest predictor of college performance proved to be SAT I scores, which explained just 12.8% of the difference (or variation) in freshman grades. SAT II scores and HSGPA each separately explained approximately 15% of the variance; each of these factors did a better job of forecasting college performance than did the SAT I.

 

The UCOP researchers also considered the value of using test scores in conjunction with high school grades, a practice commonly utilized in college admissions. Combining HSGPA and SAT II scores yielded a measure that explained 21% of the difference in freshman grades. Adding the SAT I to this equation improved the predictive ability by less than 1%, demonstrating that the SAT I adds little new information to the assessment of a student’s application. Such findings contributed in large part to UC President Richard Atkinson’s February 2001 proposal to drop the SAT I requirement for applicants (Examiner, Spring 2001).

 

The researchers discovered that the predictive power of the SAT I was further compromised when socio-economic status was taken into account. After adding parents’ education and family income into the equation, the relationships between HSGPA, SAT II scores, and college performance each remained constant, while the correlation between SAT I scores and FGPA was diminished. In other words, a great deal of the apparent relationship between SAT I scores and FGPA was actually influenced by socio-economic factors. In addition, researchers also discovered that SAT I scores were more closely associated with family income and parents’ education than were SAT II scores or high school GPA.

 

While the SAT II exams did pack more predictive power than the SAT I for UC applicants, they are still subject to many of the same flaws (see Examiner, Fall 1995). These 60-minute, multiple-choice exams test a narrow range of subject matter in areas such as American History, World History, and Biology. They are not matched with a particular curriculum, so student performance on the exams is largely dependent on the match between course content and test content. The emphasis on memorized facts rather than higher order thinking dumbs down learning and narrows high school curricula. In an interview with the Los Angeles Times, National Academy of Sciences President Bruce Alberts summed up the faults of the SAT II Biology Test: “It’s too broad and too shallow to test anything meaningful. We need to get rid of the trivial factual recall and make sure kids really understand science.”

 

Due to sizeable racial and economic score gaps, use of the SAT II tests will stifle educational equity in many of the same ways that the SAT I does. Higher income students are advantaged on the SAT II’s by access to expensive coaching classes, being able to take the exam multiple times (and thus choose which scores to send to admissions offices), advanced curricular offerings at their schools, and a disproportionately large share of extended time accommodations for learning disabilities.

 

A more equitable and sound alternative to either the SAT I or SAT II is to make all test scores optional in the admissions process. FairTest’s list of nearly 400 colleges and universities in the United States that admit a substantial number of students without regard to test scores demonstrates the effectiveness of this approach for a variety of institutions, including both large, public universities and small, private colleges. To see the list of schools, click here or send a business-sized, self-addressed stamped envelope to FairTest, 15 Court Square, Suite 820, Boston, MA 02108.