The John William Pope Center for Higher Education Policy (logo)
RSS feeds

Commentaries
A new pilot admissions program calls into question the value of standardized testing

By Jesse Saffron

Comments

October 31, 2014

From 2009 to 2013, the University of North Carolina system gradually increased its minimum admission standards. Students entering UNC schools this fall had to score at least 800 on combined math and verbal SAT tests to be admitted. 

But starting next fall, three schools will be exempted from that policy. North Carolina Central University, Elizabeth City State University, and Fayetteville State University will be allowed to admit students with SAT scores as low as 750. The three historically black universities are part of a pilot program approved at last week’s UNC Board of Governors meeting. It will place greater emphasis on applicants’ high school grade point averages.  

Vigorous discussion about the program’s potential drawbacks and unintended consequences broke out at the board’s educational planning, policies, and programs committee and again at the full board meeting on October 24. 

The pilot is based on two hotly contested premises: 1) a high school GPA is a better predictor of collegiate success than an SAT score, and 2) it is necessary to “level the playing field” by making college more accessible to those who don’t perform well on standardized tests, especially when those students come from low-income households and poorer school districts. 

At the board’s educational planning committee meeting, committee chairman J. Craig Souza aggressively supported the initiative. “I am lobbying for this, yes I am,” he joked at one point. “We have a responsibility to reach down and get these kids in schools.”  

Committee members Hari Nath and Steve Long were less sanguine. They said that grade inflation could become an issue at the high school level, as teachers and guidance counselors, attempting to “help” students on the cusp, might manipulate grades.  

And Long asked, “Are we getting into remedial education that should be at the community college? Are we not using our community colleges to prepare students for [four-year] colleges?” 

Souza quickly dismissed those questions and concerns. “I think this is a good experiment that we will learn from. I just don’t see any downside to it,” he said.  

Long and board member R. Doyle Parrish believe the pilot could be a “slippery slope.” That is, they’re worried that it could be the starting point for moving the entire UNC system in the direction of SAT-optional admissions. 

Long later said to the Pope Center, “I don’t fault the General Administration for coming up with policy proposals, but I wanted to have a debate so that board members would become more [familiar] with standardized testing and the [test-optional] issue because I predict that we will soon have another debate, and it will be more broad-based.”

If UNC system schools go test-optional, they would not be alone. More than 800 four-year colleges and universities do not require applicants to submit standardized test scores. In recent years, prestigious schools such as Sarah Lawrence, Brandeis, and Wake Forest have made the transition, too.  

Proponents of test-optional policies say that a high school grade-point average is a better indicator of college readiness than an SAT score. In fact, the SAT itself is being revamped, perhaps as a result of increased criticism of its merits. Starting in spring 2016, the essay portion will be optional, the vocabulary section will focus on "relevant" words, and other problems will be "grounded in real-world contexts."

A 2014 study by the National Association for College Admissions Counselors compared students who submitted SAT scores and those who didn’t. There was virtually no difference in the success of the students, according to the authors. The study, which gained national attention, was based on an analysis of five percent of schools with test-optional policies.  

SAT critic and Bard College president Leon Botstein, in a Time article, wrote, “High school grades adjusted to account for the curriculum and academic programs in the high school from which a student graduates [are good predictors of academic achievement in college].”  

But since distinguishing high schools and guaging the rigor of a particular school's curriculum can be a difficult guessing game, especially for top colleges or flagships where many applicants have high GPAs, some argue that the objectivity of standardized tests—particularly achievement-based tests that measure a student’s understanding of college-prep high school coursework—is preferable.  

Wake Forest sociology professor Joseph Soares, a steadfast opponent of the SAT, supported his university’s admissions shift. In his 2007 book, The Power of Privilege, he concluded from a study of the University of California system that the SAT II—which measures academic achievement—is a better predictor of college grade point averages than the SAT I, which measures general aptitude and is generally known as the SAT. 

As the Pope Center’s Jay Schalin argued, however, the correlation between SAT I (aptitude) and SAT II (achievement) is high enough to make the two tests “fairly equivalent.” Furthermore, Schalin pointed out, at least in that California sample, predictive accuracy increased by 35 percent when SAT I was combined with high school GPA, and by 44 percent when high school GPA was combined with an SAT II score. Thus, according to Schalin, using a range of measures is the best approach.  

More recently, other writers have supported the value of the SAT as a predictor of success. According to professors David Z. Hambrick and Christopher Chabris, the takeaway from a recent study of 150,000 students from 110 colleges is that “SAT scores predicted first-year college GPA about as well as high school grades did, and the best prediction was achieved by considering both factors.” 

At last week's meeting, nobody from the general administration and no board members mentioned the literature showing that analysis of both GPA and SAT provides a better indicator of academic potential than looking at GPA alone.  

The UNC system’s experiment is the brainchild of North Carolina Central’s Debra Saunders-White, who became chancellor last summer. After taking the helm at NCCU, she discovered almost 300 applicants who she believed were ready for college-level work, but who couldn’t matriculate because of low SAT scores. She then presented an alternative admissions proposal to the UNC system’s General Administration.  

Karrie Dixon, the administration’s vice president for academic and student success, led the development of the MAR pilot. It will incorporate a “sliding scale” that weighs an applicant’s high school GPA more heavily than his or her SAT score, which cannot fall below 750.  

Admitted students must have high school GPAs above the system’s 2.5 (C+) minimum, and GPAs must increase by 0.1 with each 10-point SAT score decline. For example, if an applicant has a 790 SAT, his or her GPA will need to be 2.6 or higher. At the 750 minimum, a student’s GPA must be at least 3.0. Each school will be able to admit 100 students (all in-state) each fall. Dixon predicts that the three-year pilot will result in 600 new enrollments. 

Dixon told Durham’s Herald-Sun that the administration also selected ECSU and FSU as participants because they had “shown interest in testing what national data and our UNC predictive analytics suggest about the positive correlation between HGPA and student success.”  

Monitoring of the special admits will be intensive. Resources will flow to academic mentorship, counseling, and student assessments. Board member Joan Perry, who nevertheless voted in favor of the pilot, told the educational planning committee that she is worried about the costs associated with such mentorship, and that “advising is pricey.” Steve Long went further, arguing that if a GPA is indeed a better indicator of college success, the additional academic guidance and oversight should be considered unnecessary.  

Donald Reaves, chancellor at Winston-Salem State University, commented at the board meeting that WSSU had not adhered to the admissions requirements set in 2013, which mandated a 2.5 GPA minimum and an 800 SAT score. 

In an e-mail defending that action, Chancellor Reaves told me that of the 54 students admitted without adequate SAT scores, all 54 had high school GPAs above 3.0. Forty-three of those students returned for their second year of college. Reaves pointed out that that retention rate (79.6 percent) is comparable to WSSU’s overall freshman retention rate—in fact, it is slightly higher.  

Souza, at last week’s educational planning committee meeting, said that Winston-Salem State and UNC-Pembroke may eventually join the pilot program. The chancellors at the participating universities will have to report annually to UNC system president Thomas Ross. Ross will then provide progress reports to the Board of Governors each January.  

Only four board members voted against the latest admissions pilot. 

 


Please observe the Pope Center's commenting policy.


blog comments powered by Disqus

Return to the Commentaries Archive

Copyright © 2014 The John William Pope Center for Higher Education Policy | Site Map

Website design and development by DesignHammer Media Group, LLC. Building Smarter Websites.