Different Tests, Same Flaws: A Comparison of the SAT, SAT II and ACT

Recent debate in college admissions has centered on a critique of the SAT I in favor of the SAT II and/or ACT. Proponents of these alternatives argue that the SAT I is primarily an aptitude test measuring some vague concept of "inherent ability," while the SAT II and ACT are more closely tied to what students learn in high school. However, while the origins of the exams and the rhetoric test-makers offer may differ, the SAT I, SAT II, and ACT present many of the same flaws and shortcomings. All three exams have a weak ability to predict academic performance in college, making high school grades/GPA and rigor of classes taken the best measures of student potential. All three exams are highly coachable, advantaging students who can afford to spend $800 or more on test preparation classes. All three exams have a similar format, disadvantaging groups such as females and English as a Second Language learners who tend not to perform as well on timed, multiple-choice exams. All three exams show large gaps in scores between students of different racial groups, leading to racial bias in admissions and financial aid formulas that utilize rigid test score requirements. All three exams place the financial and time burden on students rather than universities, making them low-investment sources of information for colleges but high-investment hurdles for students. Finally, all three exams assess students on a narrow range of topics, covering only a small portion of the learning students engage in over four years of high school. The conclusion: substituting one university admissions exam for another benefits neither students nor schools. Test-score optional policies are one significant step towards a sound and equitable admissions process. Every college and university in the nation can - and should - follow the lead of the nearly 400 institutions that admit a substantial number of freshman applicants without regard to SAT or ACT scores.


 

   SAT I SAT II ACT
Format Multiple-choice, 3 hr. exam with two sections: Verbal and Math. Approximately 1.2 million test-takers annually.
(format changing in 2005)
Multiple-choice (except for part of Writing), 1 hr. exams in 22 subject areas (Writing, Literature, US History, World History, Math Level I and IIC, Biology, Chemistry, Physics, & language tests). Currently required by 55 colleges. Multiple-choice, 3 ½ hr. exam with four components: Mathematics, English, Reading, and Science Reasoning. Approximately 1.1 million test-takers annually.
(optional writing test implemented in 2005)
Purpose according to the test-makers College Board: "The SAT I: Reasoning Tests are measures of abilities in quantitative and verbal reasoning that develop over a long period of time and are not closely tied to specific academic subjects." College Board: "SAT II: Subject Tests are designed to measure knowledge, and the ability to apply that knowledge, in specific subject areas. The tests are independent of any particular textbook or method of instruction. The content of the tests evolves to reflect current trends in high school curricula, but the types of questions change little from year to year." ACT, Inc.: "The ACT Assessment tests are curriculum based. The ACT Assessment is not an aptitude or an IQ test. Instead, the questions on the ACT are directly related to what is learned in high school courses in English, mathematics, and science."
Origin A direct descendant of the racist anti-immigrant Army Mental Tests of the 1920s, the SAT was first administered in 1926 but did not become fully multiple-choice until after WW II. The SAT I was developed by Carl Brigham, who believed that people of different races could be rank ordered by intelligence level based on test scores and classified into four racial strains - Nordic, Alpine, Mediterranean, and Negro. The earliest versions of the SAT II exams were administered in the early 1900s and required only by very elite institutions. These entrance examinations were essay tests in subjects such as English, French, Greek, mathematics, botany, and physics. In 1916 the College Board began to administer new comprehensive exams that were broader in scope, in addition to the essay tests. Until the mid-1990s the SAT IIs were referred to as Achievement Tests. ACT Inc. was created in 1959 in response to the SAT I. The ACT, created by E.F. Lindquist (who also designed the Iowa Test Basic Skills) and Ted McCarrel, was initially designed to be more closely tied to high school curriculum than the SAT I and to spur curricular reform. From John Katzman: "The ACT was supposed to be a combined achievement/aptitude test, whereas the SAT was pure aptitude, and it was supposed to be used mainly by state universities for placement/guidance, rather than by elite schools for selection. A much more democratic vision. But, the scores correlate very highly, it turns out, and most universities now treat them as interchangeable and let you take either one."
Testing fee  $26.00 $12.00 Writing Test, $9.00 Language Tests w/ Listening, $7.00 all other tests $25.00 ($28.00 in Florida)
Students can choose which scores to send?  No No Yes
Predictive validity evidence from test-makers College Board research claims the test predicts around 22% of the variation in first-year college grades. College Board research claims that the all SAT II exams predict on average around 16% of the variation in first-year college grades. The SAT II Writing Test predicts around 25% of the variation in first-year college grades. ACT research claims the test predicts around 17% of the variation in first-year college grades.
Validity research from independent studies The Case Against the SAT by James Crouse and Dale Trusheim points to the SAT I's poor utility in forecasting both short- and long-term success. They compared two admissions strategies, one using just the high school record and the other using high school record and SAT I scores. More than 90% of the admissions decisions were the same under both strategies. However, the SAT-based strategy led to a far greater number of rejections of otherwise academically qualified minority and low-income applicants.

Data they analyzed also demonstrated that using the high school record alone to predict who would complete a bachelor's degree resulted in "correct" admissions decisions 73.4% of the time, while using the SAT I and high school GPA forecast "correct" admissions in 72.2% of the cases.

One study (J. Baron & M. F. Norman in Educational and Psychology Measurement, Vol. 52, 1992) at the University of Pennsylvania looked at the power of high school class rank, SAT I, and SAT II in predicting cumulative college GPAs. Researchers found that the SAT I was by far the weakest predictor, explaining only 4% of the variation in college grades, while SAT II scores accounted for 6.8% of the differences in academic performance. By far the most useful tool proved to be class rank, which predicted 9.3% of the changes in cumulative GPAs. Combining SAT I scores and class rank inched this figure up to 11.3%, leaving almost 90% of the variation in grades unexplained.

A University of California study analyzing the power of the SAT series and high school grades to predict success at the state's eight public universities found that the SAT I was the weakest predictor, with the test accounting for only 12.8% of the variation in FGPA. SAT II's and HSGPA explained 15.3% and 14.5% of the variation, respectively.

One study at Chicago State University confirmed that the ACT does a poor job of predicting academic performance in college. For the vast majority of the university's graduates who scored in the middle range of the test as high school students, the ACT explained only 3.6% of the differences in cumulative college GPA. In fact, the exam over-predicted the performance of the class graduating in 1992, which had the highest average ACT score among the classes in the research study yet the poorest academic performance over four years at the university.

Another study conducted at a medium-sized regional university in the Southeast showed that the ACT explained only 6.8% of the differences in first-semester college grades for African Americans, while for Whites the ACT predicted approximately 28% of the differences. High school grades predicted equally well for both groups, demonstrating that other measures of achievement are not as subject to differences across racial groups as are test scores.

Racial score gaps (points above of below White average) Verbal + Math:
African Amer.: - 203
Amer. Indian: - 98
Asian/Asian Am.: + 10
Mexican Amer.: - 157
Other Latino: - 138
Puerto Rican: - 154
Writing:
African Amer.: - 82
Amer. Indian: - 51
Asian/Asian Am.: - 30
Mexican Amer.: - 106
Puerto Rican: - 67

Math IIC:
African Amer.: - 69
Amer. Indian: - 37
Asian/Asian Am.: + 16
Mex. Amer.: - 63
Puerto Rican: - 39

Composite (SAT equivalent):
African Amer.: - 4.9 (- 198)
Amer. Indian: - 3.1 (- 124)
Asian/Asian Am.: - 0.1 (- 4)
Mexican Amer.: - 3.5 (- 140)
Puerto Rican: - 2.5 (- 116)

(one point on ACT composite equals approximately 40 points on the SAT M+V, according to test-makers' concordance tables)

Gender score gaps Verbal: males 5 points higher

Math: males 34 points higher

Math IIC: males 35 points higher Composite score: males 0.2 points higher
Impact of coaching Princeton Review "guarantees" at least a 100 point increase in SAT I scores. None of the Above, by David Owen: "As John Katzman said…ETS' Achievement Tests [SAT IIs] are even easier to coach than the SAT [I] - not because, as some have said, their subject matter is more finite but because the method by which they are put together is more transparent." Princeton Review "guarantees" at least a 4 point increase in ACT scores, equivalent to approxmately 150 points on the SAT I.
What coaching companies have to say Kaplan, Inc.: "The SAT I lends itself to preparation because it is a predictable test concentrating on a set of specific reasoning skills that can be learned. This kind of bias is evident on all examinations. Better prepared students score higher." None of the Above, by David Owen: "Should they require ETS' SAT II Subject Tests instead of the SAT? 'That would be fine with me,' says John Katzman. 'I can prepare a kid for the math Achievement, English Achievement - three weeks, 200 points.'" Princeton Review: "The ACT tests the same information the same way, year after year…By reviewing the very specific knowledge that the people who write the ACT think is important, and by learning good test-taking strategies, it should be possible to increase your ACT score significantly."
Guessing penalty? Yes Yes No
From the College Board: "[S]cores on the SAT I and ACT are highly correlated; in the three most recent concordance tables, the correlations between individuals' SAT I and ACT scores range from .89 to .92." The correlation for SAT I and SAT II scores from one College Board study was .84.