Assessment in American Higher Education: The Role of Admissions Tests

• DOI: 10.1177/0002716219843469
• Published date: 01 May 2019
• Date: 01 May 2019
• Author: Rebecca Zwick
• Subject Matter: Assessments for Selection and Placement

843469ANN
THE ANNALS OF THE AMERICAN ACADEMYASSESSMENT IN AMERICAN HIGHER EDUCATION
research-article2019
In this article, I review the role of college admissions
tests in the United States and consider the fairness
issues surrounding their use. The two main tests are the
SAT, first administered in 1926, and the AcT, first
given in 1959. Scores on these tests have been shown to
contribute to the prediction of college performance,
but their role in the admissions process varies widely
across colleges. Although test scores are consistently
listed as one of the most important admissions factors
Assessment in in national surveys of postsecondary institutions, an
increasing number of schools have adopted “test-
American
optional” policies. At these institutions, test score
requirements are seen as a barrier to campus diversity
because of the large performance gaps among ethnic
Higher
and socioeconomic groups. Fortunately, the decentral-
ized higher education system in the United States can
Education: The accommodate a wide range of admissions policies. It is
essential, however, that the impact of admissions policy
changes be studied and that the resource implications
Role of
of these changes be thoroughly considered.
Admissions
Keywords: college admissions; standardized tests;
AcT; SAT; diversity
Tests
To understand the role of standardized tests
in the selection of college students in the
United States, it is useful to consider some his-
tory.1 At the turn of the twentieth century,
By
admission to U.S. colleges generally followed
REbEccA ZwIck
one of two paths. Applicants to elite schools
Rebecca Zwick is a distinguished presidential appointee
at Educational Testing Service and professor emerita at
the University of California, Santa Barbara. She is a fel-
low of the American Educational Research Association
and the American Statistical Association and is president
of the National Council on Measurement in Education.
Her books include Fair Game? The Use of Standardized
Admissions Tests in Higher Education (Routledge 2002)
and who Gets In? Strategies for Fair and Effective
college Admissions (Harvard University Press 2017).
NOTE: The opinions expressed here are my own and not
necessarily those of Educational Testing Service. I am
grateful to brent bridgeman, James carlson, David klieger,
and the volume editors for their thoughtful comments.
correspondence: rzwick@ets.org
DOI: 10.1177/0002716219843469
130 ANNALS,
AAPSS, 683, May 2019

---

ASSESSMENT IN AMERIcAN HIGHER EDUcATION
131
such as Harvard and Yale were subjected to stringent and lengthy examinations
imposed by the individual institutions. Many less selective institutions relied
instead on an inspection system in which applicants from approved high schools
were admitted automatically (wechsler 1977). In 1900, twelve institutions
formed the college Entrance Examinations board, which created a common
system of essay exams. Advances in testing, spurred in part by world wars I and
II, along with the invention of automated scoring machines, led to the subse-
quent development of multiple-choice admissions tests.
The United States is unique in the degree to which college admissions is
decentralized. Admissions tests are produced by private companies, rather than
government entities, and colleges are free to use them or not. why would a col-
lege choose to use a student’s test score as an admissions criterion? For many
schools, it is simply a matter of efficient resource utilization: administrators wish
to expend resources on those candidates who are likely to be able to do the
required academic work and believe that test scores are helpful in identifying
these students. A less benign reason for embracing tests is that high average test
scores and low admission rates have come to be seen as a measure of institutional
quality, increasing the school’s attractiveness to candidates.
Among schools that do use admissions test scores, the accompanying student
information, which provides the context for evaluating the test scores, varies
widely. Although high school grades are taken into account by all schools, the
array of other factors considered by some institutions is almost limitless. These
include accomplishments (such as academic awards, work experience, artistic
achievements, and athletic success), demographic factors (such as first language,
socioeconomic status, and race), and personal characteristics (such as “grit”).
The broad range of admissions policies among American institutions reflects
the deep national ambivalence about admissions testing. On one hand, reward-
ing academic excellence is considered a good thing. From this perspective, it
seems fair to give preference to high scorers when allocating university seats,
provided that the admissions tests are accepted as a valid measure of academic
skills. On the other hand, because of well-known test performance differences
among ethnic and socioeconomic groups, admissions testing is seen by some as
antithetical to the fundamental principle of equal opportunity for all. From this
point of view, test scores should be given little weight or perhaps eliminated
completely from the admissions process. Even the evidence on the predictive
value of admissions tests, which would seem to be a straightforward matter, is
viewed through lenses that seem to depend on prior beliefs, with test propo-
nents arguing that test scores are clearly predictive of college outcomes and
detractors claiming the opposite.
before delving into these complexities, it is useful to consider the specifics of
the two main college admissions tests in the United States—the SAT, first admin-
istered in 1926, and the AcT, first given in 1959. The most recent version of the
SAT, which debuted in 2016, includes mandatory components in mathematics
and in reading, writing, and language. The main scores are two section scores—
one in math and one in “evidence-based reading and writing,” as well as a total
score that is the sum of the section scores. All questions are multiple choice

---

132
THE ANNALS OF THE AMERIcAN AcADEMY
except for 22 percent of the math items, which require test-takers to supply an
answer. An optional essay yields scores in reading, analysis, and writing. In addi-
tion to the SAT itself, the SAT program includes twenty SAT subject tests, which
assess knowledge in particular areas, such as U.S. history, math, physics, French,
Spanish, and chinese. The SAT and the subject tests are owned by the college
board and administered by Educational Testing Service under a contractual
agreement.
The AcT, offered by a company also called AcT, has traditionally differed
somewhat from the SAT in that it is more closely linked to instructional goals,
though the two tests have grown more similar over time. The AcT consists of
four multiple-choice sections: English, mathematics, reading, and science.
Students receive a score in each area, along with a composite score. The AcT
recently began offering a “STEM score” as well; this is based on the student’s
performance on the math and science sections. Students who elect to take an
optional essay test in writing receive a writing test score and an English language
arts score based on the English, reading, and writing results.
In this article, I consider the role of these admissions tests in selecting college
students. In the first section, I summarize research findings on the predictive validity
of admissions tests—the degree to which the test scores can predict college out-
comes. Then, I discuss fairness issues; and in the third section, I consider the specific
role of test scores in the admissions process, including a description of test-optional
admissions and percent plans. In the final section, I offer some conclusions.
Predictive Validity of college Admissions Tests
According to the Standards for Educational and Psychological Testing, “validity
refers to the degree to which evidence and theory support the interpretations of
test scores for proposed uses of tests. … The process of validation involves accu-
mulating relevant evidence to provide a sound scientific basis for the proposed
score interpretations” (American Educational Research Association, American
Psychological Association, and National council on Measurement in Education
[AERA, APA, and NcME] 2014, 11).
what is the evidence in favor of using the SAT or AcT to select college stu-
dents? One source of validity evidence is the content of the tests, which is
intended to reflect skills needed for college success. but the primary evidence
presented in support of the utility of these tests as admissions tools is the degree
to which they predict college performance. Admissions tests have been found to
be predictive of grades throughout college and of college graduation as well (see
Sackett and kuncel [2018] and Zwick [2017, 96] for reviews).
Prediction of first-year college grade-point average
The vast majority of predictive validity studies of college admissions tests use
the first-year college grade-point average (FGPA). The FGPA criterion, though

---

ASSESSMENT IN AMERIcAN HIGHER EDUcATION
133
limited in some respects, has two clear advantages. It is less affected by student
attrition than criteria that must be assessed later on, such as graduation. Also,
because the course of study is more similar across majors in the first year of col-...