Reflections on a Century of College Admissions Tests

Educational Researcher
http://er.aera.net
http://edr.sagepub.com/content/38/9/665
The online version of this article can be found at:
DOI: 10.3102/0013189X09351981
EDUCATIONAL RESEARCHER 2009 38: 665
Richard C. Atkinson and Saul Geiser
Reflections on a Century of College Admissions Tests
Published on behalf of
American Educational Research Association
and
http://www.sagepublications.com
Additional services and information for Educational Researcher can be found at:
Email Alerts: http://er.aera.net/alerts
Subscriptions: http://er.aera.net/subscriptions
Reprints: http://www.aera.net/reprints
Permissions: http://www.aera.net/permissions
What is This?
>> Version of Record – Dec 22, 2009
Downloaded from http://er.aera.net at UNIV ARIZONA LIBRARY on August 18, 2013
Educational Researcher, Vol. 38, No. 9, pp. 665676
DOI: 10.3102/0013189X09351981
2009 AERA. http://er.aera.net december 2009 665
The College Boards started as achievement tests designed to measure students mastery of college preparatory subjects. Admissions
testing has significantly changed since then with the introduction of
the Scholastic Aptitude Test, Lindquists creation of the ACT, renewed
interest in subject-specific assessments, and current efforts to adapt
K12 standards-based tests for use in college admissions. We have
come full circle to a renewed appreciation for the value of achievement tests. Curriculum-based achievement tests are more valid indicators of college readiness than other tests and have important
incentive or signaling effects for K12 schools as well: They help
reinforce a rigorous academic curriculum and create better alignment of teaching, learning, and assessment along the pathway from
high school to college.
Keywords: achievement; admissions; assessment; colleges; educational policy; testing
Standardized testing for college admissions has seen extraordinary growth over the past century and appears to be on
the cusp of still more far-reaching changes. Fewer than
1,000 examinees sat for the first College Boards in 1901. Today
more than 1.5 million students take the SAT, 1.4 million sit for
the ACT, and many students take both. This does not count
many more who take preliminary versions of college entrance
tests earlier in school, nor does it include those who take the SAT
Subject Tests and Advanced Placement (AP) exams. Admissions
testing continues to be a growth industry, and further innovations such as computer-based assessments with instant scoring,
adaptive testing, and noncognitive assessment are poised to
make their appearance.
Despite this growth and apparent success, the feeling persists
that all is not well in the world of admissions testing. College
entrance tests and related test preparation activities have contributed mightily to what has been called the educational arms
racethe ferocious competition for admission at highly selective institutions (Atkinson, 2001). Many deserving low-income
and minority students are squeezed out in this competition, and
questions about fairness and equity are raised with increasing
urgency. The role of the testing agencies themselves has also come
into question, and some ask whether the testing industry holds
too much sway over the colleges and universities it purports to
serve. Underlying all of these questions is a deeper concern that
the current regime of admissions testing may impede rather than
advance our educational purposes.
This article reflects on the first century of admissions testing
with a view to drawing lessons that may be useful as we now contemplate the second. Our aim is not to extrapolate from the past
or to predict the specific forms and directions that admissions tests
may take in the future. Rather, our intent is to identify general
principles that may help guide test development going forward.
Putting Tests in Perspective:
Primacy of the High School Record
A first order of business is to put admissions tests in proper perspective: High school grades are the best indicator of student
readiness for college, and standardized tests are useful primarily
as a supplement to the high school record.
High school grades are sometimes viewed as a less reliable
indicator than standardized tests because grading standards differ
across schools. Yet although grading standards do vary by school,
grades still outperform standardized tests in predicting college
outcomes: Irrespective of the quality or type of school attended,
cumulative grade point average (GPA) in academic subjects in
high school has proved to be the best overall predictor of student
performance in college. This finding has been confirmed in the
great majority of predictive-validity studies conducted over the
years, including studies conducted by the testing agencies themselves (see Burton & Ramist, 2001, and Morgan, 1989, for useful
summaries of studies conducted since 1976).1
In fact, traditional validity studies tend to understate the true
value of the high school record, in part because of the methods
employed and in part because of the outcomes studied. Such
studies usually rely on simple correlation methods. For example,
they examine the correlation between SAT scores and college
grades, and the size of the correlation is taken to represent the
predictive power of the SAT. At most, these studies report multiple correlations involving only two or three variables, as, for
example, when they examine the joint effect of SAT scores and
high school grades in predicting first-year college grades (see, e.g.,
Kobrin, Patterson, Shaw, Mattern, & Barbuti, 2008).
But correlations of this kind can be misleading because they
mask the contribution of socioeconomic and other factors to the
prediction. Family income and parents education, for example,
are correlated with SAT scores and also with college outcomes, so
that much of the apparent predictive power of the SAT actually
Reflections on a Century of College Admissions Tests
Richard C. Atkinson and Saul Geiser
Features
Downloaded from http://er.aera.net at UNIV ARIZONA LIBRARY on August 18, 2013
666 educational Researcher
reflects the proxy effects of socioeconomic status. Princeton
economist Jesse Rothstein (2004) conservatively estimates that
traditional validity studies that omit socioeconomic variables
overstate the predictive power of the SAT by 150%.2 High school
grades, on the other hand, are less closely associated with students socioeconomic background and so retain their predictive
power even when controls for socioeconomic status are introduced, as shown in validity studies that employ more fully specified multivariate regression models. Such models generate
standardized regression coefficients that allow one to compare the
predictive weight of different admissions factors when all other
factors are held constant. Using this analytical approach, the predictive advantage of high school grades over standardized tests is
more evident (Geiser, 2002; Geiser & Santelices, 2007).3
The predictive superiority of high school grades has also been
obscured by the outcome measures typically employed in validity
studies. Most studies have looked only at freshman grades in college; relatively few have examined longer term outcomes such as
4-year graduation or cumulative GPA in college. A large-scale
study at the University of California (UC) that did track longterm outcomes found that high school grades were decisively
superior to standardized tests in predicting 4-year graduation and
cumulative college GPA (Geiser & Santelices, 2007). The
California findings have been confirmed in a recent national
study of college completion by William Bowen and his colleagues, Crossing the Finish Line, based on a sample of students at
a broad range of public colleges and universities: High school
grades are a far better predictor of both four-year and six-year
graduation rates than are SAT/ACT test scoresa central finding
that holds within each of the six sets of public universities that we
study (Bowen, Chingos, & McPherson, 2009, pp. 113114).
Why high school grades have a predictive advantage over standardized tests is not fully understood, as it is undeniable that
grading standards differ across high schools. Yet standardized test
scores are based on a single sitting of 3 or 4 hours, whereas high
school GPA is based on repeated sampling of student performance over a period of years. And college preparatory classes
present many of the same academic challenges that students will
face in collegeterm papers, labs, final examsso it should not
be surprising that prior performance in such activities would be
predictive of later performance.
Whatever the precise reasons, it is useful to begin any discussion of standardized admissions tests with acknowledgment that
a students record in college preparatory courses in high school
remains the best indicator of how the student is likely to perform
in college. Standardized tests do add value. In our studies at the
University of California, for example, we have found that admissions tests add an increment of about 6 percentage points to the
explained variance in cumulative college GPA, over and above
about 20% of the variance that is accounted for by high school
GPA and other academic and socioeconomic factors known at
point of admission (Geiser & Santelices, 2007). And tests can
add value in other important ways, beyond prediction, that we
shall consider later in this article.
Testing for Ability: The Saga of the SAT
The SAT, or Scholastic Aptitude Test, first made its appearance
in 1926 as an alternative to the earlier College Boards. Whereas
the older tests were written, curriculum-based examinations
designed to assess student learning in college preparatory subjects, the SAT promised something entirely new: an easily scored,
multiple-choice instrument for measuring students general ability or aptitude for learning (Lemann, 1999).
The similarity between the early SAT and IQ testing was not
coincidental, and the two shared a number of assumptions that
most now regard as problematic. The SAT grew out of the experience with IQ tests during the First World War, when 2 million
men in military service were tested and assigned an IQ based on
the results. The framers of those tests assumed that intelligence
was a unitary, inherited attribute; it was not subject to change
over a lifetime and could be measured in a single number.
Although the SAT was more sophisticated from a psychometric
standpoint, it evolved from the same questionable assumptions
about human talent and potential.
Yet especially in the years after World War II, the idea of the
SAT resonated strongly with the meritocratic ethos of American
college admissions. The SAT was standardized in a way that high
school grades were not, and it could be administered relatively
inexpensively to large numbers of students. If aptitude for learning
could be reliably measured, the SAT could help identify students
from disadvantaged circumstances who were deserving of admissionthus improving access and equity in college admissions.
Above all, the SAT offered a tool for prediction, providing admissions officers a means to distinguish between applicants who were
likely to perform well or poorly in college. It is easy to understand
why the test gained widespread acceptance in the postwar years.
The SAT has evolved considerably since that time, and both
the name of the test and the terminology describing what it is
intended to measure have changed. In an effort to alter the perception of the tests link to the older IQ tradition, in 1990 the
College Board changed the name to the Scholastic Assessment
Test and then in 1996 dropped the name altogether, so that the
initials no longer stand for anything. Official descriptions of
what the test is supposed to measure have also changed over the
years from aptitude to generalized reasoning ability and now
critical thinking, and the test items and format have been more
or less continuously revised (Lawrence, Rigol, Van Essen, &
Jackson, 2003). Throughout these changes, the one constant has
been the SATs claim to gauge students general analytic ability,
as distinct from their mastery of specific subject matter, and
thereby to predict performance in college.
By the end of the 20th century, however, the SAT had become
the object of increasing scrutiny, partly as a result of developments at our own institution, the University of California. After
Californians voted to end affirmative action in 1996, the UC
system undertook a sweeping review of its admissions policies in
an effort to reverse plummeting Latino and African American
enrollments. What we found challenged many established beliefs
about the SAT.
Far from promoting equity and access in college admissions,
we found thatcompared with traditional indicators of academic achievementthe SAT had a more adverse impact on lowincome and minority applicants.4 The SAT was more closely
correlated than other indicators with socioeconomic status and
so tended to diminish the chances of admission for underrepresented minority applicants, who come disproportionately from
Downloaded from http://er.aera.net at UNIV ARIZONA LIBRARY on August 18, 2013
december 2009 667
lower socioeconomic backgrounds. For example, when UC
applicants were rank ordered by SAT scores, roughly half as many
Latino, African American, and American Indian students
appeared in the top of the applicant pool as when the same students were ranked by high school grades (Geiser & Santelices,
2007).
Another surprise was the relatively poor predictive power of
the SAT (then also known as the SAT I) as compared not only
with high school grades but also with curriculum-based achievement tests, such as the SAT II subject tests and AP exams, which
measure students mastery of specific subjects. The SAT Is claim
to assess general analytic ability, independent of curriculum content, was long thought to give it an advantage over achievement
tests in predicting how students will perform in college.
The University of California had required applicants to take
both the SAT I and a battery of achievement tests since 1968 and
so had an extensive database to evaluate that claim. Our data
showed that the SAT I reasoning test was consistently inferior to
the SAT II subject tests in predicting student performance,
although the difference was small and there was substantial overlap between the tests. It was not the size of the difference but the
consistency of the pattern that was most striking. The subject
testsparticularly the writing examheld a predictive advantage over the SAT I reasoning test at all UC campuses and within
every academic discipline (Geiser, 2002).5,6 And in later studies
we found that the AP exams, which require the greatest depth of
subject knowledge, exhibited an even greater predictive advantage (Geiser & Santelices, 2006). Mastery of curriculum content,
it turns out, is important after all.
Another concern with the SAT I was its lack of fit with the
needs of K12 schools. After affirmative action was dismantled,
UC massively expanded its outreach to low-performing schools
throughout California in an effort to restore minority admissions
over the long term. At their height, before later state budget cuts,
UC outreach programs were serving 300,000 students and
70,000 teachers, and UC campuses had formed schooluniversity partnerships with 300 of the lowest performing schools in the
state. College admissions criteria can have a profound influence,
for good or ill, on such schoolswhat Michael Kirst has called a
signaling effect (Kirst & Venezia, 2004)and it was evident
that the SAT was sending the wrong signals.
The SAT I sent a confusing message to students, teachers, and
schools. It featured esoteric items, like verbal analogies and quantitative comparisons, rarely encountered in the classroom. Its
implicit message was that students would be tested on materials
that they had not studied in school and that the grades they
achieved could be devalued by a test that was unrelated to their
course work. Especially troubling, the perception of the SAT I as
a test of basic intellectual ability had a perverse effect on many
students from low-performing schools, tending to diminish academic aspiration and self-esteem. Low scores on the SAT I were
too often interpreted as meaning that a student lacked the ability
to attend the University of California, notwithstanding his or her
record in high school.7
These concerns prompted the first author of this article to
propose dropping the SAT I in favor of curriculum-based achievement tests in UC admissions (Atkinson, 2001).8 The University
of California accounts for a substantial share of the national
market for admissions tests, and the College Board responded to
our concerns with a revised SAT in 2005.
The New SAT (now also known as the SAT-R, for reasoning) is clearly an improvement over the previous version of the
test. The SAT II writing exam has been incorporated into the test,
and verbal analogies have been dropped. Instead of deconstructing esoteric analogies, students must now perform a task they will
actually face in collegewriting an essay under a deadline. The
old SAT featured math items, such as quantitative comparisons,
that were known for their trickery but required only an introductory knowledge of algebra; the New SAT math section is more
straightforward and covers some higher level topics in algebra.
Reports indicate that the changes have galvanized a renewed
focus on math and especially writing in many of the nations
schools (Noeth & Kobrin, 2007).
Nevertheless, as an admissions test the New SAT still falls short
in important respects. The New SAT has three sections: writing,
mathematics, and a third called critical reading. Not surprisingly,
given the University of Californias earlier findings, research by the
College Board shows that writing is the most predictive of the three
sections. Yet College Board researchers also find that, overall, the
New SAT is not statistically superior to the old test in predicting
success in college: The results show that the changes made to the
SAT did not substantially change how well the test predicts first-year
college performance (Kobrin et al., 2008, p. 1). This result was
unexpected, given the strong contribution of the writing test and the
fact that the New SAT is almost an hour longer than the old test.9
A possible explanation is provided by another study by three
economists at the University of Georgia (Cornwell, Mustard, &
Van Parys, 2008). That study found that adding the writing section to the New SAT has rendered the critical-reading section
almost entirely redundant so that it does not add significantly to
the prediction. The critical-reading section is essentially the same
as the verbal-reasoning section of the old SAT I. It appears that
the College Board was trying to have the best of both worlds. The
College Board could and did tell admissions officers that the
critical-reading and math sections of the New SAT were comparable to the verbal- and mathematical-reasoning sections of the
old SAT I. If admissions officers disliked the New SAT, they
could ignore the writing exam and then for all practical purposes
the old and new SAT would be equivalent.10
A more fundamental question is what, exactly, the new test is
intended to measure. The SATs underlying test construct has
long been ambiguous, and the recent changes have only added to
the confusion. Although the inclusion of the writing test and
some higher level math items are evidently intended to position
the New SAT as more of an achievement test, its provenance as a
test of general analytic ability remains evident as well. The verbal
and math sections continue to feature items that are remote from
what students encounter in the classroom, and the College Board
has emphasized the psychometric continuity between the old and
new versions of the test (Camara & Schmidt, 2006). In a phrase,
the New SAT appears to be a test at war with itself (Geiser,
2009), and it will be interesting to see which impulse prevails in
future iterations of the test.
Although a significant improvement over the old test, the
New SAT remains fundamentally at odds with educational priorities along the pathway from high school to college. The New
Downloaded from http://er.aera.net at UNIV ARIZONA LIBRARY on August 18, 2013
668 educational Researcher
SATs lack of alignment with high school curricula has become
especially conspicuous now that more and more states have
moved toward standards-based assessments at the K12 level.
Standards-based tests seek to align teaching, learning, and assessment. They give feedback to students and schools about specific
areas of the curriculum where they are strongest and weakest,
providing a basis for educational improvement and reform
(Darling-Hammond, 2003). Aligning admissions tests with the
needs of our schoolsespecially schools serving populations that
have been traditionally underserved by higher educationmust
be a priority as we look to the next generation of standardized
admissions tests.
Testing for Achievement: Enter the ACT
The ACT was introduced in 1959 as a competitor to the SAT.
From its inception, the ACT has reflected an alternative philosophy of college admissions testing espoused by its founder, E. F.
Lindquist (1958):
If the examination is to have the maximum motivating value for
the high school student, it must impress upon him the fact that
his chances of being admitted to college . . . depend not only on
his brightness or intelligence or other innate qualities or factors for which he is not personally responsible, but even more
upon how hard he has worked at the task of getting ready for
college. . . . The examination must make him feel that he has
earned the right to go to college by his own efforts, not that he is
entitled to college because of his innate abilities or aptitudes,
regardless of what he has done in high school. In other words,
the examination must be regarded by him as an achievement test.
(pp. 108109)
From our vantage half a century later, Lindquists vision of admissions testing seems remarkably fresh and prescient. His understanding of the signaling effect of college admissions criteria for
K12 students and schools reflects a modern sensibility, as does
his admonition that educators must not allow their standards to
be set, by default, by the tests they use. Assessment should flow
from standards, not the other way round. Lindquists concept of
achievement testing was also quite sophisticated; as against those
who would caricature such tests as measuring only rote recall of
facts, he insisted that achievement tests can and should measure
students reasoning skills, albeit those developed within the context of the curriculum.
Reflecting Lindquists philosophy, the ACT from the beginning has been tied more closely than the SAT to high school
curricula. The earliest forms of the test grew out of the Iowa Tests
of Educational Development and included four sections
English, mathematics, social studies reading, and natural sciences
readingreflecting Iowas high school curriculum. As the ACT
grew into a national test, its content came to be based on national
curriculum surveys as well as analysis of state standards for K12
instruction. In 1989 the test underwent a major revision and the
current four subject areas were introduced (English, mathematics, reading, and science), and in 2005 the ACT added an
optional writing exam in response, in part, to a request from the
University of California.
The ACT exhibits many of the characteristics that one would
expect of an achievement test. It is developed from curriculum
surveys. It appears less coachable than the SAT, and the consensus
among the test prep services is that the ACT places less of a premium on test-taking skills and more on content mastery. The
ACT also has a useful diagnostic component to assist students as
early as the eighth grade to get on and stay on track for college
another function that Lindquist believed an admissions test
should perform (ACT, 2009b).
Yet the ACT still falls short of a true achievement test in several ways. Like the SAT, the ACT remains a norm-referenced test
and is used by colleges and universities primarily to compare students against one another rather than to assess curriculum mastery. The ACT is scored in a manner that produces almost the
same bell curve distribution as the SAT. It is true that the ACT
also provides standards-based interpretations indicating the
knowledge and skills that students at different score levels generally can be expected to have learned (ACT, 2009a). But those
interpretations are only approximations and do not necessarily
identify what an examinee actually knows. It is difficult to reconcile the ACTs norm-referenced scoring with the idea of a criterion-referenced assessment or to understand how one test could
serve both functions equally.
The ACT lacks the depth of subject matter coverage that one
finds in other achievement tests such as the SAT Subject Tests or
AP exams. The ACT science section, for example, is intended to
cover high school biology, chemistry, physics, and earth/space
science. But the actual test requires little knowledge in any of
these disciplines, and a student who is adept at reading charts and
tables quickly to identify patterns and trends can do well on this
sectionunlike the SAT Subject Tests or AP exams in the sciences, which require intensive subject matter knowledge.
In a curious twist, the ACT and SAT appear to have converged over time. Whereas the SAT has shed many of its trickier
and more esoteric item types, like verbal analogies and quantitative comparisons, the ACT has become more SAT-like in some
ways, such as the premium it places on students time management skills. It is not surprising that almost all U.S. colleges and
universities now accept both tests and treat ACT and SAT scores
interchangeably.
Finally, another fundamental problem for the ACTor for
any test that aspires to serve as the nations achievement testis
the absence of national curriculum standards in the United States.
The ACT has tried to overcome this problem through its curriculum surveys, but the average curriculum does not necessarily
reflect what students are expected to learn in any given state, district, or school. The lack of direct alignment between curriculum
and assessment has led the National Association for College
Admissions Counseling (NACAC; 2008) to criticize the practice
followed by some states, such as Colorado, Illinois, and Michigan,
of requiring all K12 students to take the ACT, whether or not
they plan on attending college, and using the results as a measure
of student achievement in the schools. This practice runs counter
to the American Educational Research Associations guidelines on
testing: Admission tests, whether they are intended to measure
achievement or ability, are not directly linked to a particular
instructional curriculum and, therefore, are not appropriate for
detecting changes in middle school or high school performance
(American Educational Research Association, American
Psychological Association, and National Council on Measurement
in Education, 1999, p. 143).
Downloaded from http://er.aera.net at UNIV ARIZONA LIBRARY on August 18, 2013
december 2009 669
Of course, using the ACT to assess achievement in high school
is not the same as using it to assess readiness for college. But the
same underlying problemthe loose alignment between curriculum and assessmentis evident in both contexts. It may be that
no one test, however well designed, can ever be entirely satisfactory in a country with a strong tradition of federalism and local
control over the schools. A single national achievement test may
be impossible in the absence of a national curriculum.
Assessing Achievement in Specific
Subjects: SAT Subject Tests and AP Exams
In place of a single test, another approach taken at some colleges
and universities is to require several achievement tests in different
subjects. The assessments most often used are the SAT II subject
tests and AP exams.
During the 1930s, the College Board developed a series of
multiple-choice tests in various subject areas to replace its older,
written exams. These later became known as the SAT IIs and are
now officially called the SAT Subject Tests. In 1955 the College
Board introduced the Advanced Placement program and with it,
the AP exams. As their name indicates, the AP exams were originally intended for use in college placement: Colleges and universities used AP exam scores mainly to award course credits, allowing
high-achieving students to place out of introductory courses and
move directly into more advanced college work. Over time, however, AP has come to play an increasingly important role in admissions at selective institutions, and its role in admissions is now
arguably more important than its placement function.11
Of all nationally administered tests used in college admissions, the SAT Subject Tests and AP exams are the best examples
of achievement tests currently available. The SAT Subject Tests
are offered in about 20 subject areas and the AP exams in more
than 30. The SAT Subject Tests are hour-long, multiple-choice
assessments, whereas the AP exams take 2 to 3 hours and include
a combination of multiple-choice, free-answer, and essay questions. Students frequently sit for the tests after completing high
school course work in a given subject, so that tests often serve, in
effect, as end-of-course exams. Test prep services such as the
Princeton Review advise students that the most effective way to
prepare for subject exams is through course work, and in a telling
departure from its usual services, the Review offers contentintensive coursework in mathematics, biology, chemistry, physics, and U.S. history to help students prepare for these tests
(Princeton Review, 2009).
Until the SAT II Writing exam was discontinued and became
part of the New SAT in 2005, the University of California had
for many years required three subject tests for admission to the
UC system: SAT Writing, SAT II Mathematics, and a third SAT
II subject test of the students choosing.12 The elective test
requirement was established to give students an opportunity to
demonstrate particular subjects in which they excel and to assist
them in gaining admission to particular majors. Students can also
elect to submit AP exam scores, which, though not required, are
considered in admission to individual UC campuses.13
The idea that students should be able to choose the tests they
take for admission may seem anomalous to those accustomed to
viewing the SAT or ACT as national yardsticks for measuring
readiness for college. But the real anomaly may be the idea that
all students should take one test or that one test is suitable for all
students. Our research showed that a selection of three SAT II
subject testsincluding one selected by studentspredicted college performance better than either of the generic national assessments, although scores on all of the tests tended to be correlated
and the predictive differences were relatively small. Of the individual SAT II exams, the elective SAT II subject test proved a
relatively strong predictor, ranking just behind the SAT II Writing
test (Geiser, 2002; Geiser & Santelices, 2007). The AP exams
proved even better predictors. Although mere participation in AP
classes bore no relation to performance in college, students who
took and scored well on the AP exams tended to be very successful: AP exam scores were second only to high school grades in
predicting student performance at the University of California
(Geiser & Santelices, 2006).
Our findings in California on the superiority of achievement
tests, and especially the AP exams, have been confirmed by Bowen
et al.s (2009) recent national study of college completion. Based
on a large sample of students at public colleges and universities,
Bowen and his colleagues found that AP exam scores were
a far better incremental predictor of graduation rates than were
scores on the regular SAT/ACT and, as in the case of the SAT IIs,
including this achievement-test variable in the regression equation entirely removed any positive relationship between the SAT/
ACT scores and graduation rates. . . . It is also important to
emphasize that achievement tests are better predictors than SAT
scores for all students, including minority students and students
from low-SES backgrounds. (pp. 130131)
In the national admissions community there is growing awareness of the value of subject tests. NACAC has recently called on
colleges and universities to reexamine their emphasis on the SAT
and ACT and to expand use of subject tests in admissions.
NACACs commission on testing, which wrote the report,
included many high-profile admissions officials and was chaired
by William Fitzsimmons, dean of admissions at Harvard. The
report is unusually thoughtful and worth quoting at some length:
There are tests that, at many institutions, are both predictive of
first-year and overall grades in college and more closely linked to
the high school curriculum, including the College Boards AP
exams and Subject Tests as well as the International Baccalaureate
examinations. What these tests have in common is that they are
to a much greater extent than the SAT and ACTachievement
tests, which measure content covered in high school courses; that
there is currently very little expensive private test preparation associated with them, partly because high school class curricula are
meant to prepare students for them; and that they are much less
widely required by colleges than are the SAT and ACT. . . .
By using the SAT and ACT as one of the most important
admission tools, many institutions are gaining what may be a
marginal ability to identify academic talent beyond that indicated
by transcripts, recommendations, and achievement test scores. In
contrast, the use of . . . College Board Subject Tests and AP tests,
or International Baccalaureate exams, would create a powerful
incentive for American high schools to improve their curricula
and their teaching. Colleges would lose little or none of the information they need to make good choices about entering classes,
while benefiting millions of American students who do not enroll
in highly selective colleges and positively affecting teaching and
learning in Americas schools (NACAC, 2008, p. 44).
Downloaded from http://er.aera.net at UNIV ARIZONA LIBRARY on August 18, 2013
670 educational Researcher
The main counterargument to expanding use of such tests in college admissions is the fear that they might harm minority, lowincome, or other students from schools with less rigorous
curricula. Currently the SAT Subject Tests and AP exams are considered in admissions only at a few, highly selective colleges and
universities so that the population of test takers is smaller, higher
achieving, and less diverse than the general population that takes
the SAT or ACT. The fear is that if subject tests were used more
widely, students from disadvantaged schools might perform more
poorly than on tests less closely tied to the curriculum.
Experience at the University of California suggests that this
fear is unfounded. After introducing its Top 4 Percent Plan in
2001, which extended eligibility for admission to top students in
low-performing high schools, the university saw a significant
jump in the number of students in these schools who took the
three SAT II subject tests that the university required. Yet lowincome and minority students performed at least as well on these
tests, and in some cases better, than they did on the SAT I reasoning test or ACT. Scores on the SAT II subject tests were in most
cases less closely correlated than SAT I or ACT scores with students socioeconomic status.14 Interestingly, the elective SAT II
subject test had the lowest correlation of any exam with students
socioeconomic status, while remaining a relatively strong indicator of their performance at the University of California (Geiser,
2002).
Nevertheless, as achievement tests, the SAT Subject Tests and
AP exams do have limitations. Scoring on both tests is norm
referenced, despite the fact that colleges often treat them as proficiency tests (especially the AP exams, which are used for college
placement as well as admissions). Oddly, for tests designed to
assess curricular achievement, scores are not criterion referenced
even though they are often interpreted as such.
Another issue is how well the tests actually align with high
school curricula. The SAT Subject Tests and AP exams differ in
this regard. The latter exams are intended primarily for students
who have completed Advanced Placement courses in high school.
This arrangement has both advantages and disadvantages. The
advantage is that the exams are tied to the AP curriculum, but it
also means that the tests are not necessarily appropriate for students who have not taken AP, thus limiting the usefulness of the
exams in college admissions. Also, the AP program has come
under fire from some educators who charge that, by teaching to
the test, AP classes too often restrict the high school curriculum
and prevent students from exploring the material in depth; a
number of leading college preparatory academies have dropped
AP for that reason (Hammond, 2008).
The SAT Subject Tests, on the other hand, are not tied as
directly to particular instructional approaches or curricula but are
designed to assess a core of knowledge common to all curricula
in a given subject area: Each Subject Test is broad enough in
scope to be accessible to students from a variety of academic
backgrounds, but specific enough to be useful to colleges as a
measure of a students expertise in that subject (College Board,
2009b). This enhances their accessibility for use in admissions,
but at a cost: The SAT Subject Tests are less curriculum intensive
than the AP exams, and perhaps for that reason, they are also
somewhat less effective in predicting student success in college
(Geiser & Santelices, 2006).
Without question, the SAT Subject Tests and AP exams have
the strongest curricular foundations of any college entrance tests
now available, and more colleges and universities should find
them attractive for that reason. But both fall short of being fully
realized achievement tests.
Adapting K12 Standards-Based
Tests for Use in College Admissions
The best examples of pure achievement tests now available are
employed not in U.S. higher education but in our K12 schools:
standards-based assessments developed by the various states as
part of the movement to articulate clearer standards for what
students are expected to learn, teach to the standards, and assess
student achievement against those standards.15 The schools are
well ahead of colleges and universities in this regard. In its recent
report, NACACs commission on testing raised the possibility of
adapting K12 standards-based assessments for use in college
admissions:
As one aspect of the standards movement that has swept across
American elementary and secondary public education over the
past quarter-century, many states now require all public high
school students to take achievement-based exams at the end of
high school. These tests vary in quality; the better ones, such as
those in New York, that students take
upon completion of specific courses. Not all state high school
exams are sufficient to measure the prospect of success in postsecondary education. However, if such tests can be developed so they
successfully predict college grades as well as or better than the
SAT, ACT, AP, International Baccalaureate exams, and Subject
Tests do, and align with content necessary for college coursework,
the Commission would urge colleges to consider them in the
admission evaluation process. (NACAC, 2008, p. 44)
The idea of adapting K12 standards-based assessments for use
in college admissions has obvious attractions. In the ideal case,
students performance on end-of-course tests or exit exams could
serve the dual function of certifying both their achievement in
high school and their readiness for college. The burden on students and the amount of testing they must endure could be
greatly reduced. College entrance criteria would be aligned
directly with high school curricula, and the message to students
would be clear and unequivocal: Working hard and performing
well in ones high school course work is the surest route to college.
This is surely a compelling and worthwhile vision. At the
same time, however, there are significant obstacles to its realization. Our experience in California is not necessarily representative of other states but may help illustrate some of the difficulties
involved.
In 2000 the University of California began to explore possible
alternative assessments to the SAT and ACT that were more
closely aligned with Californias K12 curriculum yet suitable for
use in UC admissions. Some UC faculty were skeptical of this
effort in view of the volatile political environment surrounding
the states K12 assessment system, where new testing regimes
came and went with alarming frequency. In 1997, however, the
State Board of Education launched a major effort to articulate
clear curriculum standards for the schools and to align all state
tests with those standards, which seemed to promise greater stability and continuity going forward.
Downloaded from http://er.aera.net at UNIV ARIZONA LIBRARY on August 18, 2013
december 2009 671
It soon became evident, however, that most statewide tests
were inadequate for use in UC admissions. Designed to measure
achievement across the entire range of the K12 student population, the California Standards Test lacked sufficient differentiation and reliability at the high end of the achievement distribution,
from which the University of California draws its students. A
similar problem existed with the California High School Exit
Exam, then in its planning stages: An exam designed to determine whether students meet the minimum standards required for
high school graduation is unlikely to be useful in a highly selective admissions environment.
But one test did hold promise: the Golden State Examinations
(GSEs), which had been established in 1983 to assess achievement in specific academic subjects. The California Department
of Education, the states K12 administrative arm, had long
championed the GSEs as part of a broader program to improve
student achievement, similar to the national AP program. The
exams were voluntary and geared as honors-level assessments.
Matching the states test records to our own student database, we
found that GSE scores predicted first-year performance at the
University of California almost as well as the SAT I reasoning
test, although not nearly as well as the SAT II subject tests.
Although the GSEs lacked some of the technical sophistication
of the national tests, we were hopeful that those issues could be
resolved; the state had contracted with ACT, Inc., to help improve
the tests psychometric quality.16
Those hopes were dashed when funding for the GSE program
was eliminated from the states 2003 budget. The test had fallen
victim to political infighting between the California Department
of Education, which was promoting the test, and the State Board
of Education, which viewed the GSEs as a departure from its new
curriculum standards. Some state education officials also viewed
the University of Californias efforts to adapt the GSEs for use in
admissions as an incursion on the Board of Educations authority
over K12 curriculum standards.
Californias experience illustrates a more general problem
likely to confront efforts to develop standards-based assessments
that bridge the institutional divide between state university and
K12 school systems: Standards for what is expected of entering
freshmen at selective colleges and universities are different and
usually much more rigorous than K12 curriculum standards.
They overlap, to be sure, but they are not the same, and institutional conflicts over standards and testing are probably inevitable
for this reason. College and university faculty are right to be skeptical about using K12 tests in admissions if it means relinquishing control over entrance standards. And it is understandable that
secondary school educators are concerned that, in seeking to
adapt and modify K12 tests for use in admissions, colleges and
universities may exert undue influence over curriculum standards
for the schools.
A first step toward getting past this problem is for colleges and
universities to band together in articulating their own standards
for what is expected of entering freshmen, as distinct from high
school graduates. This has occurred in California. The academic
senates of the three main segments of the states higher education
systemthe University of California, the California State
University, and the California Community Collegeshave collaborated on a joint statement of specific competencies in both
English and mathematics expected of all students entering
California higher education (Intersegmental Committee of the
Academic Senates, 1997, 1998). The statements are intended to
inform students about the preparation they will need for college
beyond the minimum requirements for high school graduation,
so that students do not graduate only to find themselves unready
for college-level work. Although it is a useful first step, the standards have yet to result in any changes in admissions tests.
Nationally, the most ambitious effort to develop standards of
college readiness is Standards for Success, a project sponsored by
the American Association of Universities (AAU) and the Pew
Charitable Trusts. Led by David Conley at the Center for
Education Policy Research at the University of Oregon, the project convened representatives from AAU institutions to identify
content standards for what students need to know to succeed in
entry-level courses at those institutions. The standards covered
English, mathematics, natural sciences, social sciences, second
languages, and the arts. Then, in the most interesting phase of the
project, researchers used the standards as a reference point to
evaluate alignment of K12 standards-based tests. The project
evaluated 66 exams from 20 states, finding that although a few
were closely aligned with the standards, most bore only an inconsistent relationship to the knowledge and skills needed for college
(Brown & Conley, 2007).
Whether K12 standards-based assessments can be successfully adapted for use in college admissions may depend in part on
the response of the testing agencies. The Standards for Success
project ended in 2003, and the standards were subsequently
licensed to the College Board. The College Board has announced
that the standards are now being used in reviewing test specifications for the SAT, the Preliminary SAT/National Merit
Scholarship Qualifying Test, and AP exams. Like ACT, the
College Board has sought to have its tests adopted by the states
for assessing K12 student achievement (Hupp & Morgan,
2008), but there is as yet no indication that the standards will be
used to adapt state-level exams for admissions purposes (College
Board, 2009a).
In its call for American colleges and universities to take back
the conversation on standardized admissions testing, NACACs
(2008) blue-ribbon commission on testing had this to say about
the role of the testing agencies:
Institutions must exercise independence in evaluating and articulating their use of standardized test scores. There is also a need for
an independent forum for inter-institutional evaluation and discussion of standardized test use in admission that can provide
support for colleges with limited resources to devote to institutional research and evaluation.
While support for validity research is available from the testing agencies, the Commission does not believe that colleges and
universities should rely solely on the testing agencies for it. . . .
Rather, this Commission suggests that colleges and universities
create a new forum for validity research under the auspices of
NACAC. Such an independent discussion might begin to address
questions the Commission and other stakeholders have posed
about the tests. (pp. 21, 23)
NACACs call for independent research on admissions tests is a
useful reminder that until now most research on the SAT and
ACT has been conducted by the testing agencies themselves.
Downloaded from http://er.aera.net at UNIV ARIZONA LIBRARY on August 18, 2013
672 educational Researcher
Much of this work is published outside the academic journals,
without benefit of normal peer review, and the findings are
invariably supportive of the agencies test products. Whether or
not there is an actual conflict of interest, the appearance of a
conflict is inevitable, and the parallel with some recent issues in
medical research is troubling.
These considerations underscore the need for colleges and
universities collectively to reclaim their authority over admissions
testingand, most vitally, over the standards on which admissions tests are built. Only college and university faculty are in a
position to set academic standards for what is expected of matriculants, and this critical task can be neither delegated to the
schools nor outsourced to the testing agencies.
Shifting the Paradigm:
From Prediction to Achievement
Looking back at the arc of admissions testing over the 20th century, the signs of a paradigm shift are increasingly apparent. Ever
since the 1930s, when Henry Chauncey suggested that Carl
Brighams new Scholastic Aptitude Test could predict student
success at Harvard, the idea of prediction has captivated American
college admissions. The preoccupation continues to this day and
still drives much research on admissions testing. Yet the preoccupation with prediction has gradually given way to another idea.
Lindquists philosophical opposition to the SAT and his introduction of the ACT, the renewed interest in subject tests at some
colleges and universities, the explosion of standards-based tests in
K12 schools, and the as-yet unsuccessful efforts to adapt them
for use in college admissionsall point the way to assessment of
achievement and curriculum mastery as an alternative paradigm
for admissions testing.
Our ability to predict student performance in college on the
basis of factors known at point of admission remains relatively
limited. After decades of predictive-validity studies, our best
prediction models (using not only test scores but high school
grades and other academic and socioeconomic factors) still
account for only about 25% to 30% of the variance in outcome measures such as college GPA. This means that some
70% to 75% of the variance is unexplained. That should not
be surprising in view of the many other factors that affect student performance after admission, such as social support,
financial aid, and academic engagement in college. But it also
means that the error bands around our predictions are quite
broad. Using test scores as a tiebreaker to choose between
applicants who are otherwise equally qualified, as is sometimes
done, is not necessarily a reliable guide, especially where score
differences are small.
Moreover, there is little difference among the major national
tests in their ability to predict student performance in college.
Although the New SAT, ACT, SAT Subject Tests, and AP exams
differ in design, content, and other respects, they tend to be
highly correlated and thus largely interchangeable with respect to
prediction. It is true that subject-specific tests (in particular the
AP exams) do have a statistically significant predictive advantage
(Bowen et al., 2009; Geiser & Santelices, 2006), but the statistical difference by itself is too small to be of practical significance
or to dictate adoption of one test over another. The argument for
achievement tests is not so much that they are better predictors
than other kinds of tests but that they are no worse: The benefits
of achievement tests for college admissionsgreater clarity in
admissions standards, closer linkage to the high-school curriculumcan be realized without any sacrifice in the capacity to predict success in college (Geiser, 2002, p. 25).
For these reasons, we believe that prediction will recede in
importance, and other test characteristics will become more critical in designing standardized admissions tests in the future. We
will still need to validate our tests by demonstrating that they
are reasonably correlated with student performance in college;
validation remains especially important where tests have adverse
impacts on low-income and minority applicants. But beyond
some acceptable threshold of predictive validity, decisions about
what kinds of assessments to use in college admissions will be
driven less by small statistical differences and more by educational policy considerations.
In contrast to prediction, the idea of achievement offers a richer
paradigm for admissions testing and calls attention to a broader
array of characteristics that we should demand of our tests:
1. Admissions tests should be criterion referenced rather than
norm referenced: Our primary consideration should not
be how an applicant compares with others but whether he
or she demonstrates sufficient mastery of college preparatory subjects to benefit from and succeed in college.
2. Admissions tests should have diagnostic utility: Rather than
a number or a percentile rank, tests should provide students with curriculum-related information about areas of
strength and areas where they need to devote more study.
3. Admissions tests should exhibit not only predictive validity
but face validity: The relationship between the knowledge
and skills being tested and those needed for college should
be transparent.
4. Admissions tests should be aligned with college preparatory
coursework: Assessments should be linked as closely as possible to materials that students encounter in the classroom
and should reinforce teaching and learning of a rigorous
academic curriculum in our high schools.
5. Admissions tests should minimize the need for test preparation: Although test prep services will probably never disappear entirely, admissions tests should be designed to reward
mastery of curriculum content over test-taking skills so
that the best test prep is regular classroom instruction.
6. Finally, admissions tests should send a signal to students:
Our tests should send the message that working hard and
mastering academic subjects in high school is the most
direct route to college.
The core feature of achievement testing is criterion-referenced or
standards-based assessment. This approach to assessment is now
widely established in the nations K12 schools but has yet to take
hold in college admissions, where norm-referenced assessments
still prevail. Norm-referenced tests like the SAT or ACT are often
justified as necessary to help admissions officers sort large numbers of applicants and evaluate their relative potential for success
in college.
Once started, however, norm-referenced assessment knows no
stopping point. The competition for scarce places at top institutions drives test scores ever higher, and average scores for this
Downloaded from http://er.aera.net at UNIV ARIZONA LIBRARY on August 18, 2013
december 2009 673
years entering class are almost always higher than last years. Tests
are used to make increasingly fine distinctions within applicant
pools where almost all students have relatively high scores. Small
differences in test scores often tip the scales against admission of
lower scoring applicants, when in fact such differences have marginal validity in predicting college performance. The ever-upward
spiral of test scores is especially harmful to low-income and
minority applicants. Even where these students achieve real gains
in academic preparation, as measured on criterion-referenced
assessments, they lag further behind other applicants on normreferenced tests.17 The emphasis on picking winners makes it
difficult for colleges and universities to extend opportunities to
those who would benefit most from higher education. And the
preoccupation with test scores at elite institutions spreads outward, sending mixed messages to other colleges and universities
and to the schools.
Criterion-referenced tests, on the other hand, presuppose a
very different philosophy and approach to college admissions.
Their purpose is to certify students knowledge of college preparatory subjects, and they help to establish a baseline or floor for
judging applicants readiness for college. Along with high school
grades, achievement test scores tell us whether applicants have
mastered the foundational knowledge and skills required for
college-level work.
When we judge students against this standard, two truths
become evident. First is that the pool of qualified candidates who
could benefit from and succeed in college is larger than can be
accommodated at selective institutions. Second is that admissions criteria other than test scoresspecial talents and skills,
leadership and community service, opportunity to learn, and
social and cultural diversityare more important in selecting
whom to admit from among this larger pool. Admissions officers
often describe their work as crafting a class, a phrase that nicely
captures this meaning.
Achievement testing reflects a philosophy of admissions that
is at once more modest and more expansive than predicting success in college. It is more modest in that it asks less of admissions
tests and is more realistic about what they can do: Our ability to
predict success in college is relatively limited, and the most we
should ask of admissions tests is to certify students mastery of
foundational knowledge and skills. It is more expansive in holding that beyond some reasonable standard of college readiness,
other admissions criteria must take precedence over test scores if
we are to craft an entering class that reflects our broader institutional values. And beyond the relatively narrow world of selective
college admissions, testing for achievement and curriculum mastery can have a broader and more beneficial signaling effect
throughout all of education.
It is not our intention to try to anticipate the specific forms or
directions that admissions testing may take in the 21st century.
Yet we believe that the general principles just outlinedand the
paradigmatic idea of achievement testing that unites themwill
be useful and relevant as a guide for evaluating new kinds of
assessments that may emerge in the future. For example, these
principles lead us to be initially skeptical about efforts to develop
noncognitive assessments for use in college admissions insofar
as those efforts sometimes blur the crucial distinction between
achievement and personality traits over which the student has
little control. On the other hand, notwithstanding the many difficulties involved in adapting K12 standards-based tests for use
in admissions, we conclude that this is unquestionably a worthwhile goal if it can be realized.
It should be evident that no existing admissions tests satisfy all
of the principles we have outlined. Our purpose is not to endorse
any particular test or set of tests but to contribute to the national
dialogue about admissions testing and what we expect it to
accomplish. Two decades ago in their classic brief The Case
Against the SAT, James Crouse and Dale Trusheim (1988) argued
persuasively for a new generation of achievement tests that would
certify students mastery of college preparatory subjects, provide
incentives for educational improvement, and encourage greater
diversity in admissions tests. What is new is that today, more than
at any time in recent memory, American colleges and universities
seem open to the possibility of a fresh start in standardized admissions testing.
Notes
1The superiority of high school grade point average (GPA) over standardized test scores in predicting college outcomes is sometimes obscured
in descriptions of validity studies. For example, in a recent survey of
predictive-validity studies conducted over the past several decades,
College Board researchers described their findings this way:
The SAT has proven to be an important predictor of success
in college. Its validity as a predictor of success has been demonstrated through hundreds of validity studies. These validity
studies consistently find that high school grades and SAT scores
together are substantial and significant predictors of achievement
in college. In these studies, although high school grades typically
are slightly better predictors of achievement [italics added], SAT
scores add significantly to the prediction. (Camara & Echternacht, 2000)
2In a recent study sponsored by the College Board, Paul Sackett and
his colleagues defend the SAT, asserting that its predictive power is not
substantially diminished when controls for socioeconomic status (SES)
are introduced (Sackett, Kuncel, Arneson, Cooper, & Waters, 2009).
Sacketts study, however, examined the extent to which SES affected the
overall, bivariate correlation between SAT scores and college outcomes
(first-year college grades) but failed to consider the independent contribution of high school grades (HSGPA) and other indicators in predicting college outcomes. In real-world admissions, the key question is what
SAT scores uniquely add to the prediction of college outcomes, beyond
what is already provided by a students HSGPA and other indicators.
Looking at the unique portion of the variance in SAT scoresthe portion not shared with HSGPA or other indicatorsstudies using more
fully specified regression models have found that the predictive power of
the SAT is significantly reduced when controls for SES are introduced
(Geiser, 2002; Rothstein, 2004). Thus there is no actual conflict between
Sacketts study and others that show that the value added by the SAT is
heavily conditioned by SES, as Sackett acknowledges (personal communication, January 14, 2009). 3An example of how simple correlations can be misleading is a study
cited on the College Boards website in introducing the New SAT: In
the California study, SAT scores were slightly more predictive than high
school grade point average (HSGPA) (College Board, 2009c). The
study referred to was conducted at the University of California (UC).
The claim that the New SAT is more predictive than HSGPA was based
on the UC studys initial finding that the univariate correlation between
New SAT scores and first-year college GPA (FYGPA) was slightly greater
Downloaded from http://er.aera.net at UNIV ARIZONA LIBRARY on August 18, 2013
674 educational Researcher
than that between HSGPA and FYGPA (Agronow & Studley, 2007,
Figure 1, Models 1 and 4). The same study, however, also presented
more fully specified, multivariate regression models that allowed direct
comparison of the predictive weights of HSGPA and SAT scores when
both were included side-by-side in the same model along with other
academic and socioeconomic factors. In the more fully specified models,
HSGPA had by far the greatest predictive weight (Agronow & Studley,
2007, Table 1, Model 22). 4
It is important to be clear about what is meant by the term adverse
impact. Both the College Board and ACT go to great lengths to eliminate test bias, and we do not question those efforts. Notwithstanding
those efforts, however, it remains the case that, compared with other
admissions indicators such as high school grades and the SAT II subject
tests, SAT scores are more closely correlated with measures of socioeconomic status such as family income and parental education. As a result,
the latter test has a greater adverse statistical impact on underrepresented
minority applicants, who come disproportionately from lower socioeconomic backgrounds. 5Given the highly selective nature of UC admissions, some have
questioned whether range restriction might account for the diminished
predictive value of the SAT I as compared with high school GPA and
SAT II subject tests in the UC sample. The UC data were examined
carefully for range restriction effects, however, and there was no evidence
that this was the case. Comparing the variances in HSGPA, SAT I, and
SAT II scores in the UC applicant pool versus the pool of admitted
students, we found that HSGPAthe primary selection criterion used
in UC admissionswas the most range restricted of all admissions criteria even though it retained the greatest predictive weight. Restriction
on both SAT I and SAT II scores was less pronounced and quite similar.
Range restriction, in short, does not appear to account for the relative
predictive weights of HSGPA, SAT I, and SAT II scores found in the UC
sample (Geiser, 2002, note 4; Geiser & Santelices, 2007, note xix). 6In an independent reanalysis of the UC data, Zwick and her colleagues found the same small but consistent predictive advantage for the
SAT II subject tests (Zwick, Brown, & Sklar, 2004). The same finding
was also confirmed in a 2001 College Board study of a larger sample of
institutions that required both the SAT I and SAT II, including Barnard,
Bowdoin, Colby, Harvard, Northwestern, and Vanderbilt, as well as four
UC campuses (Bridgeman, Burton, & Cline, 2001). 7These and other conclusions about the problematic effects of the
SAT for Californias K12 schools were summarized in a policy paper,
The Use of Admissions Tests by the University of California, adopted
by the UC faculty in 2001 after intensive debate and study. The paper
was one of the first comprehensive policy statements on standardized
admissions tests to be adopted by a major U.S. university and strongly
endorsed curriculum-based achievement tests over aptitude-type
tests (University of California, 2002). 8For an account of events immediately leading up to and following
Atkinsons 2001 address to the American Council on Education, proposing elimination of the SAT at the University of California, see College
Admissions and the SAT: A Personal Perspective (Atkinson, 2004). 9College Board researchers had expected inclusion of the writing
exam in the New SAT to add modestly to the prediction of college
performance when critical reading and mathematics scores are considered (Kobrin & Kimmel, 2006, p. 7). 10In a recent article reviewing the New SAT, the authors suggested
significantly reducing or even eliminating the critical-reading section,
which would not only shorten the test but also possibly improve its
predictive validity. Along with this shortened SAT, students might be
required to take two subject tests in areas of their choosing (Atkinson &
Geiser, 2008). 11About 70% of all U.S. high schools now award bonus points for
Advanced Placement (AP) classes, according to a survey by the National
Association for College Admissions Counseling (2004). This boosts students GPAs and improves admissions profiles, and a growing number
of students now enroll in AP for this reason. 12The University of California currently requires two SAT
Subject Tests, both of which are now elective: These must be in two
different areas, chosen from the following: English, history and social
studies, mathematics (Level 2 only), science, or language other than
English. 13The UC regents have recently approved a policy change that would
appear to reverse that institutions long-standing reliance on achievement tests in admissions. As part of a broader set of changes in UC
admissions policies, in February 2009 the regents approved a proposal
to eliminate the SAT Subject Tests and require only the New SAT (or
ACT with writing) for admission to the UC system beginning in 2012.
Understandably, some have viewed the regents action as an endorsement of the New SAT and a rejection of previous UC policy favoring
achievement tests. But according to UC President Mark Yudof, this is
not the case:
It is important to note that although the subject examinations
will no longer be required, students for whom these tests represent an opportunity to demonstrate achievement in a particular
area are still encouraged to take the tests. . . . Eliminating the subject exam requirement in no way validates or confirms the use of
other tests like the SAT reasoning exam. (Letter to Asian Pacific
Islander Legislative Caucus, February 24, 2009)
14Regarding our contention that, compared with the SAT I, curriculum-based achievement measures such as the SAT II subject tests are less
affected by students socioeconomic status (SES), one reviewer of this
article objected that achievement tests are also correlated with SES. Our
point, however, is not that achievement test scores are unrelated to
SESvirtually all academic indicators are correlated with SES to one
degree or anotherbut that achievement indicators are less correlated
with SES compared with the SAT. The UC studies showed that high
school GPA had by far the lowest correlation with measures of SES such
as family income, parental education, and high school quality; the SAT
I had the strongest correlation; and the SAT II subject tests fell generally
in between (Geiser, 2002; Geiser & Santelices, 2007). College Board
researchers have also noted the stronger association between SAT I scores
and SES than between SAT II scores and SES (see Kobrin, Camara, &
Milewski, 2002, Figure 1A). 15There are substantial differences among the states in the quality of
their assessments and the extent to which their curriculum standards are
integrated with comprehensive school reform efforts. As Linda DarlingHammond (2003) has noted,
In a number of states, the notions of standards and accountability have become synonymous with mandates for student testing
that are detached from policies that might address the quality of
teaching, the allocation of resources, or the nature of schooling.
. . . States and districts that have relied primarily on test-based
accountability emphasizing sanctions for students and teachers
have often produced greater failure, rather than greater success,
for their most educationally vulnerable students. More successful
reforms have emphasized the use of standards for teaching and
learning to guide investments in better prepared teachers, higher
quality teaching, more performance-oriented curriculum and assessment, better designed schools, more equitable and effective
resource allocations, and more diagnostic supports for student
learning. (para. 3, 6)
16For an overview of the assessments used in California secondary
and postsecondary education, and the alignment (or lack thereof)
between them, see Venezia (2000).
Downloaded from http://er.aera.net at UNIV ARIZONA LIBRARY on August 18, 2013
december 2009 675
17As Darling-Hammond (2003) notes,
Use of norm-referenced tests . . . makes it impossible to gauge
progress accurately, as items are removed from the test as greater
numbers of students can answer them, thus guaranteeing continuing high rates of failure, especially for certain subpopulations
of students. (para. 9)
One of the main problems with No Child Left Behind, she argues, is
that its testing requirements push states back to the lowest common
denominator, undoing progress that has been made to improve the quality of assessments and delaying the move from antiquated norm-referenced tests to criterion-referenced systems (para. 11)
References
ACT. (2009a). College readiness standards for the ACT. Iowa City, IA:
Author. Retrieved July 26, 2009, from http://www.act.org/standard/
guides/act/index.html
ACT. (2009b). Educational planning and assessment. Iowa City, IA:
Author. Retrieved July 26, 2009, from http://www.act.org/epas/
index.html
Agronow, S., & Studley, R. (2007, November). Prediction of college GPA
from New SAT test scoresA first look. Paper presented at annual meeting of the California Association for Institutional Research, Monterey,
CA.
American Educational Research Association, American Psychological
Association, and National Council on Measurement in Education.
(1999). Standards for educational and psychological testing. Washington,
DC: American Educational Research Association.
Atkinson, R. (2001). Standardized tests and access to American universities. The 2001 Robert H. Atwell Distinguished Lecture, American
Council on Education, Washington, DC. Retrieved November 18,
2009, from http://www.rca.ucsd.edu/comments/satspch.html
Atkinson, R. (2004, April). College admissions and the SAT: A personal perspective. Invited address at the annual meeting of the American Educational
Research Association, San Diego, CA. (Republished in Journal of the
Association for Psychological Science, Observer, 18, 1522, 2005).
Atkinson, R., & Geiser, S. (2008). The new SAT: A work in progress.
Observer: A Journal of the Association for Psychological Science. 21(10),
2324.
Bowen, W., Chingos, M., & McPherson, M. (2009). Crossing the finish
line: Completing college at Americas public universities. Princeton, NJ:
Princeton University Press.
Bridgeman, B., Burton, N., & Cline, F. (2001). Substituting SAT II:
Subject tests for SAT I: reasoning tests: Impact on admitted class composition and quality (College Board Research Rep. No. 20013). New
York: College Board.
Brown, R., & Conley, D. (2007). Comparing state high school assessments and standards for success in entry-level university courses.
Educational Assessment, 12, 137160.
Burton, N., & Ramist, L. (2001). Predicting success in college: SAT studies
of classes graduating since 1980 (College Board Research Rep. No.
20012). New York: College Board.
Camara, W., & Echternacht, G. (2000). The SAT I and high school
grades: Utility in predicting success in college (College Board Rep. No.
RN-10). New York: College Board.
Camara, W., & Schmidt, A. (2006). The New SAT facts [PowerPoint
presentation]. New York: College Board. Retrieved March 7, 2009,
from http://www.collegeboard.com/prod_downloads/forum/forum06/
the-new-sat_a-comprehensive-report-on-the-first-scores.PPT
College Board. (2009a). College Board standards for college success. New
York: College Board. Retrieved March 10, 2009, from http://professionals.collegeboard.com/k-12/standards
College Board. (2009b). Frequently asked questions about SAT Subject
Tests. New York: College Board. Retrieved March 6, 2009, from
http://www.compassprep.com/subject_faq.shtml#faq2
College Board. (2009c). SAT validity studies. New York: College Board.
Retrieved July 22, 2009, from http://professionals.collegeboard.com/
data-reports-research/sat/validity-studies
Cornwell, C., Mustard, D., & Van Parys, J. (2008). How does the New
SAT predict academic performance in college? (Working paper). Athens:
University of Georgia. Retrieved November 18, 2009, from http://
www.terry.uga.edu/~mustard/New%20SAT.pdf
Crouse, J., & Trusheim, D. (1988). The case against the SAT. Chicago:
University of Chicago Press.
Darling-Hammond, L. (2003, February 16). Standards and assessments:
Where we are now and what we need. Teachers College Record.
Retrieved November 18, 2009, from http://www.tcrecord.org (ID
No. 11109)
Geiser, S. (with Studley, R.). (2002). UC and the SAT: Predictive validity and differential impact of the SAT I and SAT II at the University
of California. Educational Assessment, 8, 126.
Geiser, S. (2009). Back to the basics: In defense of achievement (and
achievement tests) in college admissions. Change, 41(1), 1623.
Geiser, S., & Santelices, M. V. (2006). The role of Advanced Placement
and honors courses in college admissions. In P. Gandara, G. Orfield, &
C. Horn (Eds.), Expanding opportunity in higher education: Leveraging
promise (pp. 75114). Albany: State University of New York Press.
Geiser, S., & Santelices, M.V. (2007). Validity of high-school grades in
predicting student success beyond the freshman year: High-school record
vs. standardized tests as indicators of four-year college outcomes. Berkeley:
Center for Studies in Higher Education, University of California,
Berkeley. Retrieved November 18, 2009, http://cshe.berkeley.edu/
publications/publications.php?id=265.
Hammond, G. (2008). Advancing beyond AP courses. Chronicle of
Higher Education, 54(34), B17.
Hupp, D., & Morgan, D. (2008). The SAT as a states NCLB assessment:
Rationale and issues confronted. Paper presented at the National
Conference on Student Assessment, Orlando, FL. Retrieved July 29,
2009, from the College Board website: http://professionals.collegeboard.com/data-reports-research/cb/other-conf/nclb-state-assmt
Intersegmental Committee of the Academic Senates. (1997). Statement
of competencies in mathematics expected of entering college students.
Sacramento: California Education Round Table. Available at http://
www.certicc.org
Intersegmental Committee of the Academic Senates. (1998). Statement
of competencies in English expected of entering college students.
Sacramento: California Education Round Table. Available at http://
www.certicc.org
Kirst, M., & Venezia, A. (Eds.). (2004). From high school to college:
Improving opportunities for success in postsecondary education. San
Francisco: Jossey-Bass.
Kobrin, J., Camara, W., & Milewski, G. (2002). The utility of the SAT I
and SAT II for admissions decisions in California and the nation (College
Board Research Rep. No. 20026). New York: College Board.
Kobrin, J., & Kimmel, E. (2006). Test development and technical information on the writing section of the SAT reasoning test (College Board
Research Rep. No. RN-25). New York: College Board.
Kobrin, J., Patterson, B., Shaw, E., Mattern, K., & Barbuti, S. (2008).
Validity of the SAT for predicting first-year college gradepoint average
(College Board Research Rep. No. 20085). New York: College Board.
Lawrence, I., Rigol, G., Van Essen, T., & Jackson, C. (2003). A historical
perspective on the content of the SAT (College Board Research Rep. No.
200303). New York: College Board.
Lemann, N. (1999). The big test: The secret history of the American meritocracy. New York: Farrar, Straus and Giroux.
Downloaded from http://er.aera.net at UNIV ARIZONA LIBRARY on August 18, 2013
676 educational Researcher
Lindquist, E. F. (1958, November 1). The nature of the problem of
improving scholarship and college entrance examinations (Paper presented at Educational Testing Service invitational conference on testing problems). Princeton, NJ: Educational Testing Service.
Morgan, R. (1989). Analysis of the predictive validity of the SAT and high
school grades from 1976 to 1983 (College Board Rep. No. 897). New
York: College Board.
National Association for College Admissions Counseling. (2004).
National school counselor survey. Alexandria, VA: Author.
National Association for College Admissions Counseling. (2008). Report
of the Commission on the Use of Standardized Tests in Undergraduate
Admissions. Arlington, VA: Author.
Noeth, J., & Kobrin, J. (2007). Writing changes in the nations K12
school system (College Board Research Rep. No. RN-34). New York:
College Board.
Princeton Review. (2009). Prep for SAT Subject Tests. Framingham, MA:
Author. Retrieved July 26, 2009, from http://www.princetonreview.
.aspx
Rothstein, J. (2004). College performance predictions and the SAT.
Journal of Econometrics, 121, 297317.
Sackett, P., Kuncel, N., Arneson, J., Cooper, S., & Waters, S. (2009).
Does socioeconomic status explain the relationship between admissions tests and post-secondary academic performance? Psychological
Bulletin, 135, 122.
University of California. (2002). The use of admissions tests by the University
of California. Oakland, CA: UC Board of Admissions and Relations
With Schools. Retrieved November 18, 2009, from http://www
.universityofcalifornia.edu/senate/committees/boars/admissionstests.pdf
Venezia, A. (2000). Connecting Californias K12 and higher education
systems: Challenges and opportunities. In E. Burr, G. C. Hayward, B.
Fuller, & M. Kirst (Eds.), Crucial issues in California education 2000:
Are the pieces fitting together? (pp. 153176). Berkeley: Policy Analysis
for California Education.
Zwick, R., Brown, T., & Sklar, J. (2004). California and the SAT:
A reanalysis of University of California admissions data. Berkeley:
Center for Studies in Higher Education, University of California,
Berkeley. Retrieved November 18, 2009, from http://cshe.berkeley
.edu/publications/publications.php?id=68
AUTHORS
RICHARD C. ATKINSON is president emeritus of the University of
California and professor emeritus of cognitive science and psychology at
the University of California, San Diego, 5320 Atkinson Hall, 9500
Gilman Drive, La Jolla, CA 92093-0436; [email protected]. His research
is on memory, perception, and cognition.
SAUL GEISER is a research associate at the Center for Studies in Higher
Education at the University of California, Berkeley, 771 Evans Hall, No.
4650, Berkeley, CA 94720-4650; [email protected]. He is a former
director of research for admissions and outreach for the University of
California system.
Manuscript received May 26, 2009
Revision received August 4, 2009
Accepted August 10, 2009
Downloaded from http://er.aera.net at UNIV ARIZONA LIBRARY on August 18, 2013

Get Professional Assignment Help Cheaply

Are you busy and do not have time to handle your assignment? Are you scared that your paper will not make the grade? Do you have responsibilities that may hinder you from turning in your assignment on time? Are you tired and can barely handle your assignment? Are your grades inconsistent?

Whichever your reason is, it is valid! You can get professional academic help from our service at affordable rates. We have a team of professional academic writers who can handle all your assignments.

Why Choose Our Academic Writing Service?

• Plagiarism free papers
• Timely delivery
• Any deadline
• Skilled, Experienced Native English Writers
• Subject-relevant academic writer
• Adherence to paper instructions
• Ability to tackle bulk assignments
• Reasonable prices
• 24/7 Customer Support
• Get superb grades consistently

Online Academic Help With Different Subjects

Literature

Students barely have time to read. We got you! Have your literature essay or book review written without having the hassle of reading the book. You can get your literature paper custom-written for you by our literature specialists.

Finance

Do you struggle with finance? No need to torture yourself if finance is not your cup of tea. You can order your finance paper from our academic writing service and get 100% original work from competent finance experts.

Computer science

Computer science is a tough subject. Fortunately, our computer science experts are up to the match. No need to stress and have sleepless nights. Our academic writers will tackle all your computer science assignments and deliver them on time. Let us handle all your python, java, ruby, JavaScript, php , C+ assignments!

Psychology

While psychology may be an interesting subject, you may lack sufficient time to handle your assignments. Don’t despair; by using our academic writing service, you can be assured of perfect grades. Moreover, your grades will be consistent.

Engineering

Engineering is quite a demanding subject. Students face a lot of pressure and barely have enough time to do what they love to do. Our academic writing service got you covered! Our engineering specialists follow the paper instructions and ensure timely delivery of the paper.

Nursing

In the nursing course, you may have difficulties with literature reviews, annotated bibliographies, critical essays, and other assignments. Our nursing assignment writers will offer you professional nursing paper help at low prices.

Sociology

Truth be told, sociology papers can be quite exhausting. Our academic writing service relieves you of fatigue, pressure, and stress. You can relax and have peace of mind as our academic writers handle your sociology assignment.

Business

We take pride in having some of the best business writers in the industry. Our business writers have a lot of experience in the field. They are reliable, and you can be assured of a high-grade paper. They are able to handle business papers of any subject, length, deadline, and difficulty!

Statistics

We boast of having some of the most experienced statistics experts in the industry. Our statistics experts have diverse skills, expertise, and knowledge to handle any kind of assignment. They have access to all kinds of software to get your assignment done.

Law

Writing a law essay may prove to be an insurmountable obstacle, especially when you need to know the peculiarities of the legislative framework. Take advantage of our top-notch law specialists and get superb grades and 100% satisfaction.

What discipline/subjects do you deal in?

We have highlighted some of the most popular subjects we handle above. Those are just a tip of the iceberg. We deal in all academic disciplines since our writers are as diverse. They have been drawn from across all disciplines, and orders are assigned to those writers believed to be the best in the field. In a nutshell, there is no task we cannot handle; all you need to do is place your order with us. As long as your instructions are clear, just trust we shall deliver irrespective of the discipline.

Are your writers competent enough to handle my paper?

Our essay writers are graduates with bachelor's, masters, Ph.D., and doctorate degrees in various subjects. The minimum requirement to be an essay writer with our essay writing service is to have a college degree. All our academic writers have a minimum of two years of academic writing. We have a stringent recruitment process to ensure that we get only the most competent essay writers in the industry. We also ensure that the writers are handsomely compensated for their value. The majority of our writers are native English speakers. As such, the fluency of language and grammar is impeccable.

What if I don’t like the paper?

There is a very low likelihood that you won’t like the paper.

Reasons being:

• When assigning your order, we match the paper’s discipline with the writer’s field/specialization. Since all our writers are graduates, we match the paper’s subject with the field the writer studied. For instance, if it’s a nursing paper, only a nursing graduate and writer will handle it. Furthermore, all our writers have academic writing experience and top-notch research skills.
• We have a quality assurance that reviews the paper before it gets to you. As such, we ensure that you get a paper that meets the required standard and will most definitely make the grade.

In the event that you don’t like your paper:

• The writer will revise the paper up to your pleasing. You have unlimited revisions. You simply need to highlight what specifically you don’t like about the paper, and the writer will make the amendments. The paper will be revised until you are satisfied. Revisions are free of charge
• We will have a different writer write the paper from scratch.
• Last resort, if the above does not work, we will refund your money.

Will the professor find out I didn’t write the paper myself?

Not at all. All papers are written from scratch. There is no way your tutor or instructor will realize that you did not write the paper yourself. In fact, we recommend using our assignment help services for consistent results.

What if the paper is plagiarized?

We check all papers for plagiarism before we submit them. We use powerful plagiarism checking software such as SafeAssign, LopesWrite, and Turnitin. We also upload the plagiarism report so that you can review it. We understand that plagiarism is academic suicide. We would not take the risk of submitting plagiarized work and jeopardize your academic journey. Furthermore, we do not sell or use prewritten papers, and each paper is written from scratch.

When will I get my paper?

You determine when you get the paper by setting the deadline when placing the order. All papers are delivered within the deadline. We are well aware that we operate in a time-sensitive industry. As such, we have laid out strategies to ensure that the client receives the paper on time and they never miss the deadline. We understand that papers that are submitted late have some points deducted. We do not want you to miss any points due to late submission. We work on beating deadlines by huge margins in order to ensure that you have ample time to review the paper before you submit it.

Will anyone find out that I used your services?

We have a privacy and confidentiality policy that guides our work. We NEVER share any customer information with third parties. Noone will ever know that you used our assignment help services. It’s only between you and us. We are bound by our policies to protect the customer’s identity and information. All your information, such as your names, phone number, email, order information, and so on, are protected. We have robust security systems that ensure that your data is protected. Hacking our systems is close to impossible, and it has never happened.

How our Assignment Help Service Works

1. Place an order

You fill all the paper instructions in the order form. Make sure you include all the helpful materials so that our academic writers can deliver the perfect paper. It will also help to eliminate unnecessary revisions.

2. Pay for the order

Proceed to pay for the paper so that it can be assigned to one of our expert academic writers. The paper subject is matched with the writer’s area of specialization.

3. Track the progress

You communicate with the writer and know about the progress of the paper. The client can ask the writer for drafts of the paper. The client can upload extra material and include additional instructions from the lecturer. Receive a paper.

4. Download the paper

The paper is sent to your email and uploaded to your personal account. You also get a plagiarism report attached to your paper.

GET A PERFECT SCORE!!!