The future of standardized testing: How well can it function
as college admissions criteria and academic performance predictor for students
in the United States
By: Lit Jhun Yeang,Benjamin
Introduction
Americans place much importance on the idea of providing
every citizen with an equal opportunity to vie for limited places in the
colleges of their choice. The only way to realize this is to have standardized
college entrance examinations which are taken by all applicants in the same
manner. Standardized college admissions testing which consists of mostly
multiple-choice questions was established in the United States in the early 20th
century by the College Entrance Examination Board to make college admissions
procedure a uniform one (Zwick, 2002). The following section shows the
comparison between hybrid tests and pure achievement tests and the predictive
effectiveness of one of the admission tests which is ACT on the academic
success of students of various ethnicities in the U.S.
Achievement
Tests versus Hybrid Tests
American high school students are very familiar with the two
types of standardized tests which are the SAT and ACT respectively. Both of
them are hybrid tests which measure a student’s inherent ability and also
academic accomplishment. They are also widely used in U.S. to predict a
freshman’s success in college. In predicting college grade, Zwick (2002) states
that “the correlation coefficient of hybrid test scores is 0.36, compared 0.39
for high school grades and 0.48 for high school grades and hybrid test scores
combined.” From the statistical measure shown, it can be deduced that college
grades has a weak and positive relationship with hybrid test scores, high
school grades or both. Meanwhile, achievement tests are tests which measure
students’ obtained skills and knowledge. They therefore measure how much
students have learned and how much they knew about a particular subject. They
are being viewed as fairer tests because they serve as a better predictor for a
colleges students’ success. It is proven that “achievement tests have a
correlation coefficient of 0.784 to what a student knows”, although their
relationship’ with students’ first year college grades are not revealed, but
they do have the potential to replace most of the aptitude tests, which only
take students’ innate talent into consideration and not their efforts made and
determination shown to do well. The current achievement tests which are
available for high schools in the U.S. to select from are the SAT IIs, the
International Baccalaureate tests and the Advanced Placement (AP) tests. The
SAT II subject tests which last for an hour, provides candidates a variety of
subjects to choose from. On the other hand, the Advanced Placement (AP) tests
could measure students’ academic performance in college as well as in high
school. The only disadvantage that they have is the long testing duration. If
they were to be used to measure students’ high school performance, a minimum of
5 subjects tests were to be taken on 5 separate days for more than 10 hours.
The next best alternatives of the achievement tests mentioned above are ACT,
PLAN and EXPLORE tests respectively which are all hybrid tests. The ACT tests
last for about 3 hours, whereas PLAN and EXPLORE tests last for about 2 hours
and consist of only multiple choice questions. The 3 tests are offered by the
same organization and hence their contents are taken from the same source. This
means that students who take 3 of them consecutively in certain duration of
time would obtain similar scores for them because the 3 tests have a mutual
relationship. In a nutshell, although achievement tests are more time
consuming, however they are able to help students to improve their academic
performance by displaying them the areas that require more attention. Most
importantly, achievement tests are based on and relevant to the high school
curricular.
The
Predictive Value of ACT and High School Grade Averages
The ACT Assessment which is vital for college admissions and
course selection decisions in the U.S. gauges students’ academic skills and
knowledge which were taught and learnt in high school. Students who received a
well-versed education in high school or did well in college-preparatory courses
will be more likely to obtain higher ACT scores, which will in turn be more
successful in their freshman year. However, it is interesting to note that
African-American and Hispanic students generally have lower scores on
standardized college admissions testing and this situation has caused an
allegation that standardized tests are “biased or unfair” (Marklein, 2000). The
fact is that they usually do not perform very well in most of the academic
achievement measures used to admit students, including high school examinations
(Zwick, 1999). This is proven by (Noble, Crouse and Schulz, 1996) who state that African American students typically
score lower on Mathematics and English in college than their white counterparts
given the same scores in high school Mathematics and English tests. According
to (Young and Kobrin, 2001), “students’ actual freshman grade point averages
(GPAs) are higher or lower than their predicted GPAs.” Prior research also show
that the freshman GPAs of African-American and Hispanic students are lower than
that of white and Asian-American students given the same admissions test scores
and high school averages. Therefore a study is conducted the investigate the
effects of using ACT aggregate scores, high school grade averages or both on
the admission rates and freshman year academic performance of African-American,
Hispanic and white students.
Results
The study shows that if ACT aggregate scores, high school
tests averages or both are used as college admissions criteria, then, without
explicitly considering ethnicity, less African-American and Hispanic students
would be admitted into colleges compared to white students. Interestingly, this
study also reveals that the ACT aggregate scores or high school grade averages
predict the likelihood of attaining academic success in college (achieve a
freshman GPA of 2.5 and above) more accurately for African-American students
than for white students. Furthermore the ACT scores and high school grade averages
serve as a less accurate academic success predictor for Hispanic students than
for white students. Also, it is important to note that the results which are
obtained from this study cannot be applied to all higher education institutions
nationwide (Noble, 2003).
Conclusion
and Implications
Based on the results shown above, it will appear that it
would be hard to maintain the diversity among student populations on college
campuses in the U.S. if standardized test scores and high school grade averages
continue to serve as college admissions criteria in the United States. However,
it is possible for tertiary institutions in the U.S. to admit students who are
less academically well-prepared, provided that they will be given extensive
remedial classes to ensure that they have a wonderful college experience too.
And most importantly, although debates about the usage of standardized college
admissions testing are still very intense, more and more SAT and ACT candidates
can be seen and about 90% of American colleges continue to use the SAT or ACT
examination as a tool to make college admissions decisions.
References
1.
Crouse, J., Noble, J. P. and Schulz,
E.M. (1996). Differential
prediction/impact in course placement for ethnic and gender groups. [pdf]
Available at: http://www.act.org/research/researchers/reports/pdf/ACT_RR96-08.pdf
[Accessed
24 October 2013]
2.
Kobrin, J.L. and Young, J. (2001). Differential validity, differential prediction,
and college admission testing: a comprehensive review and analysis. New
York: College Entrance Examination Board.
3. Marklein,
M.B. (2000). GPA jumps to head of class.
USA Today, p 10D.
4. Noble,
J. (2003). The effects of using ACT
composite scores and high school averages on college admissions decisions for
ethnic groups. [pdf] Available at: http://www.act.org/research/researchers/reports/pdf/ACT_RR2003-1.pdf
[Accessed
24 October 2013]
5.
Zwick, R. (1999). Backdoor affirmative
action. Education Week, pp. 56, 35.
6.
Zwick, R. (2002). Fair game: the use of standardized admissions tests in higher education.
New York: Routledge Falmer.
