SUMMER 2007 JOURNAL OF COLLEGE ADMISSION W W W. N A C A C .docx

| SUMMER 2007 JOURNAL OF COLLEGE ADMISSION� W
W W. N A C A C N E T. O R G
ChriStoPher erik mattSon, M.F.A., M.Ed. is the
coordinator of testing accommodations for disability
services and programs at the University of Southern
California (CA). He earned his M.Ed. from the University
of Southern California, M.F.A. from Brandeis University
(MA) and B.A. from Western Washington University (WA)
after first attending Linfield College (OR).
By Christopher Erik Mattson
SUMMER 2007 JOURNAL OF COLLEGE ADMISSION | 9W
Beyond Admission:
Understanding Pre-College Variables and the Success of At-
Risk Students
Abstract
This study examined pre-college variables from an admission-
office perspective and the
ability of these variables to predict college grade point average
(GPA) for students spe-
cially admitted into an academic support program for at-risk
students. The research was

conducted at a private, highly-selective, research university in
the southwest United
States. The primary determining factors for this special
admission program are lower-
than-average high school GPA and/or standardized test scores.
Pre-college variables that
most significantly predicted college GPA were high school
GPA, gender of student, and
leadership experience prior to applying. Scholastic Aptitude
Test (SAT) scores failed to
predict success as measured by college GPA.
Beyond Admission
Seventy-five percent of students who drop out of college do so
during their first two years, and 57 percent of students leave
their first college without graduating (Tinto, 1993). First semes-
ter grades (McGrath & Braunstein, 1997) and first year grades
(Gifford, Briceño-Perriott, & Mianzo, 2006; Reason, 2003) are
significantly linked with retention. Because these grades act as
a
quantifiable predictor of retention and because grades are
associ-
ated with academic success, this study focuses on the predicting
of first-semester and first-year GPA of at-risk students.
The use of high school GPA and rank is widely accepted as a
positive predictor of academic success (Astin, 1997; Hoffman &
Lowitzki, 2005; Schwartz & Washington, 2002; Stricker, Rock
&
Burton, 1996; Ting, 1998; Wolfe & Johnson, 1995). Standard-

ized test scores have also been found as a predictor, although
they have been questioned in recent years (Astin, 1997; Lawlor,
S., Richman, S. & Richman, C.L., 1997; Naumann, Bandalos &
Gutkin, 2003; Reason, 2001; Stricker, et. al, 1996). Student
involvement (Astin, 1984) and a variety of emotional and social
variables (Boulter, 2002; DeBerard, Spielmans & Julka, 2004;
House, Keely & Hurst, 1996; McGrath & Braunstein, 1997;
Ridgell & Lounsbury, 2004; Spitzer, 2000; Wolfe & Johnson,
1995) have also been recently demonstrated as possible predic-
tors of success.
Changes in demographics have altered studies on academ-
ic success and retention (Reason, 2001). Much of the research
has been based on the traditional view of white, 18- to 22-year-
old, full-time students, even though the number of students of
color in higher education increased 61 percent between 1984
and 1994 (Pascarella & Terenzini, 1998). These changes have
created the necessity for research that understands the new
demographics of higher education.
Literature Review
A review of the literature showed that research related to the
predicting of academic success determined by GPA in college
has been productive in recent years, but not yet progressive.
Significant findings have been made using academic-related
vari-
ables (Lawlor, et. al, 1997; Reason, 2001, 2003; Stricker, et. al,
1996), non-academic variables (DeBerard, et. al, 2004; Nau-
mann, et. al, 2003; Spitzer, 2000) and a combination of both to
predict academic success (McGrath & Braunstein, 1997; Ridgell
| SUMMER 2007 JOURNAL OF COLLEGE ADMISSION10 W

& Lounsbury, 2004; Schwartz & Washington, 2002; Ting, 1998;
Wolfe & Johnson, 1995). Despite these findings, and numerous
recommendations, the usage of pre-college variables remains
very much the same (Astin, 1975, 1984, 1997; Atkinson, 2001;
Cooper, 1999; Fleming & Garcia, 1998; Lawlor, et. al, 1997;
Organ, 2001; Pascarella & Terenzini, 1998; Speyer, 2004; Tam
& Sukhatme, 2004; Tinto, 1993).
The non-academic related factors primarily researched
included emotional health, social health and physical health.
A study by DeBerard, Spielmans and Julka (2004) examined 10
variables that encompassed academic factors, social-support,
coping methods, and health status. Their findings indicated a
correlation with their variables for 56 percent of the variance
of first year GPA. Only low high school GPA, however, could
be
significantly associated with attrition.
Positive predictors of GPA in a study of 355 full-time under-
graduates by Spitzer (2000) were academic efficacy, self-
regulation
and social support. Naumann, Bandalos and Gutkin (2003) ex-
amined first-generation college students. In their questionnaire
study of 155 students they were able to identify self-regulation
as a positive predictor. They also found that ACT scores were
positively correlated with the GPA of first-generation students.
Test scores remained under attack. The validity (Speyer,
2004; Ting, 1998), usage (Atkinson, 2001; Cooper, 1999; Rea-
son, 2001; Tam & Sukhatme, 2004) and fairness (Fleming &
Garcia, 1998) of standardized test scores, such as the SAT and
ACT were being questioned. Cooper (1999) and Reason (2001)
examined ways in which adjustments could be made to the
scores to accommodate for diversity and differences between
high schools.

Reason, who proposed a merit-index score (2001),
significantly predicted the academic achievement of white
and African-American students with an ACT-based merit-index.
Cooper (1999) addressed the “strivers” approach introduced
by the Educational Testing Service (ETS). He defined strivers
as applicants who exceed the scores of individuals from similar
backgrounds by 200 or more points. Comparable to the merit-
index
approach, the problem with this strategy is that it failed to live
up to its goal of offering more opportunities to underrepresented
populations. The students who lose spots to strivers were often
minority students attending more affluent schools.
Test scores were also used in a 1997 study by Lawlor, Rich-
man and Richman that examined SAT scores as a predictor of
achievement for white and black students at Wake Forest Uni-
versity (NC). Their findings showed a strong correlation
between
the verbal portion of the SAT and GPA for both populations.
The math portion of the SAT did not prove to be a predictor.
Furthermore, the average total SAT scores for black students
in the study was 80 points lower than the average of white
students, but there were no differences in GPA between these
two populations. This finding indicated a possible bias in the
standardized testing.
Of the studies that examined academic and non-academic
factors, McGrath and Braunstein (1997) researched coping
skills, receptivity to support and initial impressions of students.
Their findings indicated that the biggest factors affecting
retention were first semester GPA and the students’ impressions
of other students. Ridgell and Lounsbury (2004) researched
general intelligence, personality traits and work drive. General
intelligence and work drive proved significant. Extroversion,
emotional stability, agreeableness, conscientiousness, and

openness to experience, otherwise known as the “big five”
personality traits, were not found to be significant predictors.
Schwartz and Washington (2002) focused their study on the
academic achievement of African-American freshmen men at a
Historically Black College or University (HBCU). Their
research
found high school GPA and certain non-cognitive variables sig-
nificant in predicting retention and academic achievement. The
significant non-cognitive variables were attachment to the col-
lege, academic adjustment and personal-emotional adjustment.
Research by Wolfe and Johnson (1995) examined the high
school GPA, SAT score and 32 personality variables of 201 stu-
dents in an introductory psychology course. High school GPA,
with 19 percent of the variance, was identified as the most sig-
nificant predictor. Self-control, with nine percent of the
variance,
came second and was followed by SAT score with five percent.
In a study of 54 students by Ting (1998), ACT score was
found not to be a predictor of first-year grades and academic
progress. High school rank and successful leadership experience
proved to be the most effective predictors. The examination of
leadership experience as a predictive variable did not turn up in
any of the other studies.
Methodology
Design
With permission from the Institutional Review Board of the
home
university, this study examined the application materials of
more
than 900 students who entered the university through a special
admission program designed to assist students determined by
the

admission office as being academically at-risk. This determina-
tion was based on lower high school GPA and standardized test
scores than the regularly admitted school population. Students
admitted into this program receive additional support and were
required to take a first-year course focused on time
management,
college study strategies and educational psychology. Surveys
were
not needed for this study because the data was already
available.
Pre-college information for the students in the study was
obtained through admission application materials and internal
office adjustments. Choosing this viewpoint made the perspec-
tive of an admission counselor possible. The high school GPA
of
students used for this study was adjusted by the office of admis-
sion. High schools are increasingly not providing class rankings
(Ehrenberg, 2005), so institutions often independently weight
and adjust high school GPA to fairly compare applicants.
SUMMER 2007 JOURNAL OF COLLEGE ADMISSION | 11W
The students whose files were examined entered the uni-
versity as early as the fall semester of 1999 and as late as the
fall semester of 2003. Nearly one-third of the students exam-
ined were student athletes and were excluded from the study
because this group primarily received academic and social sup-
port from the department of athletics. This exclusion brought
the population to 591 students with the opportunity of at least
three academic semesters of study.
Participants

The students selected represented an accessible, ethnically
diverse,
at-risk population. Their composition consisted of 39.8 percent
white/Caucasian students, 20.5 percent black/African American,
8.8 percent Asian/Pacific Islander, 19.6 percent Hispanic or
Mexi-
can, 0.7 percent Native American, and 10.7 percent
mixed/other.
These students arrived with an average high school GPA of 3.36
and SAT of 1076. Although these numbers are respectable, they
were below the overall student averages at the university, thus
clas-
sifying the population as academically at-risk. For instance, the
fall
2004 entering freshman class at the university averaged a 4.09
GPA and middle 50 percent SAT range of 1310–1460.
Retention of these students for their second year was high
(96.3 percent) and differentiated only slightly when consider-
ing ethnicity, first-generation status, high school GPA, and
test scores. The retention rate for the university as a whole
remained between 94 percent and 96 percent during common
years. These commonalities provided another reason for this
study to remain primarily focused on academic achievement
determined by university GPA. (When it comes to the reten-
tion of at-risk students future studies might want to explore
socioeconomic status, as well rising tuition and distance from
home. Perhaps at-risk students with low college GPA are more
likely to be retained when they can afford to continue.)
Many of the students in this program were selected because
of unique characteristics that make them more desirable and
worthy of admission. Many had non-quantifiable talents and
abilities associated with music, theatre, art, engineering, busi-
ness, and architecture. Many were also from diverse
backgrounds

or offer unique perspectives that make them attractive to the
school, despite lower high school GPA or test scores.
Measures
The primary non-cognitive characteristics identified for this
study
included entry age, gender, ethnicity, first-generation status, re-
ported presence of a language spoken in the home other than
English, and reported leadership experience. Cognitive
variables
obtained from admission information included high school GPA
and SAT scores. The academic data gathered included enroll-
ment status, first-year GPA, first-semester GPA, cumulative
GPA,
and retention for the second year.
Leadership experience was defined as being peer related.
Individuals who were members of clubs and organizations
were not classified as leaders unless they held a position
clearly associated with leadership experience. Such positions
included: president, vice president, chair, vice-chair, captain,
co-captain, founder, or any other justly determined leadership
position. Of the 591 students, 287 had leadership experience
and 304 did not.
Entry age remained consistent across groups for this popu-
lation primarily because the program is designed for first-time
freshmen. For this measurement, birth dates and entry dates
were attached values according to their time of the year. For ex-
ample, June received a value of .5 because it is the sixth month
of 12. Fall cohorts received an entry value of .67 because school
began during August, the eighth month of 12. Entry age was
then calculated by subtracting the birth date from the entry date.
This allowed for a more accurate determination of entry age.

Students entering this program averaged 18.45 years. This
number did not vary much with black/African American
students
averaging 18.26 years of age, white/Caucasian students averag-
ing 18.6, and all others falling within this range. For the entire
population, the youngest student entered at 16.67 and the old-
est at 22.17.
Of the 591 students, the breakdown by gender included 55.8
percent females and 44.2 percent males. This ratio remained
relatively consistent for different ethnicities also, with females
consistently making up the majority. In measuring language,
37.2
percent reported having a language spoken in the home other
than
English. Leadership experience was identified for 48.6 percent.
As previously mentioned, average adjusted high school GPA
was
3.36 and the population had an SAT average of 1076.
When it came to succeeding academically in college,
first-year GPA averaged 2.81 and first-semester GPA 2.85.
The university average for students during the years examined
ranged from 3.08 to 3.20 for both first-semester and first-year
GPA. Although this at-risk population did not perform as well
as
their regularly admitted counterparts, they did seem to bridge
the gap when considering the differences between average high
school GPA and standardized test scores.
Results
Three variables in this study emerged as significant predictors
of academic success. High school GPA, gender and leadership
experience proved to be positive correlates at the .01 level as
predictors of first-semester GPA and first-year GPA. (See Table
4.) The significance for high school GPA is visible when

grouping
the GPA and using a one-way ANOVA the resulting first-
semester
GPA and first-year GPA demonstrated the significance. Also
see
table one for additional results for high school GPA.
Females outperformed males significantly when it came to
first-semester GPA and first-year GPA when running a one-way
ANOVA. The difference between high school GPA and SAT
scores
for females and males should also be noted. Females averaged a
3.42 high school GPA and 1059 SAT. Males averaged a 3.29
high
| SUMMER 2007 JOURNAL OF COLLEGE ADMISSION12 W
school GPA and 1098 SAT. It is unclear whether these
differences
impacted the results, although it could be argued that females
within this population outperformed males because of their
slight-
ly better high school GPA. (See Table 2 for more information.)
Students with pre-college leadership experience performed
better when it came to first-semester GPA and first-year GPA.
A one-way ANOVA shows the significance for first-semester
GPA and
first-year GPA. (For results, see Table 3.) Students with pre-
college
leadership experience tended to have similar SAT scores, 1069
to
1083, and high school GPA, 3.40 to 3.33, to those without this

characteristic. The similar high school GPA and test scores
amongst
those with and without leadership experience provides further
evi-
dence for the variables ability to predict college GPA.
Discussion
The results of this study confirmed research regarding high
school
GPA as a successful significant positive predictor (Astin, 1997;
Hoffman & Lowitzki, 2005; Schwartz & Washington, 2002;
Ting,
1998; Wolfe & Johnson, 1995). This study, investigating a
similar
population, also legitimizes the work of Tobey (1997) in show-
ing the significance of high school GPA as a positive predictor
of
academic success for at-risk students. The work of Ting (1998)
is
also validated through the significant use of leadership
experience
as a predictor. These findings also confirm the need for changes
in how SAT scores are valued (Atkinson, 2001; Cooper, 1999;
Fleming & Garcia, 1998; Reason, 2001, 2003; Speyer, 2004;
Stricker, 1996).
The ability to generalize the findings of this study were lim-
ited due to the specificity of the college and special admission
program. The presence of students in the study entering the uni-
versity during a period of different years might also draw
questions.
The findings, however, are unique in examining pre-college
factors
from an admission perspective and the resulting college GPA.

Students considered for this special admission program of-
ten had lower-than-average high school GPA or SAT scores. If a
student with a high GPA was admitted, then it usually meant
they
had a low SAT, or vice versa. This study shows that when
admitting
students with low high school GPA or low SAT scores it is more
accurate to admit based on high school GPA to predict success.
Conclusion
By identifying leadership experience, high school GPA and gen-
der as positive predictors of academic achievement this study
adds to the literature and provides further questioning as to the
heavy usage of SAT scores. Logical reasons can be found for
why leadership experience is able to predict academic achieve-
ment at both high school and college. Leadership ability can be
attached to work drive, self-regulation and other desirable per-
sonality characteristics. This finding satisfies higher education
administrators search for additional effective pre-college
predic-
tors of success (Gifford, et al, 2006). Higher education greatly
over-emphasizes certain characteristics, such as SAT scores.
Placing more of an emphasis on other characteristics, such as
pre-college leadership experience, might be more beneficial
when it comes to admitting students and ranking institutions.
Astin (1997) claimed that high school GPA, test scores, gen-
der, and race accounted for the majority of variation in
retention.
Changing demographics have blurred race, ethnicity and culture
(Pascarella & Terenzini, 1998; Reason, 2003), and higher
educa-
tion must now look deeper into the true nature of applicants
when
deciding admission. Test scores are also under scrutiny (Atkin-
son, 2001; Fleming & Garcia, 1998; Speyer, 2004) and there are

recommendations (Cooper, 1999; Reason, 2001, 2003; Stricker,
1996) as to how they should be reconsidered. The findings in
this
study encourage the development of a value system that more
ac-
curately admits, predicts and ranks success.
Recommendations
Higher education administrators need to reevaluate the
magnitude
of pre-college variables, especially when deciding which
students
are admitted and/or determined to be at-risk, and ask themselves
if SAT scores are used so heavily in the college admission
process
because they are predictors of academic success, or (Ehrenberg,
2005) because they are linked with the college ranking systems?
As race and affirmative action policies become increasingly
ques-
tioned, alternative approaches that value individuals from
diverse
backgrounds must be considered––an increase in the value of
other
variables can make possible a more accurate prediction of
success.
This study recommends questioning the weight assigned to test
scores (such as the SAT) when considering the selectivity of a
univer-
sity. Changes could place a more desirable value on more
predictive
characteristics, while also symbolizing an honest commitment
from
higher education on searching for qualities in demand.
Hs gPA sAT College gPA

By range Average 1st Semester 1st Year
2.5-2.99 1112 2.68 2.66
3.0-3.49 1091 2.83 2.78
3.5-3.99 1041 2.94 2.88
4.0 or better 1046 3.00 3.02
Table 1
(High
School
GPA)
gender sAT Hs gPA College gPA
Average Average 1st Semester 1st Year
Male 1098 3.29 2.74 2.69
Female 1059 3.42 2.94 2.90
Table 2 (Gender of Student)
LE sAT Hs gPA College gPA
Average Average 1st Semester 1st Year
Yes 1069 3.40 2.94 2.90
No 1083 3.33 2.77 2.73
Table � (Leadership Experience (LE))

SUMMER 2007 JOURNAL OF COLLEGE ADMISSION | 1�W
Astin, A.W. (1975). Preventing students from
dropping out. San Francisco: Jossey-Bass.
Astin, A. W. (1984). Student involvement: A
developmental theory for higher education.
Journal of College Student Personnel, 25,
297-308.
Astin, A. W. (1997). How ‘good’ is your insti-
tution’s retention rate? Research in Higher
Education, 38, 647–658.
Atkinson, R.C. (2001). Achievement versus
aptitude in college admissions. Issues in Sci-
ence and Technology, 18 (2) 31-6.
Boulter, L.T. (2002). Self-concept as a predic-
tor of college freshman academic adjustment.
College Student Journal, 3 6(2) 234-46.
Cooper, K.J. (1999). Admissions models for
inclusion. Black Issues in Higher Education,
16 (17), 34-35.
DeBerard, M.S., Spielmans, G.I. & Julka, D.L.
(2004). Predictors of academic achievement
and retention among college freshmen: A
longitudinal study. College Student Journal,
38 (1) 66-80.
Ehrenberg, R.G. (2005). Method or Madness? In-

side the U.S. News & World Report College Rank-
ings. Journal of College Admission, 189, 29-35.
Fleming, J. & Garcia, N. (1998). Are standard-
ized tests fair to African Americans?: Predic-
tive validity of the SAT in Black and White
institutions. The Journal of Higher Education,
69(5), 471-95.
Gifford, D.D., Briceño-Perriott, J. & Mianzo, F.
(2006). Locus of control: Academic achieve-
ment and retention in a sample of university
first-year students. Journal of College Admis-
sion, 191, 18-25.
Hoffman, J.L. & Lowitzki, K.E. (2005). Predict-
ing college success with High School grades
and test scores: Limitations for minority stu-
dents. The Review of Higher Education, 28 (4)
455-474.
House, J. D., Keely, E.J. & Hurst, R.S. (1996).
Relationship between learning attitudes, prior
achievement, and performance in a general
education course: A multi-institutional study.
International Journal of Instructional Media,
23 (3) 257-71.
Lawlor, S., Richman, S. & Richman, C.L.
(1997). The validity of using the SAT as a
criterion for Black and White students’ ad-
mission to college. College Student Journal,
31, 507-15.
McGrath, M. & Braunstein, A. (1997). The pre-

diction of freshmen attrition: An examination
of the importance of certain demographic,
academic, financial, and social factors. Col-
lege Student Journal, 31, 396-408.
Naumann, W. C., Bandalos, D. & Gutkin, T.B.
(2003). Identifying variables that predict
college success for first-generation college
students. Journal of College Admission, 181,
4-9.
Organ, D. W. (2001). Predicting success at
school and work. Business Horizons, 44 (3) 1-2.
Pascarella, E. T., & Terenzini, P. T. (1998).
Studying college students in the 21st cen-
tury: Meeting new challenges. The Review of
Higher Education, 21, 151-165.
Reason, R. D. (2001). The use of merit-index
measures to predict between-year retention of
undergraduate college students. Unpublished
doctoral dissertation, Iowa State University.
Reason, R. D. (2003). Student variables that
predict retention: Recent research and new
developments. NASPA Journal, 40 (4), 172-91.
Ridgell, S. D. & Lounsbury, J. W. (2004).
Predicting academic success: General intel-
ligence, “Big five” personality traits, and
work drive. College Student Journal, 38 (4)
607-18.
Schwartz, R. A. & Washington, C. M. (2002).

Predicting academic performance and reten-
tion among African American Freshmen men.
NASPA Journal 39 (4), 354-370.
Speyer, M. (2004). Are numbers are up! (Is
that good?). Journal of College Admission,
182, 8-15.
Spitzer, T. M. (2000). Predictors of college
success: A comparison of traditional and
nontraditional age students. NASPA Journal,
38 (1), 82-98.
Stricker, L. J., Rock, D.A. & Burton, N.W.
(1996). Using the SAT and high school record
in academic guidance. Educational and Psy-
chological Measurement, 56, 626-41.
Tam, M. S. & Sukhatme, U. (2004). How to
make better college admission decisions:
Considering high school quality and other
factors. Journal of College Admission, 183,
12-6.
Ting, S. (1998). Predicting first-year grades
and academic progress of college students
of first-generation and low-income families.
Journal of College Admission, 158, 14-23.
Tinto, V. (1993). Leaving college: Rethinking
the causes and cures of student attrition (2nd
ed.). Chicago: University of Chicago Press.
Tobey, P. E. (1997) Cognitive and non-cogni-
tive factors as predictors of retention among
academically at-risk college students: A struc-

tural equation modeling approach. Disserta-
tion Abstracts International Section A: Hu-
manities and Social Sciences 57 (7-A), 2907.
Wolfe, R. N. & Johnson, S. D. (1995). Person-
ality as a predictor of college performance.
Educational and Psychological Measurement,
55, 177-85.
REFERENCES
gender Leadership Adjusted Hs gPA sAT score First semester
gPA First year gPA
gender
Pearson Correlation 1 .080 .157(**) -.214(**) .157(**) .189(**)
Sig. (2-tailed) .052 .000 .000 .000 .000
N 591 591 591 591 591 591
Leadership
Pearson Correlation .080 1 .086(*) -.078 .132(**) .153(**)
Sig. (2-tailed) .052 .037 .059 .001 .000
N 591 591 591 591 591 591
Adjusted Hs gPA

Pearson Correlation .157(**) .086(*) 1 -.285(**) .182(**)
.198(**)
Sig. (2-tailed) .000 .037 .000 .000 .000
N 591 591 591 591 591 591
sAT score
Pearson Correlation -.214(**) -.078 -.285(**) 1 -.036 -.024
Sig. (2-tailed) .000 .059 .000 .388 .554
N 591 591 591 591 591 591
First semester gPA
Pearson Correlation .157(**) .132(**) .182(**) -.036 1 .866(**)
Sig. (2-tailed) .000 .001 .000 .388 .000
N 591 591 591 591 591 591
First year gPA
Pearson Correlation .189(**) .153(**) .198(**) -.024 .866(**) 1

Sig. (2-tailed) .000 .000 .000 .554 .000
N 591 591 591 591 591 591
** Correlation is significant at the 0.01 level (2-tailed). *
Correlation is significant at the 0.05 level (2-tailed).
Table � (Correlations)
| FALL 2011 JOURNAL OF COLLEGE ADMISSION18 W W
W. N A C A C N E T. O R G
Recasting Non-Cognitive Factors
in College Readiness as What They
Truly Are: Non-Academic Factors
by Amanda Sommerfeld
Against the backdrop of traditional measures of college
readiness (i.e., high school GPA,
standardized test scores, and high school rank), the
consideration of “non-cognitive” factors
marked a significant departure when they were first discussed as
integral aspects of college
success. Used to refer to any characteristic, ability or
disposition that was theorized to af-
fect college success or retention, the term “non-cognitive”
became a kind of a catchall for
any variables beyond those “cognitive” or “intellectual”
variables listed above. However, the
term “non-cognitive” is now and has always been a misnomer,
used to refer to a vast array
of constructs, many of which reflect cognitive process.

Now, this could be considered simply a semantic issue; “non-
cognitive,” “dispositional” or “non-academic,” what difference
does
it make if there is shared understanding of what the term refers
to?
The problem is that there isn’t shared understanding and as a
result
this essential area of study has been compromised by critiques
and
misunderstandings, causing not only confusion about how these
factors affect college readiness, but also general skepticism
about
their utility in predicting college success. It seems, then, that
for
more transparent discourses to take place about the multifaceted
nature of college readiness, this crucial area of research must
be recast, enhancing the clarity of the field in order to provide a
linguistic grounding for future research. With this in mind, as
a preliminary step the author proposes that “non-cognitive” be
replaced with the more appropriate “non-academic” term,
thereby

elucidating the true distinction that researchers are making
between
various contributing factors to college preparation and success.
Why Do “Non-Cognitive” Factors Matter?
The introduction of “non-cognitive” factors into college
readiness
discussions grew out of necessity, prompted by decades of
disparity
between the rates of college acceptance, attendance and
completion
by non-traditional college students (i.e., students of color, first-
generation college students, older students, students with
special
learning needs, etc.) in comparison to the more traditional
college-
going population (i.e., white, middle to upper-middle class
men).
These disparities were quickly attributed to higher education’s
reliance on cognitive factors (i.e., standardized test scores, high
school rank, high school GPA) as the basis of college admission
that, though shown to be effective in identifying students likely
to

succeed in higher education (e.g., Bentekoe 1992, Bridgeman,
McCamley-Jenkins and Ervin 2000, Kuncel et al. 2005, Kuncel,
Hezlett and Ones 2001, Mouw and Khanna 1993, Noble 1991)
were criticized for their inutility across more diverse student
groups.
In particular, the use of traditional college admission criteria as
the sole grounds for college access has been criticized for two
reasons: first, because traditional measures of college admission
are ineffective identifiers of students at risk of attrition
(Sedlacek
FALL 2011 JOURNAL OF COLLEGE ADMISSION | 19W W
about a field. Based on these factors, he developed the Non-
Cognitive Questionnaire (NCQ), which has been used in scores
of
studies since its development in 1976 (Sedlacek and Brooks
1976).
An equally compelling, though less-employed approach uses the
work of Owens and colleagues (e.g., Mumford and Owens 1987)
who promote the use of biodata in predicting student outcomes

such as overall achievement, disciplinary actions, college GPA,
and course withdrawals. The “non-cognitive” factors they
propose
include: knowledge and mastery of general principles
(Knowledge);
continuous learning, and intellectual interest and curiosity
(Learning); artistic and cultural appreciation (Artistic);
appreciation
for diversity (Diversity); leadership (Leadership); interpersonal
skills
(Interpersonal); social responsibility and citizenship
(Responsibility);
physical and psychological health (Health); career orientation
(Career); adaptability and life skills (Adapt); perseverance
(Persevere); and ethics and integrity (Ethics).
However, not all research on non-cognitive factors in college
success
is grounded in an overarching theory. In fact, most studies
explore
the utility of a single variable or cluster of variables in
predicting out-
comes, resulting in a seemingly limitless list of constructs that

have
fallen under the “non-cognitive” umbrella, including
metacognitive
skills (e.g., Credé and Kuncel 2008, Zeegers 2001), study
attitudes
(e.g., Zimmerman, et al. 1977), study motivation (e.g.,
Melancon
2002), academic and social integration (e.g., Milem and Berger
1997, Tinto 1993); college knowledge (Conley 2005);
personality
1993); and second because they tend to lack predictive validity
for assessing non-traditional college students’ readiness for
college,
oftentimes either resulting in a number of false negatives or
overpredicting performance (Young and Koplow 1997). For
example,
although traditional admission criteria have been routinely
criticized
for underestimating the college performance of non-traditional
students, other studies have identified the tendency for “highly
qualified” non-traditional students (e.g., students of color who
score well on SATs and have high GPAs from high-quality

schools) to
oftentimes stumble academically in the face of unwelcoming
college
environments (c.f., Linn 1990, Young 1993).
Considering “non-cognitive” factors in assessments of college
readiness was, therefore, seen as a way to improve the accuracy
of selection criteria, casting light on students’ abilities to
navigate
the multiple demands of the college environment so that they
may
have been better able to persist to graduation. In fact, research
has
consistently supported this theory, with dozens of studies across
multiple populations demonstrating that the validity of college
success predictions can be improved by including non-cognitive
factors, such as commitment to school (Tinto 1993), long-term-
goal setting (Young and Sowa 1992) and social support
(Lotkowski,
Robbins and Noeth 2004), just to name a few.
What Are “Non-Cognitive” Factors?
When we’re referring to “non-cognitive” factors, what exactly

de-
fines them? The answer, albeit troubling, is “It depends;” it
depends
Considering “non-cognitive” factors in assessments of college
readiness was, therefore, seen as a way to improve the accuracy
of selection criteria, casting light on students’ abilities to
navigate
the multiple demands of the college environment so that they
may have been better able to persist to graduation.
on who conducts the research and whether they’re basing their
work
on an established theory or exploring a new variable of interest.
For example, one of the most prolific researchers in the field of
college readiness is William Sedlacek, who theorized that there
are eight essential non-cognitive components of college
readiness:
positive self-concept regarding academics; realistic self-
appraisal;
understanding/dealing with racism; long-term goal setting;
having
an available support person; demonstrated experience and
success
with leadership; community service; and knowledge acquired in/
(e.g., Ridgell and Lounsbury 2004), student involvement (e.g.,

Astin
1993); university actions (e.g., Pascarella, Terenzini and Wolf
1986);
problem-solving skills (e.g., Le, et al. 2005), and self-efficacy,
effort
regulation, and outcome expectations (e.g., Myers 2004).
Are They Really “Not Cognitive”?
As the above section details, clearly a great deal of variability
exists
in the constructs identified as “non-cognitive,” but as the list is
considered, the appropriateness of their categorization as “non-
cognitive” becomes dubious.
From the Latin cognoscere, which means "to know" or "to
recognize,"
cognition is defined as, “The mental process of knowing,
including
aspects such as awareness, perception, reasoning, and
judgment.”
(The American Heritage Dictionary of the English Language

2000).
As such, cognitive processes refer to how people process
information, in-
cluding how they perceive, learn and consider new facts or
experiences.
Now, certainly there are some constructs included under the um-
brella of “non-cognitive” factors that truly appear to fall outside
its
definition: constructs such as personality, having an available
sup-
port person, and university actions. Aside from this short list,
the
majority of the remaining constructs are undeniably dependent
on
cognitive processes.
For example, the ability to set long-term goals is one construct
that
has received a great deal of support as an essential characteris-
tic of successful college students (e.g., Mischel and Ayduk
2004).
This construct, which is a direct manifestation of human
executive
functioning abilities seated in the prefrontal cortex, must

certainly
be considered a cognitive process; not only does it rely on an
estab-
lished awareness of consequences, but also requires the capacity
to perceive how actions affect outcomes, and the ability to
choose
to delay immediate gratification in favor of greater benefits in
the
long run.
Similarly, constructs such as adaptability, understanding and
deal-
ing with racism and problem-solving skills all require students
to be
able to analyze a situation, identify the relevant factors, and
make
choices based on reasoned logic rather than initial reactions—
all
of which fall under the purview of cognitive processes. Given
this,
the term “non-cognitive” hardly seems the appropriate choice
for
capturing this highly variable domain of inquiry, as it is not
only

phenomenologically incorrect, but unnecessarily vague as well.
Furthermore, by mislabeling these factors as “non-cognitive”
schol-
ars have succeeded in compromising the application of their
own
work, contributing to the confusion about what makes a student
ready for college and undermining possible avenues of
intervention.
Implications of Definitional Vagueness
Since their initial introduction, research on the “non-cognitive”
factors in college readiness and success has steadily grown,
with
associated implications for college programming and resource
al-
location. For example, a 2000 article in the Chronicle of Higher
Education about the Gates Millennium Scholars Program, a
schol-
arship program for minority youth, proclaimed that “The $1-
billion
scholarship program—the largest in higher education—amounts
to
an unprecedented, large-scale experiment” because of its
reliance

on “non-cognitive” variables such as “community service,
demon-
strable leadership skills, the ability to cope with racism, and
other
hard-to-quantify characteristics” rather than traditional college
ad-
mission criteria (Pulley 2000, A41).
That a scholarship program with this many resources and of this
caliber would choose to base its decisions on “non-cognitive”
characteristics is evidence of the scope and caliber of research
on
the topic, with publications numbering in the hundreds
(Thomas,
Kuncel and Credé 2007). However, despite the prolific research
and
the fact that “non-cognitive” factors have begun being
integrated
into college admission decisions at institutions such as
Louisiana
State University Medical School (LA), North Carolina State
University (NC), and Muhlenberg College (PA) (Sedlacek
2004),

critiques persist, particularly because of the varying amounts of
support that different “non-cognitive” constructs have garnered,
and the widely disparate predictive validity of those constructs
(Thomas, et al. 2007).
For example, in a study that examined the validity of non-
cognitive
factors in predicting the academic performance of African-
American
college students, Nasim and colleagues (2005) found that
predic-
tors varied in their utility across contexts. So whereas the
academic
performance of African-American students attending a predomi-
nantly white institution was associated with whether they had an
academic support person available, their capacity to understand
and deal with racism, and their espousal of humanist attitudes,
the
achievement of African-American students attending
Historically
Black Colleges or Universities was only influenced by their
degree of
positive academic self-concept.

The authors of the Noncognitive Questionnaire, the measure
most
commonly used to assess “non-cognitive” factors in the research
literature, similarly found differences in the predictive validity
of
constructs across groups. In their initial construct validation
study,
they found that of the six non-cognitive factors that were
supported
From the Latin cognoscere, which
means "to know" or "to recognize,"
cognition is defined as, “The mental
process of knowing, including aspects
such as awareness, perception,
reasoning, and judgment.”
FALL 2011 JOURNAL OF COLLEGE ADMISSION | 21W W
by their factor analysis (i.e., leadership, recognizing racism,
long-term
goal orientation, realistic self-appraisal, support for college
plans, and
self-confidence), only self-confidence, long-term goal

orientation and
realistic self-appraisal were predictive of white students’ first
semester
GPA, and only positive self-concept and realistic self-appraisal
were
predictive of GPA in African-American students.
Findings like those outlined above have been replicated in a
variety
of others studies (e.g., Ancis and Sedlacek 1997, Fuertes, Sed-
lacek and Liu 1994), indicating differential predictive validity
of
“non-cognitive” constructs. Despite this, scholars persist in
writ-
ing broadly of the importance of “non-cognitive” factors in
college
readiness. As a result, claims that “non-cognitive” factors,
overall,
are essential for college readiness have been challenged,
accompa-
nied by calls for more psychometrically sound measures and
greater
definitional clarity on the whole.
In response to these critiques, the author argues for a semantic

switch—a movement away from the broad, undefined category
of
“non-cognitive” to nomenclature that allows for greater
conceptual
clarity. By categorizing important factors of college readiness
as
“academic” v. “non-academic,” a more apparent distinction can
be
made between that which is based on formal education (i.e.,
grades,
subject matter knowledge, etc.) and those additional factors that
af-
fect a student’s ability to adapt to and meet the varying
demands of
a college environment. The category of “non-academic” could
then
be broken down into appropriate subgroups, such as
dispositions,
executive functioning abilities, habits of mind, external
resources,
and college knowledge (see Table 1), that more specifically
capture
the different domains of integral variables.

Changing the Language: Implications of the Semantic Switch on
Research and Interventions
Even if it may seem a trivial adjustment, making the linguistic
switch from “non-cognitive” to “non-academic” signals a
deeper,
meaningful change in the area of college readiness. Ever since
factors
outside of GPA and standardized test scores were initially
considered
as potentially meaningful in the collegiate experiences of
students,
Factor Definition examples
Academic Factors Factors explicitly targeted in formal
education. Subject matter knowledge, high school grades,
standardized test scores, etc.
Non-Academic Factors
Dispositions Internal characteristics that distinguish a person’s
predominant outlook or characteristic attitude.
Personality (Ridgell and Lounsbury 2004)
Temperament
Values
Attitudes (Owens et al. 1976)
Habits of Mind “Habits of thought and action that help people
manage uncertain or challenging situations…
[supporting] thoughtful and intelligent action”

(Costa and Kallick 2000, 4).
Metacognitive skills (Zeegers 2001)
Creative thinking (Owens 1976)
Study attitudes (Zimmerman, et al. 1977)
Outcome expectations (Myers 2004)
Intellectual curiosity (Owens 1976)
Appreciation for diversity (Sedlacek 1993)
Leadership (Owens 1976)
Positive self-concept (Sedlacek 1993)
Executive Functioning
Abilities
Foundational skills that allow individuals to
effectively navigate daily tasks, including the ability
to “orient, plan, program responses, and verify and
modify performances” (Denckla 1996, 263).
Study habits
Reasoning (Le, et al. 2005)
Long-term goal setting (Sedlacek 1993)
Realistic self-appraisal (Sedlacek 1993)
Decision making
Self-control (Myers 2004)
Goal commitment
External Resources External factors that the individual may be
able to
access to support college readiness/success.
University fit (Nasim, et al. 2005)
Financial stability
Family beliefs about education
Institution intervention (Pascarella, et al. 1986)
Support person (Sedlacek 1993)

College Knowledge Explicit and implicit knowledge required
for college
success; essentially knowing how to “do” college.
Knowledge of college requirements, placement test policies, and
tuition costs
(Conley 2005)
Understanding of the structure of college
Ability to recognize the systemic requirements and norms
Table 1: Factors in College Readiness and Success
Ancis, J. R., and Sedlacek,W. E. (1997). Pre-
dicting the academic achievement of female
students using the SAT and noncognitive
variables. College and Universit y, 72, 2-8.
Astin. A. W (1993). What matters in college"
Four critical years revisited. San Francisco:
Jossey-Bass.
Bontekoe, J. (1992). The ACT as a predic-
tor of college success at Trinity Christian
College. (ERIC Document Reproduction
Service No. ED 355 258).
Bridgeman, B., McCamley-Jenkins, L., and
Ervin, N. (2000). Predictions of freshman
grade point average from the revised and
recentered SAT I: Reasoning Test (College

Board Rep. No. 2000-1). New York: College
Entrance Examination Board.
Conley, D. (2005). College knowledge: What
it takes students to succeed and what we
can do to get them ready. San Francisco:
Jossey-Bass.
Crede´, M., and Kuncel, N. R. (2008). Study
habits, skills, and attitudes: The third pil-
lar supporting collegiate academic per-
formance. Perspectives on Psychological
Science, 3, 425–453.
Fuertes, J. N., Sedlacek, W. E., and Liu, W.
M. (1994). Using the SAT and noncogni-
tive variables to predict the grades and
retention of Asian-American university
students. Measurement and Evaluation in
Counseling and Development, 27, 74-84.
Kuncel, N. R., Credé, M., Thomas, L. L.,
Seiler, S. N., Klieger, D. M., and Woo, S. E.
(2005).A meta-analysis of the Pharmacy
College Admission Test (PCAT) and grade
predictors of pharmacy student success.
American Journal of Pharmaceutical Educa-
tion, 69, 339-347.
Kuncel, N. R., Hezlett, S. A., and Ones, D. S.
(2001). A comprehensive meta-analysis of
the predictive validity of the Graduate Re-
cord Examinations: Implications for gradu-
ate student selection and performance.
Psychological Bulletin, 127, 162-181.

Le, H., Casillas, A., Robbins, S. B., and
Langley, R. (2005). Motivational and skills,
social, and self-management predictors of
college outcomes: Constructing the Stu-
dent Readiness Inventory. Educational and
Psychological Measurement, 65, 482-508.
Lotkowski, V.A., Robbins, S.B., and No-
eth, R.J. (2004). The role of academic and
non-academic factors in improving college
retention: ACT policy report.
Linn, R.L. (1990). Hispanics’ education and
background: Predictors of college achieve-
ment. New York: College Entrance Exami-
nation Board.
Melancon, J. G. (2002). Reliability, struc-
ture, and correlates of Learning and Study
Strategies Inventory scores. Educational and
Psychological Measurement, 62, 1020–1027.
Milem, J. F., and Berger, J. B. (1997). A
modified model of college student persis-
tence: Exploring the relationship between
Astin's theory of involvement and Tinto's
theory of student departure. Journal of Col-
lege Student Development, 38, 387-400.
Mischel,W., and Ayduk, O. (2004). Willpower
in a cognitive-affective processing system:
The dynamics of delay of gratification. In R. F.
Baumeister and K. D. Vohs (Eds.), Handbook
of self-regulation: Research, theory, and ap-
plications (pp. 99-129). New York: Guilford.

Mouw, J., and Khanna, R. (1993). Predic-
tion of academic success: A review of the
literature and some recommendations.
College Student Journal. 27, 328-336.
Mumford, M. D., and Owens, W. A. (1987).
Methodology review: Principles, proce-
dures, and findings in the application of
background data measures. Applied Psy-
chological Measurement, 11, 1–31.
Myers, D. G. (2004). Psychology (rev. ed.).
New York: Worth.
Nasim, A., Roberts, A., Harrell, J.P., and
Young, H. (2005). Non-cognitive predic-
tors of academic achievement for African
Americans across cultural contexts. The
Journal of Negro Education, 74(4), 344-358.
Noble, J. P. (1991). Predicting college
grades from ACT assessment scores and
high school course work and grade infor-
mation (Rep. No. ACT-RR-91-3). Iowa City,
IA: American College Testing Program.
(ERIC Document Reproduction Service No.
ED344943)
Pascarella, E. T., Terenzini, P, and Wolf, L.
(1986). Orientation to college and freshman
year persistence/withdrawal decisions.
Journal of Higher Education, 57, 155-175
Pulley, J.L. (2000, June 23). A $1-billion
experiment seeks a new way to identify
talented minority students. The Chronicle

of Higher Education, pp. A41.
Ridgell, S., and Lounsbury, J. W. (2004).
Predicting collegiate academic success:
General intelligence, “Big Five” personal-
ity traits, and work drive. College Student
Journal, 38, 607–618.
Sedlacek, W. E. (1993). Employing non-
cognitive variables in admissions and
retention in higher education. In Achiev-
ing diversity: Issues in the recruitment and
retention of underrepresented racial/ethnic
students in higher education. (pp. 33-39).
Alexandria VA: National Association of Col-
lege Admission Counselors.
Sedlacek, W.E. (2004). Beyond the Big Test:
Noncognitive assessment in higher educa-
tion. San Francisco, CA: Jossey Bass.
Sedlacek, W.E., and Brooks, G.C., Jr. (1976).
Racism in American education: A model for
change. Chicago: Nelson-Hall.
The American Heritage Dictionary of the
English Language. (2009) (4th ed). Boston,
MA: Houghton Mifflin Company.
Thomas, L.L., Kuncel, N.R., and Credé, M.
(2007). Noncognitive variables in college
admissions: The case of the Non-Cognitive
Questionnaire. Educational and Psychologi-
cal Measurement, 67, 635-657.
Tinto, V. (1993). Leaving College: Rethink-

ing the Causes and Cures of Student At-
trition (2nd ed.). Chicago: University of
Chicago Press.
Young, B., and Sowa, C. (1992). Predictors
of academic success for Black student
athletes. Journal of College Student Devel-
opment, 33, 318-324.
Young, J.W. (1993). Grade adjustment
methods. Review of Educational Research,
63, 151-165.
Young, J.W. and Koplow, S.L. (1997). The
validity of two questionnaires for predict-
ing minority students’ college grades. The
Journal of General Education, 46(1), 45-55.
Zeegers, P. (2001). Approaches to learning
in science: A longitudinal study. British Jour-
nal of Educational Psychology, 71, 115–132.
Zimmerman, W. S., Parks, H., Gray, K., and
Michael, W. B. (1977). The validity of tradi-
tional cognitive measures and of scales of
the Study Attitudes and Methods Survey in
the prediction of the academic success of
Educational Opportunity Program students.
Educational and Psychological Measurement,
37, 465–470.
REFERENCES
AMANDA SoMMeRFelD, Ph.D. is an assistant
clinical faculty member at Boston University (MA).
Her current research focuses on examining the role of

economic, social and cultural capital in educational
advancement and college success for urban youth.
She is currently teaching classes in the areas of
counseling theory and cross-cultural perspectives.
research on the topic has faced (oftentimes justified) criticisms
for
its lack of clarity and rigor, as well as for overgeneralizing its
findings.
By virtue of changing the language around the constructs of
interest,
researchers can indicate an even greater attention to conceptual
clarity and an intention to pursue continued research on which
non-
academic variables impact college success and how they do so.
Furthermore, by delineating a categorization scheme that
explicitly
recognizes the variability in non-academic factors, researchers
will
be better positioned to inform interventions both within and
outside
of schools. For by more carefully delineating the different
domains
of non-academic factors, practitioners and educators could be
provided with more explicit recommendations about how

particular
non-academic factors affect college preparation and persistence,
and thereby how we might better prepare our students for their
postsecondary school years.

SUMMER 2007 JOURNAL OF COLLEGE ADMISSION W W W. N A C A C .docx

Recommended

Recommended

More Related Content

Similar to SUMMER 2007 JOURNAL OF COLLEGE ADMISSION W W W. N A C A C .docx

Similar to SUMMER 2007 JOURNAL OF COLLEGE ADMISSION W W W. N A C A C .docx (20)

More from joyjonna282

More from joyjonna282 (20)

Recently uploaded

Recently uploaded (20)

SUMMER 2007 JOURNAL OF COLLEGE ADMISSION W W W. N A C A C .docx