Learning and Individual Differences 36 (2014) 180–193

Contents lists available at ScienceDirect

Learning and Individual Differences

journal homepage: www.elsevier.com/locate/lindif

Short report (original research)

Student personality, classroom environment, and student outcomes:

A person–environment fit analysis
Dagmara K. Pawlowska a, James W. Westerman b,⁎, Shawn M. Bergman a, Timothy J. Huelsman a
a
Department of Psychology, Appalachian State University, Boone, NC 28608, United States
b
Department of Management, Appalachian State University, Boone, NC 28608, United States

a r t i c l e i n f o a b s t r a c t

Article history: This study examines whether congruence between classroom environment and student personality is associated
Received 16 September 2013 with increases in student satisfaction and performance. Data were collected from students (n = 1763) at a com-
Received in revised form 20 August 2014 prehensive state university and analyzed using mixed-level maximum likelihood polynomial regression analysis
Accepted 24 October 2014
and surface response methodology. Results indicate that student personality is a consistent predictor of student
satisfaction, classroom environment is a consistent predictor of performance, and the interplay between the two
Keywords:
Person–environment fit
is important in predicting satisfaction and performance. The personality characteristics of agreeableness and con-
Personality scientiousness and the classroom environment dimension of structure were all positively related to both satisfac-
Classroom environment tion and performance. Congruence between several personality traits and classroom environment elements was
Student satisfaction found to increase satisfaction, performance, or both. The implications of these results for instruction and research
Academic performance on person–environment fit in higher education are discussed.
© 2014 Elsevier Inc. All rights reserved.

1. Introduction
Although enrollment rates have been rising, graduation rates have
been stagnating in higher education in the U.S. (US College Dropout
Rates Spark Concern, 2006). A study performed by American College
Testing (ACT) revealed that one in every four college students drops
out before finishing their sophomore year (Whitbourne, 2010). Re-
search conducted by the National Center for Education Statistics indicat-
ed that slightly over half of first-time college students attending 4-year
institutions full-time completed a bachelor's degree or equivalent at
that institution within 6 years (U.S. Department of Education &
National Center for Education Statistics, 2010). Because there is a vast
amount of variation in individual differences among students in post-
secondary education, we must develop a better understanding of the in-
teraction between individual factors and the environment in the
learning process in order to understand the predictors of academic suc-
cess and improve student outcomes. In 1958, Pace and Stern suggested
that the “congruence between personal needs and environmental press
will be more predictive of (student) achievement, growth and change
than any single aspect of either the person or the environment”
(p. 277). However, relatively little research has been done in higher
⁎ Corresponding author at: Appalachian State University, Department of Management,
ASU Box 32089 Boone, NC 28608-2089, United States. Tel.: +1 828 262 7475; fax: +1 828
265 8685.
E-mail addresses: D.Pawlowska1@gmail.com (D.K. Pawlowska),
westermanjw@appstate.edu (J.W. Westerman), bergmans@appstate.edu (S.M. Bergman),
huelsmantj@appstate.edu (T.J. Huelsman).
education on this subject to date (Joiner, Malone, & Haimes, 2002).
This study addresses this oversight, and examines the impact of per-
son–environment congruence in the higher education classroom,
more specifically the influence of personality and classroom environ-
ment on student satisfaction and performance.
1.1. Person–environment fit
Person–environment (P–E) fit assumes that (a) meaningful and reli-
able differences can be assessed between individuals and (b) between
environments, and (c) considers that matching individuals and environ-
ments will increase the likelihood of positive outcomes (Chartrand,
1991). Since individuals and environments differ in meaningful and
reliable ways, P–E fit theory allows us to recognize important patterns
and utilize them to organize individuals and environments to achieve
optimal outcomes (Swanson & Fouad, 1999), including achievement,
performance, satisfaction, tenure, retention, and stability. And concep-
tualizing congruence as a relationship that can happen along the contin-
uum between individuals and their environments allows different
forms of congruent relationships to be evaluated. This allows for situa-
tions where positive outcomes, such as performance and satisfaction,
are highest (i.e., maximized) at different points along the congruence
continuum (i.e., both the individual and environment have a substantial
amount of complementary or mutually beneficial characteristics) than
at other points of person–environment congruence (Edwards, 1991,
1994). P–E fit research shows promise in benefitting education (Flynn
& Rapoport, 1976; Fraser & Fisher, 1983), but determining what specific

http://dx.doi.org/10.1016/j.lindif.2014.10.005
1041-6080/© 2014 Elsevier Inc. All rights reserved.
 D.K. Pawlowska et al. / Learning and Individual Differences 36 (2014) 180–193
factors will have the most influence in a higher education setting has yet
to be comprehensively examined and discussed.
1.2. Personality as the “person” factor of P–E fit
Personality research has vast potential to provide researchers and
educators information as to why so many students fail to succeed in
the current postsecondary educational system (Woszczynski, Gutherie,
& Shade, 2005). Although seldom used to study P–E fit (Ehrhart,
2006), research that spans five decades has shown that personalities re-
late systematically and predictably to a range of educational outcomes
(Furnham, Christopher, Garwood, & Martin, 2008). For example, person-
ality has been generally related to academic performance (Caspi, Chajut,
Saporta, & Beyth-Marom, 2006), college attrition and dropout rates
(Lounsbury, Saudargas, Gibson, & Leong, 2005), learning styles (Zhang,
2006), student preference for grading or evaluation method (Furnham
et al., 2008), residence hall placement, orientation outcomes, leadership
development, and advising (Lounsbury et al., 2005). Essentially, person-
ality information may be influential in nearly every college situation
where a student has to make a choice concerning commitment, involve-
ment, membership, and/or participation (Lounsbury et al., 2005).
Researchers have generally agreed that five dimensions (commonly re-
ferred to as the Big Five (Barrick & Mount, 1991; Costa & McCrae, 1995;
McCrae & Costa, 1987) can be used to represent the structure of normal
personality. The traits identified in the five factor model are the result
of decades of factor analytic research and their robustness is well-
established (Costa & McCrae, 1994; Zhang, 2006). The big five personal-
ity dimensions are extraversion, neuroticism, openness to experience,
agreeableness, and conscientiousness (Barrick & Mount, 1991)
(Table 1).
1.3. Classroom environment as the “environment” factor of P–E fit
Classroom environments have been argued to carry great influence
in student collegiate learning (Feldman, Smart, & Ethington, 2004;
Westerman & Simmons, 2007). The research foundations examining
the effects of environment on learning include Kirtz and Moos (1974),
who assert that dimensions of environmental stimuli are distinct, have
distinguishing effects on physiological processes, and suggest the use
of environmental measurement to determine which dimensions may
be beneficial (or disadvantageous) to particular groups of people. How-
ever, the empirical research on the effects of classroom environment is
less established. In one of the few such studies, Fry and Addington
(1984) examined the effects of open versus traditional classrooms and
found that participants in the open-classroom exhibited higher achieve-
ment in social problem-solving cognitions as well as higher ego-
strength and self-esteem over participants in the traditional classroom.
Fraser and Fisher (1983) found that when students had a preference for
their environment, they exhibited greater achievement than students
who had a lower preference for their environment. Furthermore,
Nielsen and Moos (1978) examined high exploration classroom envi-
ronments and found that students who preferred high exploration
Table 1
Five factor model factor descriptions.
Dimension Features
Extraversion Ambitious, sociable, gregarious, talkative, assertive,
impetuous, active
Neuroticism Depressed, anxious, angry, emotional, embarrassed,
insecure, worried
Openness to experience Imaginative, curious, cultured, broad-minded, original,
artistically sensitive, intelligent
Agreeableness Flexible, courteous, good-natured, trusting, forgiving,
cooperative, tolerant, soft-hearted
Conscientiousness Dependable, thorough, careful, organized, responsible,
achievement-oriented, hardworking, persevering
181
classroom environments were better adjusted and more satisfied than
students who preferred low exploration classroom environments. The
Classroom Environment Scale (CES) developed by Trickett and Moos
(1973), was developed to more comprehensively measure the dimen-
sions of a classroom environment from a student's perspective. The
CES measures a student's perspective of the structure and focus of the
classroom environment, the emphasis on participative learning, class-
room involvement, and student competition, and the availability of in-
structor support (Table 2).
1.4. Interaction between personality and classroom environment
Although personality and classroom environment are important as
independent factors, there is a gap in higher education research examin-
ing whether their interaction produces meaningful outcomes beyond
those they can produce individually. A study conducted by Chamorro-
Premuzic, Furnham, and Lewis (2007) found that students high in neu-
roticism were found to dislike small groups and group discussions,
while agreeable and open students had a preference for these types of
environments. Student outcomes have also been examined in research
on personality–environment fit in distance/online learning formats,
demonstrating that different personality profiles are successful in online
versus traditional college classrooms (Schniederjans & Kim, 2005;
Williamson & Watson, 2007). While these studies provide limited evi-
dence that a relationship exists between personality and classroom envi-
ronment, a more comprehensive approach is needed. Consistent with
the P–E fit paradigm, it is expected that student outcomes will be highest
(i.e., maximized) when there is congruence or fit between certain per-
sonality traits and certain elements of the classroom environment. We
next introduce the personality dimensions and classroom environment
dimensions utilized by this study, upon which we build our hypotheses.
1.5. Hypotheses
In the following sections, we review findings that suggest possible
personality–classroom environment combinations for which fit may
be associated with course outcomes.
1.5.1. Extraversion
Extraversion is associated with ambition, sociability, gregariousness,
talkativeness, assertiveness, impulsivity, and vigor. Because extraverts
crave interpersonal interaction, it has been suggested that they seek en-
vironments with a high degree of relationship orientation (Westerman
& Simmons, 2007) and affiliation (Buunk, Nauta, & Molleman, 2005)
and also have a preference for working in groups (Furnham &
Chamorro-Premuzic, 2005). Thus extraverts may respond positively to
environments with high participative learning. Further, in educational
settings, extraversion has been found to be positively related to appro-
priate assessment, clear goals, and good teaching (Nijhuis, Segers, &
Gijselaers, 2007). This indicates that extraversion may be positively as-
sociated with environments providing high structure and focus as well
as instructor support and suggests that a student with high levels of
Table 2
Classroom environment factor descriptions.
Dimension Description
Structure and focus The extent to which the classroom is organized and
course material is emphasized
Participative learning The degree to which learning is a social activity in the
classroom.
Classroom involvement The degree to which students show interest in and
provide input concerning the activities in the classroom
Instructor support The extent to which the instructor takes an interest in
the students and material unrelated to the course
Student competition The level of emphasis placed on academic competition
between students within the classroom
182 D.K. Pawlowska et al. / Learning and Individual Differences 36 (2014) 180–193
extraversion will respond positively to a highly structured and very sup-
portive instructor. These expectations are consistent with the P–E fit
paradigm, which suggests that individual outcomes will be their highest
(i.e., maximized) when congruence between the person and the envi-
ronment occurs.
Hypothesis 1a. (a): Course satisfaction and performance will be maxi-
mized when extraversion and structure and focus are each high.
Hypothesis 1b. (b): When extraversion and participative learning
are each high, course satisfaction and course performance will be
maximized.
Hypothesis 1c. (c): When extraversion and instructor support are each
high, course satisfaction and course performance will be maximized.
Westerman and Simmons (2007) found that extraversion is a signif-
icant predictor of preference for goal orientation environments. These
environments provide opportunities for challenge, success, accomplish-
ment, and maintaining task orientation. Given that the dimension of ex-
traversion includes facets such as ambition, assertiveness, and vigor it is
clear to see why this link was found. This research suggests a positive re-
lationship between extraverts and environments with high levels of
student competition, as competitive classroom environments likely
provide an opportunity for individual assertiveness and ambition.
Hypothesis 1d. (d): When extraversion and student competition
are each high, course satisfaction and course performance will be
maximized.
1.5.2. Neuroticism
The personality dimension of neuroticism is an associated character-
istic of depression, anxiety, anger, emotion, embarrassment, insecurity,
and worry. Students high in neuroticism have demonstrated higher
level of achievement in structured classrooms (Nielsen & Moos, 1978),
indicating that they may associate positively with an environment
high in structure and focus. This positive association will be associated
with individual outcomes being their highest (i.e., maximized) when
this type of congruence occurs.
Hypothesis 2a. (a): When neuroticism and structure and focus
are each high, course satisfaction and course performance will be
maximized.
Neuroticism has also been found to be negatively related to group dis-
cussion and small group participation (Chamorro-Premuzic et al., 2007),
and affiliation (Buunk et al., 2005). In educational settings, students high
in neuroticism perform better in classrooms that involve less participa-
tion (Nielsen & Moos, 1978; Williams, 1971). This research suggests
that students high in neuroticism may respond negatively to environ-
ments with high participative learning and classroom involvement.
Hypothesis 2b. (b): When neuroticism is high and participative learn-
ing is low, course satisfaction and course performance will be
maximized.
Hypothesis 2c. (c): When neuroticism is high and classroom involve-
ment is low, course satisfaction and course performance will be
maximized.
1.5.3. Openness to experience
Openness to experience is associated with imagination, curiosity,
culture, broad-mindedness, originality, artistic sensitivity, and intelli-
gence. Openness to experience may have an inverse relationship with
organization, order, and control (Westerman & Simmons, 2007), and
as a result may be negatively related to the environmental dimension
of structure and focus.
Hypothesis 3a. (a): When openness to experience is high and structure
and focus is low, course satisfaction and course performance will be
maximized.
Openness to experience has also been suggested to be positively re-
lated to relationship-oriented environments that provide high levels of
social interaction. Williams (1971) found that individuals who partici-
pated actively in the classroom were shown to have higher levels of cre-
ativity and originality of thought — characteristics of openness to
experience. This suggests that openness to experience may be positively
related to environments that provide high participative learning and
classroom involvement, and that student outcomes will be their highest
when this congruence occurs.
Hypothesis 3b. (b): When openness to experience and participative
learning are each high, course satisfaction and course performance
will be maximized.
Hypothesis 3c. (c): When openness to experience and classroom in-
volvement are each high, course satisfaction and course performance
will be maximized.
Openness to experience has also been negatively related to social
comparison (Buunk et al., 2005), suggesting that students high in open-
ness to experience will not respond well to environments high in stu-
dent competition. While this type of congruence is different than the
previously discussed hypotheses, it is consistent with the P–E fit para-
digm and suggests that student outcomes will be highest when this
type of congruence occurs.
Hypothesis 3d. (d): When openness to experience is high and student
competition is low, course satisfaction and course performance will be
maximized.
1.5.4. Agreeableness
The personality dimension of agreeableness is associated with flexi-
bility, courteousness, good nature, trust, forgiveness, cooperation, toler-
ance, and soft-heartedness. Agreeableness has been found to have a
positive relationship with system maintenance (which refers to organi-
zation, order, and clear expectations). This suggests that students high
in agreeableness may respond positively to classroom environments
high in structure and focus.
Hypothesis 4a. (a): When agreeableness and structure and focus
are each high, course satisfaction and course performance will be
maximized.
Agreeableness has also been found to be related positively with
affiliation (Buunk et al., 2005) and with environments that reinforce
relationships (Westerman & Simmons, 2007). More specifically,
agreeable students prefer group discussion and participation in
small groups (Chamorro-Premuzic et al., 2007; Furnham &
Chamorro-Premuzic, 2005). This suggests that students high in
agreeableness may respond well to environments that offer high
participative learning.
Hypothesis 4b. (b): When agreeableness and participative learning
are each high, course satisfaction and course performance will be
maximized.
1.5.5. Conscientiousness
The personality dimension of conscientiousness is associated
with dependability, thoroughness, carefulness, organization, responsibility,
 D.K. Pawlowska et al. / Learning and Individual Differences 36 (2014) 180–193 183

achievement-orientation, hard work, and perseverance. Conscientious- Table 3

ness has been related to being orderly, goal-oriented, having a need for Initially hypothesized structure of the Classroom Environment Scale.

continuous improvement (Westerman & Simmons, 2007), and having a Item Reason for being
preference for clear goals (Nijhuis et al., 2007). Chamorro-Premuzic removed
et al. (2007) found that conscientious students are achievement- Structure and focus
oriented, ambitious, organized, and hard-working. These findings sug- 1. Almost all class time is spent on discussing the
gest that conscientious students respond more effectively to environ- course material
2. Students spend more time discussing outside Step 1: failed to load
ments with high structure and focus and student competition.
activities than course material (R) at .35
3. This instructor takes class time to talk about other
Hypothesis 5a. (a): When conscientiousness and structure and focus things (R)
are each high, course satisfaction and course performance will be 4. The instructor sticks to course material and does
maximized. not get sidetracked
5. There is a clear set of rules for students to follow
6. There are very few rules to follow (R) Step 1: cross-loaded
Hypothesis 5b. (b): When conscientiousness and student competition 7. The instructor is very strict Step 1: cross-loaded
are each high, course satisfaction and course performance will be 8. The teacher makes a point of sticking to the rules Step 1: cross-loaded
maximized. he-she has made
9. This is a well-organized class
10. Assignments are usually clear so everyone knows
what to do
2. Material and methods 11. This class hardly ever starts on time (R) Step 1: cross-loaded
12. Activities and lectures in this course are clearly
and carefully planned
2.1. Participants and procedures
Instructor support
13. This instructor often spends time just talking with
Participants were 1763 undergraduate students enrolled in 41
students
separate introductory psychology courses over three consecutive 14. The instructor takes a personal interest in students
semesters. Participants were mostly White (89.4%; Black = 4.8%; 15. The instructor goes out of his-her way to help Step 3: cross-loaded
Hispanic = 7.4%), female (66.7%), and freshmen (69.4%; sophomores = students
19.8%; juniors; 7.0%; seniors; 3.7%), with average age of 18.71 years. 16. This instructor wants to know what the students Step 1: failed to load
want to learn about at .35
Students were assessed at three different times. Within the first
2 weeks of the semester students completed the personality measure. Participative learning
At the interim measurement (weeks 10 through 12), students complet- 17. New and different ways of teaching are tried very
often in this course
ed the classroom environment measure. At the final measurement oc- 18. The instructor likes students to try unusual Step 3: cross-loaded
curring within the last 2 weeks of the class, students completed the projects
course satisfaction assessment. Final grades were collected from instruc- 19. Students have very little input on how class time is
tors to assess student performance. All assessments were completed spent (R)
20. Students do the same kind of work almost every
outside of class using an online survey.
day in this class (R)

2.2. Measures
2.2.1. Personality
The Big Five personality dimensions were assessed using the
NEO-FFI Form S (McCrae & Costa, 2004), which is a 60-item inventory
where each item is rated on a 5-point scale ranging from “strongly dis-
agree” to “strongly agree.” The validity of the five factors assessed by the
NEO-FFI has been specified by a number of studies (Costa & McCrae,
1995; McCrae & Costa, 1987).
2.2.2. Classroom environment
The classroom environment was assessed using an adapted version
of the Classroom Environment Scale (CES) developed by Trickett and
Moos (1973) and refined by Deemer (2004) and Trickett and Quinlan
(1979). The CES is utilized to assess classroom environment in second-
ary school settings using a dichotomously scored (i.e., yes–no) measure-
ment approach. Thirty-two items from the CES applicable to the
collegiate classroom were selected by subject matter experts (see
Table 3). To ensure the measurement of the classroom environment
was on the same 5-point metric as the personality measure, student
subjects rated the extent to which their classroom represented the
item description on a 5-point scale ranging from “not at all like my class-
room” to “exactly like my classroom.”
2.2.3. Student performance
Student performance was assessed using each student's final grade
for the course in percentage form, with student consent.
Classroom involvement
21. Students in this class get to know each other really
well
22. Students enjoy working together on projects in
this class
23. Students enjoy helping each other with homework
24. Students don't have much of a chance to get to
know each other in this class (R)
25. Students are often clock-watching in class (R)
26. Most students in this class really pay attention to Step 1: failed to load
what the instructor is saying at .35
27. Students frequently take part in class discussions Step 1: failed to load
or activities at .35
28. Students sometimes present something they've
worked on to the class
Student competition
29. Students try hard to get the best grade
30. Grades are not very important in this class (R) Step 1: failed to load
at .35
31. Students have to work for a good grade in this
class
32. Students usually pass even if they do not do
much (R)
Note: (R) indicates that an item was reserve coded.
2.2.4. Student satisfaction
Student satisfaction was assessed using four questions from the Indi-
vidual Development and Educational Assessment (IDEA) Center course
evaluation form (Hoyt & Cashin, 1977; Hoyt & Lee, 2002). Items
consisted of “Overall, I rate this course as excellent,” “I am really excited
about this class,” “I think what we are studying in this class will be useful
184 D.K. Pawlowska et al. / Learning and Individual Differences 36 (2014) 180–193
to know,” and “I think what we are studying in this class will be impor-
tant to know.” Each item was rated on a 5-point scale ranging from
“strongly disagree” to “strongly agree.”
3. Analytic plan
3.1. Factor analyses of the Classroom Environment Scale
Given that student subjects were nested within classrooms, our data
violated the assumption of independence. Thus, we employed Muthén's
(1994) five-step procedure to examine factor structure of the classroom
environment and the multilevel structure of data. The first step used
conventional factor analytic techniques that ignored the nested struc-
ture of the data. Once a factor structure that fit the total sample covari-
ance matrix was found, the next step examined intraclass correlation
coefficients (ICC) and design effects to determine whether multilevel
techniques were needed. ICC values between .05 and .20 are common
in cross-sectional multilevel research, and design effects greater than
2.0 suggest that multilevel analyses are appropriate (cf., Muthén,
1994). Given that multilevel techniques were needed, analyses using a
weighted least-squares estimator with mean and variance adjustment
(WLSMV), a recommended method when conducting multilevel ex-
ploratory factor analyses due its computation efficiency (see Hox &
Roberts, 2011), were used in the subsequent steps. The third and fourth
steps conducted factor analyses on the sample pooled within-group
(i.e., student-level) and between-group (i.e., class-level) covariance ma-
trices, respectively.1 Once a factor structure that fit these covariance ma-
trices was found, the within- and between-group factor structures were
simultaneously modeled.
3.2. Data aggregation and descriptive statistics
rwg(j) (James, Demaree, & Wolf, 1984) calculations were examined
for the classroom environment subscales to determine if aggregation
is warranted. rwg(j) is an agreement index that compares observed vari-
ance in ratings to a uniform distribution and is preferred in situations
where multiple raters use only part of a response scale when assessing
a target or situation. This type of response set reduces variance and ar-
tificially lowers the ICC and other agreement indices despite the fact
that the raters are, in fact, in agreement (i.e., using only ratings of
three and four on a 5-point scale). The classroom environment scales
were aggregated to the classroom level based on the average agreement
indices.
3.3. Examining congruence relationships
ICC and design effects from unconditional means regression models
predicting student performance and satisfaction (i.e., models in which
the intercepts were allowed to vary across classrooms) were examined
to determine if multilevel modeling was appropriate when testing the
study's hypotheses. Personality scores were then class-mean centered
and aggregated classroom environment scores were grand-mean cen-
tered prior to being entered as level-1 and level-2 predictors (Peugh,
2010) into random intercept multilevel regression models (MLM).
To examine if and how congruence between student personality and
classroom environment were related to grades and satisfaction, the
main effects of student personality were first examined by regressing
the outcomes onto the student personality scores, Z = (b0 + μ0j) +
b1Xij, in five separate MLMs. These main effect personality MLMs deter-
mined if the student extraversion, neuroticism, openness to experience,
agreeableness, and conscientiousness were each independent predictor
1
The pooled within-group matrix used in the third step was corrected for between-
group differences by subtracting respective class means from student ratings within that
class. The between-group matrix was corrected by subtracting the grand mean from each
respective class means and examining the relationship among the random intercepts.
of performance or satisfaction before being considered alongside
classroom environment. In these personality main effect models, Z rep-
resented either performance or satisfaction, X represented the personal-
ity score of student i in j classroom, and μ0j represented the class' specific
deviation from the grand mean, b0. The main effects for classroom envi-
ronment were examined. These main effect MLMs determined if the
classroom environment factors identified in the factor analysis predict-
ed performance and satisfaction independent of student personality.
Performance and satisfaction were regressed onto aggregated class-
room environment scores, Z = (b0 + μ0j) + b1Yj, in five separate
MLMs, one model for each classroom environment sub-factor. In these
classroom environment main effect models, Y represents aggregated
classroom environment score of classroom j. The overall statistical sig-
nificance of these personality and classroom main effect models was de-
termined by change in −2 log likelihood (Δ−2LL) from the respective
unconditional model.
Next, polynomial regression models examined how personality –
classroom congruence related to course performance and satisfaction
(Edwards, 1994; Edwards & Parry, 1993). These models did not collapse
individual and environmental variables into a single index, but it
examined the joint effect that congruence between student personality
and classroom environment had on satisfaction and performance,
Z = (b0 + μ0j) + b1Xij + b2Yj + b3Xij2 + b4XijYj + b5Yj2. To test the
study's hypotheses, these congruence models were compared to the
main effect models by examining the Δ−2LL to determine which out-
come was best predicted by the congruence between personality and
the classroom environment (and not by just student personality or
classroom environment alone). If the congruence model was found to
be the best predictor, a surface response methodology (Edwards &
Parry, 1993; Kutner, Nachtsheim, Neter, & Li, 2005) was employed to
display the relationship between personality, classroom environment,
and an outcome in three-dimensional space. This surface response
graphic (SRG) allows us to determine what combination of student per-
sonality and classroom environment results in the highest levels of
course satisfaction and performance.
4. Results
4.1. Factor analyses of the Classroom Environment Scale
We first conducted the initial conventional confirmatory factory
analysis using all 32 CES items. It showed the data were a poor fit to
the initially hypothesized five-factor model, χ2(454) = 3716.11,
p b .001, Comparative Fit Index (CFI) = .58, root mean square error of
approximation (RMSEA) = .09, and standardized root mean square re-
sidual (SRMR) = .12. An examination of model's factor loadings and
modification indices revealed a number of non-significant and cross-
loading items.
Given that this was the first time this modified version of the CES has
been used to assess the collegiate classroom environment, the unknown
latent structure of the items, the number of non-significant and cross-
loading items from the CFA, a series of exploratory analyses were con-
ducted to determine the factor structure that best fit the CES items. Spe-
cifically, a series of conventional principal axis factoring (PAF) analyses
with an oblimin rotation were employed with items failing to load at or
loading higher than .35 on multiple factors being removed. This iterative
procedure was supplemented with a parallel analysis (Horn, 1965) that
generated 5000 random datasets and reported the 95th percentile of
the eigenvalues generated (Glorfeld, 1995; Hayton, Allen, & Scarpello,
2004) that helped guide the number of factors to be retained. Table 3
shows which items were removed. The final PAF solution resulted in a
five factor solution, which accounted for 45.5% of the variance.2
2
Exploratory factor analyses were also using an WLSMV estimator with the final step-
one model being similar to the PAF step-one model and with the final step-five multilevel
factor model being identical.
 D.K. Pawlowska et al. / Learning and Individual Differences 36 (2014) 180–193 185

Table 4
Multilevel exploratory factor analysis results of the Classroom Environment Scale.

Factor loadings

S&F PL IS CI SC

Item 1 2 3 4 5

9. This is a well-organized class .69 −.01 .11 .12 .14

12. Activities and lectures in this course are clearly and carefully planned .69 −.06 .11 .02 .12
4. The instructor sticks to course material and does not get sidetracked .69 .13 −.26 −.08 −.07
10. Assignments are usually clear so everyone knows what to do .56 −.05 .11 .07 .05
5. There is a clear set of rules for students to follow .45 .26 .04 −.07 −.08
1. Almost all class time is spent on discussing the course material .54 −.11 −.07 −.21 .01
22. Students enjoy working together on projects in this class −.03 .77 −.08 .04 .05
23. Students enjoy helping each other with homework .09 .63 .05 −.01 −.01
28. Students sometimes present something they've worked on to the class −.09 .62 .05 −.11 −.04
21. Students in this class get to know each other really well .05 .57 .14 .27 −.03
17. New and different ways of teaching are tried very often in this course .15 .40 .31 .07 .01
13. This instructor often spends time just talking with students .04 −.01 .63 .12 −.05
3. This instructor takes class time to talk about other things (R) −.20 .19 .52 −.13 −.02
14. The instructor takes a personal interest in students .21 .10 .43 .14 .14
24. Students don't have much of a chance to get to know each other in this class (R) .02 .04 .05 .81 −.05
20. Students do the same kind of work almost every day in this class (R) −.16 .20 −.03 .50 .05
19. Students have very little input on how class time is spent (R) −.10 .09 .08 .46 .05
25. Students are often clock-watching in class (R) .25 −.13 −.11 .47 .12
31. Students have to work for a good grade in this class .18 .08 .10 −.19 .64
32. Students usually pass even if they do not do much (R) −.09 −.09 −.21 .19 .61
29. Students try hard to get the best grade .16 .22 .20 .02 .35

Note: Multilevel exploratory factor analysis using a weighted least squares estimator with mean and variance adjustment with an oblimin rotation was employed. (R) indicates that an
item was reserve coded. Loadings presented above are from the within-group pattern matrix. PL = participative learning; S&F = structure and focus; CI = classroom involvement;
IS = instructor support; and SC = student competition.
Bold data indicate the items that load at .35 or higher on a specific factor and the items that were used to compute the five factor scores.

The ICCs and design effects of the 23 retained items from step 1
ranged from .02 to .20 with the design effects ranging from 1.33 to
4.74. This suggested that sufficient between-class variation and multi-
level exploratory analyses using an WLSMV (see Hox & Roberts, 2011)
used the third, fourth, and fifth steps. An iterative procedure where fit
indices and results from a parallel analyses were used to determine
the number of factors to be retained with items failing to load at or load-
ing higher than .35 on multiple factors being removed.
Result from the third step revealed that two additional items should
be removed due to cross-loading (see Table 3) and that a five factor solu-
tion best fit the corrected within-group matrix.3 Results from the fourth
step revealed that the saturated or unrestricted provided the best fit for
the corrected between-group matrix. The fact that no underlying factor
structure emerged between the class means (i.e., between-group covari-
ance matrix) was somewhat expected given the relatively small number
of between-level units (i.e., 41 classes) compared to the number of items
(i.e., 21) and unique covariance elements (i.e., 210) being examined. Re-
sult of the fifth step revealed that a five-factor within-class and unre-
stricted between-class model provided the best fit to the data,
χ2(115) = 140.70, p = .052, CFI = .97, RMSEA = .02, SRMRWithin =
.02, and reflected the initially hypothesized model (see Table 4).
4.2. Data aggregation and descriptive statistics
rwg(j) (James et al., 1984) calculations were examined for the
classroom environment subscales of structure and focus, participative
learning, classroom involvement, instructor support, and student
competition. These calculations provided average agreement indices
of .90, .72, .75, .75, and .85 respectively. Thus, classroom environment
data were aggregated to the classroom-level. Table 5 presents the
means, standard deviations, correlations, and reliability estimates of
the research variables.
3
The full results from steps 3 and 4 were not presented because they were (and typical-
ly are) very similar in value to the estimates of the final multilevel model.
4.3. Need for multilevel modeling and main effect models
The unconditional means models predicting student performance
and satisfaction indicated that multilevel modeling was warranted
(performance ICC = .03, design effect = 2.27; satisfaction ICC = .04,
design effect = 2.74). Thus, the hypotheses were tested using random
intercepts MLM.
4.3.1. Personality main effects
Results from the separate student personality main effect
MLMs indicated that extraversion, openness to experience, agree-
ableness, and conscientiousness were all statistically significant
independent predictors of course satisfaction (see Table 6).
These four personality dimensions had positive relationships
with course satisfaction. Agreeableness and conscientiousness
were found to be statistically significant independent predictors
of course performance (see Table 6). Both agreeableness and
conscientiousness were found to have positive relationships with
course performance.
4.3.2. Classroom environment main effects
Results from the separate classroom environment main effect MLMs
indicated that only structure and focus and student competition were
statistically significant independent predictors of course satisfaction
(see Table 6). These two classroom dimensions had positive relation-
ships with course satisfaction. Structure and focus, participative learning,
and instructor support were also found to be statistically significant in-
dependent predictors of course performance (see Table 6). These three
classroom dimensions also had positive relationships with course
performance.
4.4. Hypothesis testing
4.4.1. Hypothesis 1
Hypothesis 1 proposed that congruence between extraversion
and specific classroom dimensions would lead to higher course sat-
isfaction and grades. In predicting course satisfaction, the structure
and focus of the congruence model was the only MLM found that fit
186 D.K. Pawlowska et al. / Learning and Individual Differences 36 (2014) 180–193

Table 5
Correlations among study variables.

Variable 1 2 3 4 5 6 7 8 9 10 11 12 13 14

1. Age –
2. Gender −.06 –
3. Extraversion −.08 .07 (.71)
4. Neuroticism −.08 .21 −.32 (.85)
5. Openness .03 .01 −.04 .02 (.76)
6. Agreeableness −.01 .18 .32 −.20 .03 (.75)
7. Conscientiousness .06 .14 .18 −.22 −.18 .23 (.84)
8. CES—S&F .04 .02 −.04 .01 −.01 −.01 .06 (.77)
9. CES—PL −.06 −.01 .02 −.02 .02 .01 .03 .21 (.78)
10. CE—CI −.03 −.03 .03 −.03 .00 .01 −.04 .06 .55 (.67)
11. CE—IS −.03 −.04 .03 −.02 .05 .03 .03 .13 .51 .31 (.56)
12. CE—SC −.01 .04 .01 .02 .00 .02 .00 .26 −.13 −.09 −.04 (.56)
13. Satisfaction .00 .10 .08 .00 .10 .15 .10 .11 .03 .00 .05 .10 (.86)
14. Performance −.02 .09 .03 .02 .03 .14 .18 .11 .10 .09 .08 −.05 .19 –
Mean 18.71 1.67 2.52 1.84 2.37 2.61 2.56 3.73 2.31 3.21 2.91 3.83 4.02 85.42
Standard deviation 1.74 0.47 0.45 0.67 0.54 0.49 0.56 0.22 0.35 0.33 0.27 0.23 0.67 10.76

Note. Values on main diagonal (in parentheses) represent Cronbach's coefficient alpha. Gender is coded 0 = male, 1 = female; CES denotes Classroom Environment Scale;
PL = participative learning; S&F = structure and focus; CI = classroom involvement; IS = instructor support; and SC = student competition. Final age, gender, and performance
were assessed using single items; therefore its reliability estimate is not available.
Correlations greater than .05 in absolute magnitude are significant at p b .05, correlations greater than .07 in absolute magnitude are significant at p b .01, and correlations greater than .09
in absolute magnitude are significant at p b .001.

the data better than both the personality and classroom main effect
models (see Table 7). Examining the SRG for the extraversion–
structure and focus course satisfaction congruence model (Fig. 1)
indicated that when extraversion and structure and focus were
both high (i.e., the area in circle A), course satisfaction was maxi-
mized. However, congruence with participative learning, instructor
support, and student competition failed to predict course satisfac-
tion better than the main effect model for extraversion (Tables 6
and 7).
In predicting performance, all four congruence models were found
to fit the data better than the personality and classroom environment
main effect models (Table 7). However, the SRG revealed that a curvilin-
ear relationship between extraversion and course performance existed
in all four cases, and that the congruence between extraversion and
the classroom dimensions did not maximize performance. A summary
of which models were determined to best fit the data for each individual
hypothesis is presented in Table 8. These results provide partial support
for Hypothesis 1a, but failed to support Hypothesis 1b, Hypothesis 1c,
and Hypothesis 1d.
4.4.2. Hypothesis 2
Hypothesis 2 proposed that the congruence between neuroticism
and specific classroom dimensions would lead to higher course
Table 6
Main effect multilevel model results for personality and classroom dimensions.
Course satisfaction Course performance
Model Δ−2LL Model Δ−2LL
satisfaction and grades. None of the congruence models provided
best fit for either satisfaction or performance (Table 7). As summa-
rized in Table 8, the structure and focus was the only statistically sig-
nificant predictor of course satisfaction, while the main effect models
of structure and focus and participant learning were the only statis-
tically significant models for course performance. Hypothesis 2 was
not supported.
4.4.3. Hypothesis 3
Hypothesis 3 proposed that the congruence between openness to
experience and specific classroom dimensions would lead to higher sat-
isfaction and grades. The congruence model that included structure and
focus to predict satisfaction was the only congruence model to provide
best fit (Table 7). Hypothesis 3a proposed that when openness to expe-
rience was high and structure and focus was low, satisfaction would be
maximized. The SRG revealed that satisfaction was high when openness
to experience was high and structure and focus were low (circle A), but
was equally high at two other points (Fig. 2). Interestingly, both main
effect models for openness to experience and student competition
were significant predictors of course satisfaction, but the congruence
model failed to significantly predict better than the personality main ef-
fects model. As seen in Fig. 2, course satisfaction was maximized when
openness to experience and student competition were high (circle B).
As summarized in Table 8, the main effect model for openness to expe-
rience was a positive predictor of course satisfaction, while the structure
and focus and participant learning main effect models were statistically
significant predictors of course performance. Hypothesis 3 was not
supported.

Personality dimension 4.4.4. Hypothesis 4

Extraversion 8.81** 2.31 Hypothesis 4 proposed that the congruence between agreeableness
Neuroticism 0.04 0.42 and specific classroom dimensions would lead to higher satisfaction and
Openness 11.71*** 0.77 grades. The both agreeableness–structure and focus congruence models
Agreeableness 28.45*** 32.23***
Conscientiousness 12.65*** 55.73***
were found to provide the best fit (Table 7). The SRG (Fig. 3) indicated
that when agreeableness and structure and focus were both high, satis-
Classroom Dimension faction (circle A) and performance (circle B) were highest. While the
Structure and focus 8.57** 6.70** agreeableness–participant learning congruence model was also found
Participative learning 0.80 5.74*
to be the best fit, a curvilinear relationship between agreeableness and
Course involvement 0.05 3.63
Instructor support 1.85 5.27* satisfaction was revealed and not a congruence relationship.
Student competition 6.88** 1.05 Interestingly, both main effect models for agreeableness and partic-
Note. Model Δ− 2LL change in − 2 log likelihood from the respective unconditional
ipant learning were significant predictors of course performance but the
models; p-values computed from a chi square distribution with one degree of freedom. congruence model failed to significantly predict better than the person-
*p b .05, **p b .01, ***p b .001. ality main effect model. As seen in Fig. 3, course performance was
 D.K. Pawlowska et al. / Learning and Individual Differences 36 (2014) 180–193 187

Table 7
Congruence multilevel models results for course satisfaction and performance.

Course satisfaction Course performance

Hypothesis Personality Classroom Congruence Congruence vs. Congruence vs. Congruence Congruence vs. Congruence vs.
dimension dimension model personality classroom model personality classroom
Δ−2LL Δ−2LL Δ−2LL Δ−2LL Δ−2LL Δ−2LL

Hypothesis 1a Extra S&F 19.60*** 10.79* 11.03* 38.40*** 35.20*** 31.69***

Hypothesis 1b Extra PL 12.48* 3.67 11.68* 32.12*** 29.81*** 26.38***
Hypothesis 1c Extra IS 12.36* 3.55 10.51* 31.84*** 29.54*** 26.58***
Hypothesis 1d Extra SC 17.56** 8.75 10.68* 29.02*** 26.71*** 27.97***
Hypothesis 2a Neuroticism S&F 14.72* 14.69* 6.15 15.21** 14.80** 8.51
Hypothesis 2b Neuroticism PL 6.60 6.60 5.79 8.82 8.40 3.08
Hypothesis 2c Neuroticism CI 5.24 5.24 5.19 6.70 6.28 3.07
Hypothesis 3a Openness S&F 23.49*** 11.79* 14.92** 16.13** 15.36** 9.42
Hypothesis 3b Openness PL 14.75** 3.04 13.94** 9.01 8.24 3.27
Hypothesis 3c Openness CI 12.59* 0.88 12.54* 8.13 7.36 4.50
Hypothesis 3d Openness SC 19.75*** 8.04 12.87* 7.68 6.92 6.64
Hypothesis 4a Agreeable SF 53.41*** 24.93*** 44.84*** 44.23*** 12.00* 37.52***
Hypothesis 4b Agreeable PL 43.14*** 14.69** 42.33*** 38.52*** 6.29 32.78***
Hypothesis 5a Conscientious SF 23.98*** 11.33* 15.41** 71.62*** 15.89** 64.91***
Hypothesis 5b Conscientious SC 22.53*** 9.88* 15.65** 63.44*** 7.71 62.39***

Note. Model Δ−2LL for the congruence model is the change in −2 log likelihood from the respective unconditional models with p-values computed from a chi square distribution with five
degrees of freedom.
Model Δ−2LL for the congruence vs. personality and congruence vs. classroom are the change in −2 log likelihood respective personality and classroom main effect model from the con-
gruence models with p-values computed from a chi square distribution with four degrees of freedom.
Classroom Environment Scale; PL = participative learning; S&F = structure and focus; CI = classroom involvement; IS = instructor support; and SC = student competition.
*p b .05, **p b .01, ***p b .001.

maximized when agreeableness and participant learning were high
(circle C). Results supported Hypothesis 4a and partially supported
Hypothesis 4b.
4.4.5. Hypothesis 5
Hypothesis 5 proposed that the congruence between conscientious-
ness and specific classroom dimensions would lead to improved student
outcomes. The conscientiousness–structure and focus congruence
models provided best fit for both performance and satisfaction
(Table 7). The SRG (Fig. 4) revealed that when conscientiousness and
structure and focus were both high, course satisfaction (circle A) and
performance (circle B) were highest. The conscientiousness–student
competition congruence model for course satisfaction also provided
the best fit. As seen in Fig. 4, course satisfaction was maximized when
conscientiousness and student competition were both high (circle C).
Results supported Hypothesis 5a and partially supported Hypothesis 5b.
5. Discussion
The results of the study reveal an interesting pattern of relationships.
Personality dimensions were the predominant predictors of perfor-
mance in this study, lending support to the view that personality traits
have remarkable impacts on academic achievement. Interestingly,
only those personality dimensions framed in a positive direction were
related to course performance suggesting that these “positive” traits
have an important relation to success in the courses in this sample.
Structure and focus was the primary classroom environment dimen-
sions associated with increases in student satisfaction and performance,
while participative learning had a positive relationship with perfor-
mance. Some personality and classroom environment combinations
were also found to increase satisfaction, performance, or both.
5.1. Main effects
When considering the performance outcome, the results models
indicated that personality played a strong role in the prediction
of student classroom performance. Specifically, agreeableness and
conscientiousness had the expected positive linear relationship with
performance (Trapmann, Hell, Hirn, & Schuler, 2007). However,
extraversion's relationship with student performance was illuminated
by the surface response graphics which indicated a curvilinear relation-
ship. This suggests that a modest amount of student extraversion is best
for performance, but beyond that, student performance starts to decline.
Further, the environmental dimensions of structure and focus and
participative learning consistently predicted student classroom perfor-
mance, and may be disproportionately critical to academic success.
Providing structure and focus in a classroom provides clarity and pre-
dictability, and encourages students to concentrate on performance.
For example, classrooms rated high on structure and focus in this
study were generally well-organized, provided clearly and carefully
planned lectures, activities, and assignments, provided a clear set of
rules, and had an instructor that focused on the course material. Partic-
ipative learning also was consistently related to student performance,
indicating the importance of social activity and student engagement in
the classroom learning process. This supports contentions that partici-
pative pedagogical approaches enhance learning, perhaps through
stimulating creativity, originality in thought and enhanced problem-
solving and self-esteem (Fry & Addington, 1984; Nielsen & Moos,
1978). Finally, instructor support also provided significant improve-
ments to student performance, as taking an interest in students seemed
to enhance their performance.
In contrast to the more balanced personality and classroom environ-
ment results for student performance, student satisfaction was
disproportionately predicted by personality. In fact, four of the five per-
sonality characteristics (extraversion, openness to experience, agree-
ableness, and conscientiousness) predicted course satisfaction with
only neuroticism failing to be a significant predictor. These findings con-
tinue to suggest that personality plays an important role in academics
both in terms of what students learn and how they react to their educa-
tional experience.
However, the classroom environment dimensions of structure and
focus and student competition also had significant impacts on satisfac-
tion. These findings may have important implications for teaching
assessments. Since teaching assessments often rely on attitudinal vari-
ables (including student satisfaction) and attitudes may be more
strongly influenced by environment, these results suggest that provid-
ing satisfying classroom environments may be more beneficial for
instructors than higher grade averages in enhancing teaching evalua-
tion outcomes. The environmental dimension of structure and focus
was the only environmental dimension that transcended both the
188 D.K. Pawlowska et al. / Learning and Individual Differences 36 (2014) 180–193

Couse Satisfaction Course Performance

Hypothesis 1a

4.67
4.67
4.17 4.17

1.33 1.33
3.13 3.13

Hypothesis 1b

4.67
3.27

1.33 1.73

Hypothesis 1c

4.67
3.52

1.33
2.20

Hypothesis 1d

4.67
4.29

1.33 3.07

Fig. 1. Surface Response Graphics of the Congruence Models for Hypothesis 1.

satisfaction and performance outcomes, demonstrating the importance enhanced student attitudes, yet it did not enhance student performance.
of structure and focus in higher education courses. Student competition However, participative learning and instructor support as forms of
also provided a potentially illuminating pattern of results. Classroom engagement seemed to enhance performance, but not attitudes.
emphasis on competition between students was positively related to This demonstrates that (with the exception of structure and focus)
 D.K. Pawlowska et al. / Learning and Individual Differences 36 (2014) 180–193 189

Table 8
Summary of final multilevel models for course satisfaction and performance.

Course satisfaction Course performance

Hypothesis Personality Classroom Final model Nature of relationship Pseudo-R Final model Nature of relationship Pseudo-R
dimension dimension

Hypothesis 1a Extra S&F Congruence Positive fit .20 Congruence Extraversion: curvilinear .25
relationship
Hypothesis 1b Extra PL Main Effect- Personality Extraversion: .19 Congruence Extraversion: curvilinear .24
positive slope relationship
Hypothesis1c Extra IS Main Effect- Personality Extraversion: .19 Congruence Extraversion: curvilinear .25
positive slope relationship
Hypothesis1d Extra SC Main Effect- Personality Extraversion: .19 Congruence Extraversion: curvilinear .25
positive slope relationship
Hypothesis 2a Neuroticism S&F Main Effect- Environment S&F: positive slope .17 Main Effect- Environment S&F: positive slope .21
Hypothesis 2b Neuroticism PL None Main Effect- Environment PL: positive slope
Hypothesis 2c Neuroticism CI None None
Hypothesis 3a Openness S&F Congruence Openness and S&F: .20 Main Effect- Environment S&F: positive slope .22
both positive slopes
Hypothesis 3b Openness PL Main Effect- Personality Openness: positive .20 Main Effect- Environment PL: positive slope .21
slope
Hypothesis 3c Openness CI Main Effect- Personality Openness: positive .20 None
slope
Hypothesis 3d Openness SC Main Effect- Personality Openness and SC: .20 None
& Environment positive slope
Hypothesis 4a Agreeable SF Congruence Positive fit .26 Congruence Positive fit .25
Hypothesis 4b Agreeable PL Congruence Agreeable: curvilinear .25 Congruence Positive fit .25
relationship
Hypothesis 5a Conscientious SF Congruence Positive fit .21 Congruence Positive fit .28
Hypothesis 5b Conscientious SC Congruence Positive fit .21 Main Effect- Personality Conscientiousness: .28
positive slope

student attitudes and performance are predicted by different classroom attitudes, the results indicate a wide range of classroom environment
environment variables, and that strategies may be employed to achieve factors need to be present. The higher education community would
enhanced student performance that may not affect attitudes, and vice benefit from considering these results when attempting to effect change
versa. If the goal is to achieve both enhanced student performance and in student performance and satisfaction.

Couse Satisfaction Course Performance

Hypothesis 3a

4.83
4.17

1.83 3.13

Hypothesis 3d

4.83
4.29

1.83 3.07

Fig. 2. Surface Response Graphics of the Congruence Models for Hypothesis 3.

190 D.K. Pawlowska et al. / Learning and Individual Differences 36 (2014) 180–193

Couse Satisfaction Course Performance

Hypothesis 4a

A B

4.83 4.83
4.17 4.17

1.75 3.13 1.75 3.13

Hypothesis 4b

4.83 4.83
3.27 3.27

1.75 1.73 1.75 1.73

Fig. 3. Surface Response Graphics of the Congruence Models for Hypothesis 4.

Couse Satisfaction Course Performance

Hypothesis 5a

A B

5.00 5.00
4.17 4.17

1.67 3.13 1.67 3.13

Hypothesis 5b

5.00
4.29

1.67 3.07

Fig. 4. Surface Response Graphics of the Congruence Models for Hypothesis 5.

 D.K. Pawlowska et al. / Learning and Individual Differences 36 (2014) 180–193
5.2. Fit relationships
In terms of personality, agreeableness and conscientiousness had
the most fit relationships (all of the hypotheses for these personality
variables were at least partially supported with positive fit relation-
ships). Agreeableness was related to increased satisfaction and perfor-
mance when students were provided with an organized and focused
classroom. Agreeable students also performed better when they were
provided with social learning opportunities. The findings provide sup-
port for the notion that agreeableness is related to system maintenance
which encompasses organization, order, and expectations (Westerman
& Simmons, 2007). These results relate to the cooperative nature of in-
dividuals high on agreeableness and provide support for the belief that
agreeableness is related to relationship orientation. This is evidenced by
several studies where agreeableness has been related to interpersonal
interaction (Nikolaou, 2003) and social participation (Stevens & Ash,
2001).
Conscientiousness students also enjoyed higher grades and course
satisfaction when structure and focus in the classroom was high, and
seemed to enjoy their classes more when there was competition
among the students. This outcome relates to the finding that individuals
high on conscientiousness have a need for continuous improvement
and have achievement and goal orientations (Chamorro-Premuzic
et al., 2007; Westerman & Simmons, 2007). In terms of performance,
congruence with environment may not play an important role since
conscientiousness on its own has been related to high performance
and achievement across several studies. In particular, in their meta-
analysis, Trapmann et al. (2007) found it to be the only personality
trait that was a significant predictor of academic achievement in higher
education. These results suggest that conscientious individuals will per-
form well in most environments although they may not be as satisfied.
Overall, the results indicate variations in performance and satisfaction
with different combinations of personality and environmental dimen-
sions. This study provides evidence that personality types are more
satisfied and perform better in varying classroom environments. If dis-
ciplines or majors in higher education attract different personality
types, this research suggests there may be unique patterns of interac-
tion with classroom environments that may exist within each field.
The higher education community needs to further explore these results
in the interest of better serving their students.
More generally, in terms of the fit analyses, the results indicated that
all but one of the significant fit relationships were in the positive direc-
tion, i.e., both predictor variables needed to be high in order to produce
a positive outcome. These results may suggest that if students are
expecting certain characteristics from their classroom environments
that are indeed present, their performance and satisfaction outcomes
will likely increase. Conversely, if students are not expecting certain
characteristics from their classroom environments then it may not
make a difference in terms of student satisfaction and performance
whether the characteristics are present. Thus, classroom environment
expectations may play a key role in determining outcomes, although
further research on the role of student expectations would be helpful
in this regard.
5.3. Implications
Although initial research has shown that students learn in different
ways, at different rates, and in different environments (Weisstein &
Jacobson, 2009), we have largely disregarded this information when it
comes to establishing educational structure (Tomlinson & Kalbfleisch,
1998). Currently, there is a predominance of a “one-size-fits-all” educa-
tional theory and approach (Davis, 2010). Standards are generally uni-
form, from teaching approach, assignment options, textbook options,
grading standards, and forms of assessment. This approach to higher ed-
ucation minimizes individual differences and provides a rigid structure
to which students must conform. Thus, it is possible that traditional
191
higher education may be best suited for only a limited subset of individ-
ual characteristics. In terms of such one-size-fits-all education, our
findings are mixed. Providing a structured and focused classroom envi-
ronment seems to hold true for students in both satisfaction and perfor-
mance outcomes, and suggests that this factor should be more
comprehensively present in higher education classrooms.
However, evidence from this research also suggests that different
personality types seem to be more satisfied and perform better in vary-
ing environments lending support to the notion that one size may not fit
all. This result was found more consistently for the student satisfaction
outcome. Considering that student satisfaction is related to retention
(Elliot & Healy, 2001), this finding may have important implications.
As universities increasingly view themselves as being engaged in a ser-
vice industry, there has been a greater focus on emphasizing the impor-
tance of meeting student needs and expectations. The general finding
from this study, that environments are important predictors of student
satisfaction, may have implications for the success of university recruit-
ment and enrollment efforts. Measuring the relationships between per-
sonality traits, classroom environments, and positive outcomes may
identify these fit factors and assist higher education institutions in de-
veloping better tools for assessing what types of individuals would pros-
per on their campuses (Farsides & Woodfield, 2003), which may also
lead to the development of more effective recruitment strategies. As in-
dividuals select certain organizations and institutions in part because
they perceive them to fit well with their own needs (Harms, Roberts,
& Winter, 2006), colleges and universities could also gain information
on how to increase their applicant pools. These institutions might
enjoy the benefit of increased retention if more positive student aca-
demic performance is a consequence of congruence between personal-
ity traits and classroom environments.
In a larger sense, understanding the range of individual differences
in a given student population may guide educators towards developing
effective interventions in teaching and understanding why current in-
terventions and practices do, or do not work (Peeke, Steward, &
Ruddock, 1998). And these efforts may be focused at a college or aca-
demic major level, if differing patterns of personality amongst students
are discovered. Such efforts may provide needed guidance in how to
structure higher education to produce positive learning outcomes for
a more diverse range of students. Personality tests might also be used
to help in counseling students and helping them to see their depen-
dence on particular environments.
As personality significantly predicts performance, universities may
also consider these findings of interest to their admission processes. Re-
search indicates that personality has incremental validity over typical
measures used in college admittance such as the SAT (Trapmann et al.,
2007). Several researchers have supported the concept of using person-
ality assessments in admissions processes, as they also would likely re-
duce adverse impact for gender and race. The current study provides
support for the use of personality measures in admission processes as
four of the five personality traits measured were significant predictors
of academic performance. Using personality measures may become a
viable option for universities with large numbers of applicants and
stringent competition for admittance.
The study's findings also add support to the view that student
satisfaction is an uncertain predictor of academic performance. When
examining the fit relationships in this study, most of the personality–
environment combinations that significantly predicted satisfaction did
not predict performance. Relatedly, research in the business and organi-
zational psychology literature has found low (Iaffaldano & Muchinsky,
1985) to moderate (Judge, Thoresen, Bono, & Patton, 2001) correlations
between job satisfaction and performance. This pattern suggests that
student satisfaction and academic performance are different concepts
with different combinations of antecedents. The current findings sug-
gest that these concepts are predicted by different constructs and pro-
duce different outcomes, and supports the need for further research
on the use of teaching evaluations in higher education (which most
192 D.K. Pawlowska et al. / Learning and Individual Differences 36 (2014) 180–193
frequently rely on attitudinal responses) to achieve improved student
performance.
5.4. Limitations and directions for future research
This study was the first of its kind to use the classroom environment
survey to comprehensively assess the collegiate environment. Replica-
tion and cross validation of the studies factor analyses is needed. Addi-
tionally, some of the classroom environment scales had rather low
reliability (b.60). These results further suggest that additional work is
needed to create a more psychometrically sound measure of the colle-
giate classroom environment.
The study was limited in that it used undergraduate students in in-
troductory psychology courses. Although a diverse set of students and
majors tend to enroll in this class, it is possible this sample selection
may have produced more limited variation in student personality and
classroom environments. Further, several personality–environment fit
relationships predicted increased satisfaction, performance, or both.
These relationships need to be replicated on a broader sample of aca-
demic disciplines if they are to be used as a basis for tailoring classroom
environments. Finally, the potential influence of additional unexamined
variables (student motivation, ability, cognitive activation) on the fit re-
lationships indicated in this study could represent a promising area for
future research.
In summary, evidence from this research supports structure and
focus as an important classroom environment dimension that increases
both student satisfaction and performance. Personality dimensions
were the predominant predictors of performance and satisfaction, lend-
ing support to the view that personality traits have remarkable impacts
on academic achievement. And personality and classroom environment
fit combinations were found to increase satisfaction, performance, or
both, demonstrating the importance of person–environment fit in
higher education.
References
Barrick, M.R., & Mount, M.K. (1991). The big five personality dimensions and job perfor-
mance: A meta-analysis. Personnel Psychology, 44, 1–26.
Buunk, B.P., Nauta, A., & Molleman, E. (2005). In search of the true group animal: The ef-
fects of affiliation orientation and social comparison orientation upon group satisfac-
tion. European Journal of Personality, 19, 69–81.
Caspi, A., Chajut, E., Saporta, K., & Beyth-Marom, R. (2006). The influence of personality on
social participation in learning environments. Learning and Individual Differences, 16,
129–144.
Chamorro-Premuzic, T., Furnham, A., & Lewis, M. (2007). Personality and approaches to
learning predict preference for different teaching methods. Learning and Individual
Differences, 17, 241–250, http://dx.doi.org/10.1016/j.lindif.2006.12.001.
Chartrand, J.M. (1991). The evolution of trait-and-factor career counseling: A person
× environment fit approach. Journal of Counseling and Development, 69, 518–524.
Costa, P., & McCrae, R. (1994). Stability and change in personality from adolescence
through adulthood. In C.F. HalversonJr., G.A. Kohnstamm, & R.P. Martin (Eds.), The de-
veloping structure of temperament and personality from infancy to adulthood
(pp. 139–155). Hillsdale, NJ: Erlbaum.
Costa, P.T., & McCrae, R.R. (1995). Domains and facets: Hierarchical personality assess-
ment using the revised NEO Personality Inventory. Journal of Personality Assessment,
64, 21–50.
Davis, M. (2010). Personality and its effect on relationships and teaching and learning
styles. International Schools Journal, 29, 22–29.
Deemer, S.A. (2004). Classroom goal orientation in high school classrooms: Revealing
links between teacher beliefs and classroom environments. Educational Research,
46, 73–90.
Edwards, J.R. (1991). Person-job fit: A conceptual integration, literature review, and
methodological critique. In C.L.R.I.T. Cooper (Ed.), International review of industrial
and organizational psychology, Vol. 6. (pp. 283–357). Chichester, UK: Wiley.
Edwards, J.R. (1994). Alternatives to difference scores as dependent variables in the study
of congruence in organizational research. Organizational Behavior and Human Decision
Processes, 64, 307–324.
Edwards, J.R., & Parry, M.E. (1993). On the use of polynomial regression equations as an
alternative to difference scores in organizational research. Academy of Management
Journal, 36, 1577–1613.
Ehrhart, K.H. (2006). Job characteristics beliefs and personality as antecedents of subjec-
tive person–job fit. Journal of Business and Psychology, 21, 193–226.
Elliot, K.M., & Healy, M.A. (2001). Key factors influencing student satisfaction relate to re-
cruitment and retention. Journal of Marketing for Higher Education, 10, 1–11.
Farsides, T., & Woodfield, R. (2003). Individual differences and undergraduate academic
success: The roles of personality, intelligence, and application. Personality and
Individual Differences, 34, 1225–1243.
Feldman, K.A., Smart, J.C., & Ethington, C.A. (2004). What do college students have to lose?
Exploring the outcomes of differences in person–environment fits. The Journal of
Higher Education, 75, 528–555.
Flynn, N.M., & Rapoport, J.L. (1976). Hyperactivity in open and traditional classroom
environments. The Journal of Special Education, 10, 285–290.
Fraser, B.J., & Fisher, D.L. (1983). Use of actual and preferred classroom environment
scales in person–environment fit research. Journal of Educational Psychology, 75,
303–313.
Fry, P.S., & Addington, J. (1984). Comparison of social problem solving of children from
open and traditional classrooms: A two-year longitudinal study. Journal of
Educational Psychology, 76, 318–329.
Furnham, A., & Chamorro-Premuzic, T. (2005). Individual differences in students'
preferences for lecturers' personalities. Journal of Individual Differences, 26, 176–184,
http://dx.doi.org/10.1027/1614-0001.26.4.176.
Furnham, A., Christopher, A., Garwood, J., & Martin, N.G. (2008). Ability, demography,
learning style, and personality trait correlates of student preference for assessment
method. Educational Psychology, 28, 15–27.
Glorfeld, L.W. (1995). An improvement on Horn's parallel analysis methodology for
selecting the correct number of factors to retain. Educational and Psychological
Measurement, 55, 377–393.
Harms, P.D., Roberts, B.W., & Winter, D. (2006). Becoming the Harvard man: Person–
environment fit, personality development, and academic success. Personality and Social
Psychology Bulletin, 32, 851–865, http://dx.doi.org/10.1177/0146167206287720.
Hayton, J.C., Allen, D.G., & Scarpello, V. (2004). Factor retention decisions in exploratory
factor analysis: A tutorial on parallel analysis. Organizational Research Methods, 7(2),
191–205, http://dx.doi.org/10.1177/1094428104263675.
Horn, J.L. (1965). A rationale and test for the number of factors in factor analysis.
Psychometrika, 30, 179–185.
Hox, J.J., & Roberts, J.K. (2011). Handbook of advanced multilevel analysis (2th ed.). Hove,
East Sussex: Routledge.
Hoyt, D.P., & Cashin, W.E. (1977). Development of the IDEA System. Technical Report No. 1.
(Retrieved from http://ideaedu.org/sites/default/files/techreport-01.pdf.).
Hoyt, D.P., & Lee, E. (2002). Basic data for the revised IDEA System. Technical Report No. 12.
The Individual Development and Educational Assessment Center (Retrieved from
http://www.ideaedu.org/sites/default/files/techreport-12.pdf.).
Iaffaldano, M.T., & Muchinsky, P.M. (1985). Job satisfaction and job performance: A meta-
analysis. Psychological Bulletin, 97, 251–273.
James, L.R., Demaree, R.G., & Wolf, G. (1984). Estimating within-group interrater reliabil-
ity with and without response bias. Journal of Applied Psychology, 69, 85–98.
Joiner, K.F., Malone, J.A., & Haimes, D.H. (2002). Assessment of classroom environments in
reformed calculus education. Learning Environments Research, 5, 51–76.
Judge, T.A., Thoresen, C.J., Bono, J.E., & Patton, G.K. (2001). The job satisfaction–job perfor-
mance relationship: A qualitative and quantitative review. Psychological Bulletin, 127,
376–407.
Kirtz, S., & Moos, R.H. (1974). Physiological effects of social environments. Psychosomatic
Medicine, 36, 96–114.
Kutner, M.H., Nachtsheim, C.J., Neter, J., & Li, W. (2005). Applied linear statistical models
(5th ed.). Boston: McGraw-Hill Irwin.
Lounsbury, J.W., Saudargas, R.A., Gibson, L.W., & Leong, F.T. (2005). An investigation of
broad and narrow personality traits in relation to general and domain-specific life
satisfaction of college students. Research in Higher Education, 46, 707–729.
McCrae, R.R., & Costa, P.T. (1987). Validation of the five-factor model of personality
across instruments and observers. Journal of Personality and Social Psychology,
52, 81–90.
McCrae, R.R., & Costa, P.T. (2004). A contemplated revision of the NEO Five-Factor Inven-
tory. Personality and Individual Differences, 36, 587–596.
Muthén, B.O. (1994). Multilevel covariance structure analysis. Sociological Methods &
Research, 22, 376–398, http://dx.doi.org/10.1177/0049124194022003006.
Nielsen, H.D., & Moos, R.H. (1978). Exploration and adjustment in high school classrooms:
A study of person–environment fit. The Journal of Educational Research, 72, 52–57.
Nijhuis, J., Segers, M., & Gijselaers, W. (2007). The interplay of perceptions of the learning
environment, personality, and learning strategies: A study amongst international
business studies students. Studies in Higher Education, 32, 59–77, http://dx.doi.org/
10.1080/03075070601099457.
Nikolaou, I. (2003). Fitting the person to the organization: Examining the personality–job
performance relationship from a new perspective. Journal of Managerial Psychology,
18, 639–648.
Pace, C., & Stern, G. (1958). An approach to the measurement of psychological character-
istics of college environments. Journal of Educational Psychology, 49, 269–277.
Peeke, P.A., Steward, R.J., & Ruddock, J.A. (1998). Urban adolescents' personality and
learning styles: Required knowledge to develop effective interventions in schools.
Journal of Multicultural Counseling and Development, 26, 120–136.
Peugh, J.L. (2010). A practical guide to multilevel modeling. Journal of School Psychology,
48(1), 85–112, http://dx.doi.org/10.1016/j.jsp.2009.09.002.
Schniederjans, M.J., & Kim, E.B. (2005). Relationship of student undergraduate achieve-
ment and personality characteristics in a total web-based environment: An empirical
study. Decision Sciences Journal of Innovative Education, 3, 205–221.
Stevens, C., & Ash, R. (2001). Selecting employees for fit: Personality and preferred
managerial style. Journal of Managerial Issues, 13, 500–517.
Swanson, J.L., & Fouad, N.A. (1999). Applying theories of person–environment fit to the
transition from school to work. The Career Development Quarterly, 47, 337–347.
Tomlinson, C.A., & Kalbfleisch, M.L. (1998). Teach me, teach my brain, a call for differen-
tiated classrooms. Educational Leadership, 52–55.
 D.K. Pawlowska et al. / Learning and Individual Differences 36 (2014) 180–193
Trapmann, S., Hell, B., Hirn, J.O.W., & Schuler, H. (2007). Meta-analysis of the relationship
between the big five and academic success at university. Journal of Psychology, 215,
132–151.
Trickett, E.J., & Moos, R.H. (1973). Social environment of junior high and high school class-
rooms. Journal of Educational Psychology, 65, 93–102.
Trickett, E.J., & Quinlan, D.M. (1979). Three domains of classroom environment: Factor
analysis of the classroom environment scale. American Journal of Community
Psychology, 7, 279–291.
U.S. Department of Education, & National Center for Education Statistics (2010). The
Condition of Education 2010 (NCES 2010–028). Retrieved from. http://nces.ed.gov/
fastfacts/display.asp?id=40
“US College Dropout Rates Spark Concern.” (2006, September 27). Associated Press.
Retrieved from http://diverseeducation.com/article/6422/.
Weisstein, E., & Jacobson, D. (2009). Many sizes fit all: Considering multiple pathways
tohigher learning. The New England Journal of Higher Education, 15–17.
Westerman, J.W., & Simmons, B.L. (2007). The effects of work environment on the
personality–performance relationship: An exploratory study. Journal of Managerial
Issues, 19, 288–305.
193
Whitbourne, J. (2010). The dropout dilemma: One in four college freshmen drop out.
What is going on here? What does it take to stay in?. Careers and College. Retrieved
from. http://findarticles.com/p/articles/mi_m0BTR/is_4_22/ai_84599442/pg_7/?
tag=content;col1
Williams, R.L. (1971). Relationship of class participation to personality, ability, and
achievement variables. The Journal of Social Psychology, 83, 193–198.
Williamson, M.F., & Watson, R.L. (2007). Learning styles research: Understanding how
teaching should be impacted by the way learners learn part III: Understanding how
learners’ personality styles impact learning. Christian Education Journal, 4, 62–77.
Woszczynski, A.B., Gutherie, T.C., & Shade, S. (2005). Personality and programming.
Journal of Information Systems Education, 16, 293–299.
Zhang, L. (2006). Thinking styles and the big five personality traits revisited. Personality
and Individual Differences, 40, 1177–1187.