Content uploaded by David K Dodd
Author content
All content in this area was uploaded by David K Dodd on Dec 16, 2014
Content may be subject to copyright.
MAY /JUNE 2003 XVOL 44 NO 3 291
Student Burnout as a Function of Personality,
Social Support, and Workload
Sheri R. Jacobs David K. Dodd
Measures of social support (Multidimen-
sional Scale of Perceived Social Support),
personality (General Temperament Survey),
and workload were related to psychological
burnout (Maslach Burnout Inventory) among
149 college students (M = 20.8 yrs.). High
levels of burnout were predicted by negative
temperament and subjective workload, but
actual workload (academic and vocational)
had little to do with burnout. Low levels of
burnout were predicted by positive tempera-
ment, participation in extracurricular
activities, and social support, especially from
friends.
The term burnout was first introduced by
Freudenberger (1974), who defined it as “to
fail, to wear out, or become exhausted by
making excessive demands on energy,
strength, or resources” (p.159). The concept
of burnout was further popularized with the
development of the Maslach Burnout Inven-
tory (Maslach & Jackson, 1981). Research
on burnout originally focused on people in
various occupational groups, including
human service workers, teachers, nurses, and
psychologists. Although several studies of
burnout among college residential assistants
(RAs) have been conducted (e.g., Hardy &
Dodd, 1998), little is known about burnout
among college students in general, and that
is the focus of the current study.
Maslach and Jackson (1981) defined
burnout as a syndrome that is composed of
three dimensions: emotional exhaustion,
depersonalization, and reduced personal
accomplishment. Emotional exhaustion
refers to demands and stressors that cause
people to feel overwhelmed and unable to
give of themselves at a psychological level.
Depersonalization is the development of
negative and cynical attitudes that can create
a callous view of others, perceiving them as
deserving of their troubles. Reduced sense
of personal accomplishment is the tendency
to view oneself negatively and to be dis-
satisfied with accomplishments. Burnout is
related to various personal dysfunctions,
such as physical exhaustion, insomnia, and
increased drug and alcohol use. Some
symptoms of burnout include lower moti-
vation and satisfaction with work, increased
risk of health impairments, social conflicts,
and lower efficiency (Maslach, Jackson, &
Leiter, 1997). Many college students who
seek counseling may be experiencing burn-
out or several of its consequences. Thus,
identifying factors that affect burnout is
important in order to improve treatment and
prevention models for student burnout.
Job stress is commonly attributed to
external factors related to the work environ-
ment, such as work demands, working
conditions, and poor supervision. Maslach
and Jackson (1981) emphasized the psycho-
logical nature of the burnout syndrome rather
than the physical work environment. Sub-
sequent research has substantiated their
theory, by demonstrating the importance of
internal (e.g., personality) and interpersonal
Sheri R. Jacobs is a doctoral student of Psychology at University of South Florida. David K. Dodd is a
senior lecturer of Psychology at Washington University in St. Louis.
292 Journal of College Student Development
Jacobs & Dodd
(e.g., social support) factors, as well as
external factors (e.g., workload).
Personality characteristics have gen-
erally been related to burnout, but research
results depend upon which specific traits are
correlated with which of the dimensions of
burnout. Some researchers have found
extroversion to be correlated only with
emotional exhaustion and reduced sense of
personal accomplishment (Eastburg, Wil-
liamson, Gorsuch, & Ridley, 1994; Mills
& Huebner, 1998), whereas others have
reported extroversion to be correlated only
with depersonalization and reduced personal
accomplishment (Huebner & Mills, 1994;
Zellars, Perrewe, & Hochwarter, 2000). The
relation of neuroticism to burnout is also
varied, with Mills and Huebner reporting
significant correlations with all three dimen-
sions of burnout, but Zellars et al. (2000)
reporting a significant correlation only with
emotional exhaustion. Although this liter-
ature that relates personality to burnout is
promising, additional research is necessary
to further understand this important
connection.
Social support has also been related to
burnout (Greenglass, Fiksenbaum, & Burke,
1994; Kahill, 1986; Koniarek & Dudek,
1996), with greater support generally related
to lower levels of burnout. Like the research
on personality, however, this relationship
varies considerably, depending upon the type
of social support. Several studies of burnout
in the workplace have shown that social
support from supervisors is related to lower
levels of burnout, whereas other sources of
social support (e.g., from family, friends, and
coworkers) are less strongly related to
increased burnout (Huebner, 1994; Ross,
Altmaier, & Russell, 1989; Russell, Altmaier,
& Van Velzen, 1987).
Most researchers who have examined
the relationship between workload and
burnout have reported a positive relationship,
with greater workload associated with
greater burnout. Some studies have found
this relationship to be true only for emotional
exhaustion (Male & May, 1997, 1998), but
Greenglass, Burke, and Fiksenbaum (2001)
found that workload correlated to all three
dimensions of burnout. Inadequate measures
of workload as well as incomplete models
relating workload to burnout may have
hampered the investigation of how these two
variables interrelate (Koeske & Koeske,
1989). It is reasonable to assume that
objective workload contributes causally to
burnout, but many workers seem to cope
successfully with heavy workloads, whereas
others do not. Perhaps it is the subjective
response to workload, rather than the
workload itself, that contributes most to
burnout. Separate measures of objective
workload and subjective workload need to
be developed and differentially explored as
predictors of burnout.
The vast majority of research on burnout
has been conducted on occupational popu-
lations, including: salespeople (Sand &
Miyazaki, 2000); teachers (Greenglass et al.,
1994; Russell et al., 1987), nurses (Eastburg
et al., 1994; Koniarek & Dudek, 1996;
Zellars et al., 2000), human service workers
(Wade, Cooley, & Savicki, 1986), counselors
(Ross et al., 1989), psychologists (Kahill,
1986), and school psychologists (Huebner
& Mills, 1994; Mills & Huebner, 1998;
Sandoval, 1993). Research on burnout
among college students has been restricted
primarily to those who are in supervisory and
advisory roles, namely RAs (Benedict &
Mondloch, 1989; Fuehrer & McGonagle,
1988; Hardy & Dodd, 1998; Hetherington,
Oliver, & Phelps, 1989). The researchers in
these studies have focused almost exclu-
MAY /JUNE 2003 XVOL 44 NO 3 293
Predictors of Student Burnout
sively on individual (e.g., gender and
experience level) and situational (e.g., floor
assignment) factors, and little is known about
how intrapersonal (e.g., personality) and
interpersonal (e.g., social support) factors are
related to burnout among students. It is also
risky to generalize findings from RAs to the
general college population, because students
who become RAs may be a very select
group. For example, Hetherington et al.
(1989) reported lower scores on sense of
personal accomplishment (i.e., greater
burnout) among general students than among
RAs.
On the other hand, there has been a
substantial amount of research on stress
among general college students. Stress has
been shown to be correlated with college
students’ health behaviors (Weidner, Kohl-
mann, Dotzauer, & Burns, 1996), anxiety
concerning exams (Abouserie, 1994; Ever-
son, Tobias, Hartman, & Gourgey, 1993;
Sloboda, 1990), self-esteem (Abouserie;
Newby-Fraser & Schlebusch, 1997), and
coping strategies that students use (Dwyer
& Cummings, 2001). To the extent that stress
is an important component of burnout
(Maslach & Jackson, 1981), this research is
relevant to the development of a model of
burnout among college students. Between
classes, exams, employment, and extra-
curricular activities, students are likely to
experience high levels of stress, but do they
experience burnout? Much research is
needed to determine the prevalence of
burnout, to identify important intrapersonal
and interpersonal factors that influence
burnout, and to develop effective inter-
ventions to prevent and reduce burnout in
college students.
For the current research, we had two
important goals related to the replication and
extension of prior research on burnout: First,
we explored the relations of personality and
social support to the three components of
burnout. Second, we examined the role of
workload, as measured both objectively and
subjectively.
METHOD
Participants
Participants were 149 undergraduate stu-
dents (103 women and 46 men) enrolled in
a mid-sized, private university in the
Midwest. Consistent with the enrollment of
the university, the ethnicity of the sample
was approximately 62% Caucasian, 20%
Asian American, 6% African American, and
12% “other.” The university, located in an
urban setting, has highly selective admis-
sions criteria, and its students are generally
considered to be very motivated academ-
ically. Participants were recruited through
the psychology department’s research
participant pool.
The participant pool consists of all
students enrolled in any of approximately
eight psychology courses, including intro-
ductory psychology, introductory statistics,
developmental psychology, and abnormal
psychology, all courses which count toward
the general education requirements at the
university. Students who participate in
research or alternative options receive extra
credit in their enrolled courses. It is estimated
that 90% or more of all students in the
university take at least one of these psycho-
logy courses, and about 90% of those choose
to participate in research. By extrapolation,
it is estimated that at least 80% of all
university students participate in this
research pool, suggesting that research
samples drawn from the pool are likely to
be highly representative of the undergraduate
student population at large. Informal studies
294 Journal of College Student Development
Jacobs & Dodd
of the participation pool reveal that it closely
resembles in demographics the university’s
undergraduate student body, with one
notable exception. Women are more highly
represented in the participant pool (about
70%) than in the student body (50%),
reflecting the fact that women enroll in
psychology courses at a higher rate than
men. The proportion of women in the current
sample (69%) closely reflects the proportion
in the participation pool.
Because the focus of the study was to
examine burnout in college students who had
fully adapted to the university setting, only
students in their junior (n = 83) or senior
year (n = 66) were included in the study
(M = 20.8 yrs., SD = 1.4). Approximately
half (54%) of the sample was employed, with
weekly hours of work ranging from 2 to 27
(M = 10.4 hrs., SD = 5.8). None of the
participants worked full-time. Data were
collected during the last 4 weeks (n = 106)
of Fall semester and the first 4 weeks of the
subsequent Spring semester (n = 43).
Procedure and Measures
The research instrument, which required
about 30 minutes to complete, was admini-
stered in small groups of 4 to12 participants.
Privacy and anonymity of participants were
carefully protected. In addition to basic
biographical items, the questionnaire in-
cluded the following measures.
The Maslach Burnout Inventory (MBI)
(Maslach & Jackson, 1981) was admin-
istered to measure subjects’ level of burnout.
The MBI consists of 22 questions that are
divided into three subscales: Emotional
Exhaustion (EE), Depersonalization (DP),
and Personal Accomplishment (PA). EE is
measured by nine items (e.g., “I feel emo-
tionally drained from my work”), DP is
measured by five items (e.g., “I feel I treat
some friends as if they were impersonal
objects”), and PA is measured by eight items
(e.g., “I feel I’m positively influencing other
people’s lives through my work”). Each item
is rated on a 7-point Likert-type scale
ranging from 0 (never) to 6 (every day).
Possible score ranges are 0 to 54 for EE, 0
to 30 for DP, and 0 to 48 for PA. High
burnout is reflected in high scores on EE and
DP and in low scores on PA. The MBI
appears to be sufficiently reliable, with
reported alpha coefficients of .87 for EE, .77
for DP, and .75 for PA (Maslach & Jackson).
To make the survey more appropriate for
college students, item wording was modified
slightly (e.g., from “job” to “school” and
“coworkers” to “friends”).
The General Temperament Survey (GTS)
(Clark & Watson, 1990), a component of the
more comprehensive personality inventory,
Schedule for Nonadaptive and Adaptive
Personality (SNAP) (Clark, 1993), was
chosen for use in the current study because
it is relatively brief and has well-established
reliability and validity (Clark). The GTS is
a 90-item, true-false questionnaire that yields
measures of Negative Temperament (NT),
Positive Temperament (PT), and Disin-
hibition (DIS), dimensions that correspond
closely to neuroticism, extroversion, and
lack of conscientiousness, respectively
(Watson, Clark, & Harkness, 1994). More
specifically, NT (28 items; e.g., “I often feel
nervous and stressed”) measures negative
mood and self-concept; PT (27 items; e.g.,
“People would describe me as a pretty
enthusiastic person”) measures positive
emotionality; and DIS (35 items; e.g., “I’ll
take almost any excuse to goof off instead
of work”) measures lack of behavioral
control. Scoring is based on the number of
responses in the scored direction, with
possible score ranges of 0 to 28 for NT, 0 to
MAY /JUNE 2003 XVOL 44 NO 3 295
Predictors of Student Burnout
27 for PT, and 0 to 35 for DIS. The GTS is
considered to be highly reliable, with
reported alpha coefficients of .91 for NT, .88
for PT, and .82 for DIS for college samples
(Clark).
The Multidimensional Scale of Per-
ceived Social Support (MSPSS) (Zimet,
Dahlem, Zimet, & Farley, 1988) was used
to assess self-reported amounts of social
support. The MSPSS is a 12-item question-
naire containing three subscales measuring
perceived social support from Friends (e.g.,
“My friends really try to help me”), Family
(e.g., “I can talk about my problems with my
family”), and a Significant Other (e.g.,
“There is a special person in my life who
cares about my feelings”). Items are scored
on a 7-point Likert-type scale, ranging from
1 (very strongly disagree) to 7 (very strongly
agree) for each item. Each subscale consists
of four items and has a possible score range
of 4 to 28. High scores reflect high levels of
perceived social support. The MSPSS
appears to have very good reliability: In a
study of college students, Dahlem, Zimet,
and Walker (1991) reported alpha co-
efficients of .90, .94, and .95, respectively,
for the subscales of Friends, Family, and
Significant Other.
Subjective Workload was measured with
four items: “I think I could have handled 2
to 3 more units this semester,” “I wish that I
had enrolled in 2 to 3 fewer units this
semester,” “I think I could have handled 1
to 2 more extracurricular activities,” and “I
wish that I were involved in 1 to 2 fewer
extracurricular activities”. To maintain
consistency with the Likert-scales used in
scoring the MBI and MSPSS, a 7-point
Likert-type scale was also used to measure
level of agreement from 1 (very strongly
disagree) to 7 (very strongly agree) to the
items measuring Subject Workload. After
appropriate reverse scoring, the items were
summed, resulting in a possible score of 4
to 28, with higher scores reflective of greater
workload.
Objective Workload was measured by
three items: number of credit hours enrolled,
number of hours spent participating in
extracurricular activities each week, and
number of hours per week spent in paid
employment (with unemployed participants
coded as 0).
RESULTS
Missing data were rare, occurring for less
than 0.1% of the data set, and they were
replaced by the group mean for each item.
Cronbach’s coefficient alpha was used to
estimate the reliability of each composite
variable (see Table 1). Reliability coefficient
alphas were all above .70, except for the
Subjective Workload (.56). Although burn-
out is a global term, the three subscales of
the MBI tap separate components of burnout,
and accordingly, the intercorrelations be-
tween these subscales were only moderate:
for EE and DP, r = .49; for EE and PA,
r = –.36; and for DP and EE, r = –.29.
Therefore, separate multiple regression
procedures were used to predict each burnout
subscale. Specifically, forward stepwise-
regression was used, with alpha levels set
at .05 for inclusion and .10 for exclusion.
Level of Burnout
Descriptive data for the current sample
appear in Table 1. Relative to overall norms
for the MBI (Maslach & Jackson, 1981), the
current sample reported moderate-to-high
levels of burnout on the dimensions of EE
and PA but low-to-moderate scores on DP.
Compared to descriptive data reported for
RAs by others (Hardy & Dodd, 1998;
296 Journal of College Student Development
Jacobs & Dodd
Hetherington et al., 1989), the current sample
had slightly but nonsignificantly lower levels
of burnout on all three subscales.
Prediction of Burnout
Detailed results of the regression analyses
appear in Table 2. For the analysis using EE
as the dependent variable, total R2 was .41,
F(3, 146) = 33.44, p < .001. Higher scores
on EE were associated with higher levels of
negative temperament, higher subjective
workload, and a greater number of work
hours.
For DP, total R2 was .26, F(3, 146)
= 16.57, p < .001. Higher scores on DP were
associated with lower levels of social support
from friends, higher levels of negative
temperament, and higher subjective work-
load.
For PA, total R2 was .53, F(4, 146)
= 40.72, p < .001. Greater burnout (lower PA
scores) was associated with lower levels of
social support from friends, lower levels of
positive temperament, higher levels of
negative temperament, and fewer hours spent
in extracurricular activities.
Secondary Analyses
The multiple regression procedures used in
the primary analyses capitalize on the best
predictors, at the expense of weaker pre-
dictors. To investigate the strength of these
weaker predictors, outside of the realm of
the regression analyses, we inspected the
individual Pearson rs (criterion of p = .05,
two-tailed) for all predictors that did not
enter the regression analyses. For example,
all three measures of social support were
negatively correlated with all three measures
of burnout at statistically significant levels,
with correlations ranging from .18 to .54,
despite the fact that only support from
TABLE 1.
Descriptive Data and Coefficient Alphas for Dependent Variables and Predictors
Number Possible
of items score range MSD
αα
αα
α
Dependent Variables
Emotion Exhaustion 9 0-54 22.6 8.9 .86
Depersonalization 5 0-30 7.5 4.9 .71
Personal Accomplishment 8 0-48 33.4 6.8 .77
Predictors
Social Support-Friends 4 4-28 23.2 4.0 .93
Social Support-Family 4 4-28 21.9 4.7 .87
Social Support-Significant Other 4 4-28 23.7 4.8 .96
Subjective Workload 4 4-28 15.0 4.1 .56
Positive Temperament 27 0-27 18.0 5.8 .88
Negative Temperament 28 0-28 13.9 6.9 .90
Disinhibition 35 0-35 12.1 6.3 .85
MAY /JUNE 2003 XVOL 44 NO 3 297
Predictors of Student Burnout
friends entered into the multiple regression
equations at statistically significant levels.
In each case, greater social support was
associated with lower burnout scores. (For
conciseness, negative signs have been
dropped from the rs in this paragraph, and
the direction of the relation is indicated in
text.) Similarly, subjective workload was
negatively correlated with all three measures
(rs from .24 to .43). Extracurricular activity
was negatively correlated with EE (.17) and
positively correlated with PA (.35). Further
exploration showed that the correlation
between extracurricular activity and PA was
especially strong among men (r = .51) and
only moderate among women (r = .27). On
the other hand, academic hours enrolled,
whether a participant worked, or number of
work hours all failed to correlate signi-
ficantly with any of the burnout measures
or with any of the measures of support.
In addition, we explored the possible
effects of sex of participants, year in school
(junior vs. senior), and time of semester
(beginning vs. end) on burnout, using three-
way ANOVAs. For EE and DP, we found no
significant main effects or interactions. For
PA, those participating late in the Fall
semester had lower PA scores (M = 32.7,
SD = 7.0), and thus greater burnout, than
those participating at the beginning of the
Spring semester (M = 35.2, SD = 6.1),
F(1, 141) = 5.40, p = .02, eta = .19). None
of the remaining main effects or interactions
was significant. Follow-up analyses revealed
that those who participated at the end of Fall
semester were less conscientious (i.e., scored
higher on DIS) than were those participating
at the beginning of Spring semester, r = .20,
p = .02, but they did not differ on positive
or negative temperament.
Racial differences in burnout scores
were also explored. We used one-way
TABLE 2.
Stepwise Regression Analysis (Beta
Values) of Burnout Subscales
Burnout Subscalesa
EE DP RPA
Personality: GTS
Positive Temperament –.34
Negative Temperament .47 .25 .20
Disinhibition
Social Support: MSPSS
Friends –.32 –.36
Family
Significant Other
Subjective Workload .33 .15
Objective Workload
Academic hours enrolled
Work hours per week .13
Extracurricular hours –.19
Total R2 = .41*.26*.53*
Note. Values in table are standardized beta coefficients.
Variables with no beta values did not meet
criterion for inclusion (p < .05) or exclusion
(p < .10). GTS = General Temperament Scale.
MSPSS = Multidimensional Scale of Perceived
Social Support.
aEE = Emotional Exhaustion; DP = Depersonali-
zation; RPA = Reduced Personal Accomplishment.
* p < .001.
ANOVA, because the very small group sizes
for African Americans (n = 9) and Others
(n = 13) did not allow for including race in
the factorial analyses reported above. No
statistically significant differences were
found for EE and DP. For PA, however,
the racial groups differed significantly,
298 Journal of College Student Development
Jacobs & Dodd
F(3, 139) = 6.02, p = .001, eta = .34. Post
hoc analyses (HSD, p = .05) revealed that PA
scores for African Americans (M = 38.1,
SD = 8.2) were higher (thus lower burnout)
than for all other groups, whereas scores for
Asian Americans (M = 29.5, SD = 7.2) were
lower than for all other groups. Scores for
Caucasians (M = 34.3, SD = 5.6) and Others
(M = 34.8, SD = 9.0) were intermediate.
Finally, the relations of cumulative GPA,
a widely used global measure of academic
achievement, to burnout scores and the major
predictors were explored. GPA was signi-
ficantly correlated with EE (r = –.25,
p = .002) but not with DP (–.13) or PA (.10).
GPA was also not correlated with academic
extracurricular activity, whether participants
worked, hours of work, subjective workload,
or social support (all rs < .10).
DISCUSSION
This study was designed to evaluate the
relative influences of intrapersonal factors,
interpersonal factors, and workload on
psychological burnout. Results suggest that
personality, especially negative tempera-
ment, may predispose college students to
burnout, whereas social support, especially
from friends, may provide an important
buffer against burnout. Extracurricular
activities also appear to be important to a
student’s sense of accomplishment, thus
additionally counteracting burnout. Although
the subjective feeling of being overworked
predicted emotional exhaustion and de-
personalization, objective measures of
workload, including academic load, whether
a student was employed, and number of
hours worked, were not consistently related
to burnout.
Personality, as measured by the GTS
(Clark & Watson, 1990), was the strongest
predictor of burnout. Negative temperament
(roughly, neuroticism) especially was related
to all three aspects of burnout. According to
Clark (1993), negative temperament reflects
feelings of chronic stress and nervousness,
the experience of strong negative emotions,
and worrying, all characteristics that can
impair concentration and disrupt sleep.
Obviously, such a temperament might
contribute directly to emotional exhaustion
and, to a lesser degree, to depersonaliza-
tion and a reduced sense of personal
accomplishment.
Positive temperament was positively
correlated with personal accomplishment,
replicating previous findings (Mills &
Huebner, 1998; Zellars et al., 2000). Positive
traits such as optimism and energy may act
as a buffer to the stressors and frustration
that can lead to dissatisfaction with one’s
personal accomplishments. This inter-
pretation is supported by the simple Pearson
correlations that revealed PT to be negatively
related to EE and DP.
On the other hand, it is interesting that
disinhibition (impulsivity, lack of conscien-
tiousness, Clark, 1993) failed to predict any
aspect of burnout. This finding appears to
contradict those of Huebner and Mills
(1994), who found a positive relationship
between conscientiousness and personal
accomplishment, and of Mills and Huebner
(1998), who found a negative relationship
between conscientiousness and emotional
exhaustion. It is important to note that the
studies of Huebner and Mills measured
conscientiousness with the NEO-Five Fac-
tory Inventory (NEO-FFI) (Costa & McRae,
1985). Although Clark (1993) reported a
correlation of -.51 between the GTS DIS
scale and the NEO-FFI Conscientiousness
scale, the use of these different measures
may have caused the contradiction in
MAY /JUNE 2003 XVOL 44 NO 3 299
Predictors of Student Burnout
findings. Alternatively, population differ-
ences in these studies may better explain the
contradictory findings, especially because
the studies of Huebner and Mills were based
on samples of professional school psycho-
logists. In an undergraduate academic
setting, particularly a highly competitive
atmosphere like the one from which the
current sample was drawn, highly con-
scientious students (low scorers on DIS) may
place very demanding expectations upon
themselves and may suffer burnout as a
consequence, thus possibly counteracting the
positive elements of conscientiousness.
Given the importance of personality factors
in burnout, much additional research is
needed.
It is clear from these findings that social
support, especially from friends, is closely
related to lower levels of burnout. Speci-
fically, higher social support from friends
was associated with lower levels of de-
personalization and higher levels of personal
accomplishment, a replication of the findings
of Koniarek and Dudek (1996). In fact, all
three forms of support (friends, family, and
significant other) appear to be intercor-
related, both in the current study and as
reported by Zimet et al. (1988). Further
analysis of first-order correlations revealed
that all three measures of social support were
significantly related in a positive manner to
all three measures of burnout: Greater social
support was associated with less emotional
exhaustion, less depersonalization, and a
greater sense of personal accomplishment.
It is interesting to note that subjective
workload (i.e., feeling that one’s academic
and extracurricular load is too heavy) was
more closely related to burnout than was
objective workload (actual load of aca-
demics, extracurricular activities, and
employment). Subjective workload was
significantly related to all three aspects of
burnout, whereas work hours was related
only to emotional exhaustion and only
weakly so. Furthermore, actual academic
load did not predict any of the measures of
burnout. Clearly, students who were not
experiencing burnout felt as though they
were over committed, even though they were
enrolled in similar academic loads and were
participating in extracurricular activities less
frequently than did students who were not
experiencing burnout. This finding under-
scores the psychological nature of burnout
and the subjective experience of work
overload.
A positive relation between extra-
curricular activities and a sense of personal
accomplishment was found: The greater the
hours spent in extracurricular activities, the
greater the sense of personal accomplish-
ment. This clearly suggests that extra-
curricular involvement, rather than leading
to emotional exhaustion, promotes feelings
of achievement and self-worth, thus playing
a protective role against one aspect of
burnout. The fact that this association was
very strong for men (and only moderately
so for women) suggests that high involve-
ment in extracurricular activities may be
especially important for college men.
Secondary analysis showed that GPA
was negatively related to EE but not signi-
ficantly related to DP or PA. The cor-
relational design of this study does not
permit any conclusion about causation, but
this finding suggests that exhaustion may
contribute to lower academic performance.
Alternatively, students who are performing
below their own academic expectations,
perhaps despite considerable academic
efforts, may be more likely to experience
exhaustion. In either event, this relationship
between exhaustion and academic per-
300 Journal of College Student Development
Jacobs & Dodd
formance is potentially of substantial
importance to students, faculty, and college
administration.
The discovery of racial differences in
one aspect of burnout (reduced personal
accomplishment) is an intriguing one that
warrants further research. African Americans
reported the greatest PA scores, whereas
Asian Americans reported the lowest scores
(and thus greater burnout). Perhaps racial
groups differ in either their expectations for
accomplishment or their interpretation of
academic performance. Because this finding
is based on very small group sizes, it must
be considered very preliminary, but it calls for
more extensive study by future researchers.
Perhaps surprisingly, emotional ex-
haustion and depersonalization were no
greater among the students who participated
at the end of a semester versus those who
participated at the start of the following
semester. This contradicts the common
conception that burnout is greatest at the end
of a semester. Again however, the cor-
relational nature of this study prohibits any
firm conclusions. On the other hand, those
who participated at the beginning of a
semester scored higher on sense of accom-
plishment and reported greater conscien-
tiousness than did those participating at the
semester’s end. Participants were free to
participate at any point during the semester,
and it seems likely that there are important
personality differences between students
who pursue research participation at the very
beginning of a semester versus those who
procrastinate. This interpretation is sup-
ported by the research of Zelenski and
colleagues (Aviv, Zelenski, Rallo, & Larsen,
2002; Zelenski, Rusting, & Larsen, in press),
and it has important methodological rami-
fications for researchers who use university
participant pools.
Future Research
As with all research, the current study has
limitations that present opportunities for
future researchers. The current study was
conducted at a highly competitive, private,
primarily liberal arts college. The relations
of personality, social support, and workload
to burnout may be very different in other
academic settings, including less demanding
undergraduate settings and graduate and
professional schools. Furthermore, the data
were collected from only third- and fourth-
year undergraduate students, so the findings
may not apply to first- and second-year
college students or to high school students.
It seems clear from the findings that
college students, at least the population
sampled, experience substantial levels of
burnout. The current study is one of the first
explorations of burnout among college
students, and there is a great deal more to
learn. The consistent association between
social support and burnout is an interesting
finding that warrants further study. Does
burnout produce feelings of social isolation
and perceptions of inadequate social support,
or does social support directly moderate
burnout? Intervention studies, or at least
correlational studies with longitudinal
designs (e.g., Wade et al., 1986), are needed
to address the important question of whether
increased social support can actually reduce
or prevent burnout.
Further research is also needed to better
understand the relation between personality
and burnout. The GTS is a reliable and easily
administered measure that has potentially
great utility as both a research and a
diagnostic instrument. There are, however,
other more widely used measures of per-
sonality. Future researchers should not only
replicate these findings with the GTS but
also extend them to other measures of
MAY /JUNE 2003 XVOL 44 NO 3 301
Predictors of Student Burnout
personality.
We found that students who believe they
are over involved (high subjective workload)
experience greater burnout, but this is hardly
surprising. Workload and burnout were
measured simultaneously, and undoubtedly
most of the participants who experienced
high levels of burnout felt, perhaps after the
fact, that they were over committed. Longi-
tudinal research is important to track burnout
levels, as well as subjective and objective
workload, throughout the semester and even
from semester to semester.
Do students who experience burnout
work in less productive or less efficient ways
than students who do not experience burn-
out? Similarly, do burned out students
recreate in ways that contribute to stress or
detract from their work activities? Students
who experience burnout may have inefficient
methods of work and play that place them
at a disadvantage academically, socially, and
psychologically. Much can be learned by
systematically collecting behavioral mea-
sures of time management, including study
patterns, amount and kinds of recreation, and
employment activities.
Would providing students with detailed
information about burnout help them to
avoid burnout out at later points? Studyiing
the effects of providing information on
burnout and teaching time management
strategies during orientation for first-year
students could prove to be very valuable. In
a similar vein, the importance of social
support could be strongly emphasized to
parents, faculty and peer advisers, and
student support personnel, and the effects of
such an intervention on burnout could be
measured.
Implications for College Staff
Because we employed a correlational design
in the current study, caution must be taken
to avoid strictly causal interpretations of the
results. The results are, however, at least
consistent with causal influences of per-
sonality, social support, and subjective
workload on burnout. It is also important to
reemphasize that the current study employed
a nonrandom sample drawn from a single
college campus, and further studies are
needed before broadscale recommendations
can be made for staff intervention. None-
theless, our findings can provide several
ideas for consideration for college coun-
selors, RAs, academic advisors, and support
staff.
When counselors or advisors are faced
with a student who appears to be suffering
from burnout, it is important to recognize
that the student may be experiencing feelings
of depersonalization and reduced sense of
accomplishment, in addition to emotional
exhaustion. A commonsense prescription
might be to suggest that the student “lighten
the load” by dropping a course, cutting back
on extracurricular activities, spending less
time socializing with friends, or reducing
hours of employment. The results of this
study suggest that social support is a stronger
predictor of burnout than is workload, so
such advice may actually be very counter-
productive. Reducing extracurricular activi-
ties, or perhaps even hours of employment,
may reduce the student’s level of interaction
with supportive friends and thus exacerbate
burnout. Similarly, dropping a course might
be experienced as failure by some students
and thus contribute to a sense of reduced
personal accomplishment. A more effective
approach might be a thorough analysis of the
student’s weekly activity schedule and a
focus on effective time management strate-
gies. Efforts should be made to promote or
maintain important social relationships
302 Journal of College Student Development
Jacobs & Dodd
rather than to reduce extracurricular activi-
ties. Counselors involved in long-term
interventions with students should address
personality issues that might be directly
related to burnout.
What can friends or family members
do to help someone who experiencing
burnout?
Many college students may view burnout as
merely an “occupational hazard” that does
not warrant professional intervention or even
social support. It may be possible to train
friends and parents to be alert for signs of
student burnout and to respond effectively.
Orientation sessions and information for
parents, particularly parents of incoming
first-year students, can address burnout and
related methods of coping. It seems rea-
sonable to suggest to parents that they advise
their students to balance reasonable aca-
demic and employment schedules with
meaningful extracurricular activities rather
than assuming a strictly “academics first”
stance. Similarly, students themselves,
especially those who may become involved
in peer support activities, should be educated
about the components of burnout and
possible coping strategies.
CONCLUSION
The current study demonstrates the impor-
tance of personality as a correlate of burnout
and also the value of social support, espe-
cially from friends. It also appears to be the
subjective experience of overload, rather
than actual objective workload, that contri-
butes to burnout. Extracurricular activities,
rather than draining energy from students,
might actually be associated with less
burnout.
Considerably more research among
college students is needed to understand how
daily activities and time management skills
may be related to burnout. Greatly needed
are investigations of the effects of including
information about burnout in orientation
programs for students and parents and in
training programs for peer and professional
counselors. By gaining a greater under-
standing of predictors of burnout in college
students and effective means for providing
social support to students experiencing
burnout, student development personnel can
create programs that reduce burnout and
promote greater academic and personal
fulfillment for students.
Correspondence concerning this article should be
addressed to David K. Dodd, Department of
Psychology, Campus Box 1125, Washington
University, St. Louis, MO 63130-4899; DDodd@
artsci.wustl.edu
REFERENCES
Abouserie, R. (1994). Sources and levels of stress in relation
to locus of control and self-esteem in university students.
Educational Psychology, 14, 323-330.
Aviv, A. L., Zelenski, J. M., Rallo, L., & Larsen, R. J. (2002).
Who comes when: Personality differences in early and
later participation in a university subject pool. Personality
and Individual Differences, 33, 487-496.
Benedict, J. O., & Mondloch, G. J. (1989). Factors affecting
burnout in paraprofessional residence hall staff members.
Journal of College Student Development, 30, 293-297.
Clark, L. A., & Watson, D. (1990). The General Tempera-
ment Survey. Unpublished manuscript, Southern Method-
ist University, Dallas, TX.
Clark, L. A. (1993). SNAP: Schedule for Nonadaptive and
Adaptive Personality: Manual for administration,
scoring, and interpretation. Minneapolis: University of
Minnesota Press.
Costa, P. T., & McRae, R. R. (1985). The NEO Personality
Inventory manual. Odessa, FL: Psychological Assessment
Resourses.
MAY /JUNE 2003 XVOL 44 NO 3 303
Predictors of Student Burnout
Dahlem, N. W., Zimet, G. D., & Walker, R. R. (1991). The
Multidimensional Scale of Perceived Social Support: A
confirmation study. Journal of Clinical Psychology, 47,
756-761.
Dwyer, A. L., & Cummings, A. L. (2001). Stress, self-
efficacy, social support and coping strategies in university
students. Canadian Journal of Counselling, 35, 208-220.
Eastburg, M. C., Williamson, M., Gorsuch, R., & Ridley,
C. (1994). Social support, personality, and burnout in
nurses. Journal of Applied Social Psychology, 24,
1233-1250.
Everson, H. T., Tobias, S., Hartman, H., & Gourgey, A.
(1993). Test anxiety and the curriculum: The subject
matters. Anxiety, Stress, and Coping, 6, 1-8.
Freudenberger, H. J. (1974). Staff burn-out. Journal of Social
Issues, 30, 159-165.
Fuehrer, A., & McGonagle, K. (1988). Individual and
situational factors as predictors of burnout among resident
assistants. Journal of College Student Development, 29,
244-249.
Greenglass, E. R., Burke, R. J., & Fiksenbaum, L. (2001).
Workload and burnout in nurses. Journal of Community
and Applied Social Psychology, 11, 211-215.
Greenglass, E. R., Fiksenbaum, L., & Burke, R. J. (1994).
The relationship between social support and burnout over
time in teachers. Journal of Social Behavior and
Personality, 9, 219-230.
Hardy, S. E., & Dodd, D. K. (1998). Burnout among
university resident assistants as a function of gender and
floor assignment. Journal of College Student Devel-
opment, 39, 499-501.
Hetherington, C., Oliver, M. K., & Phelps, C. E. (1989).
Resident assistant burnout: Factors of job and gender.
Journal of College Student Development, 30, 266-269.
Huebner, E. S. (1994). Relationships among demographics,
social support, job satisfaction and burnout among school
psychologists. School Psychology International, 15,
181-186.
Huebner, E. S., & Mills, L. B. (1994). Burnout in school
psychology: The contribution of personality charac-
teristics and role expectations. Special Services in the
Schools, 8, 53-67.
Kahill, S. (1986). Relationship of burnout among profes-
sional psychologists to professional expectations and
social support. Psychological Reports, 59, 1043-1051.
Koeske, G. F., & Koeske, R. D. (1989). Workload and
burnout: Can social support and perceived accom-
plishment help? Social Work, 34, 243-248.
Koniarek, J., & Dudek, B. (1996). Social support as a buffer
in the stress-burnout relationship. International Journal
of Stress Management, 3, 99-106.
Male, D. B., & May, D. S. (1997). Burnout and workload
in teachers of children with severe learning difficulties.
British Journal of Learning Disabilities, 25, 117-121.
Male, D.B. & May, D.S. (1998). Stress and health, workload
and burnout in learning support coordinators in colleges
of further education. Support for Learning, 13, 134-138.
Maslach, C. & Jackson, S. E. (1981). Maslach Burnout
Inventory (2nd ed.). Palo Alto, CA: Consulting Psy-
chologists Press.
Maslach, C., Jackson, S. E., & Leiter, M. P. (1997). Maslach
Burnout Inventory: Third edition. In C. P. Zalaquett &
R. J. Wood (Eds.), Evaluating stress: A book of resources.
Lanham, MD: Scarecrow Press.
Mills, L. B., & Huebner, E. S. (1998). A prospective study
of personality characteristics, occupational stressors, and
burnout among school psychology practitioners. Journal
of School Psychology, 36, 103-120.
Newby-Fraser, E., & Schlebusch, L. (1997). Social support,
self-efficacy and assertiveness as mediators of student
stress. Psychology: A Journal of Human Behavior, 34,
61-69.
Ross, R. R., Altmaier, E. M., & Russell, D. W. (1989). Job
stress, social support, and burnout among counseling
center staff. Journal of Counseling Psychology, 4,
464-670.
Russell, D. W., Altmaier, E., & Van Velzen, D. (1987). Job-
related stress, social support, and burnout among
classroom teachers. Journal of Applied Psychology, 72,
269-274.
Sand, G., & Miyazaki, A. D. (2000). The impact of social
support on salesperson burnout and burnout components.
Psychology and Marketing, 17, 13-26.
Sandoval, J. (1993). Personality and burnout among school
psychologists. Psychology in the Schools, 30, 321-326.
Sloboda, J. A. (1990). Combating examination stress among
university students: Action research in an institutional
context. British Journal of Guidance and Counseling, 18,
124-136.
Wade, D. C., Cooley, E., & Savicki, V. (1986). A longitudinal
study of burnout. Children and Youth Services Review,
8, 161-173.
Watson, D., Clark, L. A., & Harkness, A. R. (1994).
Structures of personality and their relevance to psycho-
pathology. Journal of Abnormal Psychology, 103, 18-31.
Weidner, G., Kohlmann, C. W., Dotzauer, E., & Burns, L.
R. (1996). The effects of academic stress on health
behaviors in young adults. Anxiety, Stress, and Coping,
9, 123-133.
Zelenski, J. M., Rusting, C. L., & Larsen, R. J. (in press).
Consistency in the time of experiment participation and
personality correlates: A methodological note. Personality
and Individual Differences.
Zellars, K. L., Perrewe, P. L., & Hochwarter, W. A. (2000).
Burnout in health care: The role of the five factors of
personality. Journal of Applied Social Psychology, 30,
1570-1598.
Zimet, G. D., Dahlem, N. W., Zimet, S. G., & Farley, G. K.
(1988). The Multidimensional Scale of Perceived Social
Support. Journal of Personality Assessment, 52, 30-41.
