LEAIND-01122; No of Pages 6

Learning and Individual Differences xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Learning and Individual Differences

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

School burnout: Diminished academic and cognitive performance

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

Article history: Two studies examined relationships between school burnout (school related strain and stress) and indicators of
Received 4 September 2014 academic and cognitive performance. Study 1 (N = 790) investigated school burnout and grade point average
Received in revised form 2 April 2015 over three consecutive academic semesters. Hierarchical multiple regression (HMR) findings demonstrated a
Accepted 24 July 2015
consistent, negative association between school burnout and academic performance. Study 2 (N = 331)
Available online xxxx
investigated school burnout and individual differences in cognitive functioning through the assessment of
Keywords:
problem solving (serial subtraction) and attentional/inhibition processes (word-color matching Stroop task).
Academic performance HMR results indicated that increased school burnout was related to diminished attentional capacity and problem
Cognitive functioning solving success. Limitations of previous school burnout investigations were addressed by extending sampling
School burnout into American universities and utilizing analyses that controlled for related affective symptoms. These studies
are the first to show that school burnout is related to diminished academic and cognitive performance in US
tertiary education. Several future lines of research are outlined.
© 2015 Elsevier Inc. All rights reserved.

The American College Health Association-National College Health in Study 1 and the relationship between school burnout and cognitive
Assessment II suggests that maladaptive affective functioning (i.e. de- functioning in Study 2.
pression, anxiety, and psychological stress) is a widespread impediment Applied to academic populations, school burnout is conceptualized as
to collegiate academic success across the US (ACHA-NCHA, 2007). Ac- a three-dimensional affective response to school-related stress charac-
cordingly, attention is being given to understanding and ameliorating terized by exhaustion (chronic exhaustion from school-related work),
psychological risk factors that decrease academic performance, reten- cynicism (cynicism toward the meaning of school) and inadequacy (a
tion, and that negatively impact mental and physical health, particularly belief of inadequacy in school related accomplishment, Salmela-Aro
stress, depressive symptoms, and anxiety symptoms (Eisenberg, et al., 2008; Salmela-Aro et al., 2009a; Salmela-Aro, Savolainen, &
Gollust, Golberstein, & Hefner, 2007; Hamaideh, 2011; Mowbray et al., Holopainen, 2009b). There is evidence to show that school burnout is as-
2006; Taylor, Bramoweth, Grieser, Tatum, & Roane, 2013). Although sociated with physiology predictive of cardiovascular risk (i.e. increased
stress, and depression/anxiety symptoms are important risk factors blood pressure, sympathetic activity to the blood vessels, and arterial
that may negatively impact academic success in college students, stiffness see May et al., 2014a; May, Sanchez-Gonzalez, & Fincham,
there is emerging evidence to show that school burnout (school related 2014) as well as psychological and behavioral problems such as depres-
strain and stress) may be a unique and independent predictor of aca- sion, absenteeism, school dropout, and academic underperformance
demic success (Salmela-Aro, Kiuru, Leskinen, & Nurmi, 2009; (Brown, May, Sanchez-Gonzalez, Koutnik, & Fincham, 2013; Fimian &
Salmela-Aro, Kiuru, Pietikäinen, & Jokela, 2008; Walburt, 2014) as well Cross, 1986; Frydenberg & Lewis, 2004; Salmela-Aro et al., 2009a;
as cardiovascular health (May, Sanchez-Gonzalez, Brown, Koutnik, & Salmela-Aro et al., 2009b; Salmela-Aro et al., 2008; Yang, 2004).
Fincham, 2014). However, the school burnout-academic performance (Salmela-Aro et al., 2008; Salmela-Aro et al., 2009a; Salmela-Aro
association has yet to be documented among US college students and et al., 2009b; Parker & Salmela-Aro, 2011) has been instrumental in es-
research on the potential impairment of cognitive processes that may tablishing the viability of investigating burnout within a school context
contribute to the relationship between school burnout and academic and has greatly advanced understanding of the relationship between
underperformance is greatly limited. Therefore, this research explored school burnout and educational outcomes; but major limitations are ap-
the relationship between school burnout and academic performance parent. For one, Salmela-Aro et al. (2008), Salmela-Aro et al. (2009a),
Salmela-Aro et al. (2009b), Parker and Salmela-Aro, (2011) utilized pre-
dominately high-school, European student samples. Extensive research
⁎ Corresponding author.
needs to be conducted to establish the utility of school burnout among
E-mail addresses: rossmay00@gmail.com, rmay@fsu.edu (R.W. May), American college students. Also, the independence of school burnout,
kbauer1013@gmail.com (K.N. Bauer), ffincham@fsu.edu (F.D. Fincham). in relation to other related affective problems, namely depression and

http://dx.doi.org/10.1016/j.lindif.2015.07.015
1041-6080/© 2015 Elsevier Inc. All rights reserved.

Please cite this article as: May, R.W., et al., School burnout: Diminished academic and cognitive performance, Learning and Individual Differences
(2015), http://dx.doi.org/10.1016/j.lindif.2015.07.015
2 R.W. May et al. / Learning and Individual Differences xxx (2015) xxx–xxx

anxiety, in predicting indicators of academic performance has not been (GPA) in Study 1 and with individual differences in cognitive function-
clearly established. ing via assessment of problem solving and attentional/inhibition pro-
Indeed, existing data indicate that 3 of the top 6 impediments to ac- cesses in Study 2. To address limitations of previous school burnout
ademic success are affective in nature (ACH Association, 2013). Our ini- investigations (as noted earlier) the current research extends study
tial pilot data, however, showed that even though burnout, anxiety, and sampling into American universities and uses analyses that control for
depression are related, school burnout uniquely predicted key academic related affective symptomatology (anxiety and depression).
achievement outcomes (grade point average and retention); accounting
for as much outcome variance as both anxiety and depression combined 1. Study 1
(Brown et al., 2013). Although it can be argued that burnout, depression,
and anxiety can conceptually be independent constructs, empirically Study 1 was conducted to document a relationship between school
burnout shares overlapping symptomatology with other affective burnout and indicators of academic performance in the context U.S. ter-
disorders. For example, Salmela-Aro et al. (2008), Salmela-Aro tiary education. Given evidence from European counterparts that burn-
et al. (2009a) reported correlations exceeding 0.50 between depression out can lead to lower academic performance (e.g., Salmela-Aro et al.,
scores and the SBI global and subscale scores. Work burnout researchers 2008; Salmela-Aro et al., 2009a), it is prudent to explore this relation-
note the need to control for depressive and anxiety symptoms in de- ship at American universities. Today's US college student is more con-
signs focusing specifically on burnout (Melamed, Shirom, Toker, nected and more involved than previous generations of college
Berliner, & Shapira, 2006; Schaufeli & Buunk, 2004; Shirom, 2009). students. Moreover, the current competitiveness of the job market
School burnout research similarly requires the control of other related adds a great amount of pressure for students to succeed academically —
affective symptoms in order to allow a clearer understanding of whether specifically to graduate with high GPAs. The level of involvement
it is burnout, depressive, or anxiety symptoms that are the principal coupled with the added pressure of the job market suggests that
factor that is associated with poor academic outcomes. The current re- American college students may be at particular risk for burnout. Under-
search seeks to address these limitations by investigating school burnout standing the phenomenon of school burnout will allow university edu-
in American universities and through utilizing statistical analyses that cators and administrators to better assist students.
account for related affective symptoms.
To date, research examining the relationship between school burn- 2. Study 1 method
out and cognitive functioning is scarce with cognitive performance indi-
cators limited solely to grade point average (GPA). However, in the 2.1. Participants
occupational literature, research has examined relationships between
cognition and workplace burnout. In contrast to traditional theoretical Three samples of undergraduate students served as study partici-
explanations involving either motivational deficits and/or a lack of re- pants. Students that completed at least 1 collegiate semester were eligi-
source reciprocity that attempt to account for the various negative rela- ble for study participation. Sample demographics include: N = 790 (505
tionships between burnout and indicators of job performance, the females, Mage = 19.74 years, SD = 1.89), 72% Caucasian, 18% African
cognition-workplace burnout literature suggests cognitive dysfunction American, 4.0% Asian, and 6% endorsed either biracial or non-disclosed
and impairments are key factors in understanding the negative work re- ethnicity; 19% Freshmen, 24% Sophomore, 26% Junior, and 31% Senior.
lated outcomes attributable to burnout (Diestel, Cosmar, & Schmidt,
2013; Oosterholt, Van der Linden, Maes, Verbraak, & Kompier, 2012; 2.2. Measures
van der Linden, Keijsers, Eling, & van Schaijk, 2005).
Empirical evidence derived from both self-evaluations of cognitive 2.2.1. School burnout
impairments and objective cognitive tests has identified burnout as School burnout was measured using the School Burnout Inventory
being related to chronic impairments on tasks requiring executive con- (SBI: Salmela-Aro et al., 2008; Salmela-Aro et al., 2009a). The SBI con-
trol. Executive control refers to the regulation of representational, atten- sists of 9 items measuring three first-order factors of school burnout:
tional and motor processes to adaptively engage in novel, complex and (a) exhaustion at school (four items), (b) cynicism toward the meaning
changing tasks. Such processes include working memory, verbal rea- of school (three items), and (c) sense of inadequacy at school (two
soning, task switching, cognitive flexibility, abstract thinking, inhibition, items). Summed scores from the first-order factors comprise a
sequencing, planning, rule acquisition, and problem-solving. Derived second-order overall school burnout score. All the items are rated on a
from the theoretical conceptualization of executive control developed 6-point Likert-type scale ranging from 0 (completely disagree) to 5
by Miyake et al. (2000) and supporting Hacker's (2003) Action Regula- (strongly agree). Higher composite scores indicate higher burnout. Reli-
tion Theory, that purports successful efficient goal-direct behavior at ability for the present sample was α = .93.
work involves effective executive control, studies have found executive
control predicts task performance (Causse, Dehais, & Pastor, 2011; Frese 2.2.2. Depression
& Zapf, 1994). Depression was measured using the 10-item Center for Epidemio-
The current research seeks to advance the school burnout literature logic Studies Depression Scale (CES-D; Radloff, 1977; Santor & Coyne,
by examining how school burnout is related to indicators of cognitive 1997). The CES-D has been widely used as a measure of depressive
functioning. This research utilizes two general cognitive tasks, a serial symptoms in nonclinical samples. It asks participants to respond to a
subtraction task and a word-color matching Stroop task. These tasks list of ways they may have felt or behaved during the previous week.
provide an assessment of general problem solving ability and general ef- Sample items include, “I was bothered by things that usually don't both-
ficiency of attentional/inhibition cognitive processes. er me,” and “I felt hopeful about the future,” (reverse coded). Responses
Taken together, prior studies and our own research, point toward ranged from 0 = rarely or none of the time (less than one day) to 3 =
the conclusion that school burnout is potentially a critical, but often most or all of the time (5–7 days). Responses were summed into one
underappreciated factor, impacting health, cognition and academic suc- overall score, with a possible range of 0 to 30. Reliability for the sample
cess in the undergraduate student body in American colleges. Disap- was α = .77.
pointingly, research on school burnout in American universities is
lacking and is not recognized in the NCHA II assessment. Therefore in- 2.2.3. Anxiety
vestigation of the construct of school burnout in American postsecond- Anxiety was measured using the 20-item State-Trait Anxiety Inven-
ary education contexts seems necessary and timely. Accordingly we tory (STAI; Spielberger, Gorsuch, & Lushene, 1970). Participants were
explored school burnout relationships with academic performance asked to respond to anxiety items such as “upset,” “calm,” “secure,” “at

Please cite this article as: May, R.W., et al., School burnout: Diminished academic and cognitive performance, Learning and Individual Differences
(2015), http://dx.doi.org/10.1016/j.lindif.2015.07.015
 R.W. May et al. / Learning and Individual Differences xxx (2015) xxx–xxx 3

ease,” and “nervous.” Responses were scored on a 4-point Likert scale 3. Study 1 results and discussion
(1 = Not at all to 4 = Very much so). Half of the items were reverse
coded so that higher scores indicated greater anxiety. Items were then ANOVA analyses indicated that neither ethnicity F(3, 786) = 1.82,
summed to create a composite anxiety score with a possible range of p = .142, partial η2 = .022; gender F(1, 788) = 0.83, p = .363, partial
20 to 80. Reliability for the sample was α = .89. η2 = .005; nor year in school F(3, 786) = 0.74, p = .528, partial η2 =
.010 were associated with school burnout scores. Multiple regression
analyses indicated the neither ethnicity, gender, nor year in school sig-
2.2.4. Academic achievement nificantly moderated the relationship between school burnout and
Academic achievement was assessed through self-reported under- GPA (F's b 1, p N .05). Model 2 of the hierarchical regression analyses
graduate, cumulative GPA. GPA ranged from 1.50 to 4.00. showed that, after accounting for anxiety and depressive symptoms,
school burnout scores (p b .05) accounted for an additional 4% of vari-
ance in GPA values during the fall semester (see Table 1). Similarly,
2.3. Procedure school burnout scores significantly accounted for an additional 5% and
6% of GPA variance for spring and summer semesters respectively
Data collection from all eligible participants was completed via on- (Table 1). These results represent the first findings to demonstrate a
line questionnaires. Questionnaires contained demographic questions consistent, negative association between school burnout scores and
and the measurement scales described. All participants were recruited GPA while controlling for anxiety and depressive symptoms in an un-
from undergraduate classrooms as an option for voluntary class credit. dergraduate American sample.
Data for the fall (Sample 1) and spring (Sample 2) semesters were col-
lected in the middle (weeks 3–9) of the respective semester. Summer 4. Study 2
semester lasted only 6 weeks; thus, data for this semester was collected
between the 2nd and 5th weeks. All participants gave their written con- Study 1 demonstrated a relationship between school burnout and
sent prior to study participation and approval was obtained from the in- GPA and Study 2 investigates whether school burnout is related to
stitutional review board before any data were collected. more general cognitive functions. Thus it explores the relationship
between school burnout and indicators of individual differences in
cognitive functioning. In this study, a classic color-word matching Stroop
2.4. Statistical analysis task serves as a measure of general attentional/inhibition processes and
a serial subtraction task serves as a general indicator of problem solving
Univariate analysis of variance (ANOVA) evaluated ethnicity, gen- performance. Understanding the relationship between school burnout
der, and year in school associations with school burnout. Exploratory and these indicators of cognitive functioning will begin to unravel the
multiple regression analyses explored whether demographics (ethnici- potential mechanisms underlying the negative relationship between
ty, gender, and year in school) moderated the relationship between school burnout and academic performance.
school burnout and GPA. As affective disorders may have overlapping
symptomatology, investigators suggest the need to control for depres- 5. Study 2 methods
sive and anxiety symptoms in designs focusing on burnout measure-
ments (Melamed et al., 2006; Schaufeli & Buunk, 2004; Shirom, 2009). 5.1. Participants
Therefore hierarchical multiple regression (HMR) analyses were con-
structed to demonstrate the incremental contribution of school burnout Three hundred thirty one undergraduate students (N = 257 fe-
above that of anxiety and depressive symptoms in accounting for vari- males, Mage — 19.10 years, SD = 1.92) that completed at least 1 colle-
ance in GPA. A hierarchical multiple regression (HMR) analysis was con- giate semester were eligible for study participation. Annual family
ducted on each cross sectional data wave, therefore three different HMR income reported for the sample indicated 13% grossed b$30,000; 22%
analyses are reported. Model 1 of the HMR contained the anxiety and grossed $30,001 to $50,000, 33% grossed $50,001 to $100,000, and 32%
depression predictors and Model 2 introduced school burnout as a pre- grossed N$100,001. Reported academic major indicated 34% Biological
dictor. Listwise deletion was conducted for missing data on measure- Sciences (e.g. Biology, Exercise Science), 41% Social Sciences (e.g. Psy-
ment scales, occurring for 2.6% (21 out of 811) of all cases in this chology, Sociology, Communication Sciences, Criminology, Education),
sample, resulting in 790 complete data cases. and 25% Miscellaneous (e.g. Music, Information Technology, Education).

Table 1
Hierarchal multiple regression of depression, anxiety, and school burnout scores accounting for variance in undergraduate GPA cross-sectionally over 3 semesters.

Criterion (M, SD) Step Predictors (M, SD) β sr p R2 ΔR2 Model F

Fall semester S1 STAI (18.02, 4.11) −.13 −.12 .001 .13 F(2,432) = 27.91, p b .001
GPA (3.31, 0.39) CES-D (9.21, 5.12) −.28 −.26 b.001
S2 STAI −.10 −.09 .045 .17 .04 ΔF(1, 431) = 18.67, p b .001
CES-D −.16 −.13 .005
N = 435 SBI (17.11, 6.95) −.24 −.20 b.001
Spring semester S1 STAI (17.55, 4.71) −.12 −.11 .101 .14 F(2,202) = 16.99, p b .001
GPA (3.29, 0.41) CES-D (9.16, 5.42) −.31 −.28 b.001
S2 STAI −.09 −.08 .232 .19 .05 ΔF(1, 201) = 11.35, p b .001
CES-D −.17 −.14 .035
N = 205 SBI (17.53, 7.25) −.27 −.21 .001
Summer semester S1 STAI (18.46, 4.89) −.01 −.01 .929 .07 F(2, 147) = 5.26, p = .006
GPA (3.14, 0.66) CES-D (9.66, 5.71) −.26 −.25 .002
S2 STAI −.05 −.05 .528 .12 .06 ΔF(1, 146) = 9.25, p = .003
CES-D −.16 −.14 .069
N = 150 SBI (17.01, 7.45) −.27 −.24 .003

Note. sr, semi-partial correlation; GPA, grade point average; CES-D, Center for Epidemiologic Studies Depression Scale; STAI, State-Trait Anxiety Inventory; SBI, School Burnout Inventory.

Please cite this article as: May, R.W., et al., School burnout: Diminished academic and cognitive performance, Learning and Individual Differences
(2015), http://dx.doi.org/10.1016/j.lindif.2015.07.015
4 R.W. May et al. / Learning and Individual Differences xxx (2015) xxx–xxx

Additional sample demographics include: 73% Caucasian, 18% African- task (computation attempts, computation errors) outcomes. Further-
American, 4% Asian, and 5% endorsed either biracial or non-disclosed more, the HMR provides an evaluation of the incremental contribution
ethnicity; 20.5% Freshmen, 21.5% Sophomore, 30% Junior, and 28% of school burnout scores above anxiety and depressive symptomatology
Senior. in accounting of variance in the Stroop and serial subtraction task out-
comes. Listwise deletion was conducted for missing data on measure-
5.2. Measures ment scales, occurring for 0.9% (3 out of 334) of all cases in this
sample, resulting in 331 complete data cases. There were no missing
5.2.1. Measurement scales data resulting from Stroop and subtraction task measurement.
As in Study 1, school burnout was measured using the SBI, depres-
sion the CES-D, and anxiety the STAI with sample α of .93, .77, and 6. Study 2 results and discussion
.91, respectively.
As in Study 1, ANOVA analyses indicated that neither ethnicity F(3,
5.2.2. Stroop task 327) = 2.18, p = .090, partial η2 = .019; gender F(1, 329) = 0.38,
The Stroop task comprised a series of color words, each of which was p = .538, partial η2 = .002; year in school F(3, 327) = 1.63, p = .182,
presented in a color that either matched (congruent) or did not match partial η2 = .027; annual family income F(3, 327) = 1.71, p = .165, par-
(incongruent) the semantic meaning of the word. Participants were tial η2 = .027; nor academic major F(2, 328) = 0.30, p = .738, partial
instructed to identify the color of each word presented by clicking on η2 = .005 were associated with school burnout scores. Multiple regres-
a computer keyboard key that was colored (red, blue, yellow, green). sion analyses indicated that demographics (ethnicity, gender, year in
A trial consisted of a fixation cross “+” presented for 500 ms, followed school, family annual income, academic major) did not significantly
by the stimulus word presented for 200 ms. Participants completed 4 moderate the relationship between school burnout and the Stroop or
blocks, each consisting of 25 congruent trials and 25 incongruent trials. subtraction task outcomes (F's b 1, p N .05). HMR analyses of the Stroop
Two indices were recorded from this Stroop task: a summed score of the task outcomes indicate that after accounting for anxiety and depressive
total number of errors in color matching and response time latency in symptoms in Model 2, school burnout scores accounted for an addition-
color identification. al 4% of variance of congruency matches (p b .05) and an additional 9%
in variance of response time latency (p b .05; see Table 2). HMR analyses
5.2.3. Serial subtraction task of the serial subtraction task outcomes indicated that after accounting
This task was a five minute serial subtraction arithmetic task con- for anxiety and depressive symptoms in Model 2, school burnout scores
ducted through the DirectRT computer program. An instruction screen significantly accounted for an additional 6% of variance in computation
informed participants that the task was an arithmetic task in which attempts (p b .05) and an additional 3% in variance in computation er-
they would be asked to subtract 7 from a randomly selected number. rors (p b .05; see Table 2). In other words, while controlling for anxiety
Participants were not told there was a time limit of 5 min in order to and depression scores, the HMR analyses demonstrated (1) significant
eliminate time pressure as a potential confound. A practice tutorial associations between higher composite SBI scores and greater congru-
was conducted prior to data collection trials which demonstrated how ency matching errors and increased response time matching latencies
a number would appear (e.g. 1107) and how the correct answer during the Stroop task and (2) significant associations between higher
(1100) would be accepted through keystroke response and be used as composite SBI scores and a greater amount of computation errors and
the base number for the next subtraction trial. Five minutes after the solution attempts during the serial subtraction arithmetic task.
testing phase began the program ended. Two indices were collected
for analyses: the total number of computation attempts and a frequency 7. General discussion
count of the total number of computation errors.
The current studies aimed to extend school burnout research by ex-
5.3. Procedure amining the impact of burnout on academic performance (i.e., GPA) and
cognitive functioning among American university students. Together,
After completing an online questionnaire consisting of demo- the results demonstrated a consistent negative relationship between
graphics and measurement scales, eligible participants were scheduled school burnout and GPA (Study 1) as well as diminished cognitive func-
for an appointment to complete a laboratory session. The laboratory tioning (Study 2). In the remainder of the discussion, we explore these
session was comprised of the Stroop color-naming task with 100 con- findings in more detail, provide recommendations for research and
gruent and 100 incongruent trials and the five minute serial subtraction practice, and note limitations.
arithmetic task. All participants completed the serial subtraction task Consistent with prior research among European students
and Stroop task in one experimental session. Experimental task (e.g., Salmela-Aro et al., 2009a; Salmela-Aro et al., 2009b; Salmela-Aro
presentation was randomized. All participants were recruited from et al., 2008), these data suggest that increased school burnout as mea-
undergraduate classrooms as an option for voluntary class credit and sured by the School Burnout Inventory predicts less academic success.
all data was collected in the middle (weeks 3–9) of the fall semester Specifically, data collected in three semesters indicate that school burn-
the year following Study 1. All participants gave their written consent out is negatively related to concurrent GPA after controlling for depres-
prior to study participation as approved by The Florida State University sive and anxiety symptoms. To the authors' knowledge, this study is the
Institutional Review Board. first to report such a relationship in an American college student
sample.
5.4. Statistical analysis Moreover, demographic associations were not significantly related
to school burnout scores nor did they moderate the relationship be-
As conducted in Study 1, ANOVA and hierarchical multiple regres- tween school burnout and the indicators of academic and cognitive per-
sions were utilized in Study 2. ANOVA evaluated demographics (ethnic- formance measured in this research. The gender equality of burnout in
ity, gender, year in school, annual family income, and academic major) the present research deserves further consideration as this finding is
associations with school burnout. Exploratory multiple regression anal- in contrast to the robust gender differences (girls report greater burnout
yses explored whether demographics moderated the relationship be- than boys) found in the current adolescent school burnout research
tween school burnout and the Stroop and subtraction task outcomes. (see Walburt, 2014). However, it should be noted that our previous
HMR analyses tested the association between school burnout scores school burnout research using undergraduate student samples have
and Stroop (congruency errors, response latency) and serial subtraction demonstrated that increases in school burnout are associated with

Please cite this article as: May, R.W., et al., School burnout: Diminished academic and cognitive performance, Learning and Individual Differences
(2015), http://dx.doi.org/10.1016/j.lindif.2015.07.015
 R.W. May et al. / Learning and Individual Differences xxx (2015) xxx–xxx 5

Table 2
Hierarchal multiple regression of depression, anxiety, and school burnout scores accounting for variance in congruency errors, response time, computation errors, and computation
attempts.

Criterion (M, SD) Step Predictors (M, SD) β sr p R2 ΔR2 Model F

Congruency errors S1 STAI (19.11, 4.02) .12 .11 .097 .04 F(2, 328) = 4.62, p = .011
CES-D (8.84, 5.01) .11 .10 .112
S2 STAI .05 .04 .506 .08 .04 ΔF(1, 327) = 9.29, p b .002
CES-D .08 .07 .273
SBI (17.39, 6.95) .21 .19 .002
Response time S1 STAI .18 .16 .012 .03 F(2, 328) = 3.59, p = .029
CES-D .02 .02 .797
S2 STAI .07 .06 .313 .12 .09 ΔF(1, 327) = 23.59, p b .001
CES-D .07 .06 .289
SBI .33 .30 b.001
Compute errors S1 STAI .03 .03 .692 .01 F(2, 328) = 1.72, p b .181
CES-D .11 .11 .180
S2 STAI .04 .04 .560 .05 .03 ΔF(1, 327) = 7.62, p b .006
CES-D .05 .05 .454
SBI .21 .18 .006
Compute attempts S1 STAI .16 .14 .026 .05 F(2, 328) = 6.24, p = .002
CES-D .11 .10 .130
S2 STAI .07 .06 .337 .11 .06 ΔF(1, 327) = 15.92, p b .001
CES-D .06 .06 .374
SBI .27 .25 b.001

Note. N = 331. Compute; computation. Congruency errors (M = 7.23, SD = 3.74), response time (M = 603.17, SD = 76.34), computation errors (M = 0.80, SD = 1.31), computation
attempts (M = 17.07, SD = 9.407) as criterion, respectively. sr, semi-partial correlation; CES-D, Center for Epidemiologic Studies Depression Scale; STAI, State-Trait Anxiety Inventory;
SBI, School Burnout Inventory.

poorer cardiovascular function in both males and females (May et al., utilizing validated cognitive performance tasks to determine both the
2014a; 2014b). potential cognitive mechanisms affected by school burnout and the ca-
Importantly, the results demonstrated the negative effects of burnout sual relationships that may exist between school burnout and cognitive
are above and beyond those associated with more commonly assessed functioning. For example, the cognition-workplace literature indicates
maladaptive affective functioning (anxiety and depression). Given that that the strongest and most replicated relationship exists between
cross-lagged longitudinal studies of adolescents indicate school burnout burnout and working memory updating, especially in contexts where
predicts subsequent depressive symptoms, accounting for sources of executive control has been depleted by high performance demands
negative affect (e.g. depression, anxiety) is essential to help reveal the (Diestel et al., 2013; Oosterholt et al., 2012; van der Linden et al.,
unique deleterious relationship between school burnout and study out- 2005). Thus investigations into working memory would be a promising
comes of interest (Salmela-Aro et al., 2009b). As suggested by work burn- avenue for future school burnout research.
out research, doing so allows for a clearer understanding of whether it is Although our research did not indicate school burnout score differ-
burnout, depressive, or anxiety symptoms that are the principal factor ences between cohort years, research demonstrates that the symptom-
driving study results (Melamed et al., 2006; Schaufeli & Buunk, 2004; atology associated with work burnout can transfer both with and
Shirom, 2009). University educators and mental health professionals without direct or close contact among employees (Bakker, Demerouti,
should be informed of the negative impact of school burnout and provid- & Schaufel, 2006; Gonzalez-Morales, Peiro, Rodriguez, & Bliese, 2012).
ed resources for helping students exhibiting symptoms of burnout. Within organizational settings, it appears that perceived collective
Furthermore, these findings are the first to report any relationships burnout emerges as an organizational-level construct (employees'
between school burnout scores and measures of cognitive performance; shared perceptions about how burned out are their colleagues) and
specifically a measure of problem solving and a measure of general at- that this perceived collective burnout predicts individual burnout over
tention and inhibition. Results showed that school burnout was posi- and above indicators of work demands and resources. This suggests
tively related to the number of attempts as well as errors on a serial that perceived collective burnout is an important characteristic of the
subtraction arithmetic task and associated with increased congruency work environment that can be a significant factor in the development
error rates and response latency on a traditional word-color Stroop of burnout. Similarly, school burnout may be socially contagious
task. These findings provide support for the view that school burnout (Salmela-Aro et al., 2009b). While our research did not indicate school
is associated with diminished performance and with global processes burnout differences between academic majors, continued research
of cognitive functioning. into more subtle differences between academic majors as a possible an-
tecedent or predictor of school burnout transmission may prove fruitful
7.1. Limitations and directions for future research (see the distinction between hard and soft science majors used by May
& Casazza, 2012). Indeed, research indicates that certain academic ma-
The primary limitation of both studies is the cross-sectional nature jors (medical students) suffer from greater burnout prevalence
of the data. Although longitudinal research indicates the persistence of (Mazurkiewicz, Korenstein, Fallar, & Ripp, 2012; Santen, Holt, Kemp, &
school burnout over time during high school education in adolescence Hemphill, 2010). Understanding the transmission of school burnout
populations, longitudinal research of school burnout at the undergradu- will give clues to its etiology as well as inform potential interventions
ate level for emerging adulthood populations lacks sufficient investiga- aimed at ameliorating its deleterious influence on cognitive and aca-
tion (see Parker & Salmela-Aro, 2011; Salmela-Aro et al., 2009b). demic performance.
Although the present design precludes casual interpretation, the find-
ings warrant investigation of the longitudinal relations between school
burnout, cognition, and academic performance to determine direction References
of effects.
American College Health Association (2007). American College Health Association - National
Given previous cognition-workplace burnout research and our data, College Health Assessment (ACHA-NCHA) Web Summary. Updated August 2007.
it is imperative that prospective school burnout research continue Available at http://www.acha-ncha.org/data_highlights.html.

Please cite this article as: May, R.W., et al., School burnout: Diminished academic and cognitive performance, Learning and Individual Differences
(2015), http://dx.doi.org/10.1016/j.lindif.2015.07.015
6 R.W. May et al. / Learning and Individual Differences xxx (2015) xxx–xxx

ACH Association (2013). American College Health Association-National College Health Miyake, A., Friedman, N. P., Emerson, M. J., Witzki, A. H., Howerter, A., & Wager, T. D. (2000).
Assessment II: Reference Group Executive Summary Spring 2013. Hanover, MD: The unity and diversity of executive functions and their contributions to complex
American College Health Association. “Frontal Lobe” tasks: A latent variable analysis. Cognitive Psychology, 41, 49–100.
Bakker, A. B., Demerouti, E., & Schaufel, W. (2006). Crossover of burnout and engagement Mowbray, C. T., Megivern, D., Mandiberg, J. M., Strauss, S., Stein, C. H., Collins, K., et al.
in work teams. Work and Occupations, 33, 464–489. (2006). Campus mental health services: Recommendations for change. The
Brown, P. C., May, R. W., Sanchez-Gonzalez, M. A., Koutnik, A. P., & Fincham, F. D. (2013, American Journal of Orthopsychiatry, 76, 226–237.
May). The relationship of school burnout, anxiety, and depressive symptoms with ac- Oosterholt, B. G., Van der Linden, D., Maes, J. H., Verbraak, M. J., & Kompier, M. A. (2012).
ademic performance. Poster session presented at the 25th Annual Convention of the As- Burned out cognition–cognitive functioning of burnout patients before and after a pe-
sociation for Psychological Science, Washington, DC. riod with psychological treatment. Scandinavian Journal of Work, Environment &
Causse, M., Dehais, F., & Pastor, J. (2011). Executive functions and pilot characteristics pre- Health, 38, 358–369.
dict flight simulator performance in general aviation pilots. The International Journal Parker, P. D., & Salmela-Aro, K. (2011). Developmental processes in school burnout: A
of Aviation Psychology, 21, 217–234. comparison of major developmental models. Learning and Individual Differences, 21,
Diestel, S., Cosmar, M., & Schmidt, K. -H. (2013). Burnout and impaired cognitive function- 244–248.
ing: The role of executive control in the performance of cognitive tasks. Work & Stress, Radloff, L. S. (1977). The CES-D Scale: A self-report depression scale for research in the
27, 164–180. general population. Applied Psychological Measurement, 1, 385–401.
Eisenberg, D., Gollust, S. E., Golberstein, E., & Hefner, J. L. (2007). Prevalence and correlates Salmela-Aro, K., Kiuru, N., Leskinen, E., & Nurmi, J. E. (2009a). School burnout inventory
of depression, anxiety, and suicidality among university students. The American (SBI): Reliability and validity. European Journal of Psychological Assessment, 25, 48–57.
Journal of Orthopsychiatry, 77, 534–542. Salmela-Aro, K., Kiuru, N., Pietikäinen, M., & Jokela, J. (2008). Does school matter? The role
Fimian, M., & Cross, A. (1986). Stress and burnout among preadolescent and early adoles- of school context in adolescents' school-related burnout. European Psychologist, 13(1),
cent gifted students: A preliminary investigation. The Journal of Early Adolescence, 6, 12–23.
247–267. Salmela-Aro, K., Savolainen, H., & Holopainen, L. (2009b). Depressive symptoms and
Frese, M., & Zapf, D. (1994). Action as the core of work psychology: A German approach. school burnout during adolescence: Evidence from two cross-lagged longitudinal
In H. C. Triandis, M. D. Dunnette, & L. M. Hough (Eds.), Handbook of industrial and studies. Journal of Youth and Adolescence, 38, 1316–1327.
organizational psychology (pp. 271–340). Palo Alto, CA: Consulting Psychologists Press. Santen, S. A., Holt, D. B., Kemp, J. D., & Hemphill, R. R. (2010). Burnout in medical students:
Frydenberg, E., & Lewis, R. (2004). Adolescents least able to cope: How do they respond to Examining the prevalence and associated factors. Southern Medical Journal, 103(8),
their stresses? British Journal of Guidance and Counselling, 32, 25–37. 758–763.
Gonzalez-Morales, M. G., Peiro, J. M., Rodriguez, I., & Bliese, P. D. (2012). Perceived collec- Santor, D. A., & Coyne, J. C. (1997). Shortening the CES-D to improve its ability to detect
tive burnout: A multilevel explanation of burnout. Anxiety, Stress, & Coping, 25, 43–61. cases of depression. Psychological Assessment, 9(3), 233–243.
Hacker, W. (2003). Action regulation theory: A practical tool for the design of modern Schaufeli, W. B., & Buunk, B. P. (2004). Burnout: An overview of 25 years of research and
work processes? European Journal of Work and Organizational Psychology, 12, theorizing. In M. J. Schabracq, J. A. M. Winnubst, & C. C. Cooper (Eds.), The handbook of
105–130. work and health psychology (pp. 383–429) (2nd ed.). West Sussex, England: Wiley.
Hamaideh, S. H. (2011). Stressors and reactions to stressors among university students. Shirom, A. (2009). Epilogue: Mapping future research on burnout and health. Stress and
The International Journal of Social Psychiatry, 57(1), 69–80. Health, 25, 375–380.
May, R. W., & Casazza, S. P. (2012). Academic major as a perceived stress indicator: Ex- Spielberger, C. D., Gorsuch, R. L., & Lushene, R. E. (1970). Manual for the state-trait anxiety
tending stress management intervention. College Student Journal, 46(2), 264–273. inventory. Palo Alto, CA: Consulting Psychologists Press.
May, R. W., Sanchez-Gonzalez, M. A., Brown, P. C., Koutnik, A. P., & Fincham, F. D. (2014a). Taylor, D. J., Bramoweth, A. D., Grieser, E. A., Tatum, J. I., & Roane, B. M. (2013). Epidemi-
School burnout and cardiovascular functioning in young adult males: A hemodynam- ology of insomnia in college students: relationship with mental health, quality of life,
ic perspective. Stress, 1, 79–87. and substance use difficulties. Behavior Therapy, 44, 339–348.
May, R. W., Sanchez-Gonzalez, M. A., & Fincham, F. D. (2014b). School burnout: Increased van der Linden, D., Keijsers, G. J. P., Eling, P., & van Schaijk, R. (2005). Work-related stress
sympathetic vasomotor tone and attenuated ambulatory diurnal blood pressure var- and attention to action: An initial study on burnout and executive control. Work &
iability in young adult women. Stress, 1, 1–9. Stress, 19, 1–14.
Mazurkiewicz, R., Korenstein, D., Fallar, R., & Ripp, J. (2012). The prevalence and correla- Walburt, V. (2014). Burnout among high school students: A literature review. Children
tions of medical student burnout in the pre-clinical years: A cross-sectional study. and Youth Services Review, 42, 28–33.
Psychology, Health & Medicine, 17(2), 188–195. Yang, H. (2004). Factors affecting student burnout and academic achievement in multiple
Melamed, S., Shirom, A., Toker, S., Berliner, S., & Shapira, I. (2006). Burnout and risk of enrollment programs in Taiwan's technical–vocational colleges. International Journal
cardiovascular disease: Evidence, possible causal paths, and promising research di- of Educational Development, 24, 283–301.
rections. Psychological Bulletin, 132, 327–353.

Please cite this article as: May, R.W., et al., School burnout: Diminished academic and cognitive performance, Learning and Individual Differences
(2015), http://dx.doi.org/10.1016/j.lindif.2015.07.015