Content uploaded by Mary A Steinhardt
Author content
All content in this area was uploaded by Mary A Steinhardt on Mar 04, 2018
Content may be subject to copyright.
RESEARCH ARTICLE
Relationships Among Positive Emotions, Coping,
Resilience and Mental Health
‡
Christian T. Gloria
1
& Mary A. Steinhardt
2
*
†
1
Department of Health Sciences, Hawaii Pacific University, Kaneohe, HI, USA
2
Department of Kinesiology and Health Education, The University of Texas at Austin, Austin, TX, USA
Abstract
The broaden-and-build theory of positive emotions suggests that positive emotions can widen the range of potential
coping strategies that come to mind and subsequently enhance one’s resilience against stress. Studies have shown that
high stress, especially chronic levels of stress, strongly contributes to the development of anxiety and depressive
symptoms. However, researchers have also found that individuals who possess high levels of resilience are protected
from stress and thus report lower levels of anxiety and depressive symptoms. Using a sample of 200 postdoctoral
research fellows, the present study examined if (a) positive emotions were associated with greater resilience, (b) coping
strategies mediated the link between positive emotions and resilience and (c) resilience moderated the influence of
stress on trait anxiety and depressive symptoms. Results support the broaden-and-build theory in that positive
emotions may enhance resilience directly as well as indirectly through the mediating role of coping strategies—particularly
via adaptive coping. Resilience also moderated the association of stress with trait anxiety and depressive symptoms.
Although stress is unavoidable and its influences on anxiety and depressive symptoms are undeniable, the likelihood of
postdocs developing anxiety or depressive symptoms may be reduced by implementing programmes designed to increase
positive emotions, adaptive coping strategies and resilience. Copyright © 2014 John Wiley & Sons, Ltd.
Received 13 October 2013; Revised 7 April 2014; Accepted 11 May 2014
Keywords
stress; anxiety; depression; postdoc; mediation; moderation
*Correspondence
Mary A. Steinhardt, Department of Kinesiology and Health Education, The University of Texas at Austin, 1 University Station, D3700, Austin,
TX 78712, USA.
†
Email: msteinhardt@austin.utexas.edu
‡
This article was published online on 24 June 2014. Errors were subsequently identified. This notice is included in the online version to
indicate that it has been corrected [20 August 2014].
Published online 24 June 2014 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/smi.2589
Introduction
Growing research guided by the broaden-and-build
theory of positive emotions continues to find evidence
in support of the notion that positive emotions have
the ability to widen the range of potential coping
strategies that come to one’s mind during times of
stress, consequently enhancing one’s resilience against
present and future adversity (Folkman & Moskowitz,
2000; Fredrickson, 2004, 2005; Tugade, Fredrickson,
& Feldman Barrett, 2004). According to Fredrickson
(2001), the experience of positive emotions unlocks
the human cognition and encourages individuals to
think more freely, thoughtfully and creatively. These
effects, in turn, expand one’s outlook and capacity to
see the world with a broader perspective. As a result,
those who experience greater frequencies of positive
emotions have an improved ability to recognize a wider
range of possible coping strategies when faced with
adversity; thus, they are able to tackle stress more
effectively and achieve higher levels of resilience (Cohn,
Fredrickson, Brown, Mikels, & Conway, 2009;
Fredrickson, 2009; Gloria, Faulk, & Steinhardt, 2013).
In their theory of stress and coping, Lazarus and
Folkman (1984) define stress as a transactional process
between the person and the environment, whereby the
individual appraises the environmental demands as
outweighing his or her ability to meet those demands.
During stressful situations, the mind and body instinc-
tively trigger the fight-or-flight response in an effort to
diminish threat, harm or loss. This stress response
activates an array of physiological and psychological
reactions including increased heart rate, blood pressure
and respiration. In addition, one’s mindset and world-
view dramatically narrows and sharply focuses toward
the triggering stressor (Kok, Catalino, & Fredrickson,
2008). These evolutionarily adaptive reactions serve
critical purposes for survival, particularly during
145Stress and Health 32: 145–156 (2016) © 2014 John Wiley & Sons, Ltd.
threatening and stressful situations (Fredrickson,
2001). Although the fight-or-flight response can be
beneficial toward acute stressors, long-term and
chronic exposure is harmful to health. When individ-
uals are exposed to enduring forms of stress, the
physiological and psychological reactions of the fight-
or-flight response are likewise sustained over time. This
heightened and long-lasting state of stress beyond
homeostasis increases one’s risks for a number of seri-
ous health problems including cardiovascular disease
(Iso et al., 2002; Jood, Redfors, Rosengren, Blomstrand,
& Jern, 2009), obesity (de Luca & Olefsky, 2006),
diabetes (Lloyd, Smith, & Weinger, 2005), immune dis-
ease (Kemeny & Schedlowski, 2007), burnout (Lloyd,
King, & Chenoweth, 2002), anxiety (Kleppa, Sanne, &
Tell, 2008) and depression (Nielsen, Kristensen,
Schnohr, & Grønbæk, 2008; Steinhardt, Smith Jaggars,
Faulk, & Gloria, 2011).
Anxiety and depressive symptoms are particular
concerns, considering their strong associations with
stress (Newbury-Birch & Kamali, 2001; Rawson,
Bloomer, & Kendall, 1994). Studies have shown that
high levels of stress significantly contribute to the
development of subclinical symptoms of anxiety and
depression (Kleppa et al., 2008; Markou & Cryan,
2012; Melchior et al., 2007; Misra & McKean, 2000).
Anxiety and depressive symptoms can be debilitating,
not only harming the individual experiencing such
symptoms but also negatively impacting others. Those
who suffer from anxiety and depressive symptoms have
increased morbidity and mortality risks (Carney &
Freedland, 2003; Mykletun et al., 2007) and are likely
to have deteriorating interpersonal relationships (Insel
& Roth, 2012). They also have higher rates of absentee-
ism in tandem with decreased productivity at the work-
place (Stewart, Ricci, Chee, Hahn, & Morganstein, 2003).
However, there is evidence showing that individual
resilience can moderate the impact of stress on anxiety
and depressive symptoms (Aroian & Norris, 2000;
Pinquart, 2009; Wagnild, 2003; Wingo et al., 2010).
Studies found that stress had a weaker influence toward
the health of individuals who possessed higher levels of
resilience. Researchers have also demonstrated that
positive emotions can improve one’s ability to cope
with stress (Burns et al., 2008), and improved coping
subsequently enhances resilience (Tugade et al., 2004).
Cognitive-behavioural coping strategies focus on
identifying and changing the maladaptive thinking
and behaviour that create stress in an effort to prevent
or diminish threat, harm or loss (Lazarus, 1993;
Lazarus & Folkman, 1984). Adaptive coping strategies
(e.g. active coping, planning and positive reframing)
are actions and behaviours used in response to stress,
which lead to improved outcomes. In contrast,
maladaptive coping strategies (e.g. denial, venting and
substance abuse) often result in undesirable conse-
quences (Brown, Westbrook, & Challagalla, 2005;
Carver, 1997; Zeidner & Saklofske, 1996). Although
positive emotions can improve one’s ability to cope
with stress, to our knowledge, no studies have investi-
gated how positive emotions influence different types
of coping strategies—specifically, adaptive and
maladaptive. Furthermore, if one’s coping strategies
mediate the relationship between positive emotions and
resilience, such that positive emotions are positively
related to adaptive coping strategies and resilience and
inversely related to maladaptive coping strategies, it
would provide support for the broaden-and-build theory
of positive emotions (Fredrickson, 2004, 2005).
Although previous works in this field have studied a
variety of highly stressed populations, including college
students (Fredrickson, Tugade, Waugh, & Larkin,
2003), public school teachers (Steinhardt et al., 2011),
doctors (Newbury-Birch & Kamali, 2001) and military
spouses (Faulk, Gloria, Steinhardt, & Cance, 2012),
there is a lack of research toward a particularly high-
stressed population—namely postdoctoral research fel-
lows (postdocs). Postdocs, ironically, are an overlooked
and understudied population. It has been reported that
the work and life conditions of postdocs in the United
States are inundated by chronic exposure to high levels
of stress (Smaglik, 2006; Small, 2012). Often character-
ized as neither a faculty member nor a student,
postdocs tend to fall in the cracks and consequently
receive neither the recognition nor the benefits that they
feel are deserved (e.g. control over their work/funding
and health insurance for self and family; Aschwanden,
2006; Smaglik, 2006). They also often report feelings of
fear, uncertainty, pressure and lack of security due to
the impermanence of their employment, high work
expectations and extreme competitiveness of the job
market (i.e. low supply and high demand for ideal jobs
such as tenure-track professors/researchers; Kaplan,
2012; Woolston, 2002). Considering these points, it is
not surprising that postdocs describe their work and life
as extremely stressful and often filled with feelings of
anxiety and depression.
Therefore, using a sample of postdocs, the purpose
of the present study was to examine if (a) positive
emotions were associated with greater resilience, (b)
adaptive and maladaptive coping strategies mediated
the link between positive emotions and resilience and
(c) resilience moderated the influence of stress on trait
anxiety and depressive symptoms (i.e. as levels of stress
increase, individuals with higher scores of resilience
will report lower levels of trait anxiety or depressive
symptoms as compared with those with lower
resilience). Importantly, this research aimed to broadly
observe and unobtrusively explore the general relation-
ships among positive emotions, coping strategies,
resilience, stress and mental health, as these variables
occurred and interacted in the natural world and lives
of postdoctoral research fellows. Thus, this work did
not intend to clinically evaluate or diagnose partici-
pants with regard to clinical anxiety and depression;
the investigators were instead interested in studying
Positive Emotions, Resilience and Health C. T. Gloria and M. A. Steinhardt
146 Stress and Health 32: 145–156 (2016) © 2014 John Wiley & Sons, Ltd.
the participants’self-reported assessments of their own
qualities and mental health.
In the mediation analysis, it was hypothesized that
(a) positive emotions would have a positive direct effect
on adaptive coping strategies, (b) positive emotions
would have a negative direct effect on maladaptive
coping strategies, (c) positive emotions would have a
positive direct effect on resilience, (d) adaptive coping
strategies would have a positive direct effect on
resilience, (e) maladaptive coping strategies would have
a negative direct effect on resilience and (f) adaptive
and maladaptive coping strategies would mediate the
relationship between positive emotions and resilience.
It was also hypothesized that (g) the indirect effect
through adaptive coping strategies would be stronger
than through maladaptive coping strategies. As for the
moderation analysis, it was hypothesized that (a) stress
would have a positive direct effect on trait anxiety and
depressive symptoms, (b) resilience would have a
negative direct effect on trait anxiety and depressive
symptoms and (c) resilience would interact with stress
in such a manner that resilience would moderate the
effect of stress on trait anxiety and depressive symptoms.
Methods
Participants and procedures
Participants were recruited from a pool of postdocs
who were employed at a large research institution in
Texas. The email addresses of potential participants
(n= 523) were obtained from the institution’s human
resources office, and recruitment letters were sent via
email inviting postdocs to voluntarily participate in a
Qualtrics online survey that required approximately
30 min to complete. The sample selection method did
not have exclusion criteria, except that participants
must be currently employed under a postdoctoral
research fellowship appointment during the time of
data collection.
In order to enhance the survey response rate, a
variety of incentive prizes were offered (Deutskens, de
Ruyter, Wetzels, & Oosterveld, 2004). Each participant
was compensated with a $5 Starbucks gift card, a deck
of inspirational quote cards ($2 value) and an Individ-
ual Feedback Profile document that provided a
confidential report of the participant’s results as well as
an anonymous summary of the sample’s aggregate re-
sults. In addition, participants were entered into a lottery
drawing for a number of larger prizes (e.g. restaurant gift
cards valued from $10 to $50, iPod Shuffles and Amazon
Touch Kindles); one prize was awarded for every
15 surveys completed. Informed consent was obtained
from the participants, and study procedures were
approved by the Institutional Review Board.
Instruments
The online survey assessed participants’demographic char-
acteristics, positive emotions, adaptive and maladaptive
coping strategies, resilience, stress, trait anxiety and
depressive symptoms. Each of these variables is further
discussed in the following sections, and a copy of the
survey may be requested from the corresponding author.
Demographics
Participants were asked to report a variety of
personal characteristics including age, sex, race/
ethnicity, marital status, number of children, college/
school (i.e. location of employment), employment
length and nationality (i.e. country of origin). Because
these demographic characteristics may be related to
the dependent variables, the present study used them
as covariates in the regression analyses.
Positive emotions
The participants’experienced positive emotions
were measured by the 10-item positive emotions
subscale of the Modified Differential Emotions Scale
(Fredrickson et al., 2003). Each item asked participants
to recall how often they have experienced particular
sets of positive emotions during the previous 2 weeks
(e.g. ‘In the past two weeks, I have felt amused, fun-
loving, or silly.’); response options ranged on a five-
point scale from 0 (never)to4(most of the time). The
positive emotions score was calculated as the sum of
the 10 items; scores ranged from 0 to 40, with higher
scores indicating higher frequencies of experienced
positive emotions. Internal reliability for the positive
emotion scale was found to be acceptable in previous
research (α= 0.79; Fredrickson et al., 2003), and
reliability was very good in the present study (α= 0.87).
Coping strategies
The Brief Coping Orientations to Problems Experi-
enced scale was used to evaluate the participants’utility
of different coping strategies (Carver, 1997). For the
purposes of the present study, this measure included
six adaptive coping subscales (viz. active coping,
planning, positive reframing, acceptance, emotional
support and instrumental support) and six maladaptive
coping subscales (viz. self-distraction, denial, venting,
substance use, behavioural disengagement and self-
blame). Each subscale was measured by two items,
and participants were asked to report how often they
had used certain coping strategies during stressful
experiences, on a four-point response scale ranging
from 1 (not at all)to4(a lot).
Sample adaptive coping items include ‘I concentrate
my efforts on doing something about the situation I’m
in’(active coping), ‘I try to come up with a strategy
about what to do’(planning) and ‘I try to see it in a
different light, to make it seem more positive’(positive
reframing). Sample maladaptive coping items include ‘I
turn to other activities to take my mind off things’
(self-distraction), ‘I say to myself “this isn’t real”’
(denial) and ‘I say things to let my unpleasant feelings
escape’(venting). The scores for both the adaptive
C. T. Gloria and M. A. Steinhardt Positive Emotions, Resilience and Health
147Stress and Health 32: 145–156 (2016) © 2014 John Wiley & Sons, Ltd.
and maladaptive coping strategies were calculated as
the sum of the 12 items (ranging from 12 to 48); higher
scores represent greater use of the particular set of
coping strategies. The present study recorded alpha
coefficients of 0.77 and 0.66 for the adaptive and
maladaptive subscales, respectively.
Resilience
Participant resilience was assessed using the six-item
Brief Resilience Scale (Smith et al., 2008). On a five-
point scale ranging from 1 (strongly disagree)to5
(strongly agree), respondents indicated the extent to
which they agreed with statements that evaluated
their personal resilience or ability to recover from stress
(e.g. ‘I tend to bounce back quickly after hard times’,‘It
does not take me long to recover from a stressful event’
and ‘I usually come through difficult times with little
trouble’.). The resilience score was calculated as the
mean of the six items; scores ranged from 1 to 5, with
higher scores indicating higher levels of resilience. This
scale demonstrated good to excellent internal reliability
as reported by previous research with Cronbach’s
alphas ranging from 0.80 to 0.91 (Smith et al., 2008);
the present study also found the scale to be reliable at
α= 0.89.
Stress
This variable was assessed using the 10-item
Perceived Stress Scale (Cohen & Williamson, 1988),
which measured the appraised stressfulness of the
respondents’life situations. The scale items asked par-
ticipants to rate how often stressful events occurred
during the past month on a five-point scale from
0(never)to4(very often). Sample items include ‘How
often have you felt that you were unable to control
the important things in your life?’and ‘How often have
you felt difficulties were piling up so high that you
could not overcome them?’The stress score was calcu-
lated as the sum of the 10 items, ranging from 0 to 40,
with higher scores representing higher levels of
stress. Previous research found the internal reliability
of the 10-item Perceived Stress Scale to range from
acceptable (α= 0.78) to excellent (α= 0.91; Cohen &
Janicki-Deverts, 2012), and the reliability from the
present study was estimated at α= 0.86.
Trait anxiety
The 20-item trait anxiety subscale of the State-Trait
Anxiety Inventory for Adults (Spielberger, Gorsuch,
Jacobs, Lushene, & Vagg, 1968, 1977) was used to mea-
sure the participants’tendency to appraise stressful
events as threatening and thus respond with heightened
levels of state anxiety reactions (Spielberger, Gorsuch,
Lushene, Vagg, & Jacobs, 1983). Using a four-point
scale ranging from 1 (almost never)to4(almost
always), participants responded to items including,
‘I feel nervous and restless’and ‘I get in a state of
tension or turmoil as I think over my recent
concerns and interests’. Scores for this variable were
calculated as the sum of the 20 items, ranging from
20 to 80, with higher scores representing higher
levels of trait anxiety. [Correction made here after
initial online publication.] The trait anxiety subscale
demonstrated very good to excellent internal reliability,
with Cronbach’salphasrangingfrom0.89to0.91
(Spielberger et al., 1983); the present study also
recorded an excellent reliability score at α= 0.91.
Depressive symptoms
The Center for Epidemiologic Studies Depression
scale was used to assess the participants’level of experi-
enced depressive symptoms (Radloff, 1977). Consisting
of 20 items, the instrument assessed how often respon-
dents felt a variety of depressive symptoms during the
previous week. Using a four-point scale ranging from
0(rarely or none of the time; less than 1 day)to3(most
or all of the time; 5–7 days), participants responded to
statements such as ‘I was bothered by things that usu-
ally don’tbotherme’and ‘I did not feel like eating;
my appetite was poor’. The Center for Epidemio-
logic Studies Depression score was calculated as
the sum of the 20 items, ranging from 0 to 60, with
higher scores representing higher levels of experi-
enced depressive symptoms. A score of 16 or greater
is considered a moderately severe level of symptoms
and could be a marker for clinical depression
(Radloff, 1977). Previous research found the internal
consistency of the scale ranged from good to excel-
lent (α=0.85–0.90; Radloff, 1977), and the present
study also demonstrated very good reliability at
α=0.86.
Analyses
All analyses were completed using the Statistical
Package for the Social Sciences (SPSS) software ver-
sion 21 (IBM Corporation, Armonk, NY, USA).
Using the procedures detailed by Pallant (2010),
preliminary tests were performed to ensure that
the statistical assumptions of normality, linearity,
outliers, multicollinearity, independence and homo-
scedasticity were satisfied before the regression
analyses were conducted.
Descriptive statistics and correlations
Means, standard deviations and bivariate correla-
tions of all study variables were calculated using
descriptive statistics, Pearson correlations for continuous
variables, point-biserial correlations for continuous and
dichotomous variables and chi-square tests for pairs of
dichotomous variables.
Mediation analysis
In order to test the direct and indirect associations
among positive emotions, adaptive and maladaptive
coping strategies and resilience, path analysis was
performed with Preacher and Hayes’(2008) Model
Positive Emotions, Resilience and Health C. T. Gloria and M. A. Steinhardt
148 Stress and Health 32: 145–156 (2016) © 2014 John Wiley & Sons, Ltd.
INDIRECT script using bootstrap estimation. Demo-
graphic variables (viz. age, number of children, em-
ployment length, sex, marital status, college/school,
race/ethnicity and nationality) were also included in
the model as covariates.
As depicted in Figure 1, the relationship between
positive emotions and resilience was hypothesized to
be mediated by the two subscales of coping strategies
(viz. adaptive and maladaptive) while controlling for the
demographic variables. Therefore, this model examined
several types of effects: the direct effects of positive emo-
tions on resilience and each of the two subscales of coping
strategies, as well as the direct effects of the two subscales
of coping strategies on resilience; the specific indirect
effect of positive emotions on resilience through each
subscale of coping strategies (i.e. the unique mediating ef-
fect of each subscale of coping strategies while controlling
for the other subscale); the total indirect effect of positive
emotions on resilience (i.e. the sum of each of the two
specific indirect effects); and the total effect of positive
emotions on resilience (i.e. the sum of the direct and total
indirect effect). For indirect paths, this analysis produced
point estimates and three varieties of 95–99% confidence
intervals (viz. percentile, bias corrected and bias corrected
and accelerated) from 5000 bootstrap samples. Pairwise
comparison of the indirect effects was also performed to
determine if a particular mediator has a significantly
stronger unique indirect effect than the other mediator.
Moderation analysis
Guided by Aiken and West’s interaction analysis
method (1991), hierarchical multiple regression was
used to examine the moderating effect of resilience on
the association between stress and the dependent
variables (viz. trait anxiety and depressive symptoms);
each of the dependent variables was tested individually
using separate models. Demographic variables (viz. age,
number of children, employment length, sex, marital
status, college/school, race/ethnicity and nationality)
were also included in each of the models as covariates.
Prior to analysis, values of all continuous predictors
were centred to prevent potential problems associated
with multicollinearity (Aiken & West, 1991). The
hierarchical model of the multiple regression analysis
consisted of three steps. In the first step, the demo-
graphic covariates were entered in the regression of
the dependent variable. The second step involved the
addition of the focal predictors, namely stress and
resilience. In the third and final step, the interaction term
between stress and resilience—stress × resilience—was
entered into the model. This three-step process was
independently conducted for the regression of each of
the dependent variables (viz. trait anxiety and depressive
symptoms).
Results
Descriptive analysis
Data collection was conducted over a period of 2 weeks,
and the study obtained a sample size of n= 200 post-
docs (38% response rate). This response rate exceeded
expectations as previous studies with similar methods
recorded lower return rates ranging from 17% to
25% (Deutskens et al., 2004; Evans & Mathur, 2005;
Figure 1 Conceptual model of coping strategies partially mediating the relationship between positive emotions and resilience with
unstandardized (B) and standardized () coefficients (n= 196)
C. T. Gloria and M. A. Steinhardt Positive Emotions, Resilience and Health
149Stress and Health 32: 145–156 (2016) © 2014 John Wiley & Sons, Ltd.
Sax, Gilmartin, & Bryant, 2003); a meta-analysis of
56 web-based surveys from 39 studies reported an
average response rate of 35% (Cook, Heath, &
Thompson, 2000). However, due to partially missing
demographic data (viz. age and college/school), four
of the participants were dropped from the analyses,
resulting in a final sample size of n= 196; three
missing data points related to positive emotions and
one from depressive symptoms were replaced via
mean substitution.
Participants were primarily male (59.5%) with a
mean age of 32 years, ranging from 26 to 52 years. In
terms of race/ethnicity, 63.5% were non-Hispanic
White or Caucasian, 18% Asian or Pacific Islander,
8.5% Hispanic or Latino, 7% Indian or South Asian,
0.5% Black or African American and 2.5% other. The
majority were married (59.5%), 31% were single, 7%
were living with a partner and the remaining 2.5% were
either divorced or separated. Their family sizes ranged
from having zero to four children; 71% had no
children, 18.5% had one, 6% had two, 4% had three
and one participant had four children. The participants
were employed as postdocs for an average of 1.5 years
and ranged from 1 year to 6 years and 8 months. The
majority worked in the college of natural sciences
(48.5%), 20% in engineering, 7.5% in liberal arts,
5.5% in geosciences, 5% in pharmacy and the remain-
ing were in communication, education, public affairs,
social work or other. Most of the postdocs originated
from the United States (US) (50.5%), 10.5% China,
6.5% India, 4% South Korea, 3.5% United Kingdom,
3% Canada and the remaining were from 25 other
countries around the globe.
Prior to the regression analyses, multiple-category
demographic variables were collapsed into binary
variables to produce appropriately sized groups:
marital status (1 = married,0=unmarried), college/
school (1 = natural sciences,0=other), race/ethnicity
(1 = non-Hispanic White/Caucasian,0=other), and
nationality (1 = from US,0=other). Age, number of
children and employment length were retained as
continuous variables.
Table I displays the means, standard deviations and
correlations for all study variables. Positive emotions,
adaptive coping and maladaptive coping were moder-
ately correlated with resilience. Interestingly, adaptive
coping strategies were found to be unrelated to mal-
adaptive coping strategies, although previous research
has reported a moderate correlation (r= 0.30, p<0.01;
Meyer, 2001). The strongest correlations were among
stress, trait anxiety and depressive symptoms. Among
the demographic control variables, on average, US
nationals used more maladaptive coping and were
more resilient than non-US nationals, women
reported greater use of adaptive coping, being married
or having children was negatively associated with
maladaptive coping and married postdocs reported
fewer depressive symptoms.
Mediation model
In terms of the direct effects shown in Figure 1, post-
docs who experienced higher degrees of positive emo-
tions used more adaptive coping (B= 0.35, p<0.001)
and less maladaptive coping strategies (B=0.09,
p<0.05). Adaptive coping was positively related
to resilience (B=0.05, p<0.001), whereas maladaptive
coping had a negative association with resilience
(B=0.05, p<0.001). After controlling for the two
constructs of coping strategies and the set of demo-
graphic covariates, the direct relationship between
positive emotions and resilience remained significant
(B=0.03, p<0.01), indicating that coping strategies
did not completely mediate the link between positive
emotions and resilience. Among the control variables,
only employment length (B=0.09,p<0.05) and nation-
ality (B=0.21,p<0.05) had significant associations with
resilience.
As for specific indirect effects, the indirect effects of
positive emotions on resilience were significant
through both mediators: adaptive coping [B
boot
= 0.02,
Bias = 0.0001, standard error (SE) = 0.004, p<0.01
(99% bias corrected and accelerated confidence interval
(CI): 0.01, 0.03)] and maladaptive coping [B
boot
=
0.004, Bias = 0.0000, SE = 0.002, p<0.05 (95% bias
corrected and accelerated CI: 0.0002, 0.01)]. Pairwise
comparison of the two indirect effects indicated that
the mediating path through adaptive coping was
significantly stronger than through maladaptive coping
[B
boot
=0.01, Bias =0.0002, SE = 0.01, p<0.05
(95% bias corrected and accelerated CI: 0.02,
0.002)]. Combining together the direct and indirect
effects via coping strategies, the total effect of
positive emotions on resilience was estimated at
B= 0.05, p<0.001. The overall model accounted
for 34% of the total variance in resilience.
Moderation model
Trait anxiety
As displayed in Table II, the demographic control
variables were entered into Model 1, but they did
not significantly account for any variance in trait
anxiety (F
8, 187
=1.21, p>0.05). Following the addi-
tion of stress and resilience in Model 2, the total var-
iance explained was estimated at 70% (F
10, 185
=42.92,
p<0.001). In the final step, Model 3, both stress (B= 0.96,
p<0.001) and resilience (B=4.77, p<0.001) were
associated with trait anxiety. To determine the role of
resilience in moderating the association of stress on
trait anxiety, the interaction term (stress × resilience)
was also included in the final model. The analysis
revealed a significant interaction effect (B=0.31,
p<0.001), indicating that resilience moderated the
relationship between stress and trait anxiety. The
final model explained an additional 2% and
accounted for a total of 72% of the variance in trait
anxiety (F
11, 184
= 43.53, p<0.001).
Positive Emotions, Resilience and Health C. T. Gloria and M. A. Steinhardt
150 Stress and Health 32: 145–156 (2016) © 2014 John Wiley & Sons, Ltd.
Table I. Means, standard deviations (SD) and bivariate correlations for all variables (n= 196)
Variable Mean SD PE AC MC R S TA DS A NC EL F M NS W US
Positive Emotions (PE) 24.23 6.02 –
Adaptive Coping (AC) 36.53 4.87 0.44*** –
Maladaptive Coping (MC) 21.41 3.83 0.15* 0.02 –
Resilience (R) 3.62 0.70 0.38*** 0.38*** 0.29*** –
Stress (S) 16.18 5.79 0.47*** 0.19** 0.34*** 0.45*** –
Trait Anxiety (TA) 39.74 9.14 0.53*** 0.38*** 0.46*** 0.61*** 0.77*** –
Depressive Symptoms (DS) 10.61 7.59 0.49*** 0.24** 0.44*** 0.42*** 0.72*** 0.75*** –
Control
Age (A) 32.08 3.71 0.08 0.04 0.03 0.10 0.04 0.05 0.01 –
Number of Children (NC) 0.44 0.82 0.09 0.05 0.18* 0.10 0.00 0.10 0.12 0.46*** –
Employment Length (EL) 1.49 1.23 0.13 0.14 0.07 0.13 0.04 0.09 0.04 0.26*** 0.16* –
Female (F)
†
––0.02 0.22** 0.07 0.02 0.01 0.03 0.08 0.04 0.01 0.17* –
Married (M)
†
––0.01 0.04 0.19** 0.13 0.06 0.13 0.18* 0.14 0.43*** 0.15* 0.04 –
Natural Sciences (NS)
†
––0.01 0.03 0.13 0.11 0.11 0.06 0.10 0.01 0.05 0.21** 0.00 0.12 –
White (W)
†
––0.05 0.07 0.10 0.11 0.07 0.09 0.01 0.03 0.08 0.02 0.09 0.07 0.05 –
US American (US)
†
––0.02 0.06 0.15* 0.16* 0.02 0.02 0.06 0.00 0.01 0.02 0.21** 0.03 0.02 0.48*** –
†
Sex (Female = 1, Male= 0); Marital Status (Married= 1, Unmarried = 0); College/School (Natural Sciences = 1, Other = 0); Race/Ethnicity (White = 1, Other = 0); Nationality (US= 1, Other = 0).
*p<0.05; **p<0.01; ***p<0.001.
C. T. Gloria and M. A. Steinhardt Positive Emotions, Resilience and Health
151Stress and Health 32: 145–156 (2016) © 2014 John Wiley & Sons, Ltd.
Post hoc probing of the significant interaction was
conducted according to the methods of Aiken and
West (1991). Plotting the interaction was performed
to illustrate the regression of trait anxiety on varying
degrees of stress and resilience. As shown in Figure 2,
levels of stress and resilience were estimated at one
standard deviation below and above their means as
recommended by Aiken and West. Results showed that
the simple slopes from all three levels of resilience—at
1 standard deviation (SD), mean and +1 SD—were
statistically significant (p<0.001). The graph indi-
cated that increasing levels of stress were likewise
associated with increasing levels of trait anxiety. How-
ever, as suggested by the significant interaction effect,
resilience appeared to have a moderating effect on the
link between stress and trait anxiety. In other words,
postdocs with higher levels of resilience seemed to be
protected from the impact of stress and thus explaining
their lower scores of trait anxiety as compared with
those with lower levels of resilience. The protective
role of resilience was apparent across all levels of stress,
but the degree of protection was largest when stress
levels were highest. [Correction made here after initial
online publication.]
Table II. Hierarchical regression of trait anxiety on controls, focal predictors and the interaction term (n= 196)
Model 1 Model 2 Model 3
Variable BSE BBSE BBSE B
Age 0.11 0.20 0.04 0.05 0.12 0.02 0.01 0.11 0.01
Number of Children 0.39 0.99 0.04 0.79 0.56 0.07 0.64 0.54 0.06
Employment Length 1.10 0.57 0.15 0.88 0.33 0.12** 0.99 0.32 0.13**
Female
†
0.09 1.38 0.01 0.25 0.78 0.01 0.07 0.76 0.00
Married
†
2.34 1.49 0.13 0.74 0.85 0.04 0.59 0.82 0.03
Natural Sciences
†
1.50 1.34 0.08 0.33 0.76 0.02 0.29 0.74 0.02
White
†
2.16 1.56 0.11 0.67 0.89 0.04 0.92 0.85 0.05
US American
†
0.52 1.51 0.03 1.13 0.86 0.06 1.34 0.83 0.07
Stress 0.95 0.07 0.60*** 0.96 0.07 0.61***
Resilience 4.64 0.61 0.35*** 4.77 0.59 0.36***
Stress X Resilience 0.31 0.08 0.16***
Model R
2
0.05 0.70 0.72
Ffor change in R
2
1.21 199.44*** 15.65***
†
Sex (Female = 1, Male = 0); Marital Status (Married = 1, Unmarried = 0); College/School (Natural Sciences = 1, Other = 0); Race/Ethnicity
(White = 1, Other = 0); Nationality (US = 1, Other= 0).
*p<0.05; **p<0.01; ***p<0.001.
Figure 2 The moderating effect of resilience on the relationship between stress and trait anxiety
Positive Emotions, Resilience and Health C. T. Gloria and M. A. Steinhardt
152 Stress and Health 32: 145–156 (2016) © 2014 John Wiley & Sons, Ltd.
Depressive symptoms
As displayed in Table III, the demographic control
variables were entered into Model 1, but they did
not significantly account for any variance in depres-
sive symptoms (F
8, 187
= 1.49, p>0.05). Following
the addition of stress and resilience in Model 2, the
total variance explained was estimated at 58%
(F
10, 185
= 25.38, p<0.001). In the final step, Model
3, both stress (B= 0.90, p<0.001) and resilience
(B=1.40, p<0.05) were associated with depres-
sive symptoms. To determine the role of resilience
in moderating the association of stress on depressive
symptoms, the interaction term (stress × resilience) was
also included in the final model. The analysis revealed a
significant interaction effect (B=0.28, p<0.001),
indicating that resilience moderated the relationship
between stress and depressive symptoms. The final
model explained an additional 3% and accounted for a
total of 61% of the variance in depressive symptoms
(F
11, 184
= 25.81, p<0.001).
Post hoc probing of the significant interaction was
conducted according to methods of Aiken and West
(1991). Plotting the interaction was performed to illus-
trate the regression of depressive symptoms on varying
levels of stress and resilience. As shown in Figure 3,
levels of stress and resilience were estimated at one
standard deviation below and above their means.
Results showed that the simple slopes from all three
levels of resilience—at 1 SD, mean and +1 SD—were
statistically significant (p<0.001). The graph indicated
that increasing levels of stress were likewise associated
with increasing levels of depressive symptoms. How-
ever, as suggested by the significant interaction effect,
resilience appeared to have a moderating effect on the
link between stress and depressive symptoms.
Particularly when stress levels are high, postdocs
with higher levels of resilience seemed to be
protected from the impact of stress and thus
explaining their lower scores of depressive symp-
toms as compared with those with lower levels of
resilience. However, the protective role of resilience
appeared to be unimportant when stress levels were
low, but the degree of protection became more
apparent as stress levels increased. Considering the
cutoff score of 16 or higher (suggesting moderately
severe level of symptoms and a possible marker for
clinical depression), results indicated that postdocs
with high levels of resilience remained below this
criterion even when stress levels were high.
Discussion
Using a sample of n= 200 postdocs, the present study
was conducted to (a) examine if positive emotions were
associated with greater resilience, (b) test whether
coping strategies mediated the link between positive
emotions and resilience and (c) investigate if resilience
moderated the influence of stress on trait anxiety and
depressive symptoms, after controlling for a variety of
demographic variables. As hypothesized, there was a
positive association between positive emotions and
resilience, and coping strategies partially mediated the
link between positive emotions and resilience. Results
also indicated that resilience moderated the impact of
stress on trait anxiety and depressive symptoms.
With respect to the broaden-and-build theory of
positive emotions, findings from the mediation analysis
provided further support for the theory’s build hypoth-
esis (Fredrickson, 2004, 2005; Kok et al., 2008), as
Table III. Hierarchical regression of depressive symptoms on controls, focal predictors and the interaction term (n= 196)
Model 1 Model 2 Model 3
Variable BSE BBSE BBSE B
Age 0.06 0.17 0.03 0.18 0.11 0.09 0.15 0.11 0.07
Number of Children 0.70 0.82 0.08 1.09 0.55 0.12 0.95 0.54 0.10
Employment Length 0.44 0.47 0.07 0.10 0.32 0.02 0.21 0.31 0.03
Female
†
1.21 1.14 0.08 1.52 0.77 0.10* 1.22 0.75 0.08
Married
†
2.29 1.23 0.15 1.25 0.83 0.08 1.12 0.81 0.07
Natural Sciences
†
1.50 1.11 0.10 0.13 0.75 0.01 0.17 0.73 0.01
White
†
0.54 1.29 0.03 0.59 0.87 0.04 0.36 0.84 0.02
US American
†
1.32 1.25 0.09 1.34 0.85 0.09 1.53 0.82 0.10
Stress 0.88 0.07 0.67*** 0.90 0.07 0.68***
Resilience 1.27 0.60 0.12* 1.40 0.59 0.13*
Stress × Resilience 0.28 0.08 0.17***
Model R
2
0.06 0.58 0.61
Ffor change in R
2
1.49 113.77*** 13.28***
†
Sex (Female = 1, Male = 0); Marital Status (Married = 1, Unmarried = 0); College/School (Natural Sciences = 1, Other = 0); Race/Ethnicity
(White = 1, Other = 0); Nationality (US = 1, Other= 0).
*p<0.05; **p<0.01; ***p<0.001.
C. T. Gloria and M. A. Steinhardt Positive Emotions, Resilience and Health
153Stress and Health 32: 145–156 (2016) © 2014 John Wiley & Sons, Ltd.
demonstrated by the significant direct associations
of positive emotions with resilience and coping
strategies. Positive emotions’positive relationship
with adaptive coping strategies, and negative
relationship with maladaptive strategies, indicated
that positive emotions may have the ability to
enhanceadaptivecopingwhile minimizing maladap-
tive coping strategies. An increase in adaptive coping
in conjunction with a decrease in maladaptive
coping, in turn, would consequently have a building
influence on resilience.
In addition, the direct association between positive
emotions and resilience indicates that positive emo-
tions may not only have the potential to increase
resilience, but that resilience may also have the ability
to increase positive emotions, supporting the theory’s
hypothesis that a reciprocal relationship between
positive emotions and resilience could spark an
upward spiral toward increasing emotional well-being
(Fredrickson & Joiner, 2002); the same argument may
also apply for an upward spiral between positive
emotions and coping (Burns et al., 2008). Therefore,
in order to optimize resilience among postdocs, it is
important to implement programmes that would aim
to increase individual use of adaptive coping strategies,
decrease use of maladaptive coping strategies and
increase experiences of positive emotions. In turn,
enhanced levels of resilience would enable postdocs to
adapt more successfully when dealing with stressful
situations. From a practical standpoint, interventions
have been successful in eliciting enhanced positive
emotions and the accompanying resilience resources
(Emmons & McCullough, 2003; Seligman, Steen, &
Park, 2005).
As for the moderation analyses, results support
the buffering hypothesis of Fredrickson’sbroaden-and-
build theory (Kok et al., 2008). In line with expectations,
resilience demonstrated a moderating role toward the
impact of stress on trait anxiety and depressive symptoms,
as found in previous research (Aroian & Norris, 2000;
Pinquart, 2009; Wagnild, 2003). That is to say, as stress
levels increased, levels of trait anxiety and depressive
symptoms also increased; however, individuals with
higher levels of resilience exhibited some level of protec-
tion, as demonstrated by their lower scores of trait anxiety
or depressive symptoms, compared with other partici-
pants who possessed lower levels of resilience.
Regarding trait anxiety, postdocs with higher resil-
ience appeared to be protected across the full range of
stress; even when stress is low, postdocs with high resil-
ience already had lower scores on trait anxiety, and the
degree of protection—or the difference in trait anxiety
between low and high resilience—further increased as
stress magnified. In contrast, with respect to depressive
symptoms, one’s level of resilience did not seem to be
important when stress levels are low; however, having
high levels of resilience protected postdocs from
increased depressive symptoms as stress levels increased.
Importantly, those with average or low levels of
resilience were projected to have depressive symptoms
that were above the tipping point for clinical levels of
depression (Radloff, 1977) [Correction made here after
initial online publication.]. Thus, enhancing the
resilience of postdocs may help prevent an increase in
the prevalence of mental health disorders. Although
stress is unavoidable and the associations among stress,
anxiety and depression are undeniable, the link be-
tween postdocs and whether they will develop anxiety
Figure 3 The moderating effect of resilience on the relationship between stress and depressive symptoms
Positive Emotions, Resilience and Health C. T. Gloria and M. A. Steinhardt
154 Stress and Health 32: 145–156 (2016) © 2014 John Wiley & Sons, Ltd.
or depression may be ameliorated by implementing
programmes designed to increase their resilience,
adaptive coping and positive emotions.
With respect to the findings regarding the partici-
pants’nationality, we found small but significant
correlations between US postdocs and maladaptive
coping (r= 0.15, p<0.05), and between US postdocs
and resilience (r= 0.16, p<0.05). Although future
research is needed, it may be that some postdoc
stressors are not within one’s control. Given this, mal-
adaptive coping strategies might actually help postdocs
manage their stress more effectively in certain
situations. Looking at the coping means, adaptive
coping is higher than maladaptive coping, and so on
balance, one’s percentage of problem-focused coping
relative to total coping (adaptive + maladaptive) is
indicative of a healthy coping style. The small signifi-
cant relationship between US postdocs and resilience
makes intuitive sense given they are ‘at home’, and
thus, it may be easier for them to manage their
stressors; furthermore, they may have more social/
instrumental/emotional support available to them
than postdocs whose country of origin is outside of the
United States. Importantly, we also point out that we
controlled for nationality in the regression analyses,
which enhances confidence in our results.
Results from the present study should be considered
in light of the following limitations. The present study
used cross-sectional data, and thus, causality and
directionality cannot be determined from the found
associations among the variables. It is also possible that
the data may be vulnerable to inaccuracies due to
common-methods bias and the self-report nature of
the online survey instrument. The participants were
recruited from a pool of postdocs who were employed
at a large research institution in the state of Texas, USA.
There were no exclusion criteria, and all postdocs from
any college or department across the university were
allowed to participate. Due to this localized sampling,
results and implications may not be applicable to
postdocs from other institutions, locations or time. Ad-
ditional drawbacks of the present study’s methods in-
clude an increased risk for a self-selected sample and
the voluntary nature of the study.
No specific clinical evaluation of anxiety and depres-
sion was available. Therefore, it is possible that a num-
ber of factors beyond the measures of this study could
have influenced the participants’self-reported scores
(e.g. use of prescription drugs and unexpected event).
However, although such factors may affect one’s self-
assessment, this effect would not have an unexpected
influence on the relationships under examination. For
instance, if a participant were experiencing high levels
of stress and depressive symptoms, this positive corre-
lation between stress and depressive symptoms would
nonetheless remain the same even if the participant
were under the influence of an anti-stress/depression
drug and consequently feeling low levels of stress and
depressive symptoms. Likewise, if a participant with
an anxiety disorder were functioning effectively with
medication, his or her data would reflect this mindset.
Although prescription drugs may significantly change
one’s self-assessment, this effect should not have an
unexpected influence on the relationships under
examination.
In conclusion, findings from the present study
provide additional support for the build and buffering
hypotheses of the broaden-and-build theory of positive
emotions. Results suggested that positive emotions may
have the ability to fuel resilience directly, as well as
indirectly by promoting adaptive coping and demoting
maladaptive coping strategies. Although stress was
strongly associated with trait anxiety and depressive
symptoms, resilience could protect postdocs from
developing clinical levels of anxiety and depression. In
order to maintain and enhance the well-being of post-
docs, programmes should be implemented to increase
their positive emotions, adaptive coping and resilience.
Conflict of interest
The authors have declared that there is no conflict
of interest.
REFERENCES
Aiken, L. S., & West, S. G. (1991). Multiple regression:
Testing and interpreting interactions. Thousand Oaks,
CA: Sage.
Aroian, K. J., & Norris, A. E. (2000). Resilience, stress,
and depression among Russian immigrants to Israel.
Western Journal of Nursing Research,22(1), 54–67.
Aschwanden, C. (2006). Professionalizing the postdoc-
toral experience. Cell,124(3), 445–447.
Brown, S. P., Westbrook, R. A., & Challagalla, G. (2005).
Good cope, bad cope: Adaptive and maladaptive
coping strategies following a critical negative work
event. Journal of Applied Psychology,90(4), 792–798.
Burns, A. B., Brown, J. S., Sachs-Ericsson, N., Ashby
Plant, E., Thomas Curtis, J., Fredrickson, B. L., &
Joiner, T. E. (2008). Upward spirals of positive
emotion and coping: Replication, extension, and ini-
tial exploration of neurochemical substrates. Person-
ality and Individual Differences,44(2), 360–370.
Carney, R. M., & Freedland, K. E. (2003). Depression,
mortality, and medical morbidity in patients with
coronary heart disease. Biological Psychiatry,54(3),
241–247.
Carver, C. S. (1997). You want to measure coping but
your protocol’s too long: Consider the Brief COPE.
International Journal of Behavioral Medicine,4(1),
92–100.
Cohen, S., & Janicki-Deverts, D. (2012). Who’s stressed?
Distributions of psychologicalstress in the UnitedStates
in probability samples from 1983, 2006 and 2009.
Journal of Applied Social Psychology,42(6), 1320–1334.
Cohen, S., & Williamson, G. (1988). Perceived stress in
a probability sample of the United States. In S.
Spacapan, & S. Oskamp (Eds.), The social psychology
of health: Claremont Symposium on applied social
psychology. Newbury Park, CA: Sage. Retrieved from
http://www.psy.cmu.edu/~scohen/Cohen%2C%20S.%
20%26%20Williamson%2C%20G.%20(1988).pdf
Cohn, M. A., Fredrickson, B. L., Brown, S. L., Mikels, J.
A., & Conway, A. M. (2009). Happiness unpacked:
Positive emotions increase life satisfaction by building
resilience. Emotion,9(3), 361–368.
Cook, C., Heath, F., & Thompson, R. L. (2000). A meta-
analysis of response rates in web- or internet-based
surveys. Educational and Psychological Measurement,
60(6), 821–836.
de Luca, C., & Olefsky, J. M. (2006). Stressed out about obe-
sity and insulin resistance. Nature Medicine,12(1), 41–42.
Deutskens, E., de Ruyter, K., Wetzels, M., & Oosterveld,
P. (2004). Response rate and response quality of
C. T. Gloria and M. A. Steinhardt Positive Emotions, Resilience and Health
155Stress and Health 32: 145–156 (2016) © 2014 John Wiley & Sons, Ltd.
internet-based surveys: An experimental study.
Marketing Letters,15(1), 21–36.
Emmons, R. A., & McCullough, M. E. (2003). Counting
blessings versus burdens: An experimental investiga-
tion of gratitude and subjective well-being in daily life.
Journal of Personality and Social Psychology,84(2),
377–389.
Evans, J. R., & Mathur, A. (2005). The value of online
surveys. Internet Research,15(2), 195–219.
Faulk, K. E., Gloria, C. T., Steinhardt, M. A., & Cance, J. D.
(2012). Depressive symptoms among US military spouses
during deployment: The protective effect of positive
emotions. Armed Forces & Society,38(3), 373–390.
Folkman, S., & Moskowitz, J. T. (2000). Stress, positive
emotion, and coping. Current Directions in Psycholog-
ical Science,9(4), 115–118.
Fredrickson, B. L. (2001). The role of positive emotions in
positive psychology: The broaden-and-build t heory of
positive emotions. American Psychologist,56(3), 218–226.
Fredrickson, B. L. (2004). The broaden-and-build the-
ory of positive emotions. Philosophical Transactions
of the Royal Society, B: Biological Sciences,359(1449),
1367–1377.
Fredrickson, B. L. (2005). The broaden-and-build theory
of positive emotions. In F. A. Huppert, N. Baylis, & B.
Keverne (Eds.), The science of well-being (pp. 217–238).
New York, NY: Oxford University Press.
Fredrickson, B. L. (2009). Positivity. New York, NY:
Crown Publishers.
Fredrickson, B. L., & Joiner, T. (2002). Positive emo-
tions trigger upward spirals toward emotional well-
being. Psychological Science,13(2), 172–175.
Fredrickson, B. L., Tugade, M. M., Waugh, C. E., &
Larkin, G. R. (2003). What good are positive emo-
tions in crises? A prospective study of resilience and
emotions following the terrorist attacks on the United
States on September 11th, 2001. Journal of Personality
and Social Psychology,84(2), 365–376.
Gloria, C. T., Faulk, K. E., & Steinhardt, M. A. (2013). Positive
affectivity predicts successful and unsuccessful adaptation
to stress. Motivation and Emotion,37(1), 185–193.
Insel, P. M., & Roth, W. T. (2012). Connect core concepts
in health (12th ed.). New York, NY: McGraw-Hill.
Iso, H., Date, C., Yamamoto, A., Toyoshima, H.,
Tanabe, N., Kikuchi, S.,…JACC Study Group.
(2002). Perceived mental stress and mortality from
cardiovascular disease among Japanese men and
women: The Japan Collaborative Cohort study for
evaluation of cancer risk sponsored by Monbusho
(JACC study). Circulation,106(10), 1229–1236.
Jood, K., Redfors, P., Rosengren, A., Blomstrand, C., &
Jern, C. (2009). Self-perceived psychological stress
and ischemic stroke: A case-control study. BMC
Medicine,7(1), 53–60.
Kaplan, K. (2012). Postdoc or not? Nature,483(7390),
499–500.
Kemeny, M. E., & Schedlowski, M. (2007). Understand-
ing the interaction between psychosocial stress and
immune-related diseases: A stepwise progression.
Brain, Behavior, and Immunity,21(8), 1009–1018.
Kleppa, E., Sanne, B., & Tell, G. S. (2008). Working
overtime is associated with anxiety and depression:
The Hordaland Health Study. Journal of Occupational
and Environmental Medicine,50(6), 658–666.
Kok, B. E., Catalino, L. I., & Fredrickson, B. L. (2008).
The broadening, building, buffering effects of positive
emotions. In S. J. Lopez (Ed.), Positive psychology: Ex-
ploring the best of people (Vol. 3, pp. 1–19). Westport,
CT: Greenwood Publishing Company.
Lazarus, R. S. (1993). Coping theory and research: Past, pres-
ent, and future. Psychosomatic Medicine,55(3), 234–247.
Lazarus, R. S., & Folkman, S. (1984). Stress,appraisal,
and coping. New York, NY: Springer.
Lloyd, C., King, R., & Chenoweth, L. (2002). Social
work, stress and burnout: A review. Journal of
Mental Health,11(3), 255–265.
Lloyd, C., Smith, J., & Weinger, K. (2005). Stress and
diabetes: A review of the links. Diabetes Spectrum,
18(2), 121–127.
Markou, A., & Cryan, J. F. (2012). Stress, anxiety
and depression: Toward new treatment strate-
gies. Neuropharmacology,62(1), 1–2.
Melchior, M., Caspi, A., Milne, B. J., Danese, A., Poulton,
R., & Moffitt, T. E. (2007). Work stress precipitates
depression and anxiety in young, working women and
men. Psychological Medicine,37(8), 1119–1129.
Meyer, B. (2001). Coping with severe mental illness: Re-
lations of the Brief COPE with symptoms, function-
ing, and well-being. Journal of Psychopathology and
Behavioral Assessment,23(4), 265–277.
Misra, R., & McKean, M. (2000). College students’aca-
demic stress and its relation to their anxiety, time
management, and leisure satisfaction. American
Journal of Health Studies,16(1), 41–51.
Mykletun, A., Bjerkeset, O., Dewey, M., Prince, M.,
Overland, S., & Stewart, R. (2007). Anxiety, depres-
sion, and cause-specific mortality: The HUNT study.
Psychosomatic Medicine,69(4), 323–331.
Newbury-Birch, D., & Kamali, F. (2001). Psychological
stress, anxiety, depression, job satisfaction, and
personality characteristics in preregistration house
officers. Postgraduate Medical Journal,77, 109–111.
Nielsen,N. R., Kristensen, T. S.,Schnohr, P., & Grønbæk,
M. (2008). Perceived stressand cause-specific mortality
among menand women: Results froma prospective co-
hort study. American Journal of Epidemiology,168(5),
481–491.
Pallant, J. (2010). SPSS survival manual: A step by
step guide to data analysis using the SPSS program
(4th ed.). New York, NY: Open University Press.
Pinquart, M. (2009). Moderating effects of dispositional
resilience on associations between hassles and
psychological distress. Journal of Applied Developmen-
tal Psychology,30 (1), 53–60.
Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and
resampling strategies for assessing and comparing
indirect effects in multiple mediator models. Behavior
Research Methods,40(3), 879–891.
Radloff, L. S. (1977). The CES-D scale: A self-report
depression scale for research in the general
population. Applied Psychological Measurement,1
(3), 385–401.
Rawson,H.E.,Bloomer,K.,&Kendall,A.(1994).
Stress, anxiety, depression, and physical illness in
college students. The Journal of Genetic Psychology,
155(3), 321–330.
Sax, L. J., Gilmartin, S. K., & Bryant, A. N. (2003).
Assessing response rates and nonresponse bias in
web and paper surveys. Research in Higher Education,
44(4), 409–432.
Seligman, M. E. P., Steen, T. A., Park, N. & Peterson, C.
(2005). Positive psychology progress: Empirical vali-
dation of interventions. American Psychologist,60(5),
410–421.
Smaglik, P. (2006, February). Stress management: Uni-
versities take a look at postdocs’mental-health issues.
Nature,439, 629.
Small, G. (2012). The postdoc dilemma. Nat ure,
483, 235.
Smith, B. W., Dalen, J., Wiggins, K., Tooley, E., Christopher,
P., & Bernard, J. (2008). The brief resilience scale:
Assessingthe ability to bounce back. International Jour-
nal of Behavioral Medicine,15,194–200.
Spielberger, C. D., Gorsuch, R. L., Jacobs, G. A.,
Lushene, R., & Vagg, P. R. (1968, 1977). State-Trait
Anxiety Inventory for adults. Retrieved from http://
www.mindgarden.com/products/staisad.htm
Spielberger, C. D.,Gorsuch, R. L., Lushene, R., Vagg,P. R.,
& Jacobs, G. A. (1983). State-Trait Anxiety Inventory
for adults: Manual and sample. Retrieved from http://
www.mindgarden.com/products/staisad.htm
Steinhardt, M. A., SmithJaggars, S. E., Faulk,K. E., & Glo-
ria, C. T. (2011). Chronic work stress and depressive
symptoms: Assessing the mediating role of teacher
burnout. Stress and Health,27(5), 420–429.
Stewart,W.F.,Ricci,J.A.,Chee,E.,Hahn,S.R.,&
Morganstein, D. (2003). Cost of lost productive work time
among US workers with depression. The Journal of the
American Medical Association,289(23), 3135–3144.
Tugade, M. M., Fredrickson, B. L., & Feldman
Barrett, L. (2004). Psychological resilience and
positive emotional granularity: Examining the
benefits of positive emotions on coping and
health. Journal of Personality,72(6), 1161–1190.
Wagnild, G. (2003). Resilience and successful aging.
Comparison among low and high income older
adults. Journal of Gerontological Nursing,29(12),
42–49.
Wingo, A. P., Wrenn, G., Pelletier, T., Gutman, A. R.,
Bradley, B., & Ressler, K. J. (2010). Moderating effects
of resilience on depression in individuals with a
history of childhood abuseor trauma exposure. Journal
of Affective Disorders,126(3), 411–414.
Woolston, C. (2002). Perpetual postdocs. Retrieved
January 2013 from http://chronicle.com/article/
Perpetual-Postdocs/46038/
Zeidner, M., & Saklofske, D. (1996). Adaptive and mal-
adaptive coping. In M. Zeidner, & N. S. Endler (Eds.),
Handbook of coping (pp. 505–531). New York, NY:
John Wiley & Sons, Inc.
Positive Emotions, Resilience and Health C. T. Gloria and M. A. Steinhardt
156 Stress and Health 32: 145–156 (2016) © 2014 John Wiley & Sons, Ltd.
