The

Australian
Psychological

Australian Psychologist, December 2010; 45(4): 274-282

Society no

On being mindful, emotionally aware, and more resilient; Longitudinal

pilot study of police recruits
VIRGINIA WILLIAMS1, JOSEPH CIARROCHI1, & FRANK PATRICK

DEANE12

School of Psychology and 2lllaivana Institute for Mental Health, University of Wollongong, Wollongong, New South Wales,
Australia
1

Abstract
Police officers are at particular risk of stress when compared to people in other occupational groups. A compounding factor is
that police are prone to the use of avoidant coping strategies when attempting to deal with this stress. Evidence suggests that
anti-avoidance strategies, of acceptance, mindfulness and emotional awareness, are more effective ways of coping, and are
linked to both mental health and personal effectiveness. This study followed 60 trainee police officers from the recruit phase
into the workplace to determine if these processes predicted more positive mental health and wellbeing in police recruits after
1 year of service. Mindfulness predicted depression at follow-up, while emotion identification skill predicted general mental
health. These results suggest that police officers and police organisations may benefit from interventions aimed at developing
and promoting mindfulness and emotion identification.

Key words: Acceptance, emotion identification, mental health, mindfulness, police.

Researchers have found policing to be among the

most stressful of all occupations (Burke, 1993; Hart

& Cotton, 2002; He, Zhao, & Archbold, 2002;

Violanti, Marshall, & Howe, 1985). The range of
stressors that are likely to be experienced by officers
is almost limidess and can vary gready even within
one shift. For instance, an officer may find them
selves involved in a violent confrontation with an
offender, and then within die space of minutes may
be called upon to calm and console the family of a
road-trauma victim. Policing is unique in that
officers are required to take charge in situations
that are quite likely inconceivable to others in the
community. In addition to these occupational
stressors, police face many organisational stressors,
which include pressure from the government and
community to reduce crime, lack of supervisor
support, internal investigations, organisational
change, and reorganisation of departments (Brown
& Campbell, 1994; Kohan & Mazmanian, 2003).
What determines how well an officer deals with
these stressors? Recent theory suggests that people
who chronically avoid their internal experiences and

who are less aware of them also tend to adjust more

poorly to stress (Campbell-Sills, Barlow, Brown, &
Hofmann, 2006; Gross, 2002; John & Gross, 2004).
Little of this research has focused on the police and
in particular on the ability of police recruits to adjust
to police life. The present study followed police
recruits from the trainee phase into the workplace.
We assessed the extent that several variables related
to avoidant coping and low awareness predicted
future wellbeing in police officers. Our Time 1
variables included measures of experiential avoid
ance (avoiding negative private experience), thought
suppression (actively suppressing aversive thoughts),
difficulty identifying emotions (low awareness of
feelings), and mindfulness (curious awareness and
acceptance of experience).
Experiential avoidance and thought suppression
Avoidance behaviours can be thought of as any
action engaged in by an individual in an attempt to
escape from, suppress, change, forget about or in any
other way, avoid, parts of their private experience

Correspondence: Dr J. Ciarrochi, School of Psychology, University of Wollongong, Northfields, Wollongong, NSW 2522, Australia. E-mail: joec@uow.edu.au

ISSN 0005-0067 print/ISSN 1742-9544 online The Australian Psychological Society Ltd
Published by Taylor & Francis
DOI: 10.1080/00050060903573197

Mindfulness and police resilience

with which they are uncomfortable (Hayes et al.,
2004). Private experiences include thoughts,
emotions, memories, images, and bodily sensations
(Hayes et ah, 2004). These constitute a combinauon
of psychological content and autonomic nervous
system responses that take place within the indivi
dual. Experienual avoidance is a natural defensive
tendency for many people in that it seems logical that
one should try to avoid a sumulus that causes pain or
discomfort (Ciarrochi, Robb, & Godsell, 2005). This
kind of avoidance predicts negauve outcomes for
substance abuse and depression (Hayes, Wilson,
Gifford, Follette, & Strosahl, 1996) and can predis
pose individuals to higher levels of anxiety (Feldner,
Zvolensky, Eifert, & Spira, 2003) and poorer well
being (Bond & Bunce, 2000). In a study involving
>2,400 subjects, Hayes et ah (2004) found that
higher levels of experiential avoidance were linked to
higher levels of general psychopathology, depression,
anxiety, phobias, trauma and a lower quality of life.
In trying to avoid private experiences, the indivi
dual is trying to avoid something that is part of them
and, as such, experiential avoidance may (at least
sometimes) be futile, a suggestion with some
empirical support. For example, in a study looking
at the effects of thought suppression, participants
were instructed not to think about something
nonsensical for a short period of time (Wegner &
Zanakos, 1994). Even though the target was some
thing they were highly unlikely to think of ordinarily
(i.e., a white bear), after they were instructed not to
think about this they experienced a pre-occupation
with white bears (Wegner & Zanakos, 1994). It was
argued that individuals become primed or hyper
sensitive toward the suppressed stimuli due to
subconscious and conscious monitoring processes.
Individuals attempting to suppress a particular
thought tend to experience an intensification of the
very thing they are trying to avoid (Wegner, Erber, &
Zanakos, 1993). This state of mental hypervigilance
has been shown to translate into harmful arousal
of autonomic nervous system functioning (Pennebaker, 1995). Richards and Gross (1999) found that
the cardiovascular activity (i.e., heart rate) of subjects
who attempted to suppress inappropriate thoughts
increased significandy from baseline during the
period of attempted suppression. Similarly, research
suggests that mood control can be, at best, difficult
and potentially harmful (Hayes et al., 1996).
Participants who were primed for a particular mood
state by having them evoke a sad memory, were then
instructed not to dunk about the evoked memory.
Those in the thought-suppression group experienced
greater interference from emotionally salient words
(e.g., unhappy) when completing an incidental
computer task in comparison to subjects who did not
receive the instruction to suppress (Wegner et al.,

275

1993). Deliberate attempts to suppress a particular

mood may actually lead to an increase in die emotion
in a kind of self-amplifying loop (e.g., Feldner et al.,
2003).

Organisational climate and avoidance in police

Research has implicated the climate or culture of

police organisations as a factor that predisposes
officers to use avoidant coping strategies. As sum
marised by Lennings (1997), The police culture
inhibits the expression of emotion (p. 560). The
organisational climate determines what behaviours
are appropriate and therefore likely to be used by
police when dealing with stressful or emotionally
charged situations (Hart & Cotton, 2002). If being
rational, non-emotional and in control is the
behaviour modelled by senior officers, it is likely
that these attitudes and behaviours will become the
norm for junior police. Violanti (1993) found that
maladaptive coping (especially avoidance) was so
cialised among a sample of recruits in training. It has
also been found that police officers become more
emotionally detached and hardened within the first
18 months of service (Robinson, Sigman, & Wilson,
1997). Police are therefore at risk of developing
personality styles that are unsentimental, detached
(Evans, Coman, Stanley, & Burrows, 1993), and
cynical (Violanti et al., 1985). They may tend to deal
with emotions by blocking them out, even when it
is not appropriate (e.g., non-crisis situations, perso
nal relationships).
Difficulty identifying feelings
Blocking emotions and detachment may be protec
tive during crises, but may not be so beneficial to
officers in other situations (e.g., regular workplace
interactions, personal relationships). Amaranto,
Steinberg, Castellano, and Mitchell (2003) reported
that the police subculture gives rise to a tendency
toward emotional restraint and the suppression of
emotions. This tendency to suppress and avoid
emotions and emotional content tends to indicate
that police are likely to have more difficulty identify
ing and using emotions to appropriately guide their
behaviour.
The ability to process emotional information is
hypothesised to influence an individuals ability to
cope during a stressful period. For example, in
dividuals who have difficulty identifying feelings are
more likely to experience distress and poorer
adjustment to the demands of the stressor (Salovey,
Mayor, Goldman, Turvey, & Palfai, 2002). Those
who are able to recognise and differentiate between
emotional responses to stress are less distressed in
comparison to others who are confused or at a loss

276

V. Williams et al.

about their emotional reactions (Kerr, Johnson,

Cans, & Krumrine, 2004). Difficulty identifying
feelings is likely to predispose an individual to poorer
mental health (Ciarrochi, Chan, Caputi, & Roberts,
2001). Those with clinically significant deficits in
emotional functioning are said to have a condition
called alexithymia, which literally means no
words for moods (Sifneos, 1973, p.255). Clinically,
alexithymia is defined as
an inability to identify
and describe ones own feelings (Bagby, Parker, &
Taylor, 1994). Alexithymia has been associated with
psychosomatic disorders (e.g., hypertension and
gastrointestinal complaints), panic attacks and eating
disorders (Taylor, 2000). Alexithymia is theorised to
be a continuous variable, such that alexithymic
behaviours can be seen in non-clinical populations,
particularly in response to stressful life circumstances
(Kerr et al., 2004).

..

Acceptance and mindfulness as alternatives to avoidance

One alternative to experiential avoidance is psycho

logical acceptance. Acceptance can be described as
an active taking in of thoughts, feelings, memories
(i.e., private experiences) and an openness to private
experiences as they occur, without trying to avoid,
escape, or change them (Hayes, Strosahl, & Wilson,
1999). Acceptance is a deliberate openness, mind
fulness or psychological embracing of experience
and involves deliberate action to heighten contact, or
remove barriers to private experiences (Hayes, 200 1,
p. 27). Acceptance has been linked to improved
mental health, wellbeing and personal effectiveness
in both clinical and workplace studies (see Ciarrochi
& Godsell, 2005, for a review).
Mindfulness is a concept related to acceptance and
refers to an active state of consciousness, which
involves being open to and engaging with all aspects
of ones moment-to-moment experience (Ciarrochi
& Godsell, 2005). Mindful awareness involves
paying attention to the present reality, in a nonjudgmental manner (Kabat-Zinn, 1990). It is the
antithesis of mindlessness, which is characterised
by compulsive or automatic behaviour that is
disengaged with present consciousness (Brown &
Ryan, 2003). Where mindfulness is an open, active
mental state, mindlessness can be defensively moti
vated and may involve refusal to acknowledge a
thought or ruminative posturing (Brown & Ryan,
2003). In this sense, mindfulness shares similarities
with the concept of acceptance because it is
represented by openness and welcoming of psycho
logical content without prescriptive judgment of its
nature (e.g., thats a bad thought).
Ryan and Deci (2000) posited that mindfulness is
associated with wellbeing enhancement because it
may help individuals disengage from unhealthy,

automatic patterns of behaviour. For example,

individuals diagnosed with major depressive disorder
indicated that they experienced a significant decrease
in the relapse of depressive episodes after having
participated in an 8-week group intervention pro
gram based on mindfulness-based cognitive therapy
(Segal, Teasdale, Williams, & Gemar, 2002; Teasdale, Segal, & Williams, 1995). One of the major
skills that was taught in the program was attentional
control, whereby participants observe their thoughts
and feelings non-judgmentally and view them as
mental events that come and go. In a separate study,
Brown and Ryan (2003) recruited 41 cancer patients
to participate in an 8-week mindfulness-based stress
reduction program. It was found that individuals
with higher levels of mindfulness were more likely to
experience lower levels of stress and mood distur
bance, which contributed significantly to their overall
wellbeing.
Much of the prior research involving police has
focused on trying to identify sources of stress and the
ways officers try to cope with their distress (Evans
et al., 1993). Brown and Campbell (1994) found that
some groups of police within the organisation are
more prone to stress and stress-related illness. Newly
appointed police officers are likely to be one such
group. Very little research, however, has examined
this transition period and studies involving trainee
police are also scant (Violanti, 1993).
The aim of the present study was therefore to
investigate whether the processes of acceptance,
mindfulness and emotion identification predict
mental health and wellbeing of police recruits after
their first year on the job. We should note that
although we have implied that acceptance is the
opposite of avoidance, the relationship is not so
simple. One may not be avoiding thoughts actively,
but also not accepting the thoughts. Different
measures of acceptance and avoidance may correlate
only modestly, suggesting that they predict unique
variance. Our plan is therefore not to lump all
measures together into a single dimension, but rather
to examine the extent that each of the measures is a
longitudinal predictor. Our general hypothesis is that
trainee police who score high on avoidance-related
indices will have poorer mental health and wellbeing
in comparison to their colleagues, following their
transition into the workplace.
Method

Participants
The sample was derived from employees of the New
South Wales Police (NSW Police), who were
trainees at initial testing (Time 1) and probationary
constables at the time of the second data collection

Mindfulness and police resilience

(Time 2). Ethics clearance was obtained via the
University of Wollongong Human Ethics Commit
tee. The potential participant pool at Time 1 was all
592 personnel enrolled in the Bachelor of Policing
offered at a major university, who completed the
baseline questionnaire. A total of 60 participants
(10%) from Time 1 (w = 592) completed the ques
tionnaire on the second administration 10-12
months later. Forty-four (73%) were male and 16
(27%) were female. When asked to indicate their
relationship status, 50% of respondents reported
being married or in a relationship, 47% were single,
2% reported being divorced and the remaining 1%
did not specify. There were no differences on any of
the demographic or main variables for those who
completed both time periods and those who did not.

Measures
Acceptance and Action Questionnaire. The Accep
tance and Action Questionnaire (AAQ) (Hayes et al.,
2004) is a 19-item inventory that measures low
willingness to be present to psychological content
(including thoughts, emotions, images and mem
ories) whether positive, negative or otherwise, while
behaving in a manner that is consistent with tire
pursuit of valued goals. It features items such as,
When I feel depressed or anxious, I am unable to
take care of my responsibilities and reverse-scored
items such as I am able to take action on a problem
even if I am uncertain of what the right thing to do
is. Items are rated on a 7-point scale, ranging from
1 (never true) to 7 (always true). Higher scores on the
AAQ indicate a propensity toward an unwillingness
to accept psychological experiences and commit to
valued action (experiential avoidance). The AAQ has
demonstrated substantial incremental and criterion
related validity, correlating significantly and moderately-highly with measures of general psychopathol
ogy, anxiety, depression and trauma (Hayes et al.,
2004). The AAQ demonstrated modest reliability in
the present study (Cronbach a = .64).

White Bear Suppression Inventory. The White Bear

Suppression Inventory (WBSI) (Wegner & Zanakos,
1994) is a seven-item scale that measures the
tendency to suppress unwanted thoughts (i.e.,
avoidance). Whereas the AAQ measures general
avoidance, the WBSI focuses on one facet of
avoidance, namely thought suppression. The instru
ment consists of statements such as, I often do
things to distract myself from thoughts (Cronbach
a = 0.90). Respondents rate their level of agreement
with each statement using a 5-point scale ranging
from strongly disagree (1) to strongly agree (5). Higher
overall scores indicate a higher tendency toward
thought suppression. The WBSI was used as a

277

reverse measure of acceptance in that it targets the

opponent process of avoidance.

Toronto Alexithymia Scale. The Toronto Alexithymia Scale (TAS-20) (Bagby, Parker et al., 1994;
Bagby, Taylor, & Parker, 1994) is a 20-item
instrument designed to measure the extent to which
individuals have difficulty identifying and describing
their feelings. The scale has three subscales:
Difficulty Identifying Feelings (DIF)> Difficulty
Describing Feelings, and Externally Oriented Think
ing. The DIF subscale was used to measure the
construct of Emotion Identification Skill (EIS).
As mentioned previously, an individual who pos
sesses emotional identification skills is able to
recognise and differentiate between emotional re
sponses to stress and is more likely to be less
distressed in comparison to others who are confused
about their emotional reactions. A low score on the
DIF subscale indicates that an individual is high in
EIS. A statement typical of this scale is, I am often
confused about what emotion I am feeling (Cronbach = .85).

Mindful Attention Azvareness Scale. The Mindful

Attention Awareness Scale (MAAS) (Brown & Ryan,

2003) is a 15-item instrument that assesses in
dividual differences in the frequency of mindful
states over time (Brown & Ryan, 2003). The
measure consists of items such as It seems I am
running on automatic without much awareness of
what I am doing (Cronbach a =.72). Participants
respond on a 6-point scale regarding their level of
agreement to each statement (1 almost always,
6 = almost never). Although a relatively new measure,
the MAAS has demonstrated good retest reliability
(Pearson r=0.81, p < .001), as well as adequate
convergent and discriminant validity when compared
to other measures of mindfulness. Higher scores on
the MAAS indicate a greater tendency toward
mindful awareness, which has been associated with
greater wellbeing (Brown & Ryan, 2003).
General Health Questionnaire-12. The General
Health Questionnaire- 12 (GHQ-12) (Goldberg,
1978) contains 12 items that measure general mental
health (Cronbach a = .70). An example item is,
Have you recently .. . felt constantly under strain?
Responses are made on a Likert scale ranging from 0
( not at all) to 3 (much more than usual). Higher scores
are indicative of greater mental ill-health.
Depression Anxiety and Stress Scale. The Depression
Anxiety and Stress Scale (DASS) (Lovibond &
Lovibond, 1995) consists of three 14-item subscales
designed to measure depression (I couldnt seem to
experience any positive feeling at all, Cronbach

278

V. Williams el al.

a = .95); anxiety (I was worried about situations in

which I might panic and make a fool of myself,
Cronbach a = .89); and stress (I found it hard to
relax, Cronbach a = .92). Participants rated the
extent to which each statement applied to them over
the past 2 weeks using a 4-point rating scale ranging
from did not apply to me at all (0) to applied to me
much or most of the time (3).
Procedure
At the initial data collection point (Time 1),
participants were enlisted from four consecutive
graduating recruit classes. Participants were in their
final week of recruit training.
For the Time 2 administration, an electronic copy
of die self-report survey utilised in the first admin
istration was forwarded to participants via email.
This survey included all measures administered to
participants at Time 1. Each participant was con
tacted by email 10-12 months after graduating from
the recruit training program. A follow-up email was
sent to each Time 1 participant approximately 1
month after the electronic survey was disseminated
Table I. Mentat health outcome measures: Time 1 vs. Time 2
(iV= 60)
Time 2

Time 1

Measure
Outcome
General mental
health problems
Depression
Anxiety
Stress

Af

SD

SD

0.58

0.24

0.70

0.30

-2.47*

0.037

0.08
0.08
0.36

0.11
0.07
0.28

0.21

-2.57*

0.04
0.28

0.15
0.35

-1.63

1.67

0.68

1.63

0.63

0.46

4.71
3.21
2.36

0.65
0.44
0.74

4.70
3.11
2.39

0.72
0.44
0.96

0.03
1.65
-0.23

-0.08

Process
Difficulty identifying
feelings
Mindfulness
Avoidance
Thought suppression
Note. *p < .05.

to capture participants who did not receive the first

email (c.g., were on annual leave) in an attempt to

promote higher response rates.

Results

Of the 60 respondents only two had a small amount

of missing data (<10%), and scaled mean scores
were used in these cases. As can be seen in Table I,
mean scores for the four measures of mental health
(i.e., GHQ and each of the three DASS subscales)
were calculated and compared by way of a pairedsample -test to determine whether the withinsubjects scores differed significantly from Time 1 to
Time 2. There was a significant difference in means
on both the GHQ and the depression subscale of the
DASS, indicating that depression and general mental
health worsened from Time 1 to Time 2. There were
no other significant differences.
Table I also provides the mean scores, standard
deviations and differences between means for the
process measures from Time 1 to Time 2. As can be
seen, there were no significant differences in mean
scores of any of the key variables.
Table II presents the correlations between Time 1
variables. There were correlations in the expected
direction between the process variables and each of
the four measures of mental health (i.e., GHQ and
DASS subscales). In each case, the correlations
between mindfulness (MAAS) and the measure of
mental health were negative, indicating that a lower
mindfulness score was related to higher mental ill
health. Difficulty identifying feelings, low accep
tance, and thought suppression were all associated
with poorer mental health.
Table III provides the correlations between all
variables at Time 2. Significant correlations were
found between each of the process measures and at
least one measure of mental health at Time 2.
Specifically, a significant correlation was found
between difficulty identifying feelings and general
mental health (GHQ- 1 2), such that lower scores on
this measure were linked to poorer mental health at

Table II. Interrelations between process and outcome measures at Time 1 (N= 60)

1. Difficulty identifying feelings

2. Mindfulness
3. Avoidance
4. Thought suppression
5. Mental health problems
6. Depression
7. Anxiety
8. Stress

Note. *p < .05, **p < .01.

5.

2.

3.

4.

-.56**
1.00

0.35**
-.38**

0.49**

.21

- .53**

-.40**
.36**
.55**
1.00

1.00

.23
1.00

6.
-.09

-.34**
.11
.15
.16
1.00

7.

.18
-.22
-.09
.19
-.08
.02
1.00

8.

.22

- .43**
0.03
.23
.15

.42**
.36**
1.00

Mindfulness and police resilience

Time 2. Significant relationships were found to exist
between mindfulness and all four measures of mental
health at Time 2, the strongest (negative) correlation
being between mindfulness and depression. Table III
shows interrelations between measures of acceptance
and die outcome variables of depression, stress (for
WBSI only) and general mental health. Acceptance
was not significantly correlated with anxiety at
Time 2.
Table IV presents the link between Time 1 and
Time 2 variables. Generally, all the process measures
had moderate test-retest correlations. In contrast,
the only significant test-retest correlation for the
mental health variables involved anxiety. Thus,
the process variables were moderately stable across
the very different contexts (participants at the
training academy [Time 1] vs. participants working
as a police officer [Time 2]), whereas the mental
health variables were not.
Concerning the link between the process variables
and the outcome variables, difficulty identifying
feelings at Time 1 was found to correlate moderately
with both general mental health (GHQ-12) and
depression (DASS-D). Mindfulness scores at Time 1
were found to be related to depression at Time 2,
such that lower mindfulness was correlated with
higher depression scores. Additionally, acceptance

279

(WBSI) was found to be moderately correlated with

depression at Time 2.
Regression analyses
Regression analyses were conducted to further
explore each of the significant relationships as
described in Table IV. Time 1 baseline measures of
wellbeing were controlled when predicting Time 2
variables. This analysis examines whether, on aver
age, people high in avoidance (or related variable)
tend to show greater decreases in wellbeing relative
to those low in avoidance with the same baseline
levels of wellbeing (Wainer, 1991).
As can be seen in Table V, mindfulness predicts
depression scores at Time 2, when baseline (Time 1)
measures of depression were controlled for. Simi
larly, thought suppression predicted depression at
Time 2, controlling for baseline depression scores.
Finally, difficulty identifying feelings predicted de
pression and general mental health at Time 2. No
other effects were significant.
A stepwise multiple regression analysis was con
ducted to determine which of the three variables that
predicted depression (i.e., mindfulness, acceptance
or emotion identification skills) had the strongest
influence and was the most unique predictor.

Table III. Interrelations between process and outcome measures at Time 2 (iV=60)

1. Difficulty identifying feelings

2. Mindfulness
3. Avoidance
4. Thought suppression
5. Mental health problems
6. Depression
7. Anxiety
8. Stress

2.

3.

4.

5.

-.62**
1.00

.63**
-.58**

.62**
-.58*
.62**

.42**
-.34**
.44**
.32*

1.00

1.00

6.

7.

8.

.59**

.23

.42**
-.52**

.48**
.44**
.41**

-.27*
.13
.00
.03
.42*

- .50**

1.00

1.00

1.00

.17
.30*
.17

.51**
.22
1.00

Roie. *p < .05, **p < .01.

Table IV. Process and mental health outcome measures from Time 1 to Time 2 (iV =60)

Time 2
Time 1

1. Difficulty identifying feelings

2. Mindfulness
3. Avoidance
4. Thought suppression
5. Mental health problems
6. Depression
7. Anxiety
8. Stress

1.

2.

.56**

- .42**
.35**
.30*

.23
13
.14
-.03

71**
-.24

- .54**
-.46**

-.04

-.26*
-.22

3.

4.

5.

6.

.46**
-.40**
36**
37**
17

.48**
-.53**

.35**
-.21
.24
.14
.14
-.02
.18
.00

.33*
-.34*
.15
.33*

.09
-.15
-.03
.11

.26
-.02

.06
-.09

-.13
18
-.07

.18

.52**
.19
.17
.14

.30*

.31*
-.04

7.

.39**
.11

8.
.17
-.24
-.03
.20
.24
.09
.29*
.15

Note. *p < .05, **p < ,01.

!

280

V. Williams et al.

Table V. Process variables predicting depression and mental

health at Time 2 (iV= 60)

SE

R2

-.13
.10
.11

0.04
0.04
0.04

-.39**
.34*
.33*

.14**
.11*
.11*

General Mental health problems (T2)

-.09
Mindfulness (Tl)
Thought suppression (Tl)
.04
Difficulty Identifying
.15

.07
.07
.06

.20
.09

.03
.01
.11*

Variable

Depression (T2)
Mindfulness (Tl)

Thought suppression (Tl)

Difficulty identifying
feelings (Tl)

.34*

Feelings (Tl)

Notes. After controlling for Time 1 (Tl) levels of depression and

mental health.
*/> < .05, **p < .01.

Baseline depression was entered in Step 1 . Then, the

three process variables were submitted to a stepwise
entry procedure, with p to enter the model set at .05
and p to be removed set at .10. Mindfulness was the
only variable that uniquely predicted depression at
Time 2, when controlling for baseline (Time 1)
depression. Mindfulness was found to account for
approximately 14% of variance in depression scores
at Time 2.
Discussion

Policing is an emotionally challenging occupation, as

is evidenced by our finding diat police recruits
experienced increases in depression and mental
health problems as they entered the police workforce.
But not all officers responded badly to entering the
workforce. After controlling for baseline levels of
wellbeing, officers who tended to be mindful, could
identify their feelings, and did not suppress their
thoughts tended to experience smaller increases in
depression. In addition, difficulty identifying feelings
predicted decreases in mental health. When the
variables in a regression analysis were all entered to
see which was the best predictor, it was found that
mindfulness predicted depression over and above the
other variables.
Research has shown that die enhancement of an
individuals mindfulness can lead to increased well
being on a variety of measures (Kabat-Zinn, 1990).
In the present research, mindfulness was found to be
the best predictor of depression in recendy appointed
police. Those recruits who tended to be less open to
present experiences at entry experienced higher
levels of depressive symptomatology following their
transition into the workplace.
This study demonstrated partial support for the
premise that acceptance plays a role in mental health
and wellbeing (i.e., Hypothesis 1). Higher scores on
the measure of thought suppression were predictive

of higher self-reported depression. The predictive

value of thought suppression, however, was not
unique after controlling for mindfulness. Thus, it
may be that avoidance is subsumed by the mind
fulness construct. This may suggest that by increas
ing mindfulness there will be a decrease in
experiential avoidance.
An unanticipated finding of this study was that
avoidance (as measured by the AAQ) was not found
to predict mental health of police recruits after
entering the workplace. Other workplace research
suggests that the AAQ does predict future mental
health (Bond & Bunce, 2003). One explanation
relates to the characteristics of the AAQ in the
current research context. This scale includes items
such as, When I feel depressed or anxious, I am
unable to take care of my responsibilities and other
items that link emotions with behavioural outcomes.
The use of the word unable in the sample item
above may have a connotation of weakness in the
policing context. Police officers tend to perceive
themselves as strong, in control and non-emotional
(Evans et al., 1993). As such, the AAQ may have
poor face-validity in a population that is particularly
resistant to the notion that they cannot take care of
responsibilities. Consistent with this view, the inter
nal reliability of the AAQ in the present study was
somewhat low (.64). The AAQ is a relatively new
measure and future research should be focused on
determining its relevance in a variety of non-clinical
populations.
There was a significant increase in depression and
mental health problems from Time 1 to Time 2, but
there was no increase in stress and anxiety. While
there was no a priori suggestion that any one aspect
of mental health was more likely to be affected, it
might be expected that stress and anxiety would
increase as trainee police come to terms with the role
and responsibilities of their new profession. It may be
that individuals who choose to join the police do so
because they believe that they will enjoy, or are well
suited to, the inherent stressful ness of the job
(Hart & Cotton, 2002). Just as people who choose to
skydive enjoy an experience that many others may
find terrifying, perhaps individuals who choose to
become police officers are self-selecting on the basis
that they will be suited to what others would find a
stressful and anxiety provoking career.
The absence of significant findings in relation to
avoidance (as measured by the AAQ) and anxiety
and stress may also in part be due to the small sample
size, which was highly self-selected. That is, only
10% of the original sample completed the Time 2
measures, potentially creating unknown sample bias
and limiting generalisability. A goal for future
research could be to replicate this study, using an
alternate method of data collection at Time 2 to

Mindfulness and police resilience

reduce the dropout rate (e.g., region-based pen and
paper administrations and/or incentives). The ethical
need, however, to emphasise that participation is
voluntary is particularly important in an organisa
tional context in which there is a highly hierarchical
system.

The present results suggest that difficulty identify

ing feelings and mindfulness are key variables in
predicting the wellbeing and mental health of police
in their first year of service. Recruits who are better
able to identify their feelings and are more able to be
present to moment-to-moment experience are more
likely to make a transition into die workplace that is
characterised by lower levels of depressive symptoms
and better mental health. We should note that
avoidance of feelings and experience are not ex
pected to be unhelpful to people in all contexts.
There are likely to be occasions in which avoidance
or suppression of emotions may serve useful pur
poses in the short term (e.g., allowing for focused
behavioural responses in crisis situations). Rather,
we expect the greatest problems to arise when police
rigidly attempt to avoid all unpleasant feelings and
experience.
These findings have important implications in the
selection of recruits, and in their pre-workplace
training. Policing organisations may consider taking
baseline measures of each applicants mindfulness
and difficulty identifying feelings. Although it is not
argued that individuals should be screened out in this
process, there is evidence to suggest that both
emotion identification skills (e.g., Salovey et al.,
2002) and mindfulness (e.g., Ciarrochi & Blackledge, 2006) can be taught. Policing organisations
may find it useful to include interventions (e.g.,
workshops) that target these processes in the recruit
training program as a resilience strategy.
Broad implementation of an intervention that
promotes mindfulness and emotional awareness
might address the specific aspects of the police
culture that predispose officers to avoidant coping
strategies. Violanti (1993) found that maladaptive
coping strategies were modelled by senior police to
their more junior counterparts. More effective
coping strategies could be similarly modelled
throughout the organisation, if they were introduced
and then implemented by senior police.

References
Amaranto, E Steinberg,]., Castellano, C., & Mitchell, R. (2003).
Police stress interventions. Brief Treatment and Crisis Interven
tion y 3y 47-53.
Bagby, R. M., Parker, J. D. A., & Taylor, G. J. (1994). The
twenty-item Toronto Alexithymia Scale: I Item selection and
cross-validation of the factor structure. Journal of Psychosomatic
Research, 383 23-32.

281

Bagby, R. M., Taylor, G. J., & Parker, J. D. A. (1994). The

twenty-item Toronto Alcxichymia Scale: II Convergent, dis
criminant and concurrent validity. Journal of Psychosomatic
Research, 38, 3340.
Bond, F. W., & Bunce, D. (2000). Mediators of change in
emotion-focused and problem-focused worksite stress manage
ment interventions. Journal of Occupational Health Psychology,

5, 156-163.
Bond, F. W., & Bunce, D. (2003). The role of acceptance and job
control in mental health, job satisfaction, and work perfor
mance. Journal of Applied Psychology, 88, 1057-1067.
Brown, J. M., & Campbell, E. A. (1994). Stress and policing: Sources
and strategies. Chichester: John Wiley & Sons.
Brown, K. W., & Ryan, R. M. (2003). The benefits of being
present: Mindfulness and its role in psychological well-being.
Journal of Personality and Social Psychology, 84, 822-848.
Burke, R. J. (1993). Work-family stress, conflict, coping and
burnout in police officers. Psychological Reports, 65, 3-12.
Campbell-Sills, L., Barlow, D. H., Brown, T. A., & Hofmann, S.
G. (2006). Effects of suppression and acceptance on emotional
responses of individuals with anxiety and mood disorders.
Behaviour Research and Therapy, 44, 1251-1263.
CiarTochi, J., & Blackledge, J. T. (2006). Mindfulness-based
emotional intelligence training: A new approach to reducing
human suffering and promoting effectiveness. In J. Ciarrochi,
J. P. Forgas, & J. D. Mayer (Eds.), Emotional intelligence in
everyday life (2nd ed., pp. 206-228). New York: Psychology

Press.

Ciarrochi, J., Chan, A., Caputi, P., & Roberts, R. (2001).

Measuring emotional intelligence. In J. Ciarrochi, & J. P.
Forgas (Eds.), Emotional intelligence in everyday life: A scientific
inquiry (pp. 25-45). Philadelphia: Psychology Press.
Ciarrochi, J., & Godsell, C. (2005). Mindfulness-based emotional
intelligence: A theory and review of the literature. In
R. Schulze, & R. D. Roberts (Eds.), Emotional intelligence: An
international handbook (pp. 69-90). Gottingen: Hogrefe &
Huber Publishers.
Ciarrochi, J., Robb, H., & Godsell, C. (2005). Letting a little non
verbal air into the room: Insights from acceptance and
commitment therapy Part 1: Philosophical and theoretical
underpinnings. Journal of Rational-Emotive and CognitiveBehavior Therapy, 23, 79-106.
Evans, B. J., Coman, G. J., Stanley, R. O., & Burrows, G. D.
(1993). Police officers coping strategies: An Australian police
survey. Stress Medicine, 9, 237-246.
Feldner, M. T., Zvolensky, M. J., Eifert, G. H., & Spira, A. P.
(2003). Emotional avoidance: An experimental test of
individual differences and response suppression using biologi
cal challenge. Behaviour Research and Therapy, 41, 403-411.
Goldberg, D. P. (1978). Manual of the General Health Questionnaire.
Windsor: National Foundation for Educational Research.
Gross, J. J. (2002). Emotion regulation: Affective, cognitive, and
social consequences. Psychophysiology, 39, 281-291.
Hart, P. M., & Cotton, P. (2002). Conventional wisdom is often
misleading: Police stress within an organisational health
framework. In M. F. Dollard, A. H. Winefietd, & H. R.
Winefield (Eds.), Occupational stress in the service professions (pp.
103-139). London: Taylor and Francis.
Hayes, S. C. (2001). Acceptance and commitment therapy,
relational frame theory, and the third wave of behavior therapy.
Behavior Therapy, 35, 639-665.
Hayes, S. C., Strosahl, K., & Wilson, K. G. (1999). Acceptance and
commitment therapy: An experiential approach to behavior change.
New York: Guilford Press.
Hayes, S. C., Strosahl, K., Wilson, K. G., Pistorello, J., Toarmino,
D., Polusny, M. A., et al. (2004). Measuring experiential
avoidance: A preliminary test of a working model. Psychological
Record, 54, 553-578.

282

V. Williams el al.

Hayes, S. C., Wilson, K. G Gifford, E. V., Follette, V. M &

Strosahl, K. (1996). Experiential avoidance and behavioral
disorders: A functional dimensional approach to diagnosis and
treatment.Journal of Consulting and Clinical Psychology, 64, 1 1 52
1168.

He, N., Zhao, J., & Archbold, C. A. (2002). Gender and police
stress: The convergent and divergent impact of work environ
ment, work-family conflict, and stress coping mechanisms of
female and male police officers. Policing, 25, 687-708.
John, O. P., & Gross, J. J. (2004). Healthy and unhealthy emotion

regulation: Personality processes, individual differences, and

life span development. Journal of Personality. 72, 1301-1333.
Kabat-Zinn, J. (1990). Full catastrophe living: Using the wisdom of
your body and mind to face stress, pain, and illness. New York:

Delacorte.
Kerr, S., Johnson, V. K Gans, S. E., & Krumrinc, J. (2004).
Predicting adjustment during the transition to college: Alex
ithymia, perceived stress and psychological symptoms. Journal
of College Student Development, 45, 593-607.
Kohan, A., & Mazmanian, D. (2003). Police work, burnout, and
pro-organizational behavior. Criminal Justice and Behavior, 30,
559-583.
Lennings, C. J. (1997). Police and occupationally related violence:
A review. Policing: International Journal of Police Strategies and
Management, 20, 555-566.
Lovibond, S. H., & Lovibond, P. F. (1995). Manual for the
Depression Anxiety Stress Scales (DASS). Sydney: University of
New South Wales.
Pennebaker, J. W. (1995). Emotion, disclosure and health: An
overview. In J. Pennebaker (Ed.), Emotion, disclosure and health
(pp. 3-10). Washington, DC: American Psychological Asso
ciation.
Richards, J., & Gross, J. J. (1999). Composure at any cost? The
cognitive consequences of emotion suppression. Personality and
Social Psychology Bulletin, 25, 1033-1044.
Robinson, H., Sigman, M., & Wilson, J. (1997). Duty-related
stressors and PTSD symptoms in suburban police officers.
Psychological Reports, 81, 835-845.

Ryan, R. M., & Dcci, E. L. (2000). Self-determination theory and

the facilitation of intrinsic motivation, social development, and
well-being. American Psychologist, 55, 68-78.
Salovey, P., Mayor, J. D., Goldman, S. L., Turvey, C., & Palfai,
T. P. (2002). Emotional attention, clarity, and repair:
Exploring emotional intelligence using the Trait Meta-Mood
scale. In J. W. Pennebaker (Ed.), Emotion, disclosure, and health
(pp. 125-154). Washington, DC: American Psychological
Association.
Segal, Z. V., Teasdale, J. D., Williams, J. M. G., & Gemar, M. C.
(2002). The Mindfulness-Based Cognitive Therapy adherence
scale: Inter-rater reliability, adherence to protocol and treat
ment distinctiveness. Clinical Psychology and Psychotherapy, 9,
131-138.
Sifneos, P. E. (1973). The prevalence of alexithymic symptoms
in psychosomatic patients. Psychotherapy and Psychosomatics,
22, 255-262.
Taylor, G. T. (2000). Recent developments in alexithymia
theory and research Canadian Journal of Psychiatry, 45, 134142.
Teasdale, J. D., Segal, Z. V., & Williams, J. M. G. (1995). How
does cognitive therapy prevent depressive relapse and why
should attentional control (mindfulness) training help? Beha
vior Research Therapy, 33, 25-39.
Violand, J. M. (1993). What does high stress policing teach
recruits? An analysis of coping. Journal of Criminal Justice, 21,

411-417.

Violanti, J. M., Marshall, J. R., & Howe, B. H. (1985). Stress,

coping and alcohol use: The police connection. Journal of Police
Science and Administration, 13, 106-110.
Wainer, H. (1991). Adjusting for differential base rates: Lords
paradox again. Psychological Bulletin, 109, 147-151.
Wegner, D. M., Erber, R., & Zanakos, S. (1993). Ironic
processes in the mental control of mood and mood-related
thought. Journal of Personality and Social Psychology, 65, 1093-

1104.
Wegner, D. M., & Zanakos, S. (1994). Chronic thought
suppression. Journal of Personality, 62, 615-640.

Copyright of Australian Psychologist is the property of Taylor & Francis Ltd and its content may not be copied
or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission.
However, users may print, download, or email articles for individual use.

Received 12/15/09
Revised 05/11/10
Accepted 07/20/10

Research
Managing Stress and Maintaining
Well-Being: Social Support, Problem-Focused
Coping, and Avoidant Coping
Ruth Chu-Lien Chao
study tested a model that links stress, social support, problem-focused coping, and well-being. First, it looks at
how high support significantly moderated the association between stress and well-being. Next, the students problemfocused coping was seen as mediating this moderated association. Finally, a 3-way interaction of stress, social support,
and avoidant coping revealed that only frequent use of avoidant coping accelerated the association between stress
and well-being in a negative way at both low and high support.

Life is full of stress for college students (Roberti, Harrington,

& Storch, 2006). For students to manage their stress, positive
social support and useful coping are essential. Bear in mind
that students manage stress differently; they assess stress, seek
support from families and friends, and execute their coping
all in their own ways.
However, although many researchers report that social sup
port and coping are positively associated with well-being (BenZur, 2009), there are two pitfalls in applying these associations
to students. One, social support is unstable or even decreasing
for some students in the recent decade (Arria et al., 2009), so it
is important to see how low support may change the association
between stress and students well-being. Two, not all coping
strategies are effective; some are functional and useful, whereas
others are less so (Carver, Scheier, & Weintraub, 1989). This
study aims to examine the moderating roles of social support
and coping on the association between stress and well-being.
From both theoretical and practical perspectives, there
is lack of clear understanding of the factors that can predict
the well-being of college students surrounded by a multitude
of stressors (Skowron, Wester, & Azen, 2004). Thus, in this
article I seek to address this gap by providing theoretical,
empirical, and practical insight into the conditions under
which stress is linked to the well-being of college students.
I integrate theories of social support and coping to propose
that the association between stress and well-being will be
buffered when students perceive high support that enhances
their problem-focused coping; in contrast, the association will
be deteriorated when students frequently use avoidant coping
when they perceive low support.
A particularly useful framework for understanding the po
tential protective nature of social support and coping is Lazarus

and Folkmans (1984) stress, appraisal, and coping theory. This

theory focuses on perceived stress (a relationship between the
person and environment that is appraised as exceeding available
resources), appraisal (ones perception and assessment of social
support in the situation), and coping (effortful or purposeful
thoughts and actions to manage or overcome stressful situations;
Frydenberg, 1997; Lazarus & Folkman, 1984). Lazarus and
Folkman characterized coping as problem-focused, referring
to attempts to engage, act on, or change the perceived stress
(Carver et al., 1989). This overarching conceptual framework .
has often served as a useful framework for understanding col
lege students stress experiences.

Stress and Well-Being of

College Students

Because college students are in transition from living at home

to living in a new city, in a new school, or with new room
mates, their developmental needs could be specifically related
to their college life. Their perceived stress can be attributed
to a multitude of social, academic, and emotional stressors
(Allgower, Wardle, & Steptoe, 2001), ranging from academic
difficulties to uncertainty about the future, conflicts with
friends, or problems with dating. Moreover, daily constant
stressors in college life are more detrimental to well-being
than episodic or change-related stressors (Lepore, Miles, &
Levy, 1997). What comes out as a major issue is how much
the students themselves feel their lives are unpredictable,
uncontrollable, and overloaded (S. Cohen, Kamarck, &
Mermelstein, 1983; Weinstein, Brown, & Ryan, 2009). S.
Cohen et al. (1983) added that the stress can be viewed as
assessing a state that places people at risk of, i.e., is anteced-

Ruth Chu-Llen Chao, Counseling Psychology Program, University of Denver. Correspondence concerning this article should be
addressed to Ruth Chu-Lien Chao, Counseling Psychology Program, Morgridge College of Education, University of Denver, 1999
East Evans Avenue, Denver, CO 80208 (e-mail: cchao3@du.edu).

2011 by the American Counseling Association. All rights reserved.

338

Journal of Counseling & Development

Summer 2011

Volume 89

Managing Stress and Maintaining Well-Being

Support, Problem-Focused
Coping, and Well-Being of College
Students

ent to, clinical psychiatric disorder (p. 394) and encouraged

scholars to "determine whether social support protects one
from the pathogenic effects of stressful events . . . appraised
stress causes an illness outcome (p. 393). Lazarus and Folkman (1984) suggested that the stress perceived as harmful
or threatening may exacerbate the impact of stress and may
be associated with lower well-being. Thus, the present study
measures college students perceived stresses that involve a
variety of stressors and are negatively associated with well
being (Schwitzer, 2008; Wang & Castaneda-Sound, 2008;
Weinstein et al., 2009).

Social support is negatively related to mental problems

(Brown et al., 1987). In addition, although many students live
stressful lives, some appear to handle stressors better than
their peers. For instance, some seek and obtain support from
families and/or friends to manage stress to maintain well
being, whereas others may lack support and feel devastated
(Hefner & Eisenberg, 2009; Lee, Keough, & Sexton, 2002).
Uneven levels of social support cannot always buffer stress and
facilitate growth; low social support is incapable of buffering
the negative impact of stress on well-being. Among the few
studies on low social support that are negatively associated
with well-being (e.g., Vermeulen & Mustard, 2000), individu
als with low social support were found more likely to engage
in less healthy activities such as sedentary behavior, alcohol
use, too much or too little sleep, and fatigue (Thorsteinsson
& Brown, 2008). Worse, people with low social support were
found linked to life dissatisfaction, even suicidal behavior
(Allgower et al., 2001). In interviews with 1,249 college
students, 6% reported having suicidal ideation, and low so

Support and Problem-Focused

Coping

According to Brown, Brady, Lent, Wolfert, and Hall (1987), sat

isfaction with social support is a function of the match between
the strength of ones interpersonal needsand the social resources
provided to fulfill those needs. It was further hypothesized that
dissatisfied interpersonal environment would be accompanied
by behavioral, emotional, and physiological strain. In short, the
present study focuses on how much college students are satis
fied with the social support they receive (Brown et al., 1987).
Moreover, research on social support and coping has shown
that satisfaction with social support is related to problemfocused coping (Sarid, Anson, Yaari, & Margalith, 2004).
Indeed, social support has been found to be related to coping
behaviors in college female students, and perceived satisfac
tory support will positively associate with students use of
problem-focused coping (Asberg, Bowers, Renk, & McKin
ney, 2008). I thusdraw on theories of social support (Brown et al.,
1987) and coping (Carver et ai., 1989) to propose that students
satisfied support may be related to their use of problem-focused
coping. Figure 1 shows four hypotheses of the study.

cial support was a predictor of suicidal ideation (Arria et al.,

2009). Taken together, data show that when people perceive
insufficient social support, they would lack a buffer against
life stress that deteriorates well-being (Hefner & Eisenberg,
2009). That is, when encountering stress, the college students
who have high social support may have a buffer to moder
ate the association between stress and well-being, whereas
those with low social support lack the buffer against stress. It
seems likely that during limes of increased stress associated
with the transition to college, social support may be a useful
way of insulating the individual from the harmful impact of
stress. According to Lazarus and Folkman (1984), a mismatch
between appraisal of available social support and perceived
stress can result in negative outcomes, whereas the appraisal
of adequate social support can serve as a buffer against per
ceived stress on well-being.

Hypothesis 1: Social support is positively associated with

students problem-focused coping.
(I report each hypothesis after reviewing relevant literature.
Concretely, Hypothesis 1 follows the literature review on
social support and problem-focused coping.)
H1

Social

ProblemFocused
Coping

Support

H4

Perceived
Stress

Avoidant
Coping

H2

H3b

H3a

Psychological
Well-Being

FIGURE 1

Model of Four Hypotheses

Wofe.The model links perceived stress, social support, problem-focused
coping, and avoidant coping to psychological well-being. H = hypothesis.

Journal of Counseling & Development

Hypothesis 2: Social support moderates the relation

between stress and well-being. High social support
may buffer against perceived stress that negatively
associates with well-being.

Summer 201 1

I then propose that by strengthening students use of

problem-focused coping, social support will further protect
the students well-being. Past research on college students has
revealed positive relationships between the use of problemfocused coping and well-being (Sarid et al., 2004). Specifi
cally, students who use problem-focused coping tend to be
more optimistic and persistent (Sarid et al., 2004). Because
stressed college students struggle to maintain their well-being,

Volume 89

339

Chao
their ability to achieve well-being is heavily dependent on
their use of problem-focused coping. Students with lower
levels of stress, in contrast, may be less concerned about us
ing problem-focused coping to buffer the association between
stress and well-being. Thus, problem-focused coping main
tains the well-being of stressed college students by enabling
them to deal with stressors with effective coping (Carver et al.,
1989; Sarid et al., 2004). These arguments suggest that social
support is likely to protect the well-being of stressed college
students by enhancing their use of problem-focused coping.
When people perceive stress, they appraise social support
and seek out coping resources. Lazants and Folkmans (1984)
theory characterized such coping as problem-focused, that is,
engaging, acting on, or changing the perceived stress (Carver
et al., 1989) so as to manage and/or overcome it.
Lazarus and Folkman (1984) suggested that coping effec
tiveness plays an important role in the impact of perceived
stress on psychological outcomes: Effective coping strate
gies such as problem-focused coping result in [managing]
situations in a way such as to mitigate stress when it occurs
(p. 198). Carver et al. (1989) further explained that problemfocused coping aims at problem-solving or doing something to
alter the perceived stress. One might expect that, on the basis
of Lazarus and Folkmans theoretical model, problem-focused
coping would buffer the impact of stress by influencing indi
viduals accurate appraisals of available coping resources and
using specific coping efforts that mitigate stress.
Hypothesis 3a: Problem-focused coping moderates the re
lationship between stress and well-being. The stronger
the problem-focused coping is, the greater the buffer
effect is on the relationship.
Next, I propose that social support buffers the association
between stress and well-being by enhancing students use of
problem-focused coping. Lopez, Mauricio, Gormley, Simko,
and Berger (2001) found that there are mediating roles for
coping in the relationship between attachment orientations
and current distress. Furthermore, problem-focused coping
was found to significantly mediate the relationship between
social support and well-being (Chen, Ma, & Fan, 2009). Thus,
I go further than Lazarus and Folkmans (1984) theory and
Lopez et al.s (2001) finding to predict that the moderating
effect of support on the relationship between stress and well
being is mediated by problem-focused coping. According to
Frazier,Tix, and Barron (2004), mediated moderation refers to
instances in which a mediator variable (i.e., problem-focused
coping) explains the relation between an interaction term (i.e.,
social support) in a moderator model and an outcome (i.e.,
well-being). Because mediated moderation is present when a
moderating effect is explained by a mediating process (e.g.,
Edwards & Lambert, 2007), my arguments suggest the fol
lowing hypothesis:

340

Hypothesis 3b: The moderating effect of social support

on the relationship between stress and well-being is
mediated by problem-focused coping.

Coping
Lazarus and Folkman (1984) stated that the effectiveness of a
coping strategy depends on the extent to which it is appropriate
to internal and/or external demands of the situation. Carver et
al. (1989) elaborated that certain responses to stress may tend
to be maladaptive. Specifically, the tendency to focus only on
venting frustration may be less useful to meet the demands of
the situation. Avoidant coping refers to the strategies with little
or no effectiveness (Roth & Cohen, 1986). Avoidant coping in
cludes three aspects: (a) focusing on and venting of emotions,
(b) behavioral disengagement, and (c) mental disengagement
(Carver et al., 1989). Focusing on and venting of emotions
indicates how distress is central in emotions without adaptive
behaviors. Behavioral disengagement stops ones struggling
to deal with stress, and the stress still remains. For instance,
sleeping away stress is an avoidant coping behavior (Carver
et al., 1989). Mental disengagement puts stress out of sight
by various activities (out of sight, out of mind).
Avoidant coping may also serve as an important source
of information about college students well-being. Coping
theorists have long argued that in addition to relying on social
support to manage stress, college students sometimes use
avoidant coping (Ben-Zur, 2009; Brown et al., 1987; Lopez
et al., 2001). Thus, it is important to understand avoidant
coping because it is prevalent among current college students
(Brougham, Zail, Mendoza, & Miller, 2009). For example, to
reduce stress, the five most frequent coping strategies among
students were browsing the Internet, sleep and rest, using
instant messaging, complaining, and watching television or
movies (Sideridis, 2008).
Individual differences in avoidant coping the tendency to
distract from stressors (Carver et al., 1989) appear to play a
critical role in shaping how students process the social support
they receive and translate that information into the association
between stress and well-being. Drawing on theories of coping,
I propose that high avoidant coping may deteriorate the effect
of perceived social support or further weaken the association.
General support for this proposition appears in Carver et al.s
(1989) model of coping. I predict that, when students lack
support, high avoidant coping may further deteriorate the re
lationship between stress and well-being. The logic underlying
this idea is that students who lack social support may also be
deprived of resources, protection, and role modeling, which
are crucial buffers to stress. When social support is low, the
buffer is also low, and students are more vulnerable to stress,
especially when they frequently use avoidant coping. Besides,
students who perceive low support and frequently use avoidant
coping may have the lowest scores in well-being.

Journal of Counseling & Development

Summer 2011

Volume 89

Managing Stress and Maintaining Well-Being

It is interesting to note that even with high social support,

some college students may still use their own avoidant cop
ing to manage their stress. Carver et al. (1989) suggested
that some individuals habitually use avoidance or distraction
to cope with stress. Lopez et al. (2001) found that avoidant
coping is convenient, easy, and quick to use. Thus, I predict
that even in the presence of sufficient social support, high
avoidant coping would exacerbate the association between
stress and well-being, and low avoidant coping would make
no difference on the association because students did not have
many avoidant behaviors.
Hypothesis 4: There will be a three-way interaction of
stress, support, and avoidant coping in predicting
well-being, such that the association will be weakest
when the support is low but avoidant coping is high.

Present Study
In this study, 1 constructively tested the four hypotheses (see
Figure 1) with a sample of nonclinical college students. This
study examines how college students handle stress and aims
to determine (a) whether social support relates to problemfocused coping; (b) whether social support moderates the
association between stress and well-being; (c) whether
problem-focused coping mediates the moderation of sup
port on the association; and (d) whether there is a three-way
interaction among stress, support, and avoidant coping. It is
my hope that the findings will assist college students to deal
with stress and help advance knowledge on avoidant coping
in the face of life stress. Because college students well-being
could be confounded with three covariates (age, year in col
lege, and psychotherapy), I controlled these covariates. Note
that, first, previous studies reported that age might influence
the study results if it was not controlled (e.g., Nilsson, Leppert, Simonsson, & Starrin, 2010; Noor, 2008). Next, senior
students have more college years and are more familiar with
college life, and so may have higher levels of well-being
(Bettencourt, Charlton, Eubanks, Kemahan, & Fuller, 1999).
Finally, psychotherapy undergone by students might be a
confounding variable for their well-being (Colbert, Jefferson,
Gallo, & Davis, 2009) because they may know more than
students without therapy how to increase well-being. For
these reasons, I examine these three variables as covariates.

Participants
A total of 459 college students responded to the online sur
vey; 390 (85%) were Caucasian White, 32 (7%) were African
American, 23 (5%) were Latino, and 14 (3%) were Asian
American. Among these participants, 239 (52%) were men
and 220 (49%) were women at a large public university, with
a mean age of 20.23 years old (SD = 3.45, range = 1 8-35

Journal of Counseling & Development

Summer 2011

years old). These participants included 120 (26%) freshmen,

115 (25%) sophomores, 110 (24%) juniors, and 114 (25%)
seniors. Among them, 12% ( = 55) reported having had
therapy or counseling.

Instruments
Perceived Stress Scale. The Perceived Stress Scale (PSS;
S. Cohen et al., 1983) is a 10-item measure, with a 5-point
Likert scale ranging from 1 (not at all satisfied) to 7 (very
satisfied), to assess the degree to which individuals perceive
their lives as stressful. Higher scores indicate greater percep
tions of life stress, and lower scores reflect lower perceptions
of stress. The PSS showed adequate coefficient alphas, .84
and .85 for two college samples (S. Cohen et al., 1983); in
this study, the coefficient alpha was .85. The PSS has been
positively correlated with life-event scores, depressive and
physical symptomatology, social anxiety, and maladaptive
health-related behaviors such as increased smoking (S. Cohen,
Sherrod, & Clark, 1986). These pieces of evidence indicate
the construct and concurrent validity of the PSS. Further
more, Kuiper, Olonger, and Lyons (1986) found the PSS to
be associated with peoples greater vulnerability to depressive
symptoms related to stressful life events.
Social Support Inventory. The Social Support Inventory
(SSI; Brown et al., 1987) is a 39-item questionnaire that
assesses satisfaction with support and help received from
others over the previous month. Ratings are made on 7-point
Likert-type scales ranging from 1 (not at all satisfied) to 7
(very satisfied). The SSI has five subscales: Acceptance and
Belonging, Appraisal and Coping Assistance, Behavioral and
Cognitive Guidance, Tangible Assistance and Material Aid,
and Modeling. Brown et al. (1987) suggested that the total
score can be obtained by summing up ratings over 39 items.
Higher scores indicate higher perception of and satisfaction
with support received. Brown et al. conducted a factor analysis
of SSI among 340 college students from three universities and
reported satisfactory coefficient alphas of the five subscales,
from .79 to .91. For the total score, the coefficient alpha was
.95 in Brown et al. and .90 for the present sample. The SSI had
appropriate construct validity, being positively related to other
social support measures and negatively related to depression
measures (Brown et al., 1987).
Coping. Problem-focused coping and avoidant coping were
measured with the COPE Inventory (Carver et al., 1989),
which assesses different ways of responding to stress. The 60item self-report measure uses a 4-point Likert scale ranging
from 1 (/ usually don V do this at all) to 4 (/ usually do this a
lot) to assess 13 patterns of coping, including problem-focused
coping and avoidant coping. Among these 1 3 patterns of cop
ing, problem-focused coping is composed of five scales: Ac
tive Coping, Planning, Suppression of Competing Activities,
Restraint, and Use of Instrumental Social Support. Avoidant
Coping has three scales: Focus on and Venting of Emotions,

Volume 89

341

Chao
Behavioral Disengagement, and Mental Disengagement. High

scores of particular patterns indicate high tendency to use

these coping patterns in stress. Carver et al. (1989) reported
that coefficient alphas of the scales were adequate for both
problem-focused coping (.83) and avoidant coping (.80); my
sample had a coefficient alpha of .85 for problem-focused
coping and .84 for avoidant coping. Convergent validity has
been demonstrated with numerous personality measures, in
cluding optimism, control, self-esteem, intemality, hardiness,
self-monitoring, and anxiety. The COPE Inventory has also
been used to assess strategies for coping with specific life
events (Carver & Scheier, 1994). Previous studies have shown
that the COPE Inventory fit the original factor structures well,
with adequate convergent and discriminant validity. The COPE
Inventory is correlated with various measures, including
hassles and uplifts, physical symptoms, degree of satisfaction
with life, positive affectivity, and negative affectivity (Clark,
Borman, Cropanzano, & James, 1995).
Mental Health Inventory. The Mental Health Inventory
(MHI; Veit & Ware, 1983) is a 38-item measure assessing both
distress and well-being among adults. Ratings are made on a
5-point Likert scale. The MHI has two scales: Psychological
Distress and Psychological Well-Being. The present study used
the Psychological Well-Being scale, which Veit and Ware (1983)
reported as appropriate as an outcome measure. The MHI has
been used extensively in studies of nonpsychiatric samples, and
its psychometric adequacy has been well established (Siegel,
Karus, Raveis, & Hagen, 1998; Veit & Ware, 1983). My sample
was found to have a coefficient alpha of .89. Concurrent and
convergent validity for the MHI-Psychological Well-Being
scale have been established through positive correlations with
measures on positive affect (Siegel et al., 1998).
Demographic information. The demographic informa
tion includes questions on participants ethnicity/race, age,
sex, year in college, and whether they have had therapy or
counseling.

Procedure
Participants completed answering an online questionnaire
package containing the PSS, SSI, COPE problem-focused and
COPE avoidant scales, MHI-Psychological Well-Being scale,
and a demographic questionnaire. Immediately affer comple
tion of the questionnaire, participant data were stored in a
password-protected data file on a networked computer. Par
ticipants login details were immediately placed in a separate
file; these files could not be combined to identify individual
participants responses. However, the login data file was used
to verify individual student participation; these data were also
used to verify that participants only participated once.
Before the survey began, a research coordinator announced
the study project in psychology classes, such as introductory
psychology, psychological assessment, personality theory,
experimental psychology, and statistics.The research coordi
nator then e-mailed 500 students an invitation and a reminder.

342

Four hundred and eighty-five students agreed to participate,

479 students completed the survey, and 20 students responded
to a validity item inaccurately. So, the completed data of 459
participants (a response rate of 92% [459/500 = 92%]) were
used in the analyses. Those who agreed to participate received
an invitation e-mail with a survey link. Participants were in
formed that the goal of the study was to identify factors that
can change well-being among college students. The online
survey indicated to the participants that by completing the
survey, they had consented to participate in the study. After
the survey was completed, a debriefing form was provided.
Also, in addition to receiving their research credit, partici
pants could provide their contact information, which would
be stored in a separate data file, to enter a drawing for one of
ten $20 cash prizes.

Preliminary Analyses and Descriptive Statistics

The descriptive statistics, internal consistency reliability
estimates, and correlations among variables of interest were
examined for the study. The zero-order correlations among
the variables indicated that both perceived stress and avoid
ant coping were related negatively to overall psychological
well-being, but social support and problem-focused coping
were related positively to well-being. A series of analyses of
variance (ANOVAs) were then conducted on well-being to
examine whether the dependent variable varied as a function
of participants sex and ethnicity. None of these demographic
variables had significant effects on well-being: sex, F(l, 458)
= 0.69, p > .05, and ethnicity, F(4, 455) = 0.78, p > .05. Thus,
in general, the dependent variable did not differ significantly
in terms of these demographic variables (all ps > .05). Yet, to
college students, stress may be perceived differently according
to differences in age, year of college, and psychotherapy (Thorsteinsson & Brown, 2008), therefore it was critical to control
these factors before conducting multiple regression analyses.
Another issue with multiple regression analysis is nor
mality, because substantial departures from normality can
adversely affect the analyses, and thus the data need to meet
regression assumptions of normality, linearity, and homoscedasticity (J. Cohen, Cohen, West, & Aiken, 2003). Results of
multiple regression analyses indicated that the skewness of
residuals ranged from -0.17 (Z = -l.33,p> .05) to -0.19 (Z
= -1.54, p > .05) and the kurtosis of residuals ranged from
0.54 (Z= 1.65,p> .05) to 0.57 (Z= 1.68,p> .05); the results
indicated no statistical departure from normality.

Two-Way Interaction of Stress and Social Support

To test the four hypotheses (see Figure 1) in the link between
perceived stress and psychological well-being, I followed
Dawson and Richters (2006) recommendation and used
hierarchical multiple regression analyses to test moderator
effects. I followed Aiken and Wests (1991) suggestion and

Journal of Counseling & Development

Summer 2011

Volume 89

Managing Stress and Maintaining Well-Being

used centered variables (i.e., mean deviation scores) to reduce

multicollinearity between interaction terms and main effects
when testing for moderator effects.
In support of Hypothesis 1, social support was positively
associated with problem-focused coping (B = 1.58, SE =
0.32, p = .23, t = 4.97, p < .001). The results of analyzing
Hypothesis 2 are displayed in Table 1, Steps 1, 2, and 3.
Step 1 indicated that no covariate (i.e., age, year in college,
and psychotherapy) significantly accounts for the variance
of psychological well-being in college students. In Step 2,
perceived stress and social support accounted for 40% of
well-being. In Step 3, the two-way interactions significantly
predicted well-being (AR} = .01,/? < .05). In addition, the
regression coefficient for the two-way interaction of Per
ceived Stress x Social Support was statistically significant
(see Table 1). Champoux and Peters (1987) comprehensively
reviewed social science literature and reported that interac
tion terms typically account for approximately 1% to 3% of
the variance, although J. Cohen (1992) indicated that an R1
value of .01 indicates a small effect size.
After a significant two-way interaction effect was found, the
next step was to interpret the interaction by plotting social support
scores for stress scores of one standard deviation aboveand below
the mean (Aiken & West, 1991). To check whether the slopes of
simple regression lines at high and low social support significantly
differed from zero, I conducted simple regression analyses outlined
by Aiken and West (1991). This regression models criterion vari
able (i.e., well-being) is regressed on the predictor (i.e., stress), the
moderator (i.e., social support) at a conditional value (e.g., high or
low), and the Predictor x Moderator interaction.The t test for the
regression coefficient of the predictor variable in this equation did
reflect tire significance of the simple slope (i.e., whether the slope
is significantly different from zero).

The results indicated that the slope with high social support
was insignificantly different from zero (B = -0.52, SE = 0.6 1,
P = -.05, t = -0.86, p = .39), but the slope with low social
support was significantly different from zero (B = -2.80, SE
= 0.58, P = -.25, t = -4.85, p < .001). These results indicate
that high social support served as a buffer for college students
to keep up their well-being, whereas low support was not a
buffer against stress. The difference between these two regres
sion lines was also significant, as indicated by the significant
regression coefficients found for the interaction terms in the
tests of the moderator effects, according to the procedure
recommended by Aiken and West (1991).

Test of Mediated Moderation by Problem-Focused

Coping
Next, I sought to examine whether problem-focused cop
ing mediated the moderating effect of social support on the
relationship between stress and well-being. To do so, I used
the path analysis framework developed by Edwards and
Lambert (2007). Because my previous results supported the
link between social support and problem-focused coping, I
turned to Hypothesis 3a, which stated that problem-focused
coping would moderate the association between stress and
well-being. Following recommendations from Edwards and
Lambert, I tested this hypothesis with a hierarchical regression
analysis in which I first controlled for stress, support, and the
interaction product of stress and support (see Table 1 , Step 3)
and then entered problem-focused coping and the interaction
product of stress and problem-focused coping (see Table 1,
Steps 4 and 5). The result in Table 1, Step 5, showed that, as
hypothesized, the interaction of Perceived Stress x ProblemFocused Coping significantly predicted psychological well
being of college students.

TABLE 1

Hierarchical Multiple Regression Analysis Predicting Psychological Well-Being:

Test of Hypotheses 2 and 3a
B

Step and Variable

Step 1
Age

College year
Previous counseling
Step 2

Perceived stress
Social support
Step 3
Perceived Stress x
Social Support
Step 4
Problem-focused coping
Step 5
Perceived Stress x
Problem-Focused
Coping

SEB

0.24
-0.63
2.46

0.19
0.53
2.77

.06
-.06
.04

1.25
-1.19
0.89

-1.32
6.98

0.47
0.44

-.12
.60

-3.97*"
15.89***

-0.01

0.00

-.57

-2.80

1.54

0.45

.13

3.45

0.96

0.45

-.08

R1

AR2

.01

.01

AF
1.03

(1,426)

.41

.40

140.25*

(2, 424)

.42

.01

7.86*

(3, 423)

.43

.01

11.87

(1,422)

.44

.01

4.25*

(1, 421)

d(

-2.06*

Note. N = 459.

*p < .05. p < .01. ***p < .001.

Journal of Counseling & Development

Summer 2011

Volume 89

343

Chao
I interpreted the interaction by plotting the simple slopes
for the relationships between stress, problem-focused coping,
and well-being at one standard deviation above and below
the mean, The simple slopes indicated that stress was more
negatively associated with well-being when college students
reported low rather high problem-focused coping. Similar to
the pattern observed for social support, the simple slopes for
the relationship between stress and well-being were negative
and differed significantly from zero at low levels of problemfocused coping (B = -3.40, SE = 0.79, p = -.30, t = -4.28,
p < .001) but did not differ significantly from zero at high
levels of problem-focused coping (B = -0.7 1 ,SE = 0.72, p =
-.06, t = -0.98, p = .33). These results indicate that, in sup
port of Hypothesis 3a, problem-focused coping did buffer
the relationship between students stress and well-being, and
students who had high problem-focused coping had signifi
cantly higher well-being than those with low problem-focused
coping. Hypothesis 3b predicted that problem-focused coping
would mediate the moderating effect of social support on
the association between stress and well-being. I completed
the test of mediation following the recommendations from
Edwards and Lambert (2007) and estimated the indirect
effects of the Perceived Stress x Social Support interaction
through problem-focused coping. I used the coefficients from
the prior analyses and then applied bootstrapping methods to
construct bias-corrected confidence intervals on the basis of
1,000 random samples with replacement from the full sample.
Mediation occurs when the size of an indirect effect differs
significantly from zero (MacKinnon, Fairchild, & Fritz, 2007).
The size of the indirect effect from the full sample was -3.24
(1.58 x -2.05), and the 95% confidence interval [-5.33, -1.15]
excluded zero. Thus, in support of Hypothesis 3b, perceived
task significance mediated the moderated effect of social
support on the relationship between stress and well-being.

Three-Way Interaction of Perceived Stress, Social

Support, and Avoidant Coping
I tested Hypothesis 4 by examining the three-way interaction
among perceived stress, social support, and avoidant cop
ing. All terms (all predictors and two-way interactions) were
entered, followed by the three-way interaction term, in the
regression analyses. Data in Table 2 show that, in Step 4, a
three-way interaction was found to significantly contribute to
the variance of well-being (AR2 = .01,p< .05). After seeing
that Step 4 showed significant three-way interaction, I then
used the same procedure described earlier to plot the three-way
interaction. As illustrated in Figure 2, among students who
reported high levels of social support, the association between
stress and well-being was not significantly different from zero
for those who only infrequently used avoidant coping (B =
0.78, SE = 0.96, p = .07, t = 0.81 p > .05) but was significantly
different for those who frequently used avoidant coping (B
= -1.86, SE = 0.73, p = -.16, l--2.54,p < .05). The results
indicate that even with high social support, frequent use of
avoidant coping would still affect the association between
stress and well-being in a negative way (see Figure 2).
Lower social support relates to lower well-being than does
higher support, and a similar finding appears on the patterns
of interaction of stress and avoidant coping (see Figure 2).
When students perceived low support, those with frequent
avoidant coping had the lowest scores on well-being (B =
-3.88, SE = 0.95, p =-.33, t = -4.02,/? < .001). In contrast,
even perceiving low support, those who used less avoidant
coping exhibited a pattern in which the association between
stress and well-being was not significantly different from zero
(B = 0.40, SE = 0.79, P = .04, t = 0.51 , p > .05). These results
indicated that college students who reported low social support
and frequent use of avoidant coping were most vulnerable to

TABLE 2
Hierarchical Multiple Regression Analysis Predicting Psychological Well-Being From Perceived Stress,
Social Support, Dysfunctional Coping, and Their Interaction: Test of Hypothesis 4
Step and Variable
Step 1
Ago
College year
Previous counseling
Step 2
Perceived stress
Social support
Avoidant coping

Step 3
Perceived Stress x Social Support
Perceived Stress x Avoidant Coping
Social Support x Avoidant Coping
Step 4
Perceived Stress x Social Support x
Avoidant Coping

SE B

0.24
-0.63
2.46

0.19
0.53
2.77

.06
-.06

-1.32

0.47
0.44
0.47

-.12"
.60"*
-.08*

-0.01

0.00
0.40
0.44

.57**
.02
.02

6.98
-0.96

0.23
0.21
-0.79

0.38

.04

Ft1

AF

.01

Afl2
.01

1.03

df
(1, 426)

.41

.40

95.58*"

(3, 423)

.42

.01

2.88*

(3, 420)

.43

.01

4.32*

(1,419)

-.09*

Note. N= 459.
*p < .05. **p<.01.***p<.001.

344

Journal of Counseling & Development

Summer 2011

Volume 89

Managing Stress and Maintaining Well-Being

.....

= High Support/Low Avoidance

= High Support/High Avoidance
= Low Support/Low Avoidance
= Low Support/High Avoidance

60

B = 0.78

55-

| 50I 45S

CD

40

B= 1.86**
-O

B = 0.40

S = -3.88***

3530 -I

Low (1 SD below)

High (1 SD above)
Perceived Stress

FIGURE 2
Three-Way Interaction of Stress,
Social Support, and Avoidance Coping on
_ Psychological Well-Being
**p<.01.***p<.001.

stress, and infrequent use of avoidant coping would not worsen

the association between stress and well-being.

The results provide initial support for Hypotheses 1, 2, 3a, 3b,

and 4. Social support appears to buffer the relationship between
perceived stress and psychological well-being of college students,
and this moderating effect is mediated by problem-focused cop
ing. In addition, as predicted in the three-way interaction, avoidant
coping deteriorates the association between stress and well-being,
especially when there is low social support. Specifically, students
using high avoidant coping in a low social support environment
may have the lowest well-being when they are in stress. Taken to
gether, the present study offers important contributions to research
on stress, social support, coping, and well-being in three ways.
First, the study extends a growing body of research on
stress and well-being. Rather than assuming that stress is
always negatively associated with well-being, I examined the
conditions under which stress would be more and less likely
to predict these outcomes. In doing so, my findings provide
novel insights into the role of social support and a significant
two-way interaction between stress and social support. When
in stress, college students need assurance that others (e.g.,
friends, family) are willing to listen and talk. Unfortunately,
some students may feel dissatisfied with the social support
they receive (Arria et al., 2009). Not surprisingly, students
with unsatisfactory support in their social environment were
quite vulnerable, showing a low level of well-being.

Journal of Counseling & Development

Summer 2011

Moreover, many college students live at the crossroads of

various stressors, struggling for independence, problems with
roommates or friends, worries about dating, and concerns
about grades. All these stressors can make students feel as
if things happen to them unpredictably, unexpectedly, and
overwhelmingly. For instance, who could provide them sup
port by serving as models on how to bounce back to Grade
A after they fail a midterm exam? How could they see hope
and love after breaking up with a loved one? Students face
a long list of daily stressors. Rather than detailing stressors,
scholars, educators, and counselors must understand that low
levels of support diminish well-being, and as I have specified,
particular components of support moderate the association
between stress and well-being.
Second, problem-focused coping in the present study
serves two roles regarding stress and well-being. On the
one hand, a high level of problem-focused coping helped
students maintain their well-being when in stress, whereas a
low level of problem-focused coping affected this association
in a negarive way. Simply put, when students face stressors,
dealing with the problems may be a good way to keep peace
of mind while postponing or not solving the problem may
take a mental or emotional toll on their well-being. On the
other hand, problem-focused coping significantly mediates the
moderation of support on the association between stress and
well-being. In other words, when individuals feel supported
by family and friends, concentrating on solving problems may
further help them maintain their well-being.
Third, a significant three-way interaction of Perceived
Stress x Social Support x Avoidant Coping significantly con
tributes to the variance of college students well-being. High
avoidant coping reduces well-being at both high and low social
support (see Figure 2); however, low avoidant coping does
not affect the association between stress and well-being in a
negative way at high and low support. Thus, taken together,
the findings highlight the significance of avoidant coping.
Specifically, even as students perceive high levels of social
support, continuing avoidant coping will lower their well
being. In other words, avoidant coping may overpower social
support to reduce well-being. Worse, in recent years, avoidant
coping has been used by some college students (Brougham et
at., 2009) and appears to weaken well-being, but to date, not
much scholarly attention is paid to this problem.
These findings have advanced the knowledge on college
students mental health in two ways. First, social support is
indispensable to the students. Students well-being wanes
without social support buffering harmful stress, and the
moderation effect of support is mediated by problem-focused
coping. Second, high divorce rate, unsteady family income,
and absentee parents are indexes to low social support that
fail to alleviate students stress at school and in relationships
with others (Bulduc, Caron, & Logue, 2007). The findings
suggest that when support is low, students are more vulnerable

Volume 89

345

Chao

to avoidant coping. The studys results of the exacerbating

effects of low social support and avoidant coping on stress
and well-being contribute to the understanding of college life.
Limitations
The present study has four limitations. One, the findings
describe current college students social support, problemfocused coping, avoidant coping, and well-being based on the
theories of Lazarus and Folkman (1984), Brown et al. (1987),
and Carver et al. (1989) and cannot be generalized to other
populations. People of different ages and status of learning
may have different or even contrasting perceptions on stress
and well-being. For example, seniors may embrace the value
of toughing out stress, whereas some young adults may
prefer to have fun (Sideridis, 2008). Two, the study examined
two types of coping, problem-focused and avoidant, to better
describe that people may use more than one type of coping
to manage stress; however, the study does not focus on the
relation between these two types of coping. The findings did
not examine when and how students use a particular type of
coping over the other, although it is understood that students
may simultaneously use problem-focused coping academically
and avoidant coping (e.g., venting emotions) for other issues.
Three, the findings may only explain the avoidant coping of
students in the United States and do not demonstrate that
avoidant coping is cross-culturally universal in deteriorating
the well-being of all students. Fortunately, this limitation
could be resolved by future studies. Four, there is a potential
monomethod bias based on using self-report measures alone.
The participants were recruited in classrooms and then fol
lowed up with e-mails, and so the sampling process could be
limited in external validity.

Future Research Directions

There are four recommendations for future research. First, I
recommend replicating this study among students in different
cultures. For example, I found that social support is critical
for students, but students of different cultures may need other
types of support, such as financial help (Olson, Garriott,
Rigali-Oiler, & Chao, 2009). Appreciating the role of a specific
component of social support in a specific cultural framework
is essential for counselors who attempt to provide effective
help to students of various cultures. Second, it may be benefi
cial to examine how college students in different cultures use
problem-focused and/or avoidant coping. Specifically, some
Asian students were found to use behavioral disengagement
more often than U.S. students did, whereas U.S. students used
problem-focused coping more often than students in other
countries (Connor-Smith & Flachsbart, 2007; Cross, 1995)
did. Thus, it is necessary to know how problem-focused and
avoidant coping interweave with culture. Third, the present
study relies on self-report measures, so the connections
between the variables need to be further evaluated by other

346

types of research. I suggest that future studies examine social

support and avoidant coping via other methodologies (such as
qualitative research). For example, students subjective reflec
tion may prove or disprove my results, offer information I may
not be aware of, or extend my results to provide a better under
standing how college students cope with stress. It is equally
important to further evaluate interventions based on the model
in the present study. For instance, future study could examine
the effectiveness of interventions, such as social support as a
buffer or problem-solving coping as a mediator between stress
and well-being. Fourth, how to maintain well-being in the face
of stress has been a challenge for college students (Schwitzer,
2008; Wang & Castaneda-Sound, 2008). The present study
examines a nonclinical sample, and so it appears appropriate
to take well-being, not distress, as an outcome variable. Future
studies may need to include distress as an outcome variable to
see how social support, problem-solving coping, and avoidant
coping relate to psychological distress.

Implications for Counseling

This study has three implications for counseling. To begin
with, as the literature review shows, when students perceive
low levels of social support, they may have less of a buffer
against stress. So, counselors would do well to invite the
students to talk about how they perceive their social sup
port rather than assuming that most students enjoy a highly
supportive life. For students with low social support or lack
of a buffer against stress, counselors can help them explore
which specific component(s) of social support was so low as
to render them quite vulnerable.
Next, counselors can facilitate students problemfocused coping. Enhancing problem-focused coping could
help students in stress in crucial ways. For example, coun
selors can offer psychoeducation to help students make
plans, such as breaking problems into workable steps or
having mentors to guide in academic work. Learning how to
solve problems to work through stress could very possibly
increase their well-being. Or students can learn to cultivate
self-assurance and hope, thereby buffering their stress.
Shapiro, Oman, Thoresen, Pante, and Flinders (2008) re
ported that increased mindfulness mediated reductions in
perceived stress and rumination.
Furthermore, counselors can allow college students to
describe what, why, how, and when they use avoidant coping.
The more counselors understand the function of this type
of coping in students life, the more effective counselors
interventions would become. For instance, some students
may struggle with academic stress without knowing how to
take notes or make reading plans, to the point that they avoid
problems by resorting to listening to music, complaining, or
mentally disengaging. Counselors resources on study skills
could enable the students to reduce avoidant coping. Thus,
low avoidant coping, especially combined with high levels

Journal of Counseling & Development

Summer 2011

Volume 89

Managing Stress and Maintaining Well-Being

of support, will help students maintain their well-being and
make them less vulnerable to stress (see Figure 2).
In sum, in addition to the moderating effect of social
support on the association between stress and well-being,
this study advanced knowledge on coping. Problem-focused
coping significantly mediates the moderating effect of social
support on the association between stress and well-being, but
avoidant coping has a negative effect on association.

Aiken, L., & West, S. G. (1991). Multiple regression: Testing and

interpreting interactions. Newbury Park, CA: Sage.
Allgower, A., Wardle, J., & Steptoe, A. (2001). Depressive symptoms, so
cial support, and personal health behaviors in young men and women.
Health Psychology, 20. 223-227. doi:I0.1037/0278-6133.20.3.223
Arria, A. M OGrady, K. E Caldeira, K. M., Vincent, K. B., Wilcox,
H. C & Wish, E. D. (2009). Suicide ideation among college
students: A multivariate analysis. Archives of Suicide Research,
13, 230-246. doi:10.1080/1381 1 1 10903044351
Asberg, K. K., Bowers, C., Renk, K., & McKinney, C. (2008). A
structural equation modeling approach to the study of stress and
psychological adjustment in emerging adults. Child Psychiatry &
Human Development, 39, 481-501. doi:10.1007/s10578-008-0102-0
Ben-Zur, H. (2009). Coping styles and affect. International Journal
of Stress Management, 16, 87-101. doi:10.1037/a0015731
Bettencourt, B. A., Charlton, K., Eubanks, J., Kemahan, C., &
Fuller, B. (1999). Development of collective self-esteem among
students: Predicting adjustment to college. Basic and Applied
Social Psychology, 21, 213-222.
Brougham, R. R., Zail, C. M., Mendoza, C. M., & Miller, J. R.
(2009). Stress, sex differences, and coping strategies among
college students. Current Psychology, 28, 85-97. doi:10.1007/

S12144-009-9047-0
Brown, S. D Brady, T., Lent, R. W., Wolfert, J., & Hall, S. (1987).
Perceived social support among college students: Three studies
of the psychometric characteristics and counseling use of the
Social Support Inventory. Journal of Counseling Psychology,
34, 337-354. doi: 10.1037/0022-0167.34.3.337
Bulduc, J. L., Caron, S. L., & Logue, M. E. (2007). The effects of
parental divorce on college students. Journal of Divorce & Re
marriage, 46, 83-104. doi:10.1300/J087v46n03_06

Carver, C. S., & Scheier, M. F. (1994). Situational coping and

coping dispositions in a stressful transaction. Journal of Per
sonality and Social Psychology, 56, 267-283. doi:10. 1 037/00223514.56,2.267

Carver, C. S., Scheier, M. F., & Weintraub, J. K. (1989). Assessing

coping strategies: A theoretically based approach. Journal of Per
sonalityand Social Psychology, 56, 267-283. doi:10.1037/0022-

3514.56.2.267
Champoux, J. E & Peters, W. S. (1987). Form, effect size and power
in moderated regression analysis. Journal of Occupational Psy
chology, 60, 243-255.

Journal of Counseling & Development

Summer 2011

Chen, L.-F., Ma, S.-B., & Fan, C.-X. (2009). A structural equation
model of stress, coping style, social support and subjective
well-being for college students. Chinese Journal of Clinical
Psychology, 17, 266-268.
Clark, K. K., Borman, C. A., Cropanzano, R. S., & James, K. (1995).
Validation evidence for three coping measures. Journal of Person
ality Assessment, 65, 434-455. doi:10. 1 207/sl 5327752jpa6503_5
Cohen, J. (1992). A power primer. Psychological Bulletin, 112,
155-159. doi:10.1037/0033-2909.112.1.155
Cohen, J., Cohen, P., West, S. G., & Aiken, L. S. (2003). Applied
multiple regression/correlation analysisfor the behavioral sci
ence (3rd ed.). Mahwah, NJ: Erlbaum.
Cohen, S., Kamarck, T., & Mermelstein, R. (1983). A global measure
of perceived stress, Journal of Health and Social Behavior, 24,
385-396. doi:10.2307/2136404
Cohen, S., Sherrod, D. R., & Clark, M. S. (1986). Social skills and
the stress-protective role of social support. Journal of Person
ality and Social Psychology, 50, 963-973. doi:10.1037/0022-

3514.50.5.963
Colbert, L. K., Jefferson, J. L., Gallo, R., & Davis, R. (2009). A study
of religiosity and psychological well-being among African Ameri
cans; Implications for counseling and psychotherapeutic process.
Journal of Religion and Health, 48, 278-289. doi: 10. 1007/

S10943-008-9195-9
Connor-Smith, J. K & Flachsbart, C. (2007). Relations between
personality and coping: A meta-analysis. Journal of Personal
ity and Social Psychology. 93, 1080-1107. doi:10.1037/00223514.93.6.1080
Cross, S. E. (1995). Self-construals, coping, and stress in crosscultural adaptation. Journal of Cross-Cultural Psychology, 26,
673-697. doi: 10. 1 177/002202219502600610
Dawson, J. E, & Richter, A. W. (2006). Probing three-way interactions
in moderated multiple regression: Development and application
of a slope difference test. Journal of Applied Psychology, 91,
917-926. doi:10.1 037/0021-9010.91.4.917
Edwards, J. R., & Lambert, L. S. L. (2007). Methods for integrating
moderation and mediation: A general analytical framework us
ing moderated path analysis. Psychological Methods, 12, 1-22.
doi:10.1037/1 082-989X.12.1.1
Frazier, P A.,Tix, A. P, & Barron, K. E. (2004). Testing moderator and
mediator effects in counseling psychology research JournalofCoun
seling Psychology, 51, 115-134. doi:10.1037/0022-0167.51.1.1 15
Frydenberg, E. (1997). Adolescent coping: Research and theoretical
perspectives. London, England: Routledge.
Hefner, J., & Eisenberg, D. (2009). Social support and mental health
among college students. American Journal of Orthopsychiatry,
79, 491-499. doi:10.1037/a0016918
Kuiper, N. A., Olonger, J., & Lyons, L. M. (1986). Global per
ceived stress level as a moderator of the relationship between
negative life events and depression. Journal of Human Stress,
12, 149-153.
Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, and coping.
New York, NY: Springer.

Volume 89

347

Chao
Lee, R. M., Keough, K. A., & Sexton, J. D. (2002). Social connect
edness, social appraisal, and perceived stress in college women
and men. Journal of Counseling & Development, 80, 355-361.
Lepore, S., Miles, H., & Levy, J. S. (1997). Relation of chronic
and episodic stressors to psychological distress, reactivity, and
health problems. International Journal of Behavioral Medicine,
4, 39-59. doi:10.1207/s 1 5327558ijbm0401_3
Lopez, F. G., Mauricio, A. M., Gormley, B., Simko, T., & Berger,
E. (2001). Adult attachment orientations and college student
distress: The mediating role of problem coping styles. Journal
of Counseling & Development, 79, 459-464.
MacKinnon, D. P Fairchild, A. J., & Fritz, M. S. (2007). Me
diation analysis. Annual Review of Psychology, 58, 593-614.
doi: 10. 1 146/annurev.psych.58.1 10405.085542
Nilsson, K. W, Leppert, J., Simonsson, B., & Starrin, B. (2010).
Sense of coherence and psychological well-being: Improvement
with age. Journal of Epidemiology and Community Health, 64,
347-352. doi:10.1 136/jech.2008.081 174
Noor, N. M. (2008). Work and womens well-being: Religion and
age as moderators. Journal of Religion and Health, 47, 476-490.
doi: 1 0. 1 007/s 1 0943-008-9 188-8
Olson, A., Garriott, P. O., Rigali-Oiler, M. E., & Chao, R. (2009,
August). Debt, class, and coping:A modelfor counseling psychol
ogy trainees. Poster presented at the Annual Convention of the
American Psychological Association, Toronto, Ontario, Canada.
Roberti, J. W., Harrington, L. N., & Storch, E. A. (2006). Further
psychometric support for the 10-item version of the Perceived
Stress Scale. Journal of College Counseling, 9, 135-147.
Roth, S., & Cohen, L. J. (1986). Approach, avoidance, and
coping with stress. American Psychologist, 41, 813-819.
doi: 1 0. 1 037/0003-066X.4 1 .7.8 1 3
Sarid, O., Anson, O., Yaari, A., & Margalith, M. (2004). Coping styles
and changes in humoural reaction during academic stress. Health
and Medicine, 9, 85-98.

348

Schwitzer, A. M. (2008). College student health, mental health, and

well-being. Journal of College Counseling, 11, 99-100.
Shapiro, S. L., Oman, D.,Thoresen, C. E., Pante.T. G., & Flinders, T.
(2008). Cultivating mindfulness: Effects on well-being. Journal
of Clinical Psychology, 64, 840-862. doi: 10. 1002/jclp.20491
Sideridis, G. D. (2008). The regulation of affect, anxiety, and stress
ful arousal from adopting mastery-avoidance goal orientations.
Stress and Health, 24, 55-69. doi: 1 0. 1 002/smi. 1 1 60
,
Siegel, K., Karus, D., Raveis, V. H., & Hagen, D. (1998). Psy
chological adjustment of women with HIV/AIDS: Racial and
ethnic comparisons. Journal of Community Psychology, 26,
439-455. doi:10.1002/(SICI)1520-6629( 199809)26:5<439::AIDJCOP4>3.0.CO;2-M
Skowron, E. A., Wester, S. R., & Azen, R. (2004). Differentiation of
self mediates college stress and adjustment. Journal of Counseling
<& Development, 82, 69-78.
Thorsteinsson, E. B., & Brown, R. F. (2008). Mediators and modera
tors of the stressor-fatigue relationship in nonclinical samples.
Journal of Psychosomatic Research, 66, 21-29.
Veit, C. T., & Ware, J. E. ( 1983). The structure of psychological distress
and well-being in general populations. Journal of Consulting and
Clinical Psychology, 51, 730-742. doi:10.1037/0022-006X.51.5.730
Vermeulen, M., & Mustard, C. (2000). Gender differences in
job strain, social support at work, and psychological distress.
Journal of Occupational Health Psychology, 5, 428-440.
doi:10.1037/1076-8998.5.4.428
Wang, C.-C. D. C., & Castaneda-Sound, C. (2008). The role of
generational status, self-esteem, academic self-efficacy, and
perceived stress in college students psychological well-being.
Journal of College Counseling, 11, 101-118.
Weinstein, N., Brown, K. W.,& Ryan, R. M. (2009). A multi-meth
od examination of the effects of mindfulness on stress attribu
tion, coping, and emotional well-being. Journal of Research
in Personality, 43, 374-385. doi: 10.101 6/j.jrp.2008.1 2.008

Journal of Counseling & Development

Summer 2011

Volume 89

Copyright of Journal of Counseling & Development is the property of American Counseling Association and its
content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for individual use.