Journal of Occupational Health Psychology © 2014 American Psychological Association

2014, Vol. 19, No. 2, 259 –267 1076-8998/14/$12.00 DOI: 10.1037/a0036267

Healthy Eating at Different Risk Levels for Job Stress:

Testing a Moderated Mediation

Daniel P. Fodor and Conny H. Antoni Amelie U. Wiedemann

Universität Trier Freie Universität Berlin

Silke Burkert
Charité – Universitätsmedizin Berlin
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Health behavior, like fruit and vegetable consumption (FVC), is affected by unfavorable job conditions.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

However, there is little research to date that combines job stress models and health-behavior change
models. This longitudinal study examined the contribution of risk factors associated with job stress to the
intention–planning–FVC relationship. In the context of the Health Action Process Approach, action
planning (when-where-how plans) and coping planning (plans to overcome anticipated barriers) have
been shown to be successful mediators in the translation of health-related intentions into action. Risk
factors for job stress are operationalized as the interaction of job demands and job resources in line with
the Job Demands–Resources (JD-R) model. Two hundred seventy-two employees (mean age 41.2 years,
73.9% female) from different jobs completed measures of intention at baseline (t1), action planning and
coping planning 2 weeks later (t2), and FVC another 2 weeks later (t3). Job demands and job resources
were assessed at t1 and t2. A moderated mediation analysis indicated that risk factors for job stress
moderate the translation of intention into action planning (B ⫽ ⫺0.23, p ⬍ .05) and coping planning
(B ⫽ ⫺0.14, p ⬍ .05). No moderation effect of the planning-FVC relationship by risk factors for job
stress was found. However, coping planning directly predicted FVC (B ⫽ 0.36, p ⬍ .001). Findings
suggest that employees intending to eat healthily use action planning and coping planning when job
demands exceed job resources. For increasing FVC, coping planning appears most beneficial.

Keywords: job demands-resources model, healthy diet, work strain, planning, health behavior change

In recent years, research on the relationship between conditions
of job stress and health behaviors, like fruit and vegetable con-
sumption (FVC), has increased. However, results are inconsistent
(e.g., Siegrist & Rödel, 2006). We bring forward two assumptions
that may add to the understanding of how risk factors for job stress
are related to FVC. First, the majority of studies on that topic
focused on health or risk behaviors simply as outcomes of job
stress. We think that risk factors for job stress need to be linked to
the volitional steps of the behavior change process according to the
Health Action Process Approach (HAPA) (Schwarzer, 2008).
Here, the realization of an intention to eat healthy can be supported
by action planning and coping planning as personal resources.
Whereas action planning is a goal-oriented strategy to achieve an
intended behavior (Gollwitzer, 1999), coping planning is a barrier-
focused strategy to deal with obstacles that may arise while acting
upon an intention (Sniehotta, 2009). Second, in almost all studies,
Daniel P. Fodor, Universität Trier, Trier, Germany; Conny H. Antoni,
Work and Organizational Psychology, Universität Trier; Amelie U. Wie-
demann, Division of Health Psychology, Freie Universität Berlin, Berlin,
Germany; Silke Burkert, Institute of Medical Psychology, Charité - Uni-
versitätsmedizin Berlin, Berlin, Germany.
Correspondence concerning this article should be addressed to Daniel P.
Fodor, Insitute of Medical Psychology, Charité – Universitätsmedizin
Berlin, Luisenstr. 57, 10117 Berlin, Germany. E-mail: daniel.fodor@
charite.de
the demand– control model (DCM; Karasek, 1979) was used as a
framework model for job strain. With respect to the complexity of
the HAPA, we think that the Job Demands–Resources (JD-R)
model (Bakker & Demerouti, 2007), which has not been linked to
health behavior so far, more appropriately specifies risk factors
associated with job demands and job resources. By combining the
JD-R model and the HAPA, the present study aims at investigating
how risk factors for job stress relate to FVC and its antecedents. In
particular, we will consider action planning and coping planning as
potential mediators of the relationship between intentions and
FVC. Additionally, the interaction of job demands and job re-
sources with these two planning types will be assessed as addi-
tional moderators of this mediation.
Predicting Healthy Eating
Consuming a sufficient amount of fruit and vegetables is a key
factor for a healthy diet and minimizes the risk of developing
cardiovascular disease and other health problems (World Health
Organization [WHO], 2003). The WHO recommends consuming
at least 400 g or approximately 5 servings of fruit and vegetables
each day. However, most people do not meet the recommended
levels (e.g., Blanck, Gillespie, Kimmons, Seymour, & Serdula,
2008). To explain why some people successfully adopt a healthy
diet and others fail to do so, the HAPA (Schwarzer, 2008) differ-
entiates between a motivational and a volitional phase. The moti-
vational phase results in the formation of an intention which can be

259
 260 FODOR, ANTONI, WIEDEMANN, AND BURKERT
defined as an individual’s motivation to show a specific behavior
to reach a desired goal. Meta-analyses have identified fundamental
correlations between intentions and different health behaviors,
including a healthy diet (e.g., Armitage & Conner, 2001). Never-
theless, intentions still appear insufficient to change behavior in
experimental studies (Webb & Sheeran, 2006). According to the
HAPA, an individual who has formed an intention steps into the
volitional phase, which can be further subdivided into a preactional
phase and an action phase. In the preactional phase, individuals,
who are intending to perform but have not (yet) performed a health
behavior, prepare for the implementation of their healthy actions.
In this phase, forming action plans and coping plans is acknowl-
edged as an effective self-regulatory strategy to overcome voli-
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
tional barriers to action. In creating action plans, individuals spec-
This document is copyrighted by the American Psychological Association or one of its allied publishers.
ify when, where, and what action they will carry out to achieve a
certain goal (Sniehotta, Scholz, & Schwarzer, 2005). By mentally
linking environmental cues to a behavioral response, goal-
fulfilling behavior becomes more likely (Gollwitzer, 1999). In the
work context, action planning is assumed to be particularly nec-
essary. Highly demanding jobs are characterized by time pressure,
goal conflicts, and instable intentions. Under such conditions,
individuals may favor time-effective and convenient fast-food
solutions over healthy meals rich in fruits and vegetables. To adapt
a healthier diet anyway, an individual may specify an action
plan—“If I am checking my e-mails in the morning (when) at my
workplace (where), then I will eat an apple (what).” However,
despite realistic and meaningful action planning, the implementa-
tion of health behavior often requires dealing effectively with
barriers for action. In the context of the present study, barriers may
arise from negative occupational conditions (e.g., time pressure)
and a lack of resources for counterregulation (e.g., time control).
Coping planning helps to deal with such barriers by mentally
linking a suitable coping response to an anticipated risk situation
(Sniehotta, 2009). For instance, an employee may anticipate over-
time hours and form a coping plan to still consume fruits or
vegetables—“If overtime hours prevent me from eating fruit salad
at lunch-time, then I will buy and consume fruit salad take-out on
my way back home.” Several studies demonstrate that action
planning and coping planning explain the intention-behavior rela-
tionship, and thus are effective personal resources (e.g., Scholz,
Schüz, Ziegelmann, Lippke, & Schwarzer, 2008; Schwarzer et al.,
2007; Sniehotta et al., 2005).
Nevertheless, forming good intentions, action plans, and coping
plans still do not guarantee the adoption and maintenance of a
healthy diet. Because the majority of employed adults in Germany
spend most of their waking hours at work (Federal Statistical
Office of Germany, 2006), the occupational situation, especially
resources and barriers at work, should play a crucial role in both
translating an intention into planning strategies and translating
these strategies into actual behavior.
Healthy Eating at Challenging Job Conditions
The JD-R model (Bakker & Demerouti, 2007) acknowledges that
every occupational setting has specific physical, psychological, social,
or organizational factors associated with job stress, which can be
either classified as “job demands” or “job resources.” Thus, job
demands and job resources may interact in predicting job stress,
depending on the specific occupational circumstances. An imbalance
of high job demands and low job resources increases the likelihood of
job stress. The JD-R model also proposes that high job resources can
buffer the negative impact of high job demands on the development
of job stress (Bakker & Demerouti, 2007). Evidence for the buffering
effects of several job resources (e.g., “autonomy,” “social support”)
on the impact of various job demands (e.g., “emotional demands,”
“work-home conflict”) on symptoms of burnout, for example, support
this buffering hypothesis (Bakker, Demerouti, & Euwema, 2005). The
second hypothesis of the JD-R model, the coping hypothesis, which
states that job demands moderate the motivating effect of job re-
sources (Bakker & Demerouti, 2007), was confirmed by the finding
that job resources mainly foster employees’ motivation when job
resources and job demands are high (e.g., Hakanen, Bakker, & De-
merouti, 2005). Previous studies on the JD-R model investigated the
interaction effects of job demands and job resources on outcomes
such as burnout, organizational commitment, or work engagement. To
our knowledge, there are only a few studies combining assumptions
from job stress models and health-behavior theory so far.
Payne, Jones, and Harris (2002) showed detrimental effects of job
conditions with high job demands and low job control on health
behavior change (i.e., exercise). However, they found no influence of
job conditions on FVC (Payne et al., 2005). In an experimental study
with a planning intervention they investigated direct effects of job
demands on exercise behavior and their interaction with psychosocial
variables (Payne, Jones, & Harris, 2010). Unexpectedly, the planning
intervention did not help to overcome work related barriers. However,
job demands moderated the relationship between intention and exer-
cise. Job resources, which might have buffered the job demands,
however, were not investigated.
Referring to the JD-R model, the present study aims at exam-
ining whether self-administered planning helps to enact intentions,
and takes unfavorable job conditions, that is, the imbalance of
diverse job demands and job resources, into account.
The Present Study
The purpose of this study is to investigate how challenging job
conditions relate to the health-behavior change process. In line
with the HAPA, we assume that intention is the prerequisite of any
action like FVC. Furthermore, we acknowledge action planning
and coping planning as powerful strategies to concretize the in-
tention, which is assumed to be especially helpful if intention
implementation is impeded. With the buffering hypothesis, the
JD-R model provides a rationale for such impeding conditions in
the work context. When job resources are insufficient to buffer
high job demands, unfavorable job conditions may arise, which
potentially lead to job stress and which also impede the achieve-
ment of health goals. Accordingly, we combine the buffering
hypothesis of the JD-R model and the HAPA in our moderated
mediation model (see Figure 1). In that, we test whether action
planning and coping planning are successfully used to implement
an intention for a proper FVC. “Used” refers to the transfer of an
intention into action planning or coping planning (a-path), and
“successfully used” refers to the transfer of these plans into actual
FVC (b-path). Additionally, we consider a moderating effect of high
and low risk factors for job stress on both paths (see Figure 1).
In the work context, planning strategies can be considered as
additional, job-independent resources. Their application might
equilibrate the ratio of job demands and job resources if healthy
 HEALTHY EATING AND JOB STRESS 261

Job resources t1 Job resources t2

Job demands t1 Job demands t2

Planning t2

a b
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Intention t1 FVC t3
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Figure 1. Moderated mediation model with job demands and job resources moderating the a-path and the
b-path according to moderated mediation Model 48 (Hayes, 2013). Planning t2 ⫽ action planning or coping
planning at t2, respectively; FVC t3 ⫽ fruits and vegetable consumption at t3.

eating should be implemented at work. Therefore, in our first
hypothesis, we predict that action planning and coping planning
are applied to support FVC-intentions especially under unfavor-
able job conditions, that is, high job demands and low job re-
sources. This hypothesis is tested by a three-way interaction be-
tween a) intention, b) job demands, and c) job resources predicting
action planning and coping planning, respectively, which refers to
the a-path.
However, even if employees have generated plans to support
their FVC-intention, the successful translation of these plans into
FVC may still be challenged. For example, in the presence of high
job demands the implementation of the planned behavior could be
impeded if job resources (e.g., time control) are low. Therefore, in
our second hypothesis, we assume that employees with high job
demands are more likely to translate their plans into FVC when
they experience high job resources (e.g., high time control) than
employees with an insufficient amount of job resources (e.g., low
time control). Again, this hypothesis is tested by a three-way
interaction between a) action planning and coping planning, re-
spectively, b) job demands, and c) job resources predicting FVC,
which refers to the b-path.
Both hypotheses constitute a moderated mediation model
with action planning or coping planning as mediators between
intention and FVC and job stress conditions as moderators,
depicted in Figure 1.
Method
Participants and Procedure
To cover a broad variety of job characteristics, participants were
recruited from three different sectors, namely the IT and marketing
sector (service providers: atypical working hours, high time pres-
sure), the research sector (researchers: variation in time pressure,
higher degree of time and decision latitudes), and the public sector
(administrative employees: fixed working hours, narrow time and
decision latitudes). E-mail addresses were acquired through elec-
tronic trade and research directories and business contacts; 4,177
employees were invited to participate in a questionnaire study by
emails that detailed study’s aims and procedures and contained an
Internet link to the first online questionnaire (t1). Individuals
providing their e-mail address at the end of the t1 questionnaire
received a second (t2) and third (t3) questionnaire in two-week
intervals via e-mail. The study was conducted in line with the
National Psychological Society’s ethical guidelines.
A total of n ⫽ 668 employees participated in the baseline
questionnaire, and 272 of them completed the longitudinal survey
until t3 (40.7% of baseline). This met the required sample size,
calculated in an a priori power analysis with an assumed power of
.80 and p value of .05 (n ⫽ 227), which was based on similar
studies (Payne et al., 2010; Sniehotta et al., 2005); 73.9% of the
participants were female, and the average age was 41.2 years
(SD ⫽ 12.5; range 20 to 87 years). The average body mass index
(BMI) of female participants was 23.3 (SD ⫽ 4.2) and of male
participants 25.6 (SD ⫽ 4.2), which are both slightly below the
average BMI of employed adults in Germany (female ⫽ 24.1 and
male ⫽ 26.1; Federal Statistical Office of Germany, 2011). Work-
ing hours per week ranged from 5 to 80 with an average of 38.2
hours per week (SD ⫽ 12.4). Most participants were employed in
the research sector (44.5%), followed by administration/public
services (18.4%), IT/marketing service providers (16.2%), and
other sectors (19.9%).
Drop-out analyses using analyses of variance (ANOVAs)
showed that those participants who completed the survey had
significantly lower working hours per week (p ⬍ .01), lower job
demands (p ⬍ .05), and a lower BMI (p ⫽ .05) than dropouts.
They also reported higher baseline levels of action planning (p ⬍
.001).
Measures
If not otherwise stated, items were answered on 5-point Likert
scales either ranging from 1 ⫽ very rarely to 5 ⫽ very often or
from 1 ⫽ completely disagree to 5 ⫽ completely agree and were
framed consider the participant’s previous week.
 262 FODOR, ANTONI, WIEDEMANN, AND BURKERT

Job demands and job resources. According to the JD-R
model (Bakker & Demerouti, 2007), an imbalance between job
demands and job resources can lead to job stress. Therefore, job
demands were assessed with a five-item scale “time pressure”
(e.g., How often did you work under time pressure?) from the
validated Instrument for Stress-oriented Job Analysis (Instrument
zur Stressbezogenen Tätigkeitsanalyse; ISTA; Semmer, Zapf, &
Dunckel, 1999) and three additional items (e.g., In the last week,
time pressure at work was so high that work quality was suffering
once in a while; Semmer, Kälin, Elfering, & Tschan, 2008). To
consider specific nutrition-related job demands, two self-
constructed items were added similarly to Semmer et al. (2008). A
sample item is In the last week, time pressure at work was so high
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
fruit or vegetable (to eat), with Cronbach’s alpha ⫽ .84 at t2.
Coping planning was measured by two items such as Last week I
made specific plans regarding how to consume 5 servings of fruit
and vegetable in the face of difficult situations, with Cronbach’s
alpha ⫽ .95 at t2. Means, standard errors, and correlations of all
variables are provided in Table 1.
Analytical Procedure
Hypotheses were tested by means of regression analyses. In
accordance with the HAPA, two regression analyses represent the
health-behavior change process from intention through planning
strategies to behavior. Additionally, according to the buffering

that I hardly had time to eat. Cronbach’s alpha was .92 at t1 and
assumption of the JD-R model, risk factors for job stress are
This document is copyrighted by the American Psychological Association or one of its allied publishers.

.93 at t2. Job resources were measured with two five-item scales
from the ISTA, “time-control” (e.g., How much were you able to
determine your pace of work on your own?) and “control at work”
(e.g., How much influence did you have on the type of work that
was assigned to you?). The response format of two items was a
6-point Likert scale (from 1 ⫽ not at all to 6 ⫽ more than one hour
and from 1 ⫽ less than five minutes a day to 6 ⫽ more than two
hours a day, respectively). These items were transformed into
5-point scales (see Semmer et al., 1999). The response format of
one item was a 5-point Likert scale (1 ⫽ not at all to 5 ⫽ all
results of work). Cronbach’s alpha was .87 at both t1 and t2.
Fruit and vegetable consumption (FVC). FVC was assessed
with an open response item (e.g., Wiedemann, Lippke, & Schwar-
zer, 2012): How many servings of fruit and vegetables did you eat
on a typical day of the last week? A serving was defined as one
handful; for example, one handful of salad. A statement was also
provided that rice and products made of potatoes should not be
regarded as fruit or vegetables.
Social– cognitive variables. Social– cognitive variables were
assessed in accordance with the 5-a-day recommendation for fruit
and vegetables consumption and had been used previously (e.g.,
Wiedemann et al., 2012). Response formats were 6-point Likert
scales (1 ⫽ completely disagree to 6 ⫽ completely agree). Behav-
ioral intention was measured at baseline by the item, Last week I
intended to eat at least 5 servings of fruit and vegetable each day.
Action planning was assessed by three items: Last week I made
specific plans regarding [when to eat, where to eat, what kind of]
considered as moderators (i.e., the interaction of job demands and
job resources). Hypotheses were tested separately for action plan-
ning and coping planning as a mediator, respectively. For this
purpose, the PROCESS macro for SPSS (Hayes, 2013) was used,
which allows testing moderation and mediation models as well as
indirect effects on different levels of continuous moderators.
Model 48 in this macro refers to a mediation model with both
mediation-paths being moderated (see Figure 1). Hypothesis 1
refers to a moderated a-path; an interaction of intention (t1) with
job demands and job resources (t1) predicting planning strategies
(t2). Hypothesis 2 refers to a moderated b-path; an interaction of
the planning strategies (t2) with job demands and job resources
(t2) predicting FVC (t3). Each regression analysis was controlled
for age, sex, and baseline FVC. Predictors that built the interaction
terms were mean centered. Given a significant three-way interac-
tion, the size of the conditional effect of intention on planning
strategies, and the conditional effect of planning strategies on
FVC, respectively, will be analyzed at combinations of high and
low values of job demands and job resources (i.e., at the 10th, 50th
and 90th percentiles). Conditional effects based on percentiles are
in line with simple slope analyses (Cohen, Cohen, West, & Aiken,
2003), but this procedure guarantees that all quantile values are
within the range of the observed data, even if the distribution is
skewed (Hayes, 2013). Moreover, it allows observing extreme
values of job stress condition (the 10th percentile stands for very
low; the 90th for very high).

Table 1
Means, Standard Deviations, and Correlations for All Study Variables (n ⫽ 272)

Variable M SD 1 2 3 4 5 6 7 8 9 10

1. JD t1 2.55 .90
2. JD t2 2.47 .87 .76ⴱⴱ
3. JR t1 3.88 .62 ⫺.37ⴱⴱ ⫺.28ⴱⴱ
4. JR t2 3.82 .63 ⫺.31ⴱⴱ ⫺.36ⴱⴱ .73ⴱⴱ
5. INT t1 2.07 1.38 .02 .07 ⫺.12 ⫺.11
6. AP t2 3.16 1.28 .11 .08 ⫺.09 ⫺.13ⴱ .12
7. CP t2 1.91 .99 .17ⴱ .13ⴱ ⫺.16ⴱ ⫺.11 .46ⴱⴱ .41ⴱⴱ
9. FVC t1 2.55 1.52 ⫺.11 ⫺.04 .07 ⫺.01 .46ⴱⴱ .07 .16ⴱ
8. FVC t3 2.79 1.43 .01 ⫺.02 .01 .01 .38ⴱⴱ .10 .35ⴱⴱ .51ⴱⴱ
10. Sexa 1.25 .43 ⫺.05 .01 .17ⴱⴱ .16ⴱⴱ ⫺.18ⴱⴱ ⫺.22ⴱⴱ ⫺.23ⴱⴱ ⫺.16ⴱⴱ ⫺.16ⴱⴱ
11. Age 41.20 12.45 .04 .05 .15ⴱ .14ⴱ .03 .01 .09 .13ⴱ .09 .13ⴱ
Note. JD ⫽ Job demands; JR ⫽ Job resources; INT ⫽ Intention; AP ⫽ Action planning; CP ⫽ Coping planning; FVC ⫽ Fruit and vegetable consumption.
a
1 ⫽ male, 2 ⫽ female.
ⴱ
p ⬍ .05. ⴱⴱ p ⬍ .01.
 HEALTHY EATING AND JOB STRESS 263

Taking hypothesis 1 and 2 together, they depict the volitional
phase in the HAPA from intention to FVC through the mediator
action planning or coping planning, respectively, and risk factors
for job stress as additional moderators in line with the JD-R-model.
Therefore, we finally test this moderated mediation model. If the
strength of the indirect effect of intention on FVC via action
planning or coping planning, respectively, changes as a result of
one of the two assumed moderation effects, this indicates moder-
ated mediation. If a moderated mediation is indicated, the condi-
tional indirect effect of the mediator on the intention-FVC asso-
ciation at combinations of different levels (at the mean and ⫾ one
standard deviation) of job demands and job resources at t1 and/or
t2 will be tested with bias-corrected bootstrap analysis. These
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
upper part of Table 2. Three-way interactions of intention, job
demands, and job resources were significantly associated both with
action planning and coping planning.
The a-path three-way interaction is exemplarily displayed for
the prediction of coping planning in Figure 2. Conditional effects
of intention on coping planning for all combinations of low,
average, and high values of job demands and job resources were
tested and support hypothesis 1. Under all combinations of job
demands and job resources intention was significantly associated
with coping planning, with the exception of the high job demands-
high job resources combination (B ⫽ 0.15, ns) and the low job
demands-low job resources combination (B ⫽ 0.18, ns). In line
with hypothesis 1, the highest effects were found for the high job

asymmetric bootstrap intervals are robust against irregularly demands–low job resources condition (B ⫽ 0.42, p ⬍ .001) and the
This document is copyrighted by the American Psychological Association or one of its allied publishers.

shaped sampling distributions (Hayes, 2013).
All analyses were run with the 272 completed surveys after item
and unit missings (t2) were imputed using the Expectation Maxi-
mization algorithm in SPSS 20 (Enders, 2001).
Results
low job demands– high job resources condition (B ⫽ 0.43, p ⬍
.001). Similar patterns were found for the three-way interaction
predicting action planning.
Moderation of the B-Path Between Planning Strategies
and FVC

Moderation of the A-Path Between Intention and
Planning Strategies
To test hypothesis 1, regression analyses both for action plan-
ning and coping planning were conducted and are depicted in the
With regard to the b-path, results did not support hypothesis 2.
Risk factors for job stress had no effect on the b-path, the three-
way interaction of job demands, job resources, and neither action
planning nor coping planning reached significance (see lower part
of Table 2).

Table 2
Moderated Mediation Model With A-Path and B-Path Moderation for Each Mediator Action
Planning and Coping Planning (PROCESS Model 48; Hayes, 2013)

Me: Action Me: Coping

planning t2 planning t2
Moderation B SE B SE

A-path moderation: planning t2 (Me) as outcome

Sex ⫺.56ⴱⴱ .19 ⫺.34ⴱⴱ .13
Age .00 .01 .01ⴱ .00
Past behavior t1 .00 .06 ⫺.06 .04
Intention t1 (a-path) .01 .07 .29ⴱⴱⴱ .05
Job demands t1 .07 .09 .10 .06
Job resources t1 ⫺.13 .14 ⫺.12 .10
Intention ⫻ job demands t1 ⫺.01 .07 ⫺.01 .05
Intention ⫻ job resources t1 .00 .10 .00 .07
Job demands ⫻ job resources t1 .21 .13 ⫺.13 .09
Intention ⫻ job demands ⫻ job resources t1
(moderated a-path) ⫺.23ⴱ .09 ⫺.14ⴱ .06
R2 .09ⴱⴱ .29ⴱⴱⴱ
B-path moderation: behavior t3 (DV) as outcome
Sex ⫺.21 .19 ⫺.07 .18
Age .00 .01 .00 .01
Past behavior t1 .40ⴱⴱⴱ .06 .42ⴱⴱⴱ .05
Intention t1 .18ⴱⴱ .06 .07 .07
Planning t2 (b-path) .03 .06 .36ⴱⴱⴱ .09
Job demands t2 ⫺.02 .09 ⫺.05 .09
Job resources t2 .08 .14 .09 .13
Planning ⫻ job demands t2 ⫺.06 .07 ⫺.09 .09
Planning ⫻ job resources t2 .10 .10 .07 .14
Job demands ⫻ job resources t2 .00 .13 .01 .12
Planning ⫻ job demands ⫻ job resources t2
(moderated b-path) .03 .08 ⫺.02 .11
R2 .30ⴱⴱⴱ .35ⴱⴱⴱ
Note. Me ⫽ mediator.
ⴱ
p ⬍ .05. ⴱⴱ p ⬍ .01. ⴱⴱⴱ
p ⬍ .001.
 264 FODOR, ANTONI, WIEDEMANN, AND BURKERT
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Figure 2. Moderation effect of low, average, and high job resources on the transfer of intention (t1) into coping
planning (t2) under high, average, and low job demands (t1). ⴱⴱⴱ p ⬍ .001.

Conditional Indirect Effect of Intentions on FVC resources (⫹1 SD). Only if employees experienced high job de-
Through Planning Strategies mands (⫹1 SD), which were buffered by high job resources (⫹1
SD), the indirect effect of intention on FVC through coping plan-
Finally, both paths were tested within a moderated mediation ning was not significant.
model. In our second regression analysis action planning was not
related to FVC, neither directly nor in interaction with risk factors
for job stress. Therefore, action planning is no mediator of the Discussion
intention–FVC association. In contrast, coping planning mediated
The present study tested the effects of risk factors for job stress
the intention–FVC association. Thus, the conditional indirect ef-
on health-behavior change. New insights into the role of stress-
fect of intention on FVC through coping planning at interactions of
related job conditions on health-behavior change were obtained by
different levels of job demands and job resources was tested by
including job stress antecedents as a moderator of the intention–
means of bootstrap analyses. As Table 3 shows, coping planning
planning–FVC association.
mediated the intention–FVC association at almost all levels of job
A significant moderation of the intention–planning association
demands and job resources with the strongest indirect effects under
by risk factors for job stress (a-path) was found, both for action
conditions of high job demands (⫹1 SD) and low job resources
planning and coping planning. This conditional effect sheds further
(⫺1 SD) as well as of low job demands (⫺1 SD) and high job
light on health-behavior change under different job stress condi-
tions. Supporting our first hypothesis, particularly those employees
with the intention to eat recommended amounts of fruits and
Table 3
vegetables and who are working under conditions associated with
Bootstrapped Indirect Effects of Intention (t1) on Change in
job stress, engaged in action planning and coping planning. In line
FVC (t1–t3) Through Coping Planning at Interactions of Job
with the buffer hypothesis of the JD-R model, employees also
Demands and Job Resources t1 at Specific Values (With Means
abstained from forming action plans or coping plans for intention
for Job Demands and Job Resources t2)
implementation when they perceived to have sufficient levels of
Job Job job resources to buffer their job demands. For instance, when
demands resources B SE LL CI UL CI employees perceived enough time control for FVC even in highly
demanding situations, they refrained from furnishing their inten-
⫹1 SD ⫺1 SD .13 .05 .06 .24
⫹1 SD Mean .10 .05 .04 .23 tions with action plans or coping plans. Moreover, especially
⫹1 SD ⫹1 SD .07 .07 ⫺.01 .26 coping plans helped to enact intentions for FVC at all combina-
Mean ⫺1 SD .11 .04 .05 .19 tions of job demands and job resources (b-path) in our study. The
Mean Mean .11 .04 .05 .20 mediation of coping planning itself is in line with previous re-
Mean ⫹1 SD .11 .05 .04 .23
⫺1 SD ⫺1 SD .08 .04 .02 .19
search (e.g., Scholz et al., 2008). Results suggest that coping
⫺1 SD Mean .11 .04 .05 .20 planning is a personal work-independent resource that compen-
⫺1 SD ⫹1 SD .14 .05 .06 .26 sates missing work-related resources in highly demanding jobs.
Note. n ⫽ 5000 bootstrapping resamples; LL CI ⫽ Lower level and UL
Findings also tentatively support the coping hypothesis of the
CI ⫽ upper level of bias corrected bootstrapped confidence intervals for JD-R model (Bakker & Demerouti, 2007): Employees make am-
␣ ⬍ .05. plified use of their self-regulatory resources when experiencing
 HEALTHY EATING AND JOB STRESS
risk of job strain to prevent a loss of resources. Under unfavorable
job conditions, they may apply coping planning as a resource to
follow a healthy diet which in turn is a key condition for successful
stress management at work.
However, contrary to our second hypothesis, risk factors for job
stress did not moderate the coping planning-FVC relationship
(b-path), that is, they did not impede the beneficial effects of
coping plans for FVC: Employees’ coping plans appear to be
helpful for the adoption or maintenance of the intended FVC,
irrespective of current job stress conditions. This perfectly sup-
ports the purpose of coping planning: Anticipated barriers (e.g.,
unforeseen working hours over lunch time) can be overcome by
the use of ex-situ planning strategies (e.g., consumption of pre-
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
produced fruit salad). Furthermore, the lack of direct effects of risk
This document is copyrighted by the American Psychological Association or one of its allied publishers.
factors for job stress on FVC indicates that job demands are not per
se a threat to FVC at work.
Our findings confirm a moderated mediation model for coping
planning, combining HAPA and JD-R model assumptions, in that
the buffering assumption of the JD-R model adds to the under-
standing of the volitional phase of the HAPA. Nevertheless, in
contrast to previous studies (e.g., Schwarzer et al., 2007; Wiede-
mann, Schüz, Sniehotta, Scholz, & Schwarzer, 2009), action plan-
ning did not mediate the relationship between intentions and
behavior. Neither a mean association between intentions and ac-
tion planning (a-path) nor an association between action planning
and FVC (b-path) were identified. The nonsignificant mean a-path
may be attributable to the conditionality of the intention–action
planning relationship under certain conditions of job stress (three-
way interaction). Intentions further had a significant direct effect
on FVC. Also, the nonsignificant association between action plan-
ning and FVC is explicable in context. Action plans may not have
been flexible enough to prevail under challenging job conditions.
Previous research points to further factors like self-efficacy beliefs,
which may also play a crucial role in this context (Lippke, Wie-
demann, Ziegelmann, Reuter, & Schwarzer, 2009).
Limitations and Future Directions
Limitations of the present study and subsequent implications for
future research should be mentioned. This longitudinal study cov-
ers only a short period of health-behavior change. Two-week
intervals allow to document changes in frequent health behavior
like FVC. Nevertheless, the causality of the association between
risk factors for job stress and health-behavior change remains
unclear. Therefore, future research should investigate long-term
influences of risk factors for job stress on health-behavior change
via randomized controlled trials, as well as daily fluctuations in the
association via diary studies. In contrast to online-surveys, with an
excessive reduction of the baseline number of participants during
the survey period attributable to the nonbinding and anonymous
nature of this type of study (Galesic, 2006), surveys with face-to-
face contacts will likely lead to higher commitment.
Dropout analysis points to a selective sample. Active partici-
pants reported more action planning, a lower BMI, and less job
demands as well as less working hours than those who dropped
out. These factors limit generalizability of our findings. Besides
the effectiveness of coping planning, this may partly explain why
risk factors for job stress were not associated with the planning-
FVC-association. As questionnaire measures were used in this
265
study, the dropout rate also points to the possibility of response
biases.
Moreover, a single-item measure was used for FVC instead of
objective measures of actual fruit consumption because of practi-
cal reasons in this online-study. However, several aspects point to
the validity of this self-reported single-item measure. First, recent
studies have shown high correlations of very similar single-items
and fruit and vegetable measures (Chapman, Armitage, & Norman,
2009). Second, a similar single-item measure of FVC has been
validated against dietary biomarkers (Steptoe et al., 2003). Third,
because of anonymity, these self-reports are less prone to be biased
(Armitage, 2007). Nevertheless, distortions in measuring FVC
cannot be ruled out, which may possibly explain why action
planning did not mediate the intention–FVC relationship. Future
studies on this topic should consider more reliable measures of
FVC as provided in the Compendium of Surveys for F&V con-
sumptions (California Department of Public Health, 2012).
Generalizability was also limited with respect to job types.
Future research on this topic should therefore include samples
from other branches (e.g., handcraft or social services). Within the
limits of a quantitative online study, our operationalization of risk
factors for job stress does not sufficiently comply with the theo-
retical approach of the JD-R model. According to Bakker and
Demerouti (2007) each job type has its specific job demands and
job resources. Our sample, which included a broad range of
different job types, “time pressure” as job demand and “time-
control” and “control of work” as job resource, respectively, may
not have represented all characteristics for each job type under
investigation. Consequently, risk factors for job stress may have
been underestimated in the present sample. Future research would
benefit from investigating job-specific demands and resources.
Complementarily, expert ratings (e.g., see Demerouti, Bakker,
Nachreiner, & Schaufeli, 2001) or external assessments (e.g., see
Bakker, Demerouti, & Verbeke, 2004; Salanova, Agut, & Peiró,
2005) could be used. Subsequent studies on the role of risk factors
for job stress on the volitional processes of health-behavior change
might also consider health behaviors other than healthy eating.
Practical Implications and Conclusion
The present findings suggest that considering risk factors for job
stress further qualifies under which conditions intentions lead to
behavior change through planning strategies. Taking work-related
conditions into account when investigating health behavior is
crucial given the large amount of time individuals spend working.
Further research should investigate at which levels of job stress
action planning and coping planning are beneficial or whether they
can be complemented by other strategies. To preserve health and
obviate sickness, practical implications of our research primarily
refer to coping planning. Employees working under unfavorable
job conditions should anticipate individual obstacles that hamper
their FVC at work and equip themselves with detailed plans how
to overcome these obstacles. Furthermore, health promotion pro-
grams in the work context (e.g., stress management) could be
supplemented with coping planning strategies for successful be-
havior change. From an organizational point of view, managers
should provide their employees with appropriate decision latitudes
that enable them dealing with job demands but also to decide when
it is time to interrupt the work process, for example, to act upon the
 266 FODOR, ANTONI, WIEDEMANN, AND BURKERT
intention to consume an apple in that very moment. Combining job
stress models such as the JD-R model and health-behavior change
models such as the HAPA may be a fruitful avenue for future
research.
References
Armitage, C. J. (2007). Effects of an implementation intention-based
intervention on fruit consumption. Psychology & Health, 22, 917–928.
doi:10.1080/14768320601070662
Armitage, C. J., & Conner, M. (2001). Efficacy of the theory of planned
behaviour: A meta-analytic review. British Journal of Social Psychol-
ogy, 40, 471– 499. doi:10.1348/014466601164939
Bakker, A. B., & Demerouti, E. (2007). The job demands-resources model:
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
State of the art. Journal of Managerial Psychology, 22, 309 –328.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
doi:10.1108/02683940710733115
Bakker, A. B., Demerouti, E., & Euwema, M. C. (2005). Job resources
buffer the impact of job demands on burnout. Journal of Occupational
Health Psychology, 10, 170 –180. doi:10.1037/1076-8998.10.2.170
Bakker, A. B., Demerouti, E., & Verbeke, W. (2004). Using the job
demands-resources model to predict burnout and performance. Human
Resource Management, 43, 83–104. doi:10.1002/hrm.20004
Blanck, H. M., Gillespie, C., Kimmons, J. E., Seymour, J. D., & Serdula,
M. K. (2008). Trends in fruit and vegetable consumption among U.S.
men and women, 1994 –2005. Preventing Chronic Diseases, 5, A35.
California Department of Public Health. (2012, October 1). Compendium
of surveys for nutrition education and obesity prevention. Retrieved
from http://www.cdph.ca.gov/programs/cpns/Documents/Compendium%
20of%20Surveys.pdf
Chapman, J., Armitage, C. J., & Norman, P. (2009). Comparing imple-
mentation intention interventions in relation to young adults’ intake of
fruit and vegetables. Psychology & Health, 24, 317–332. doi:10.1080/
08870440701864538
Cohen, J., Cohen, P., West, S. G., & Aiken, L. S. (2003). Applied multiple
regression/correlation analyses for the behavioural sciences (3rd ed.).
Mahwah, NJ: Erlbaum.
Demerouti, E., Bakker, A. B., Nachreiner, F., & Schaufeli, W. B. (2001).
The job demands-resources model of burnout. Journal of Applied Psy-
chology, 86, 499 –512. doi:10.1037/0021-9010.86.3.499
Enders, C. K. (2001). A primer on maximum likelihood algorithms avail-
able for use with missing data. Structural Equation Modeling: A Mul-
tidisciplinary Journal, 8, 128 –141. doi:10.1207/S15328007SEM0801_7
Federal Statistical Office of Germany. (2006, December 7). Zeitbudgets –
Tabellenband I–2001/2002 [time budgets–list of tables I–2001/2002].
Wiesbaden, Germany: Federal Statistical Office of Germany. Retrieved
from https://www.destatis.de/DE/Publikationen/Thematisch/
EinkommenKonsumLebensbedingungen/Zeitbudgeterhebung/
ZeitbudgetsTabellenband1.html
Federal Statistical Office of Germany. (2011, January 24). Mikrozensus -
Körperma␤e der Bevölkerung 2009 [micro census – body measurements
of the population 2009]. Wiesbaden, Germany: Federal Statistical Office
of Germany. Retrieved from https://www.destatis.de/DE/Publikationen/
Thematisch/Gesundheit/Gesundheitszustand/Koerpermasse5239003099004
.pdf?__blob⫽publicationFile
Galesic, M. (2006). Dropouts on the web: Effects of interest and burden
experienced during an online survey. Journal of Official Statistics, 22,
313–328.
Gollwitzer, P. M. (1999). Implementation intentions: Strong effects of
simple plans. American Psychologist, 54, 493–503. doi:10.1037/0003-
066X.54.7.493
Hakanen, J. J., Bakker, A. B., & Demerouti, E. (2005). How dentists cope
with their job demands and stay engaged: The moderating role of job
resources. European Journal of Oral Sciences, 113, 479 – 487. doi:
10.1111/j.1600-0722.2005.00250.x
Hayes, A. F. (2013). Introduction to mediation, moderation, and condi-
tional process analysis: A regression-based approach. New York, NY:
Guilford Press.
Karasek, R. A. (1979). Job demands, job decision latitude, and mental
strain: Implications for job redesign. Administrative Science Quarterly,
24, 285–308. doi:10.2307/2392498
Lippke, S., Wiedemann, A. U., Ziegelmann, J. P., Reuter, T., & Schwarzer,
R. (2009). Self-efficacy moderates the mediation of intentions into
behavior via plans. American Journal of Health Behavior, 33, 521–529.
doi:10.5993/AJHB.33.5.5
Payne, N., Jones, F., & Harris, P. (2002). The impact of working life on
health behavior: The effect of job strain on the cognitive predictors of
exercise. Journal of Occupational Health Psychology, 7, 342–353. doi:
10.1037/1076-8998.7.4.342
Payne, N., Jones, F., & Harris, P. R. (2005). The impact of job strain on the
predictive validity of the theory of planned behavior: An investigation of
exercise and healthy eating. British Journal of Health Psychology, 10,
115–131. doi:10.1348/135910704X14636
Payne, N., Jones, F., & Harris, P. R. (2010). A daily diary investigation of
the impact of work stress on exercise intention realisation: Can planning
overcome the disruptive influence of work? Psychology & Health, 25,
111–129. doi:10.1080/08870440903337622
Salanova, M., Agut, S., & Peiró, J. M. (2005). Linking organizational
resources and work engagement to employee performance and customer
loyalty: The mediation of service climate. Journal of Applied Psychol-
ogy, 90, 1217–1227. doi:10.1037/0021-9010.90.6.1217
Scholz, U., Schüz, B., Ziegelmann, J. P., Lippke, S., & Schwarzer, R.
(2008). Beyond behavioral intentions: Planning mediates between inten-
tions and physical activity. British Journal of Health Psychology, 13,
479 – 494. doi:10.1348/135910707X216062
Schwarzer, R. (2008). Modeling health behavior change: How to predict
and modify the adoption and maintenance of health behaviors. Applied
Psychology, 57, 1–29. doi:10.1111/j.1464-0597.2007.00325.x
Schwarzer, R., Schüz, B., Ziegelmann, J. P., Lippke, S., Luszczynska, A.,
& Scholz, U. (2007). Adoption and maintenance of four health behav-
iors: Theory-guided longitudinal studies on dental flossing, seat belt use,
dietary behavior, and physical activity. Annals of Behavioral Medicine,
33, 156 –166. doi:10.1007/BF02879897
Semmer, N. K., Kälin, W., Elfering, A., & Tschan, F. (2008, November).
When time pressure really hurts: The case of performance impairment.
Paper presented at the 6th Conference of the European Academy of
Occupational Health Psychology, Valencia, Spain.
Semmer, N. K., Zapf, D., & Dunckel, H. (1999). Instrument zur streßb-
ezogenen Tätigkeitsanalyse [Instrument for stress-related job analysis].
In H. Dunckel (Ed.), Handbuch psychologischer Arbeitsanalysever-
fahren [Analytical Instrumentation Handbook] (pp. 179 –204). Zürich,
Switzerland: VdF Hochschulverlag.
Siegrist, J., & Rödel, A. (2006). Work stress and health risk behavior.
Scandinavian Journal of Work, Environment & Health, 32, 473– 481.
doi:10.5271/sjweh.1052
Sniehotta, F. F. (2009). Towards a theory of intentional behaviour change:
Plans, planning, and self-regulation. British Journal of Health Psychol-
ogy, 14, 261–273. doi:10.1348/135910708X389042
Sniehotta, F. F., Scholz, U., & Schwarzer, R. (2005). Bridging the
intention-behavior gap: Planning, self-efficacy and action control in the
adoption and maintenance of physical exercise. Psychology & Health,
20, 143–160. doi:10.1080/08870440512331317670
Steptoe, A., Perkins-Porras, L., McKay, C., Rink, E., Hilton, S., & Cap-
puccio, F. P. (2003). Psychological factors associated with fruit and
vegetable intake and with biomarkers in adults from a low-income
neighborhood. Health Psychology, 22, 148 –155. doi:10.1037/0278-
6133.22.2.148
Webb, T. L., & Sheeran, P. (2006). Does changing behavioral intentions
engender behavior change? A meta-analysis of the experimental evi-
 HEALTHY EATING AND JOB STRESS 267

dence. Psychological Bulletin, 132, 249 – 68. doi:10.1037/0033-2909 World Health Organization. (2003). Fruit, vegetables and NCD disease
.132.2.249 prevention. Geneva, Switzerland: WHO. Retrieved from http://
Wiedemann, A. U., Lippke, S., & Schwarzer, R. (2012). Multiple plans and whqlibdoc.who.int/trs/WHO_trs_916.pdf
memory performance: Results of a randomized controlled trial targeting
fruit and vegetable intake. Journal of Behavioral Medicine, 35, 387–392.
doi:10.1007/s10865-011-9364-2
Wiedemann, A. U., Schüz, B., Sniehotta, F., Scholz, U., & Schwarzer, R.
(2009). Disentangling the relation between intentions, planning, and Received June 25, 2013
behavior: A moderated mediation analysis. Psychology & Health, 24, Revision received January 21, 2014
67–79. doi:10.1080/08870440801958214 Accepted February 12, 2014 䡲
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Showcase your work in APA’s newest database.

Make your tests available to other researchers and students; get wider recognition for your work.
“PsycTESTS is going to be an outstanding resource for psychology,” said Ronald F. Levant,
PhD. “I was among the first to provide some of my tests and was happy to do so. They will be
available for others to use—and will relieve me of the administrative tasks of providing them to
individuals.”
Visit http://www.apa.org/pubs/databases/psyctests/call-for-tests.aspx
to learn more about PsycTESTS and how you can participate.
Questions? Call 1-800-374-2722 or write to tests@apa.org.
Not since PsycARTICLES has a database been so eagerly anticipated!