See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/281674105

The role of hindrance stressors in the job demand–control–support model of

occupational stress: A proposed theory revision

Article in Journal of Organizational Behavior · September 2015

DOI: 10.1002/job.2049

CITATIONS READS

137 5,233

3 authors, including:

Kevin M. Dawson Kimberly O'Brien

Central Michigan University Central Michigan University
2 PUBLICATIONS 153 CITATIONS 28 PUBLICATIONS 1,070 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Kevin M. Dawson on 22 November 2017.

The user has requested enhancement of the downloaded file.

 Journal Code Article ID Dispatch: 06.08.15 CE: Palang, Olivia
J O B 2 0 4 9 No. of Pages: 19 ME:

1 Journal of Organizational Behavior, J. Organiz. Behav. (2015)

2 Published online in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/job.2049

Research Article
3
4
5 The role of hindrance stressors in the job demand–
6
7 control–support model of occupational stress: A
8
9
proposed theory revision
10
11
KEVIN M. DAWSON*, KIMBERLY E. O’BRIEN AND TERRY A. BEEHR
12
Central Michigan University, Mt. Pleasant, Michigan, U.S.A.
13
14
15
16
Summary Previous research on the job demand–control–support (JDCS) model of occupational stress has generally
17
been inconsistent at best regarding a key issue: the interaction of demands, control, and support in predicting
18 employee health and well-being. However, the model continues to be tested in a variety of studies and aca-
19 demic journals owing to its intuitive appeal. By incorporating conservation of resources theory with knowl-
20 edge from the challenge–hindrance stressor framework, we proposed that hindrance stressors, not the
21 challenge stressors commonly assessed when testing JDCS theory, will provide validation for the model. A
22 two-wave panel study of 228 employees in a variety of occupations provided support for three-way interac-
23 tions between hindrance demands, control, and support predicting job-related anxiety and physical symptoms.
24 Three-way interactions using a challenge demand (forms of workload) were not significant, consistent with
25 our propositions. In summary, this study supports the buffering effect of control and support on the relation-
26 ship between job demands and strain only when job demands reflect hindrance stressors, thereby proposing to
alter the JDCS model by specifying that it applies primarily to hindrance stressors in a job hindrance–control–
27
support model. Copyright © 2015 John Wiley & Sons, Ltd.
28
29 Keywords: stressor; strain; challenge; hindrance; JDCS
30
31 Job stressors have considerable mental, physical, behavioral, and performance consequences (Matteson &
32 Ivancevich, 1987) that cost American businesses over $300bn a year (Cynkar, 2007). Thus, investigating the com-
33 plexities of stressor–strain relationships may prove invaluable to both the organization and its employees. Several
34
occupational stress models exist, but the underlying principle is that during interactions with certain parts of the
35
work environment (job stressors), job strain (employee poor health or well-being) develops, and this relationship
36
37 is influenced by perceptions of the environment (Spector, 1998).
38 Work situations that would normally seem stressful are not always accompanied by dire consequences, however,
39 leading to a search for moderators marking boundary conditions for the negative effects of job stressors. The job
40 demand–control–support model (JDCS; Karasek & Theorell, 1990) has been a key anchoring point for research
41 on the impact of work characteristics on employee health and well-being for the past three decades (van Veldhoven,
42 Taris, de Jonge, & Broersen, 2005). Its central tenet is the buffer hypothesis, which posits that job control and social
43 support interact with job stressors to reduce levels of employee strain. Despite the popularity and prevalence of the
44 JDCS model, empirical evidence supporting the seminal buffer hypothesis of this model has been marginal at best.
45 Meta-analyses consistently fail to show adequate support for the buffer hypothesis of the JDCS (e.g., Häusser,
46
Mojzisch, Niesel, & Schulz-Hardt, 2010; van der Doef & Maes, 1999), even when controlling for methodological
47
rigor (de Lange, Taris, Kompier, Houtman, & Bongers, 2003). Although empirical evidence indicates dubious va-
48
49 lidity for the JDCS model, the buffer hypothesis continues to pervade the literature and has led some to coin the
50 JDCS a zombie theory, a theory that continues to haunt stress research despite having died from a lack of empirical
51
52
53 *Correspondence to: Kevin M. Dawson, Central Michigan University, Sloan 100, Mt. Pleasant, Michigan 48858, U.S.A. E-mail:
54 Dawso3km@cmich.edu
55
56
57 Received 01 July 2014
58 Copyright © 2015 John Wiley & Sons, Ltd. Revised 13 July 2015, Accepted 23 July 2015
59
60
61
 2 K. M. DAWSON ET AL.
1
2
3 support (Taris, 2006). Despite a lack of empirical evidence, literature continues to advocate the use of the JDCS
4 model, proclaiming validity and clamoring for further investigation of the model (e.g., Häusser et al., 2010).
5 The conceptualization of job demands in JDCS research provides an avenue for investigation. Empirical tests of
6 the buffer hypothesis commonly measure work load and/or time pressure as indicators of job demands (Häusser
7 et al., 2010), but stress researchers have since made distinctions between types of stressors: those appraised as po-
8 tentially promoting growth and achievement (i.e., challenge stressors) and those appraised as thwarting goal attain-
9 ment (i.e., hindrance stressors; Cavanaugh, Boswell, Roehling, & Boudreau, 2000). We contend that the failure to
10 distinguish types of demands has led to inconsistent JDCS results. Incorporating knowledge from conservation of
11 resources (COR) theory (Hobfoll, 1989), we propose that job control and social support should buffer the strain as-
12
sociated with hindrance demands, but not employee strain associated with challenge demands.
13
We make three contributions to the literature. First, although the JDCS model is perhaps the most prominent
14
15 framework in the stressor–strain literature, researchers have been unable to produce sufficient empirical support
16 for the model’s seminal buffer hypothesis. By differentiating job demands based on the challenge–hindrance frame-
17 work, and integrating the competing tenants of the COR theory (Hobfoll, 1989), we attempt to clarify the applica-
18 bility of the JDCS model. Second, by demonstrating that challenge and hindrance demands lead to differential
19 effects in the JDCS model, we explain the pattern of previous null effects that have plagued JDCS research. Instead
20 of previous suggestions faulting methodological issues, we provide a theoretical rationale as to why strain is not
21 buffered by control and support in the face of challenge demands. Third, replacement model proposals for the JDCS
22 tend to have large numbers of narrower variables tailored to specific situations, but the current model retains only the
23 few categories of variables of the JDCS that are applicable to a broad variety of situations, thereby fulfilling the need
24 for parsimonious job stress models (van Veldhoven et al., 2005).
25
26
27
28 The Job Demand–Control–Support Model
29
30
31 The JDCS model has three components: job demands, job control, and social support (Karasek & Theorell, 1990).
32 Job demands were originally defined as “psychological stressors involved in accomplishing the workload” (Karasek
33 & Theorell, 1990, p. 291). Job control (originally decision latitude) is the extent to which an employee has authority
34 to make decisions and utilize skills concerning the job, while social support is characterized by helpful relations with
35 supervisors and coworkers. The buffer hypothesis proposes a three-way interaction effect between job demands,
36 control, and support on employee strain, in which high levels of control and support reduce the strength of the pos-
37
itive relationship between demands and strain. It is assumed that the combination of control and support facilitates
38
39 coping which, in turn, enhances employee health and well-being (Daniels & Harris, 2005). This assumption may be
40 premature, as a meta-analysis reviewing 20 years of JDCS research reported support rates of less than 15 percent,
41 concluding that “the body of evidence paints a gloomy picture of the JDCS interaction” (Häusser et al., 2010, p.
42 30). We provide and test an explanation for these “gloomy” results and propose a revision to the JDCS theory, in
43 which it applies to one category of stressors, hindrance stressors.
44 Several researchers have offered explanations for the lack of JDCS support, often focusing on methodological is-
45 sues such as insufficient sample size (e.g., de Lange et al., 2003), use of overly broad demand–control–support mea-
46 sures (e.g., Beehr, Glaser, Canali, & Wallwey, 2001), and a lack of longitudinal studies (e.g., van der Doef & Maes,
47 1999). However, even studies with larger sample sizes, more specific demand–control–support constructs and lon-
48
gitudinal designs report null effects (review by de Lange et al., 2003). A possible explanation for the lack of support
49
is the common conceptualization and operationalization of job demands (van der Doef & Maes, 1999; Brough &
50
51 Biggs, 2013) as workload and time pressure. Researchers have since discovered the value in distinguishing between
52 two broader categories of stressors: challenge and hindrance (e.g., LePine, LePine, & Jackson, 2004). Through this
53 distinction, we expect to find meaningful buffer effects while providing an explanation for the abundance of past null
54 findings.
55
56
Copyright © 2015 John Wiley & Sons, Ltd. J. Organiz. Behav. (2015)
57
58 DOI: 10.1002/job
59
60
61
 HINDRANCE STRESSORS IN THE JDCS MODEL 3
1
2
3 Challenge and Hindrance Stressors: Resource Conservation Versus Resource
4 Acquisition
5
6
7 The joint buffer effect of job control and social support may depend on the type of stressor or demand, with respect
8 to the way they are typically appraised and managed by employees (i.e., as challenges or hindrances). Stressors that
9 are appraised as challenging (and potentially beneficial) are differentiated from stressors appraised as hindrances
10 (and thus threatening and potentially harmful), a distinction, thus far, overlooked in JDCS research. Challenge
11 stressors include demands such as workload, time pressure, and job scope, which tend to be perceived as obstacles
12
that can be overcome and help facilitate the opportunity for goal achievement and personal growth. Hindrance
13
stressors tend to be perceived as unmanageable factors that unnecessarily thwart personal goals and include demands
14
15 such as organizational constraints, interpersonal conflict, role ambiguity, and red tape (Lepine, Podsakoff, & Lepine,
16 2005). Empirical research has since supported this distinction, revealing differential effects on job attitudes such as
17 work self-efficacy (Webster, Beehr, & Christiansen, 2010), job satisfaction (Cavanaugh et al., 2000), and organiza-
18 tional commitment (Podsakoff, Lepine, & Lepine, 2007).
19 Despite the obverse relationships between challenge and hindrance stressors on a variety of outcomes, both types
20 of stressors result in the depletion of energy resources (i.e., time and effort) and thus have been found to relate pos-
21 itively to strain (Lepine et al., 2004; Podsakoff et al., 2007). In spite of the positive relationship of both challenge
22 and hindrance stressors with strains, individuals are likely to believe that successfully meeting challenge demands
23 will result in valuable outcomes. In fact, research suggests that challenge stressors may boost employee performance
24 through increased motivation and engagement (e.g., Lepine et al., 2004; Lepine et al., 2005). Thus, there is a positive
25
motivational response to challenge demands that does not occur for hindrance demands. For hindrance demands, in-
26
27
dividuals typically believe that no amount of effort can successfully deal with hindrance stressors and have been
28 found to reduce employee motivation and performance (e.g., Lepine et al., 2004; Lepine et al., 2005).
29 In addition to differential motivational responses to challenge and hindrance demands, we propose that job re-
30 sources, including job control and social support in the COR theory, may play different roles in buffering the pos-
31 itive relationship between challenge versus hindrance demands and strain. According to the COR theory (Hobfoll,
32 1989), individuals strive to protect, maintain, and accumulate such resources because they are valued by the individ-
33 ual (Hobfoll, 1989, p. 516). The COR theory maintains that individuals experience strain when there is either actual
34 or threatened resource loss, or an insufficient resource gain following resource investment.
35 In order to protect or acquire resources, the model posits two competing tenets, resource conservation and resource
36 acquisition, to help explain how and why individuals manage resources to cope with stressors. The resource-
37
conservation tenet contends that “resource loss is disproportionately more salient than resource gain” (Hobfoll, 1998,
38
39 p. 62), and individuals are unlikely to invest their current resources into situations where gains are improbable. For ex-
40 ample, Halbesleben and Bowler (2007) claimed that when employees feel they cannot accomplish a demand, they seek
41 to conserve resources by reducing the energy toward behaviors that would consume their resources (e.g., reducing effort
42 toward task performance). In contrast, the resource-acquisition tenet argues that “people must invest resources in order
43 to protect against resource loss, recover from losses, and gain resources” (Hobfoll, 1998, p. 63). Thus, employees are
44 expected to invest personal resources (e.g., time, effort, and energy) into meeting demands that offer the potential for
45 resource gains. As discussed later, we argue that job demands (i.e., challenge and hindrance demands) and resources
46 (i.e., control and support) uniquely combine to determine resource allocation strategies and subsequent strain responses.
47
48
49
50
Hindrance Demands, Control, and Support Three-Way Interaction
51
52
53 On the basis of the challenge–hindrance framework, and consistent with the COR theory, we propose that hindrance
54 demands prompt employees to protect themselves from strain by conserving their resources as they believe attempts
55
56
Copyright © 2015 John Wiley & Sons, Ltd. J. Organiz. Behav. (2015)
57
58 DOI: 10.1002/job
59
60
61
 4 K. M. DAWSON ET AL.
1
2
3 to meet these demands will be inadequate and therefore unlikely to result in further resource gains (Lepine et al.,
4 2005). Such resource allocation strategies include escape/avoidance, emphasizing the positive, and resignation
5 (e.g., believing that the situation will take care of itself) responses (Cordes & Dougherty, 1993; Havlovic & Keenan,
6 1991). These efforts allow an individual to protect resources by preserving energetic resources (i.e., time and effort)
7 and reducing the threat of resource loss.
8 Job control and social support are valuable resources that are expected to enhance conservation motives and fur-
9 ther protect resources from threatening situations. Accordingly, the minimax principle states that job control reduces
10 the maximum threat posed by stressors (Thompson, 1981). For example, if an employee experiences conflicting role
11 demands, having control over when and where work is performed will enable the individual to choose what demand
12
to undertake and how it is performed. The perception of conflict would thus be seen as less threatening, and the sub-
13
sequent emotional response less severe (Spector, 2002). Indeed, job control has been found to bolster resource-
14
15 conservation attempts, helping to attenuate employee strain (Daniels & Harris, 2005; Daniels, 1999). In the face
16 of threat or uncertainty, employees are also likely to seek social support from others to validate the appropriateness
17 of their emotions and reduce their levels of fear and anxiety (Schachter, 1959). Hobfoll (1989) himself designated
18 social support as a salient resource in the COR theory that reassures a heightened sense of security and sense of self,
19 which facilitates resource preservation. Thus, we propose that the JDCS buffer hypothesis will be valid for hindrance
20 stressors.
21 Consistent with previous studies (e.g., Lepine et al., 2005), we operationalized hindrance stressors as interper-
22 sonal conflict, role conflict, and organizational politics.
23
24 Hypothesis 1: The three-way interaction between interpersonal conflict, control, and support will be related to (i)
25
emotional exhaustion, (ii) job-related anxiety, and (iii) physical symptoms, such that higher control and support
26
27
will result in weaker relationships between interpersonal conflict and strains.
28
29 Hypothesis 2: The three-way interaction between role conflict, control, and support will be related to (i) emotional
30 exhaustion, (ii) job-related anxiety, and (iii) physical symptoms, such that higher control and support will result in
31 weaker relationships between role conflict and strains.
32
33 Hypothesis 3: The three-way interaction between organizational politics, control, and support will be related to (i)
34 emotional exhaustion, (ii) job-related anxiety, and (iii) physical symptoms, such that higher control and support
35 will result in weaker relationships between organizational politics and strains.
36
37
38
39 Challenge Demands, Control, and Support Three-Way Interaction
40
41
42 A situation with high levels of both challenge demands and resources tends to result in a different resource allocation
43 strategy. Studies have demonstrated that despite the strain incurred when dealing with challenge stressors, em-
44 ployees are able to recognize opportunities for meaningful resource gains, typically in the form of personal growth
45 and achievement (Cavanaugh et al., 2000; Podsakoff et al., 2007). Challenge demands have thus been shown to in-
46 crease task motivation (e.g., LePine et al., 2005), especially for employees with an abundance of job resources
47 (Bakker, Hakanen, Demerouti, & Xanthopoulou, 2007; Crawford, LePine, & Rich, 2010). For example, employees
48
are likely to believe that time pressure demands can be met with an increase in effort (LePine et al., 2005; Sonnentag,
49
2003), especially when accompanied with high levels of control and support (Daniels & de Jonge, 2010). Indeed, in
50
51 response to challenge demands, empirical studies show that job control and social support increase employee
52 resource-acquisition behaviors (Ito & Brotheridge, 2003; Ohly & Fritz, 2010). Therefore, we argue that when facing
53 challenge demands, employees with high levels of control and support will be prompted to invest their resources to
54 meet the demand.
55
56
Copyright © 2015 John Wiley & Sons, Ltd. J. Organiz. Behav. (2015)
57
58 DOI: 10.1002/job
59
60
61
 HINDRANCE STRESSORS IN THE JDCS MODEL 5
1
2
3 In contrast with resource-conservation efforts that attempt to alter the way a situation is interpreted, resource-
4 acquisition strategies attempt to directly alter the situation (Perrewé & Zellars, 1999). Despite the potential benefits
5 of this strategy, the investment required for resource acquisition actually erodes the very resources it relies upon for
6 investment (Bickerton, Miner, Dowson, & Griffin, 2014). Thus, “simply because one is able to manage a stressful
7 situation with the resources at one’s disposal does not mean that it is not stressful to utilize those resources in han-
8 dling the stressor” (Meurs & Perrewé, 2011, pp. 1047–1048). Although control and support inspire attempts to meet
9 challenge demands, the effort and energy expended acts as an impeding factor in the challenge stressor–strain rela-
10 tionship. For example, workers may utilize job control to work on tasks during lunch breaks, by staying late after
11 work, or by taking work home with them. Because the total number of resources is finite (Hobfoll, 1989), this strat-
12
egy gradually depletes personal resources and cancels out the potential benefit of subsequent resource gains. Accord-
13
ingly, Daniels and Harris (2005) discovered that despite increased task achievement, the use of control to accomplish
14
15 demands failed to reduce strain. Taken together, employees facing challenge demands, with high levels of control
16 and support, sacrifice their resources in exchange for meeting demands.
17 In contrast, those with low levels of workplace control and support are likely to believe they lack the resources
18 necessary to accomplish high levels of challenge demands (Hobfoll, 1989). Consequently, employees forgo costly
19 resource-acquisition behaviors and instead seek to preserve their resources via resource-conservation strategies
20 (Halbesleben & Bowler, 2007). Despite an initial lack of resources (i.e., control and support), conservation motives
21 should thus help reduce any further accumulation of strain by preserving personal resources (e.g., time and effort)
22 that would otherwise be invested in meeting the demand. Accordingly, we propose no buffering effect for control
23 and support in the relationship between challenge stressors and strain.
24 Ultimately, we conclude that the null findings plaguing JDCS research is due to the fact that researchers have
25
overwhelmingly conceptualized and operationalized job demands as challenge stressors. To rule out the possibility
26
27
that our results are conditional upon specific measurement characteristics, the traditional “job demands” construct
28 measured by the Job Content Questionnaire (JCQ; Karasek & Theorell, 1990) was included as a challenge stressor,
29 along with quantitative workload.
30
31 Hypothesis 4: The three-way interaction between workload (i.e., the two challenge demands of job demands and
32 quantitative workload), control, and support will not be related to (i) emotional exhaustion, (ii) job-related anxi-
33 ety, and (iii) physical symptoms, so that no joint buffering effect of control and support will be found.
34
35
36
37
Method
38
39
40 Participants and procedure
41
42 Participants were recruited using Qualtrics, a third-party survey and market research company that provides an on-
43 line recruiting system with access to panelists from a variety of countries. This recruitment process was chosen based
44 on its use in previous studies (e.g., DeCelles, DeRue, Margolis, & Ceranic, 2012; Long, Bendersky, & Morrill,
45 2011; Strauss, Griffin, & Parker, 2012). We recruited only employed panelists from the United States working a
46 minimum of 30 hours a week. To help reduce common method variance (CMV) and demonstrate stability of the ef-
47 fect (Podsakoff, Mackenzie, & Podsakoff, 2012), a two-part data collection strategy was used. All variables were
48
assessed in each of two survey administrations, with a 4-week lag. A 4-week time lag was chosen based on the rec-
49
ommendation of Daniels and Guppy (1994), who state that a 1-month time interval is long enough to allow changes
50
51 in employee strain but is also short enough to allow stability in one’s environment (e.g., job control and support).
52 Initially (Time 1), 522 people completed the survey. Of these, 248 (48 percent) participants responded 4 weeks later
53 at Time 2. Twenty participants were removed for non-purposeful responding and missing data. The hypotheses were
54 therefore tested with a final sample of 228 participants (114 men and 114 women). Participants were mainly
55
56
Copyright © 2015 John Wiley & Sons, Ltd. J. Organiz. Behav. (2015)
57
58 DOI: 10.1002/job
59
60
61
 6 K. M. DAWSON ET AL.
1
2
3 Caucasian (72.4 percent), with fewer African-American (11.4 percent), Asian American (10.1 percent), Hispanic
4 (4.3 percent), and other (1.8 percent) ethnic groups. Their average age was 46.89 (standard deviation (SD) = 11.23),
5 and the mean tenure in their current organization was 10.47 years (SD = 9.42). Participants worked an average of
6 42.85 hours/week (SD = 6.59) and were employed in a variety of industries (e.g., manufacturing, retail, and profes-
7 sional) and occupations (e.g., teacher, manager, and engineer).
8
9
10 Measures
11
12 Hindrance demands
13 Interpersonal conflict. Interpersonal conflict (Time 1 α = .70; Time 2 α = .74) was measured with the four-item
14 Spector and Jex (1998) Interpersonal Conflict at Work Scale. Participants rated how often they experienced conflict,
15
with items such as “how often do you get into arguments with others at work” on a frequency scale ranging from 1
16
(never) to 5 (very often).
17
18
19 Role conflict. Work-related role conflict (Time 1 α = .89; Time 2 α = .87) was assessed using Rizzo, House, and
20 Lirtzman’s (1970) eight-item measure. A sample item is “I receive incompatible requests from two or more people”
21 rated on a 7-point scale from 1 (strongly disagree) to 7 (strongly agree).
22
23 Organizational politics. Organizational politics (Time 1 α = .87; Time 2 α = .85) were assessed using the 15-item
24 Perceptions of Organizational Politics Scale (Kacmar & Carlson, 1997). An example item is “People in this orga-
25 nization attempt to build themselves up by tearing others down” rated on a 7-point scale ranging from 1 (strongly
26
disagree) to 7 (strongly agree).
27
28
29 Challenge demands
30 Job demands. Consistent with much past JDCS research, job demands were measured with Karasek and Theorell’s
31 (1990) job demands (Time 1 α = .75; Time 2 α = .79) measure from the JCQ. A sample item is “My job requires
32 working very fast” rated on a 5-point scale ranging from 1 (strongly disagree) to 5 (strongly agree).
33
34 Quantitative workload. Quantitative workload (Time 1 α = .86; Time 2 α = .86) was assessed with the five-item
35 Quantitative Workload Inventory (Spector & Jex, 1998). A sample item is “How often does your job require you
36 to work very hard” rated by indicating the frequency of such an occurrence, from 1 (less than once per month) to
37 5 (several times per day).
38
39
40 Moderators
41 Control. Job control (Time 1 α = 87; Time 2 α = .88) was assessed using nine items from the JCQ (Karasek &
42 Theorell, 1990). An example item is “On my job, I am given a lot of freedom to decide how I do my work” rated
43 on a 5-point scale from 1 (strongly disagree) to 5 (strongly agree).
44
45 Support. Social support at work (Time 1 α = .95; Time 2 α = .93) was measured using eight items from the JCQ
46 (Karasek & Theorell, 1990), with four items tapping managerial support and four items tapping coworker support.
47 Example items are “My supervisor pays attention to what I am saying” and “People I work with are helpful in get-
48 ting the job done,” rated on a 5-point scale ranging from 1 (strongly disagree) to 5 (strongly agree). Consistent with
49
past JDCS research, managerial and coworker support items were combined to create one social support scale.
50
51
52 Employee strain
53 Three commonly studied strains were examined as criteria in the study: emotional exhaustion, job-related anxiety,
54 and physical symptoms.
55
56
Copyright © 2015 John Wiley & Sons, Ltd. J. Organiz. Behav. (2015)
57
58 DOI: 10.1002/job
59
60
61
 HINDRANCE STRESSORS IN THE JDCS MODEL 7
1
2
3 Emotional exhaustion. Emotional exhaustion (Time 1 α = .95; Time 2 α = .95) was assessed using the nine-item emo-
4 tional exhaustion sub-scale from the Maslach Burnout Inventory (Maslach & Jackson, 1981). An example item is “I
5 feel used up at the end of the workday” rated on a 7-point frequency scale ranging from 1 (never) to 7 (every day).
6
7 Job-related anxiety. Job anxiety (Time 1 α = .82; Time 2 α = .85) was assessed using Warr’s (1990) anxiety–
8 contentment scale. Participants were asked to rate, in the past month, how much of the time their job had made them
9 feel anxious, tense, and worried, and not feel relaxed, calm, and contented, using a 5-point frequency scale from 1
10 (never) to 5 (all of the time).
11
12
13
Physical symptoms. Physical symptoms were assessed using the 18-item Physical Symptoms Inventory (Spector &
14 Jex, 1998), examining somatic health issues often related to stressors, such as backaches, heartburn, and insomnia.
15 Participants indicated, in the past month, whether they had experienced 18 health issues, with the options of 1 (no), 2
16 (yes—has occurred but I did not see a doctor), or 3 (yes—has occurred and I have seen a doctor about it). This scale
17 is considered a causal indicators model, rather than effect model (Spector & Jex, 1998). This proscribes the interpre-
18 tation of a reliability coefficient, which we list out of tradition with that caveat (Time 1 α = .83; Time 2 α = .85). The
19 scale met guidelines for acceptable normality and was analyzed with parametric statistics, consistent with previous
20 research (e.g., Liu, Spector, & Shi, 2007).
21
22
23
24
Results
25
26
27 Table 1 displays the means, SDs, reliabilities, and zero-order correlations of all study variables. Hierarchical mod-T1
28 erated regression analyses were conducted to test all study hypotheses. Based on the recommendations set forth by
29 Aiken and West (1991), all predictor variables were standardized. Strain measured at Time 2 was then regressed
30 onto the predictor variables in the following steps: (i) Time 1 strain, job demands, control, and support; (ii) all
31 two-way interactions; and (iii) the three-way interaction (Tables 2–4). Each significant three-way interaction wasT2 T3 T4
32 then graphed by inserting the high (1 SD above the mean) and low (1 SD below the mean) values for the three var-
33 iables into the regression equation. Simple slopes for the three-way interactions are reported in Table 5. T5
34
35
36 Hindrance demands
37
38
39 Regarding Hypotheses 1a–c, the three-way interactions between interpersonal conflict, control, and support were
40 tested against the strains of emotional exhaustion, job-related anxiety, and physical symptoms. The predictors’
41 three-way interaction was not significantly related to emotional exhaustion (H1a; β = !.01, p = .84, Table 2) but
42 was significantly related to job-related anxiety (H1b; β = !.15, p < .05; Table 3) and physical symptoms (H1c;
43 β = !.17, p < .05; Table 4). To further examine the nature of these moderation effects, the significant interactions
44 were plotted (Figures 1 and 2), and significance tests of the simple slopes and slope differences were conductedF1 F2
45 (Table 5) as per the recommendations of Dawson and Richter (2006). As illustrated in the table, the relationship be-
46 tween interpersonal conflict and job-related anxiety, along with interpersonal conflict and physical symptoms, was
47 most strongly buffered by employees with high levels of both control and support. Thus, Hypotheses 1b and 1c were
48
supported.
49
The three-way interaction proposed in Hypothesis 2 for role conflict, control, and support was not significant for
50
51 emotional exhaustion (H2a; β = !.02, p = .65; Table 2) or physical symptoms (H2c; β = !.10, p = .10; Table 4).
52 However, the three-way interaction between role conflict, control, and support predicting job-related anxiety was
53 significant (H2b; β = !.13, p < .05; Table 3). Thus, only Hypothesis 2b was supported, such that the strongest buff-
54 ering effect was found among employees with both high levels of control and support (Figure 3 and Table 5). F3
55
56
Copyright © 2015 John Wiley & Sons, Ltd. J. Organiz. Behav. (2015)
57
58 DOI: 10.1002/job
59
60
61
9
8
7
6
5
4
3
2
1

11

61
60
59
58
57
56
55
54
53
52
51
50
49
48
47
46
45
44
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
10
8

Table 1. Means, standard deviations, reliabilities, and zero-order correlations.

K. M. DAWSON ET AL.

M SD 1 2 3 4 5 6 7 8 9 10 11 12 13

1. Interpersonal 5.10 1.90 .70

conflict T1
2. Role 28.60 11.22 .39** .89
conflict T1

Copyright © 2015 John Wiley & Sons, Ltd.

3. Organizational 58.33 16.38 .41** .55** .87
politics T1
4. Job 15.91 3.73 .14* .36** .18* .75
demands T1
5. Quantitative 14.98 5.82 .28** .41** .18* .68** .90
workload T1
6. Control T1 30.41 6.90 !.22* !.10 !.40** .02 .02 .87
7. Support T1 28.14 6.54 !.31** !.43** !.60** !.23** !.16* .47** .94
8. Emotional 31.96 14.37 .42** .54** .56** .47** .45** !.27** !.44** .95
exhaustion T1
9. Job!related 17.15 4.51 .36** .47** .50** .46** .39** !.30** !.45** .69** .82
anxiety T1
10. Physical 22.02 3.71 .10 .24** .28** .22** .20* !.20* !.22** .37** .47** .83
symptoms T1
11. Emotional 30.58 14.24 .43** .52** .52** .45** .45** !.21** !.40** .86** .66** .37** .95
exhaustion T2
12. Job-related 16.48 4.52 .38** .46** .53** .45** .38** !.32** !.44** .72** .74** .43** .74** .85
anxiety T2
13. Physical 22.19 4.06 .21** .26** .29** .17* .15* !.23** !.25** .39** .44** .70** .45** .44** .85
symptoms T2
Note: N = 228. Cronbach’s alphas are italicized and on the diagonal.
*p < .05;
**p < .001.

DOI: 10.1002/job
J. Organiz. Behav. (2015)
 HINDRANCE STRESSORS IN THE JDCS MODEL 9
1
2 Table 2. Hierarchical multiple regression analyses predicting Time 2 emotional exhaustion from Time 1 emotional exhaustion,
3 hindrance and challenge stressors, control, and support.
4
5 Type of stressor
6
7 Interpersonal Role Organizational Quantitative
8 conflict conflict politics Job demands workload
9 (hindrance) (hindrance) (hindrance) (challenge) (challenge)
10
Predictor ΔR2 β ΔR2 β ΔR2 β ΔR2 β ΔR2 β
11
12 Step 1 .75** .75** .74** .75** .75**
13 Emotional exhaustion .82** .81** .83** .82** .81**
14 Stressor .08* .07 .05 .06 .08*
15 Control !.01 !.03 !.01 !.03 !.03
16 Support !.01 .00 .00 !.02 !.02
17 Step 2 .00 .00 .01 .00 .00
18 Stressor × Control !.04 !.04 .02 !.02 .00
Stressor × Support .02 !.02 .02 .00 !.04
19
Control × Support !.07 !.06 !.02 !.06 !.05
20 Step 3 .00 .00 .00 .00 .00
21 Stressor × Control × Support !.01 !.02 !.06 .01 .02
22 Total R2 .75** .75** .75** .75** .75**
23
24 Note: N = 228. All predictor variables were assessed at Time 1. Standardized betas are reported.
*p < .05;
25 **p < .001.
26
27
28
29
30
31 Table 3. Hierarchical multiple regression analyses predicting Time 2 job-related anxiety from Time 1 job-related anxiety, hin-
32 drance and challenge stressors, control, and support.
33 Type of stressor
34
35 Interpersonal Role Organizational Quantitative
36 conflict conflict politics Job demands workload
37 (hindrance) (hindrance) (hindrance) (challenge) (challenge)
38
39 Predictor ΔR2 β ΔR2 β ΔR2 β ΔR2 β ΔR2 β
40
41 Step 1 .58** .58** .59** .59** .58**
42 Job-related anxiety .65** .63** .63** .60** .62**
Stressor .11* .12* .18* .16* .13*
43
Control !.06 !.09 !.05 !.10* !.09
44 Support !.09 !.06 !.03 !.09 !.10
45 Step 2 .00 .01 .00 .01 .01
46 Stressor × Control !.05 !.06 !.03 .12* .09
47 Stressor × Support !.03 .01 !.01 !.02 !.05
48 Control × Support .01 .04 .04 .06 .07
49 Step 3 .01* .01* .01* .00 .00
50 Stressor × Control × Support !.15* !.13* !.13* !.09 !.07
51 Total R2 .59** .60** .60** .60** .59**
52 Note: N = 228. All predictor variables were assessed at Time 1. Standardized betas are reported.
53 *p < .05;
54 **p < .001.
55
56
Copyright © 2015 John Wiley & Sons, Ltd. J. Organiz. Behav. (2015)
57
58 DOI: 10.1002/job
59
60
61
 10 K. M. DAWSON ET AL.
1
2 Table 4. Hierarchical multiple regression analyses predicting Time 2 physical symptoms from Time 1 physical symptoms,
3 hindrance and challenge stressors, control, and support.
4
5 Type of stressor
6
7 Interpersonal Organizational Quantitative
8 conflict Role conflict politics workload Job demands
9 (hindrance) (hindrance) (hindrance) (challenge) (challenge)
10
Predictor ΔR2 β ΔR2 β ΔR2 β ΔR2 β ΔR2 β
11
12 Step 1 .51** .50** .50** .50** .50**
13 Physical symptoms .67** .66** .66** .67** .67**
14 Stressor .12* .07 .06 .01 .01
15 Control !.04 !.06 !.05 !.06 !.06
16 Support !.05 !.04 !.05 !.07 !.07
17 Step 2 .01 .02 .01 .01 .01
18 Stressor × Control !.09 !.09 !.08 .07 .05
Stressor × Support .01 !.05 !.02 !.02 !.09
19
Control × Support !.11 !.10 !.12 !.06 !.05
20 Step 3 .01* .01 .01* .00 .00
21 Stressor × Control × Support !.17* !.10† !.15* .04 .01
22 Total R2 .53** .53** .52** .51** .51**
23
24 Note: N = 228. All predictor variables were assessed at Time 1. Standardized betas are reported.
*p < .05;
25 **p < .001;
26 †
p = .10.
27
28
29 Finally, the three-way interaction between organizational politics, control, and support was not significant for
30 emotional exhaustion (β = !.06, p = .28; Table 2); thus, Hypothesis 3a was not supported. The three-way interaction
31 was, however, significantly related to both job-related anxiety (H3b; β = !.13, p < .01; Table 3) and physical symp-
32 toms (H3c; β = !.15, p < .01; Table 4). In support of Hypotheses 3b and 3c, high levels of both control and support
33 most strongly buffered the relationship between role conflict and these outcomes (Figures 4 and 5 and Table 5). F4 F5
34 Overall, there were six differences between slopes for the hindrance demands (Table 5). Half of them (three) were
35 between the high control–high support slope and the low control–high support slope, but the high control–high sup-
36 port slope was significantly different from all of the slopes in at least one comparison.
37
38
39 Challenge demands
40
41 The study proposed that the relationship between hindrance stressors and employee strain would be buffered by job
42 control and support. The contrasting assumption was that this would not be true for challenge stressors (i.e., job de-
43 mands and workload). Consistent with reasoning reviewed earlier, the three-way interactions between challenge
44 stressors, control, and support predicting strains were not significant for any combination of demands (i.e., job de-
45
mands or quantitative workload) and the strains of emotional exhaustion (Table 2), job-related anxiety (Table 3), or
46
physical symptoms (Table 4). The results regarding challenge stressors in the present sample are thus consistent with
47
48 past JDCS findings and our predictions in Hypothesis 4.
49
50
51 Supplementary analyses
52
53 To provide a traditional test of the JDCS model, perceptions of managerial and coworker support were com-
54 bined to operationalize social support. While some researchers have suggested that coworker and managerial
55
56
Copyright © 2015 John Wiley & Sons, Ltd. J. Organiz. Behav. (2015)
57
58 DOI: 10.1002/job
59
60
61
 HINDRANCE STRESSORS IN THE JDCS MODEL 11
1
2 Table 5. Simple slopes for three-way interactions between hindrance stressors (interpersonal conflict, role conflict, and organi-
3 zational politics), control, and support in predicting job-related anxiety and physical symptoms.
4
5 Job-related anxiety Physical symptoms
6
7 Group Slope t Slope t
8
Interpersonal conflict
9
1 (high control, high support) !.04 !.50 !.16 !.46
10 2 (high control, low support) .20 2.03* 1.04 1.76
11 3 (low control, high support) .13 1.67 .86 1.36
12 4 (low control, low support) .15 3.38** 1.13 2.43*
13 Slope difference
14 1 and 2 !1.71 !1.64
15 1 and 3 !1.88 !2.39*
16 1 and 4 !3.00** !2.26*
17 2 and 3 .52 !.11
2 and 4 .58 .19
18
3 and 4 !.35 .41
19 Role conflict
20 1 (high control, high support) !.01 !.07 — —
21 2 (high control, low support) .17 2.13* — —
22 3 (low control, high support) .24 3.11* — —
23 4 (low control, low support) .14 1.79 — —
24 Slope difference
25 1 and 2 !1.90 —
26 1 and 3 !2.25* —
27 1 and 4 !1.32 —
2 and 3 !.69 —
28
2 and 4 .24 —
29 3 and 4 1.14 —
30 Organizational politics
31 1 (high control, high support) .07 .95 !.43 !1.10
32 2 (high control, low support) .20 2.76* .35 .77
33 3 (low control, high support) .21 2.98* .75 1.58
34 4 (low control, low support) .17 2.42* .51 1.49
35 1 and 2 !2.02* !1.47
36 1 and 3 !1.61 !2.08*
37 1 and 4 !.95 !1.70
38 2 and 3 !.15 !.58
2 and 4 .27 !.33
39 3 and 4 .63 .54
40
41 Note: Group numbers correspond to groups listed in the figures.
42 *p < .05;
**p < .001.
43
44
45 support are equally important, empirical studies regarding their differential influence have been mixed (Ng &
46 Sorensen, 2008). Therefore, we repeated all analyses by investigating the buffering effects for managerial and
47 coworker support separately. The results and conclusions (based on plots of all significant interactions) re-
48
main unchanged for 13 out of the 15 original interactions, with the other two interaction results described
49
later.
50
51 The three-way interaction between role conflict, control, and managerial support in the prediction of physical
52 symptoms was significant (p = .047), compared with the present non-significant result of the combined support
53 construct (p = .100). When coworker support was included, the interaction effect was no longer significant
54 (p = .131).
55
56
Copyright © 2015 John Wiley & Sons, Ltd. J. Organiz. Behav. (2015)
57
58 DOI: 10.1002/job
59
60
61
 12 K. M. DAWSON ET AL.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19 Figure 1. The interaction of interpersonal conflict, control, and support on job-related anxiety
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39 Figure 2. The interaction of interpersonal conflict, control, and support on physical symptoms
40
41 The three-way interaction between organizational politics, control, and combined support was significant when
42 predicting physical symptoms. The interaction involving coworker support was also significant (p = .021); however,
43 the interaction with managerial support was non-significant (p = .055).
44
45
46
47 Discussion
48
49
50
51
Previous research on the JDCS model has largely revealed null findings for the model’s crucial buffer hypothesis
52 (Häusser et al., 2010). The present study contributes to the literature by providing a theoretical explanation for
53 the abundance of null findings by classifying demands based on the challenge–hindrance model and guided by
54 the COR theory; it provides support for a revised version of the JDCS, which may be labeled JHCS for job
55
56
Copyright © 2015 John Wiley & Sons, Ltd. J. Organiz. Behav. (2015)
57
58 DOI: 10.1002/job
59
60
61
 HINDRANCE STRESSORS IN THE JDCS MODEL 13
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19 Figure 3. The interaction of role conflict, control, and support on job-related anxiety
20
21
22
23
24
25
Colour online, B&W in print

26
27
28
29
30
31
32
33
34
35
36
37
38
39 Figure 4. The interaction of organizational politics, control, and support on job-related anxiety
40
41
42 hindrance–control–support. Most research on the JDCS model has operationalized demands as workload (amount of
43 work in a given time), a challenge stressor, but the theory is likely accurate mainly for hindrance stressors. Of the 62
44 studies testing the JDCS buffer hypothesis, we only identified one study measuring the effects of hindrance demands
45 (Schaubroeck & Fink, 1998). In line with the new JHCS model, high levels of job control and supervisor support
46 were found to buffer the strain associated with job demands classified as hindrance stressors (e.g.,, role conflict
47 and skill underutilization), but not with demands classified as challenge stressors (e.g., responsibility for others
48
and workload). The type of stressor is thus a boundary condition for the original model.
49
Based on the COR theory, employees experiencing hindrance stressors see little hope for task success and likely
50
51 use the resources of control and support for coping strategies aimed at reducing their strain. Employees experiencing
52 challenge stressors, however, believe that they can succeed and cope by using their resources for personal achieve-
53 ment, in the hope of gaining more resources. This resource-acquisition approach does not alleviate strain while the
54 resources are being used, however. Five of the nine 3-way interactions for the hindrance stressors in the study were
55
56
Copyright © 2015 John Wiley & Sons, Ltd. J. Organiz. Behav. (2015)
57
58 DOI: 10.1002/job
59
60
61
 14 K. M. DAWSON ET AL.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19 Figure 5. The interaction of organizational politics, control, and support on physical symptoms
20
21 significant, but none of the six 3-way interactions for challenge stressors were significant. For each significant inter-
22
action, the buffering effect was dramatic, with the slope for the high control–high support condition being the only
23
negative slope of the four conditions. Furthermore, this was true for both physical strain and one of the psycholog-
24
25 ical strains. The only caveat regarding the new JHCS model of occupational stress is that the interactions including
26 hindrances demands for one strain variable, emotional exhaustion, were not significant. This could be because T1
27 emotional exhaustion accounted for a large proportion of the variance (74 percent), reflecting the relative stability
28 of this construct. In fact, we found that emotional exhaustion does fit the trend of significant interactions with hin-
29 drance stressors and non-significant interactions with challenge stressors when T1 strain is not included in the re-
30 gression, suggesting that the time lag (4 weeks) in the present study is too short for the effect to occur. Future
31 research could examine the relationship between variables in the JHCS model and exhaustion over a time period lon-
32 ger than 4 weeks. Based on recent longitudinal evidence (Dunford, Shipp, Boss, Angermeier, & Boss, 2012), we
33 suggest 6 months as a minimum time lag for future researchers studying emotional exhaustion over time.
34
35
36 Theoretical and practical implications
37
38
39 The present study contributes to the literature by introducing theoretical refinements of the JDCS to make it a JHCS
40 model. Our revision is explained based on the COR theory (Hobfoll, 1989) and on the importance of distinguishing
41 between challenge and hindrance stressors in the JDCS model. Job control and social support buffer the negative
42 impact of hindrance stressors on employee strain, but they do not buffer effects of the commonly assessed challenge
43 demands.
44 In an effort to enhance the utility of research findings, stressor–strain research has traditionally attempted to iden-
45 tify a broad array of universally applicable, job characteristics, but a more parsimonious, less complex set of vari-
46 ables might be just as useful for understanding occupational stress (Van Veldhoven et al., 2005). The JDCS
47 model has the advantage of parsimony, proposing only a set of three categories of variables to explain occupational
48
stress. Broader theories like COR (e.g., Hobfoll, 1989) and also the job demands-resources theory (Bakker &
49
Demerouti, 2007) are more inclusive as they can accommodate many more resources than just control and support;
50
51 if only a subset of variables can adequately explain the occupational stress process, however, it would be easier to
52 understand and use in practice. Therefore, it is worth discovering whether the JDCS theory can be useful, and the
53 present study suggests that it may be quite useful for explaining the large set of workplace stressors characterized
54 as hindrances.
55
56
Copyright © 2015 John Wiley & Sons, Ltd. J. Organiz. Behav. (2015)
57
58 DOI: 10.1002/job
59
60
61
 HINDRANCE STRESSORS IN THE JDCS MODEL 15
1
2
3 One such model that elaborates the JDCS tradition of increasing specificity is the demand-induced strain compen-
4 sation model (de Jonge & Dormann, 2006). Known as the triple-match principle, it argues that the likelihood of find-
5 ing moderating effects may be enhanced when all study variables are based on qualitatively identical dimensions.
6 However, very specific matching hypotheses are less useful across jobs and all elements of a job and are less easily
7 examined than more generally applicable constructs. The present study’s approach is more generally applicable, be-
8 cause every demand, support, control, and strain variable do not need to be matched. Instead, our findings suggest
9 that more general types of control and support may be explanatory if one simply classifies the demands into two cat-
10 egories, hindrance and challenge.
11 Another form of specificity is to examine occupation-specific measures. For the JDCS, that would mean measur-
12
ing demands, control, and support that are specific to the sample. This approach has been taken several times in the
13
past (e.g., Beehr, Jex, Stacy, & Murray, 2000; Brough & Biggs, 2013; Brough & Williams, 2007). Measuring spe- Q1
14
15 cific demands such as encountering violence in a law enforcement job or tutoring struggling students might be ex-
16 amples of stressors that are specific to a given job or situation, and they might be more strongly related to strains than
17 more generic measures that can apply to a wider variety of jobs. The trade-off is that more specific measures might
18 provide stronger research results, provide deeper insight into the specific job situation, and help in the design of spe-
19 cific interventions, but the sample-specific measures need to be developed and redeveloped for every sample.
20 From a theoretical perspective, our findings suggest that job stress models need not increase complexity and speci-
21 ficity in order to garner validity. Instead, we proposed and found that more parsimonious models are possible by merely
22 distinguishing job demands based on the challenge–hindrance framework. Through this broad approach, job resources
23 such as control and support are expected to be important moderators for an array of hindrance demands, ultimately en-
24 hancing the comparability of findings across a wide array of contexts and models. This is especially important for JDCS
25
research, as the buffer hypothesis has not been supported very often, and yet it has been cited over 3700 times in a va-
26
27
riety of academic journals and applied settings (Daniels & de Jonge, 2010). The revision to the model suggested in the
28 present study is relatively easy to understand and use in research and application. Consequently, we encourage re-
29 searchers to further test the efficacy of our revised model and validate the findings obtained in the present study.
30 Our findings indicate that an increase in control and support may mitigate only the deleterious effects associated
31 with hindrance stressors, and not challenge stressors. Therefore, organizations hoping to alleviate employee strain
32 should increase worker control and social support in the face of hindrance demands such as interpersonal conflict
33 and organizational politics but should not expect such changes to working conditions to help ease employee strains
34 much when challenge stressors are the issue. Ultimately, of course, an organization should reduce the occurrence of
35 hindrance demands, as they thwart goal attainment and lead to a variety of negative outcomes through their effects
36 on strains and job performance (LePine et al., 2004; LePine et al., 2005), possibly moderated by their effects on em-
37
ployees’ cognitive appraisals and coping styles (Taris, 2006).
38
39 Although job control and social support are expected to enhance the ability to cope with and meet challenge de-
40 mands, our findings indicate that these resources do not reduce employee strain perceptions. While control and sup-
41 port may help promote task performance (e.g., individuals are better able to accomplish their workload),
42 organizations primarily concerned with worker health and well-being should focus on reducing the energy and effort
43 employees expend on task duties. For example, in lieu of directly reducing workload, organizations could provide
44 training to help employees maximize work efficiency and productivity.
45
46
47 Limitations and directions for future research
48
49
The results strongly argue that future research based on the JDCS theory should focus on hindrance stressors, for
50
51 which the three-way interaction is more likely to be found. Based on the COR theory, we propose that an employee’s
52 resources are used differently for these two types of stressors, sometimes primarily for self-protection from strains
53 (when encountering hindrance stressors) and other times primarily for personal accomplishment (when encountering
54 challenge stressors). The distinction between hindrance and challenge stressors is an appraisal in the eye of the
55
56
Copyright © 2015 John Wiley & Sons, Ltd. J. Organiz. Behav. (2015)
57
58 DOI: 10.1002/job
59
60
61
 16 K. M. DAWSON ET AL.
1
2
3 beholder of course, and any one stressful situation is probably perceived as more hindrance or challenge rather than
4 only one or the other (as shown by Webster, Beehr, & Love, 2011). Future research and theory development on oc-
5 cupational stress would still benefit from finding a similarly parsimonious and simple model to explain effects of
6 resources on challenge stressors.
7 All of the six differences in slopes in the present study involved the high control–high support slopes, and all three
8 of the other slopes were different from the high–high slope at least once. Half of these differences also involved the
9 low control–high support slope, and future research should seek to determine whether this is a consistent finding
10 across samples. If the primary difference is between the high control–high support and low control–high support sit-
11 uations, it may indicate that control is the primary factor accounting for the buffering effect. That is, control may
12
buffer the effects of hindrance stressors more effectively than support.
13
The present study used all self-report measures, raising concerns about CMV effects. CMV is systematic error
14
15 variance due to using a single reporting method and can either attenuate or disattenuate relationships found in a
16 study (Spector, 2006). We minimized CMV by temporally separating the predictor and criterion variables using a
17 4-week time lag as recommended by writers on methods (e.g., Podsakoff et al., 2012). Furthermore, as CMV often
18 inflates correlations (Spector, 2006), it greatly reduces the likelihood of finding significant interaction effects (Wall,
19 Jackson, Mullarkey, & Parker, 1996), suggesting that the present study’s results may be conservative—the interac-
20 tion effects may be stronger than they appeared in the data. Siemsen, Roth, and Oliveira (2010) demonstrated that
21 interaction effects cannot be artifacts of CMV. Nevertheless, to further probe concerns of CMV, based on the rec-
22 ommendations of Podsakoff et al. (2012), all analyses in the study were repeated controlling for negative affectivity,
23 age, sex, and hours worked per week (results available from the first author). These analyses suggested CMV is
24 likely not a concern, as the results and conclusions remained unchanged after including these control variables.
25
Although the use of multiple time points was a strength of this study, by reducing effects of CMV and mood or
26
27
other temporary effects, temporal precedence is not sufficient to show causality. Thus, we only argue that hindrance
28 stressors, control, and support directly affected strain based on strong a priori theory, as this study only revealed that
29 these variables were related. Future studies investigating the challenge–hindrance framework within the JDCS
30 should employ longitudinal, experience sampling, or preferably experimental designs.
31 In addition, Hypothesis 4 predicted no moderating effect of control and support on forms of challenge stressors
32 and strains. While the use of null hypotheses has often been criticized (e.g., Cortina & Folger, 1998; Wainer & Rob-
33 inson, 2003), a null hypothesis may be accepted if a boundary effect is found (Frick, 1995), suggesting that it is pos-
34 sible to find an effect if it truly exists. The fact that we found buffering effects for hindrance demands suggests that
35 we would have been able to find an effect for challenge demands if the null hypothesis were in fact false, thus
36 supporting the use of a null hypothesis in the present study. Additionally, science also advances when effects are
37
not obtained (Platt, 1964) as a greater understanding of a phenomenon is obtained when one identifies “both when
38
39 a phenomenon occurs and when it does not” (Cortina & Folger, 1998, p. 335). Accordingly, establishing the
40 challenge–hindrance boundary effect in the JDCS provides an important step toward the potential usefulness of fur-
41 ther investigation of these mechanisms in occupational stress research.
42
43
44
45 Author biographies
46
47
48
Kevin M. Dawson is an I/O Psychology PhD candidate at Central Michigan University. His research interests in-
49
clude employee well-being, personality, and employee selection.
50
51 Kimberly E. O’Brien received her doctoral degree from the University of South Florida in 2008. Her research pro-
52 gram centers on organizational citizenship behaviors and counterproductive work behaviors, from an occupational
53 health psychology perspective. She has also published on positive and negative mentoring, interpersonal conflict,
54 and psychometric/statistical issues.
55
56
Copyright © 2015 John Wiley & Sons, Ltd. J. Organiz. Behav. (2015)
57
58 DOI: 10.1002/job
59
60
61
 HINDRANCE STRESSORS IN THE JDCS MODEL 17
1
2
3 Terry Beehr is the Director of the PhD Program in I/O Psychology at Central Michigan University. He earned his
4 PhD in organizational psychology from the University of Michigan and has conducted research on occupational
5 stress, employees’ retirement decisions, leadership, and careers, among other topics.
6
7
8
9 References
10 Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting interactions. Thousand Oaks, CA: Sage.
11 Bakker, A. B., & Demerouti, E. (2007). The job demands-resources model: State of the art. Journal of Managerial Psychology,
12 22, 309-328. doi: 10.1108/02683940710733115
13 Bakker, A. B., Hakanen, J. J., Demerouti, E., & Xanthopoulou, D. (2007). Job resources boost work engagement, particularly
14 when job demands are high. Journal of Educational Psychology, 99, 274-284. doi: 10.1037/0022-0663.99.2.274
15 Beehr, T. A., Glaser, K. M., Canali, K. G., & Wallwey, D. A. (2001). Back to basics: Re-examination of demand–control theory
16 of occupational stress. Work & Stress, 15, 115-130. doi: 10.1080/02678370110067002
17 Beehr, T. A., Jex, S. M., Stacy, B. A., & Murray, M. A. (2000). Work stressors and coworker support as predictors of individual
strain and job performance. Journal of Organizational Behavior, 21, 391-405.
18 Bickerton, G. R., Miner, M. H., Dowson, M., & Griffin, B. (2014). Spiritual resources and work engagement among religious
19 workers: A three-wave longitudinal study. Journal of Occupational and Organizational Psychology, 87, 370-391. doi:
20 10.1111/joop.12052
21 Brough, P., & Biggs, A. (2013). Job demands × job control interaction effects: Do occupation specific job demands increase their
22 occurrence? Stress and Health. Advance online publication. doi: 10.1002/smi.2537
23 Brough, P., & Williams, J. (2007). Managing occupational stress in a high-risk industry: Measuring the job demands of correc-
24 tional officers. Criminal Justice and Behavior, 34, 555-567. doi: 10.1177/0093854806294147
25 Cavanaugh, M. A., Boswell, W. R., Roehling, M. V., & Boudreau, J. W. (2000). An empirical examination of self-reported work
26 stress among U.S. managers. Journal of Applied Psychology, 85, 65-74. doi: 10.1037/0021-9010.85.1.65
Cordes, C. L., & Dougherty, T. W. (1993). A review and an integration of research on job burnout. The Academy of Management
27
Review, 18, 621-656. doi: 10.2307/258593
28 Cortina, J. M., & Folger, R. G. (1998). When is it acceptable to accept a null hypothesis: No way Jose? Organizational Research
29 Methods, 1, 334-350. doi: 10.1177/109442819813004
30 Crawford, E. R., LePine, J. A., & Rich, B. L. (2010). Linking job demands and resources to employee engagement and burnout:
31 A theoretical extension and meta-analytic test. Journal of Applied Psychology, 95, 834-848. doi: 10.1037/a0019364
32 Cynkar, A. (2007). Whole workplace health. Monitor on Psychology, 38, 28–31.
33 Daniels, K. (1999). Coping and the job demands–control–support model: An exploratory study. International Journal of Stress
34 Management, 6, 125-144. doi: 10.1023/A:1022932427319
35 Daniels, K., & de Jonge, J. (2010). Match making and match breaking: The nature of match within and around job design. Jour-
36 nal of Occupational and Organizational Psychology, 83, 1-16. doi: 10.1348/096317909X485144
37 Daniels, K., & Guppy, A. (1994). Occupational stress, social support, job control, and psychological well-being. Human Rela-
tions, 47, 1523-1544.
38 Daniels, K., & Harris, C. (2005). A daily diary study of coping in the context of the job demands–control–support model. Journal
39 of Vocational Behavior, 66, 219-237. doi: 10.1016/j.jvb.2004.10.004
40 Dawson, J. F., & Richter, A. W. (2006). Probing three-way interactions in moderated multiple regression: Development and ap-
41 plication of a slope difference test. Journal of Applied Psychology, 91, 917-926. doi: 10.1037/0021-9010.91.4.917
42 de Jonge, J., & Dormann, C. (2006). Stressors, resources, and strain at work: A longitudinal test of the triple-match principle.
43 Journal of Applied Psychology, 91, 1359-1374. doi: 10.1037/0021-9010.91.5.1359
44 de Lange, A. H., Taris, T. W., Kompier, M. A. J., Houtman, I. L. D., & Bongers, P. M. (2003). “The very best of the millennium”:
45 Longitudinal research and the demand–control-(support) model. Journal of Occupational Health Psychology, 8, 282-305.
46 DeCelles, K. A., DeRue, D. S., Margolis, J. D., & Ceranic, T. L. (2012). Does power corrupt or enable? When and why power
47 facilitates self-interested behavior. Journal of Applied Psychology, 97, 681-689. doi: 10.1037/a0026811
Dunford, B. B., Shipp, A. J., Boss, R. W., Angermeier, I., & Boss, A. D. (2012). Is burnout static or dynamic? A career transition
48
perspective of employee burnout trajectories. Journal of Applied Psychology, 97, 637-650. doi: 10.1037/a0027060
49 Frick, R. W. (1995). Accepting the null hypothesis. Memory & Cognition, 23, 132-138.
50 Hackman, J. R., & Oldham, G. R. (1980). Work redesign. Reading, MA: Addison-Wesley. Q2
51 Halbesleben, J. R. B., & Bowler, W. M. (2007). Emotional exhaustion and job performance: The mediating role of motivation.
52 Journal of Applied Psychology, 92, 93-106. doi: 10.1037/0021-9010.92.1.93
53 Häusser, J. A., Mojzisch, A., Niesel, M., & Schulz-Hardt, S. (2010). Ten years on: A review of recent research on the job
54 demand–control (–support) model and psychological well-being. Work & Stress, 24, 1-35. doi: 10.1080/02678371003683747
55
56
Copyright © 2015 John Wiley & Sons, Ltd. J. Organiz. Behav. (2015)
57
58 DOI: 10.1002/job
59
60
61
 18 K. M. DAWSON ET AL.
1
2
Havlovic, S. J., & Keenan, J. P. (1991). Coping with work stress: The influence of individual differences. Journal of Social Be-
3
havior & Personality, 6, 199-212.
4 Hobfoll, S. E. (1989). Conservation of resources: A new attempt at conceptualizing stress. American Psychologist, 44, 513–524.
5 Hobfoll, S. E. (1998). Stress, culture, and community. New York: Plenum.
6 Ito, J. K., & Brotheridge, C. M. (2003). Resources, coping strategies, and emotional exhaustion: A conservation of resources per-
7 spective. Journal of Vocational Behavior, 63, 490-509. doi: 10.1016/S0001-8791(02)00033-7
8 Kacmar, K. M., & Carlson, D. S. (1997). Further validation of the perception of politics scale (POPs): A multiple sample inves-
9 tigation. Journal of Management, 23, 627-658.
10 Karasek, R., & Theorell, T. (1990). Healthy work: Stress, productivity and the reconstruction of working life. New York: Basic
11 Books.
12 LePine, J. A., LePine, M. A., & Jackson, C. L. (2004). Challenge and hindrance stress: Relationships with exhaustion, motivation
to learn, and learning performance. Journal of Applied Psychology, 89, 883-891. doi: 10.1037/0021-9010.89.5.883
13
LePine, J. A., Podsakoff, N. P., & LePine, M. A. (2005). A meta-analytic test of the challenge stressor–hindrance stressor frame-
14 work: An explanation for inconsistent relationships among stressors and performance. Academy of Management Journal, 48,
15 764-775. doi: 10.5465/amj.2005.18803921
16 Liu, C., Spector, P. E., & Shi, L. (2007). Cross-national job stress: A quantitative and qualitative study. Journal of Organizational
17 Behavior, 2, 209-239. doi: 10.1002/job.435
18 Long, C. P., Bendersky, C., & Morrill, C. (2011). Fairness monitoring: Linking managerial controls and fairness judgments in
19 organizations. Academy of Management Journal, 54, 1045-1068. doi: 10.5465/amj.2011.0008
20 Maslach, C., & Jackson, S. E. (1981). The measurement of experienced burnout. Journal of Organizational Behavior, 2, 99–113.
21 doi: 10.1002/job.4030020205
22 Matteson, M. T., & Ivancevich, J. M. (1987). Controlling work stress: Effective human resource and management strategies. San
Francisco, CA: Jossey-Bass.
23
Meurs, J. A., & Perrewé, P. L. (2011). Cognitive activation theory of stress: An integrative theoretical approach to work stress.
24 Journal of Management, 37, 1043-1068. doi: 10.1177/0149206310387303
25 Ng, T. W. H., & Sorensen, K. L. (2008). Toward a further understanding of the relationships between perceptions of support and
26 work attitudes: A meta-analysis. Group & Organization Management, 33, 243-268. doi: 10.1177/1059601107313307
27 Ohly, S., & Fritz, C. (2010). Work characteristics, challenge appraisal, creativity, and proactive behavior: A multi-level study.
28 Journal of Organizational Behavior, 31, 543-565. doi: 10.1002/job.633
29 Perrewé, P. L., & Zellars, K. L. (1999). An examination of attributions and emotions in the transactional approach to the orga-
30 nizational stress process. Journal of Organizational Behavior, 20, 739-752.
31 Platt, J. R. (1964). Strong inference. Science, 146, 347–353.
32 Podsakoff, N. P., LePine, J. A., & LePine, M. A. (2007). Differential challenge stressor–hindrance stressor relationships with job
attitudes, turnover intentions, turnover, and withdrawal behavior: A meta-analysis. Journal of Applied Psychology, 92, 438-
33
454. doi: 10.1037/0021-9010.92.2.438
34 Podsakoff, P. M., MacKenzie, S. B., & Podsakoff, N. P. (2012). Sources of method bias in social science research and recom-
35 mendations on how to control it. Annual Review of Psychology, 63, 539-569. doi: 10.1146/annurev-psych-120710-100452
36 Rizzo, J. R., House, R. J., & Lirtzman, S. I. (1970). Role conflict and ambiguity in complex organizations. Administrative Science
37 Quarterly, 15, 150-163.
38 Schachter, S. (1959). The psychology of affiliation. Palo Alto, CA: Stanford Univ. Press.
39 Schaubroeck, J., & Fink, L. S. (1998). Facilitating and inhibiting effects of job control and social support on stress outcomes and
40 role behavior: A contingency model. Journal of Organizational Behavior, 19, 167-195.
41 Siemsen, E., Roth, A., & Oliveira, P. (2010). Common method bias in regression models with linear, quadratic, and interaction
42 effects. Organizational Research Methods, 13, 456-476. doi: 10.1177/1094428109351241
Sonnentag, S. (2003). Recovery, work engagement, and proactive behavior: A new look at the interface between nonwork and
43
work. Journal of Applied Psychology, 88, 518-528. doi: 10.1037/0021-9010.88.3.518
44 Spector, P. E. (1998). A control model of the job stress process. In C. L. Cooper (Ed.), Theories of organizational stress (pp. 153–
45 169). London: Oxford University Press.
46 Spector, P. E. (2002). Employee control and occupational stress. Current Directions in Psychological Science, 11, 133-136. doi:
47 10.1111/1467-8721.00185
48 Spector, P. E. (2006). Method variance in organizational research: Truth or urban legend? Organizational Research Methods, 9,
49 221-232.
50 Spector, P. E., & Jex, S. M. (1998). Development of four self-report measures of job stressors and strain: Interpersonal conflict at
51 work scale, organizational constraints scale, quantitative workload inventory, and physical symptoms inventory. Journal of
52 Occupational Health Psychology, 3, 356-367. doi: 10.1037/1076-8998.3.4.356
Strauss, K., Griffin, M. A., & Parker, S. K. (2012). Future work selves: How salient hoped-for identities motivate proactive career
53
behaviors. Journal of Applied Psychology, 97, 580-598. doi: 10.1037/a0026423
54
55
56
Copyright © 2015 John Wiley & Sons, Ltd. J. Organiz. Behav. (2015)
57
58 DOI: 10.1002/job
59
60
61
 HINDRANCE STRESSORS IN THE JDCS MODEL 19
1
2
Taris, T. W. (2006). Bricks without clay: On urban myths in occupational health psychology. Work & Stress, 20, 99-104.
3
Thompson, S. C. (1981). Will it hurt less if I can control it? A complex answer to a simple question. Psychological Bulletin, 90,
4 89-101. doi: 10.1037/0033-2909.90.1.89
5 van der Doef, M., & Maes, S. (1999). The job demand–control (–support) model and psychological well-being: A review of 20
6 years of empirical research. Work & Stress, 13, 87-114. doi: 10.1080/026783799296084
7 van Veldhoven, M., Taris, T. W., de Jonge, J., & Broersen, S. (2005). The relationship between work characteristics and em-
8 ployee health and well-being: How much complexity do we really need? International Journal of Stress Management, 12,
9 3-28.
10 Wainer, H., & Robinson, D. H. (2003). Shaping up the practice of null hypothesis significance testing. Educational Researcher,
11 32, 23-31.
12 Wall, T. D., Jackson, P. J., Mullarkey, S., & Parker, S. K. (1996). The demands–control model of job strain: A more specific test.
Journal of Occupational and Organizational Psychology, 69, 153-166. doi: 10.1111/j.2044-8325.1996.tb00607
13
Warr, P. (1990). The measurement of well-being and other aspects of mental health. Journal of Occupational Psychology, 63,
14 193-210.
15 Webster, J. R., Beehr, T. A., & Christiansen, N. D. (2010). Toward a better understanding of the effects of hindrance and chal-
16 lenge stressors on work behavior. Journal of Vocational Behavior, 76, 68-77. doi: 10.1016/j.jvb.2009.06.012
17 Webster, J., Beehr, T. A., Love, K. G. (2011). Extending the challenge–hindrance model of occupational stress: The role of ap-
18 praisal. Journal of Vocational Behavior, 79, 505-516. doi: 10.1016/j.jvb.2011.02.001
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
Copyright © 2015 John Wiley & Sons, Ltd. J. Organiz. Behav. (2015)
57
58 DOI: 10.1002/job
59
60
61
 Author Query Form

Journal: Journal of Organizational Behavior

Article: job_2049
Dear Author,

During the copyediting of your paper, the following queries arose. Please respond to these by annotating your proofs
with the necessary changes/additions.
• If you intend to annotate your proof electronically, please refer to the E-annotation guidelines.
• If you intend to annotate your proof by means of hard-copy mark-up, please use the standard proofing marks. If
manually writing corrections on your proof and returning it by fax, do not write too close to the edge of the paper.
Please remember that illegible mark-ups may delay publication.
Whether you opt for hard-copy or electronic annotation of your proofs, we recommend that you provide additional
clarification of answers to queries by entering your answers on the query sheet, in addition to the text mark-up.

Query No. Query Remark

Q1 AUTHOR: The sentence “Measuring specific demands such as encountering

violence in a law enforcement job or tutoring struggling students might be
examples of stressors” is not clear. Please provide updated text.
Q2 AUTHOR: Reference “Hackman & Oldham (1980)” is not cited in the text.
Please indicate where it should be cited; or delete from the reference list.
USING e-ANNOTATION TOOLS FOR ELECTRONIC PROOF CORRECTION

Required software to e-Annotate PDFs: Adobe Acrobat Professional or Adobe Reader (version 7.0 or
above). (Note that this document uses screenshots from Adobe Reader X)
The latest version of Acrobat Reader can be downloaded for free at: http://get.adobe.com/uk/reader/

Once you have Acrobat Reader open on your computer, click on the Comment tab at the right of the toolbar:

This will open up a panel down the right side of the document. The majority of
tools you will use for annotating your proof will be in the Annotations section,
pictured opposite. We’ve picked out some of these tools below:

1. Replace (Ins) Tool – for replacing text. 2. Strikethrough (Del) Tool – for deleting text.

Strikes a line through text and opens up a text Strikes a red line through text that is to be
box where replacement text can be entered. deleted.

How to use it How to use it

Highlight a word or sentence. Highlight a word or sentence.
Click on the Replace (Ins) icon in the Annotations Click on the Strikethrough (Del) icon in the
section. Annotations section.
Type the replacement text into the blue box that
appears.

3. Add note to text Tool – for highlighting a section 4. Add sticky note Tool – for making notes at
to be changed to bold or italic. specific points in the text.

Highlights text in yellow and opens up a text Marks a point in the proof where a comment
box where comments can be entered. needs to be highlighted.

How to use it How to use it

Highlight the relevant section of text. Click on the Add sticky note icon in the
Click on the Add note to text icon in the Annotations section.
Annotations section. Click at the point in the proof where the comment
Type instruction on what should be changed should be inserted.
regarding the text into the yellow box that Type the comment into the yellow box that
appears. appears.
 USING e-ANNOTATION TOOLS FOR ELECTRONIC PROOF CORRECTION

5. Attach File Tool – for inserting large amounts of 6. Add stamp Tool – for approving a proof if no
text or replacement figures. corrections are required.

Inserts an icon linking to the attached file in the Inserts a selected stamp onto an appropriate
appropriate pace in the text. place in the proof.

How to use it How to use it

Click on the Attach File icon in the Annotations Click on the Add stamp icon in the Annotations
section. section.
Click on the proof to where you’d like the attached Select the stamp you want to use. (The Approved
file to be linked. stamp is usually available directly in the menu that
Select the file to be attached from your computer appears).
or network. Click on the proof where you’d like the stamp to
Select the colour and type of icon that will appear appear. (Where a proof is to be approved as it is,
in the proof. Click OK. this would normally be on the first page).

7. Drawing Markups Tools – for drawing shapes, lines and freeform

annotations on proofs and commenting on these marks.
Allows shapes, lines and freeform annotations to be drawn on proofs and for
comment to be made on these marks..

How to use it
Click on one of the shapes in the Drawing
Markups section.
Click on the proof at the relevant point and
draw the selected shape with the cursor.
To add a comment to the drawn shape,
move the cursor over the shape until an
arrowhead appears.
Double click on the shape and type any
text in the red box that appears.

For further information on how to annotate proofs, click on the Help menu to reveal a list of further options:

View publication stats