Exposure to extraorganizational stressors: Impact on mental health and organizational perceptions for police workers

Page 1

Exposure to Extraorganizational Stressors: Impacton Mental Health and Organizational Perceptionsfor Police Officers

Amanda Biggs, Paula Brough, and Jennifer P. BarbourGriffith University

Disasters, whether natural or human-initiated, occur beyond organizationalboundaries and affect organizational functioning. This research investigatedthe impact of a natural disaster on the health and work attitudes of policeofficers. Structural equation modeling was employed to test whether exposureto a natural disaster intensified job demands and diminished job resources,which, in turn, negatively influenced work outcomes. The research sampleconsisted of 1,623 police officers who completed electronic surveys collectedapproximately 10 months prior to, and 1 month after, a natural disaster.Exposure to certain aspects of a natural disaster was significantly associatedwith work culture support, which, in turn, was associated with job satisfac-tion, work engagement, psychological strain, and turnover intentions(�2[1094] � 2484.03; p � .001; standardized root mean square residual �.04; Tucker-Lewis index � .97; comparative fit index � .98; parsimony-adjusted comparative fit index � .87; root mean square error of approxi-mation � .03). Job resources in particular had a significant impact upon theoutcome variables, supporting theoretical models that emphasize their crit-ical role in the stressor–strain process (e.g., Hobfoll, 1989). This researchsuggests that positive work-related outcomes for organizations directly in-volved with major disasters may be attained through (a) the provision of a

This article was published Online First June 30, 2014.Amanda Biggs, Paula Brough, and Jennifer P. Barbour, School of Applied Psychology

and Griffith Health Institute, Griffith University.The authors acknowledge the assistance provided by the Queensland Police Service. The

views expressed in this material are those of the authors and are not those of the QueenslandPolice Service. Responsibility for any errors of omission or commission remains with theauthors. The Queensland Police Service expressly disclaims any liability for any damageresulting from the use of the material contained in this publication and will not be responsiblefor any loss, howsoever arising, from use of or reliance on this material. This research is fundedby a Queensland Smart State Senior Researcher Fellowship Grant and this support is gratefullyacknowledged.

Correspondence concerning this article should be addressed to Amanda Biggs, School ofApplied Psychology, Griffith University, Mount Gravatt Campus, 176 Messines Ridge Road,Mt Gravatt QLD 4122 Queensland 4111, Australia. E-mail: [email protected]

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

International Journal of Stress Management © 2014 American Psychological Association2014, Vol. 21, No. 3, 255–282 1072-5245/14/$12.00 http://dx.doi.org/10.1037/a0037297

255

Page 2

supportive work culture, (b) targeted supportive organizational responses toemployees personally affected by disasters, and (c) adequate recognition forthe work performed by employees involved in disaster relief efforts.

Keywords: disasters, stress, police officers, work culture, work engagement

Employee well-being and organizational performance may be influencedby stressors originating from multiple sources within and external to anorganization. Extraorganizational stressors are acute or chronic stressors thatoccur externally to the organization and have a negative impact on bothindividual and organizational outcomes (Hochwarter, Laird, & Brouer,2008). Although organizational psychology research has focused consider-ably on stressors derived from within the work context (e.g., job demands,insufficient control, and inadequate support; Brough, O’Driscoll, Kalliath,Cooper, & Poelmans, 2009), the impact of extraorganizational stressors onwork-related outcomes are seldom investigated. Notable exceptions includeresearch focused on chronic extraorganizational stressors such as work–lifeconflict and recovery, which has demonstrated that events occurring outsideof work can influence employee and organizational outcomes, including jobsatisfaction, absenteeism, and job performance (Brough, O’Driscoll, &Biggs, 2009; O’Driscoll & Brough, 2010).

In contrast, the influence of acute extraorganizational stressors on work-related outcomes has received less attention. A recent review called forresearch assessing the impact of acute extraorganizational stressors on orga-nizational outcomes, particularly in relation to natural disasters (James,2011a). Knowledge of the impact of these stressors is valuable for informingboth theory and practice of stress management in organizational settingsduring times of disaster (Hochwarter et al., 2008). This research directlyaddresses this issue by examining the impact of exposure to an acuteextraorganizational stressor (i.e., a natural disaster) on work-related individ-ual and organizational outcomes in a sample of police officers.

MAJOR DISASTERS AND THEIR IMPACT ONWORK-RELATED OUTCOMES

Naturally occurring and human-initiated (intentional or accidental) di-sasters are classified as acute extraorganizational stressors, as they occurbeyond organizational boundaries and adversely influence employee andorganizational performance (Byron & Peterson, 2002; Hochwarter et al.,2008). Disasters occur suddenly, are substantially novel, and are highlydestructive, affecting large portions of the community, causing substantialpersonal and societal damage, and placing a considerable burden upon

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

256 Biggs, Brough, and Barbour

Page 3

economic and public service resources (James, 2011a; Morren, Dirkzwager,Kessels, & Yzermans, 2007). Although their occurrence is rare comparedwith chronic stressors, natural and human-initiated disasters are becomingincreasingly prevalent throughout the world. Recent scholarly discussionshave focused on appropriate organizational preparations and responses toadequately manage employees’ exposure to major disasters (James, 2011b).To date, limited research has studied the impact of major disasters onindividual and organizational functioning; consequently, urgent calls havebeen issued to address this research deficit (Hochwarter et al., 2008; James,2011b).

RESEARCH CONTEXT

From December 2010 to January 2011, a major flood disaster occurred inthe state of Queensland, Australia. Seventy-eight percent of Queensland wasdeclared a disaster zone (consisting of an area exceeding the size of Franceand Germany combined), 2.5 million people were affected, and 29,000homes and businesses were inundated with flood waters (Queensland FloodsCommission of Inquiry, 2012). Police officers assisted in the flood reliefeffort in a variety of ways, including specialist, logistic, and administrativesupport; public order, traffic, and emergency responses; evacuee registration;and public communications. Numerous officers assisted in the flood reliefeffort in an informal, voluntary capacity, and some officers were also per-sonally impacted by the floods because of damage to their own property orpersonal losses.

The present study extends previous research into the effects of majordisasters in four important ways. First, we examine the work-related conse-quences of an acute, extraorganizational stressor (natural disaster), and ex-tend existing research that primarily focused on individual mental healthoutcomes (Freedy, Shaw, Jarrell, & Masters, 1992). Second, we comparehow exposure to different aspects of a natural disaster (i.e., deployment tofrontline operations, communications, voluntary support, and personal loss orproperty damage) influences individual psychological health and work-related outcomes. Third, baseline measures of the research variables collectedapproximately 10 months prior to the onset of the disaster will be controlledin the analysis. The lack of inclusion of baseline controls is a widelyacknowledged limitation of the existing research (e.g., Bacharach & Bam-berger, 2007; Byron & Peterson, 2002). Finally, we examine an indirecteffects model to test whether exposure to aspects of a natural disasterimpacted upon employee well-being directly or indirectly through the inten-sification of job demands and reduced job resources, in accordance with the

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

257Exposure to Extraorganizational Stressors

Page 4

conservation of resources theory (COR; Hobfoll, 1989). The findings of thisresearch can be applied to inform how organizations can most effectivelyminimize the negative consequences of disaster exposure.

THEORETICAL FRAMEWORK AND RESEARCH HYPOTHESES

Research has clearly demonstrated the individual consequences of ex-posure to major disasters, including posttraumatic stress disorder (PTSD),anxiety, depression, and substance abuse (Freedy et al., 1992; Stephens,Long, & Miller, 1997). Fewer studies have focused on the impact of majordisasters on work-related well-being, attitudes, and performance. Byron andPeterson (2002) surveyed 108 employed university students to investigate theimpact of the September 11, 2001, terrorist attacks on individual and work-related outcomes. Exposure to the terrorist attack was associated with higherlevels of event-related psychological strain and organizational supportiveresponses. Event-related psychological strain was also associated with higherlevels of absenteeism from work, after controlling for job dissatisfaction.Morren et al. (2007) compared sick-leave records for rescue workers exposedto a disaster in the Netherlands with a control group of rescue workers, from6 months prior until 3 years after the disaster (factory explosion). Sick leavesignificantly increased for the rescue workers exposed to the disaster com-pared with the control group. Finally, across five unique samples, Hochwarteret al. (2008) demonstrated that hurricane-induced job stress was significantlyassociated with job-related strain. An interaction effect between hurricane-induced stress and perceived resources was observed for job satisfaction,job-related effort, and job-related strain, demonstrating that the deleteriouseffects of hurricane-induced stressors were exacerbated by low resources.

The impact of exposure to a natural disaster on individual and work-related outcomes is a particular concern for emergency services personnel.These employees are required to cope with their own personal circumstances,in addition to being deployed to frontline emergencies, handling increasedcommunity demands for information and services, working long hours, andexperiencing increased emotional demands as a result of the compassionateresponses required by victims (Morren et al., 2007; Waldman, Carmeli, &Halevi, 2011). Although it may be argued that emergency services personnelare specifically trained to respond to such events, research has demonstratedthat they are not impervious to adversity arising from exposure to traumaticevents (Alexander & Wells, 1991). Furthermore, little is known about howdifferent forms of personal or work-related exposure to natural disastersimpacts upon employee well-being and organizational performance.

In this research, we examine how exposure to a natural disaster (flood)impacts upon psychological and work-related outcomes in a sample of police

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

258 Biggs, Brough, and Barbour

Page 5

officers. In particular, we compare four specific aspects of exposure to thedisaster: (a) communications and major projects, (b) deployment to frontlineoperations, (c) volunteering, and (d) personal property damage or loss.Exposure to any aspect of the natural disaster is expected to increasepsychological strain, job dissatisfaction, disengagement, and turnover inten-tions (Byron & Peterson, 2002; Hochwarter et al., 2008; James, 2011b;Kastenmüller et al., 2011).

Hypothesis 1: Exposure to aspects of a natural disaster will be signifi-cantly associated with job satisfaction and work engagement (negativeassociations), and with psychological strain and turnover intentions(positive associations).

Furthermore, we apply COR theory (Hobfoll, 1989) to explain themechanisms underlying the adverse effect of exposure to a natural disaster onwork-related outcomes. Theoretical models of work-related stress typicallyfocus on the mechanisms linking chronic stressors occurring within organi-zational boundaries to strain, and are less conducive to the study of acuteextraorganizational stressors (Kleber & van der Velden, 2009). However,COR theory has been applied in studies investigating the impact of bothchronic and acute stressors on individual and organizational outcomes. Thismodel is adopted in this research to explain how exposure to a natural disasterproduces adverse individual and organizational outcomes via the intensifi-cation of work pressure and reduction in the availability of resources at workfollowing a natural disaster (Hochwarter et al., 2008).

COR theory (Hobfoll, 1989) contends that people purposefully strive toaccrue and protect valued resources that facilitate the achievement of goalsand desired states, and help individuals successfully adapt to stressors(Freedy et al., 1992; Hobfoll, 1989). Within work settings, employees investresources to adapt to chronic job demands. Over time, unsuccessful adapta-tion to chronic job demands depletes resources, increases the likelihood offuture stressors being appraised as threats, limits coping options, enhancesstrain, and promotes reliance on counterproductive strategies to conserveresources (Freedy et al., 1992; Hobfoll, 2001). Despite possessing a suddenand finite period of exposure, disasters also severely tax resource reservesand produce significant adverse outcomes (Beehr, Jex, Stacy, & Murray,2000; Brough, 2004). Therefore, the magnitude of an individual’s strainresponse is dependent on the degree of resource loss incurred by the stressor,rather than the nature of the stressor itself (Hobfoll, 1989).

We expect work pressures will be exacerbated, availability of workculture and supervisor support will be reduced, and employees will have lesscontrol over their work tasks as a result of exposure to a natural disaster. Thisincrease in job demands and reduced job resources will, in turn, be associated

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

259Exposure to Extraorganizational Stressors

Page 6

with increased levels of psychological strain and turnover intentions, andreduced levels of job satisfaction and work engagement (Biggs, Brough, &Barbour, 2014b; Brough et al., 2013; Demerouti, Bakker, Nachreiner, &Schaufeli, 2001). First, in accordance with COR theory (Hobfoll, 1989), jobresources that assist in coping with chronic and acute stressors, such as jobcontrol and work support, are likely to be (temporarily) reduced duringexposure to acute, extraorganizational stressors. Second, intensified workdemands during natural disasters will result from the combination of in-creased job tasks and reduced staffing numbers, as employees personallyaffected by disasters are likely to be absent from work (Hochwarter et al.,2008). Third, disasters may physically compromise actual workplaces, leav-ing employees to manage physical repairs to their own workplaces while alsomaintaining expected levels of service (Hochwarter et al., 2008). Thus,exposure to an extraorganizational stressor (flood) will produce adverseoutcomes via reduced job resources and intensified job demands.

Hypothesis 2: Exposure to aspects of a natural disaster will have anindirect effect on outcomes (job satisfaction, work engagement, psycho-logical strain, and turnover intentions), via perceived job demands andjob resources (job control, supervisor support, and work culture support).

METHOD

Procedure and Participants

Approval to conduct this research was obtained from the ethics commit-tees of the university and participating organization. Two electronic, self-report questionnaires were distributed to all employees of an Australian statepolice service via their internal e-mail system. Time 1 data collection oc-curred in February 2010 (approximately 10 months prior to the Queenslandflood disaster) and Time 2 data collection occurred in February 2011 (ap-proximately one month after the Queensland flood disaster). This researchwas conducted as a component of a longer-term research program, and thetiming of the questionnaire administrations was specified at the commence-ment of the program, rather than being timed to coincide with the potentialeventuation of natural disasters.

The Time 1 and Time 2 questionnaires included demographic items,standardized measures of work characteristics and work outcomes, andoccupation-specific measures designed for the participating police service.Additional questions asking respondents to indicate their involvement withthe Queensland flood disaster were included at Time 2.

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

260 Biggs, Brough, and Barbour

Page 7

Overall, 5,021 (38%) of the 14,907 questionnaires administered at Time 1,and 4,477 (34%) of the 14,976 questionnaires administered at Time 2, werecompleted and returned to the researchers. Respondents were retained in thesample for this study if they were sworn police officers and completed question-naires at both Time 1 and Time 2 (N � 1,662; 40%). Thirty-nine respondentswere deleted from the sample because of missing responses for more than 50%of items within any single research variable. This produced a final sample of1,623. Maximum likelihood estimation applying the expectation-maximizationalgorithm was utilized to impute the remaining missing data. Little’s MissingCompletely at Random test was nonsignificant, suggesting a random pattern ofmissing data.

The average age of the respondents was 41.44 years (SD � 7.93), andaverage organizational tenure was 16.53 years (SD � 10.10). The majority of therespondents were male (n � 1,276; 79%), were employed on a full-time basis(n � 1,581; 97%), were shift workers (n � 1,039; 64%), worked an average of43.44 hr per week (SD � 8.07 hr), and were either senior constables andconstables (n � 794; 49%) or senior sergeants and sergeants (n � 699; 43%).Compared with the general population of police officers from this police service,the research respondents reported a longer organizational tenure (general popu-lation mean � 13.36 years). This is largely because police from higher rankswere overrepresented in the sample (8% of the respondents were inspectors andthose of higher rank, compared with 4% of the general population), whereaspolice of lower ranks were underrepresented in the sample (18% of the respon-dents were constables compared with 31% of the general population). Theproportion of mid-level ranks was equivalent, however, as was the proportion ofmale and female police officers (75% male in general population) and theaverage age of police officers (39.71 years in general population).

Measures

Job Demands and Control

Job demands and job control were measured using a 19-item scale byWall, Jackson, and Mullarkey (1995). Job demands were measured with nineitems (sample item, “Are you required to deal with problems, which aredifficult to solve?”) and job control with 10 items (sample item, “Can youdecide how to go about getting your job done?”) on a 5-point scale rangingfrom 1 (not at all) to 5 (a great deal). High scores represented greaterperceived job demands or job control. This measure demonstrated a highlevel of internal consistency (Cronbach’s alpha) for job demands (.88 at Time1 and .89 at Time 2) and job control (.93 at Time 1 and .94 at Time 2).

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

261Exposure to Extraorganizational Stressors

Page 8

Supervisor Support

Emotional and instrumental support from supervisors was assessed witha four-item measure by Caplan, Cobb, French, Van Harrison, and Pinneau(1980). A sample item is, “How much can [your supervisor] be relied on toprovide you with practical advice, information, or assistance when things gettough at work?” Responses were rated on a 5-point scale from 0 (don’t havesuch a person) to 4 (very much). High scores indicate higher levels ofsupervisor support. This measure demonstrated a high level of internalconsistency (Cronbach’s alpha): .92 (Time 1 and Time 2).

Individual Perceptions of Work Culture Support

A four-item, unidimensional measure assessed individual perceptions ofthe supportiveness of the police organizational culture in relation to bothchronic and acute traumatic stressors, and whether the supportive valuesespoused by the service were reflective of actual practices and policies(Gracia, 2007). The items for this scale were derived from qualitativeinterviews with police officers, and were checked by three subject-matterexperts to ensure the face validity of the items. Subsequent testing of thescale has confirmed its dimensionality and criterion validity (Biggs, Brough,& Barbour, 2014a, in press). A sample item is “[The Service] is sufficientlypeople-focused in its day-to-day management approach.” Responses weremade according to a 5-point scale ranging from 1 (strongly disagree) to 5(strongly agree). High scores indicate positive perceptions of the support-iveness of the police organizational culture. This measure demonstrated ahigh level of internal consistency in the present research (Cronbach’s alpha):.81 (Time 1) and .84 (Time 2).

Intrinsic Job Satisfaction

Intrinsic job satisfaction was measured using a seven-item scale by Warr,Cook, and Wall (1979). Respondents were asked to indicate their level ofsatisfaction with seven different aspects of work on a 7-point scale rangingfrom 1 (I’m extremely dissatisfied) to 7 (I’m extremely satisfied). A sampleitem asks about satisfaction with “the attention paid to suggestions youmake.” High scores indicate high levels of intrinsic job satisfaction. Thismeasure demonstrated a high level of internal consistency (Cronbach’salpha): .86 (Time 1) and .88 (Time 2).

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

262 Biggs, Brough, and Barbour

Page 9

Work Engagement

The nine-item Utrecht Work Engagement Scale (Schaufeli, Bakker, &Salanova, 2006) contains three subscales that assess the three theoreticalcomponents of engagement: Vigor (e.g., “When I get up in the morning, I feellike going to work”), Dedication (e.g., “My job inspires me”), and Absorp-tion (e.g., “I feel happy when I am working intensely”). Responses weremade according to a 7-point response scale ranging from 0 (never) to 6(everyday), with higher scores indicating higher levels of work engagement.The composite measure of work engagement demonstrated a high level ofinternal consistency (Cronbach’s alpha) in this research: .92 (Time 1) and .93(Time 2).

Turnover Intentions

Brough and Frame’s (2004) three-item measure was included to assessintention to leave the organization over the past 12 months (sample item, “Inthe last 12 months, I have seriously considered quitting my job”). Responseswere made on a 5-point scale ranging from 1 (strongly disagree) to 5(strongly agree). Higher scores indicate higher turnover intentions. Thismeasure demonstrated a high level of internal consistency (Cronbach’salpha): .89 (Time 1 and Time 2).

Psychological Strain

The 12-item General Health Questionnaire (GHQ-12; Goldberg, 1972)was employed as an assessment of police officers’ mental health. Items wereprefaced with the stem, “Have you recently experienced the following in thepast few weeks . . . ,” and a sample item is “felt constantly under strain.”Items were measured on a 4-point scale from 0 (not at all) to 3 (much morethan usual), with higher scores indicating higher psychological strain. Themeasure demonstrated a high level of internal consistency (Cronbach’salpha): .90 (Time 1) and .91 (Time 2).

Exposure to a Natural Disaster

Four dichotomous questions devised by this research were included todetermine exposure to different aspects of flood damage and relief efforts: (a)major incident room or communications, (b) deployment to frontline opera-

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

263Exposure to Extraorganizational Stressors

Page 10

tions, (c) volunteer (e.g., helping clean or repair damaged property), and (d)personal property damage or loss. Respondents were asked to indicatewhether the items were relevant to them, by selecting either “yes” (coded 1)or “no” (coded 0).

Data Analysis

Measurement Model

The 20-factor measurement model incorporating each of the Time 1and Time 2 research variables was evaluated using confirmatory factoranalysis (CFA; Anderson & Gerbing, 1988). For the four aspects ofnatural disaster exposure, single-item latent variables were constructedwith the factor loadings set to 1 and residual variances set to equal thevariance multiplied by 1 minus the internal consistency. As internalconsistency cannot be calculated in single-item scales, we set the internalconsistency to .90 (Demerouti, Shimazu, Bakker, Shimada, & Kawakami,2013). Individual items were used as indicators for work culture support,supervisor support, and turnover intentions. To reduce the complexity ofthe model, item parcels and composite scores were used as indicators forvariables containing greater than four items. Subscale scores were used asindicators for multifactor variables (i.e., work engagement, job demands,and job control), and item parcels were constructed for variables with asingle underlying dimension (i.e., intrinsic job satisfaction and psycho-logical strain), using the balancing method (Little, Cunningham, Shahar,& Widaman, 2002; Little, Rhemtulla, Gibson, & Schoemann, 2013). Theunidimensionality of intrinsic job satisfaction and psychological strainwas verified via exploratory factor analysis prior to constructing theparcels. Finally, the residuals for each indicator at Time 1 were correlatedwith their corresponding Time 2 residual to account for the systematicvariance shared by identical items over time (Pitts, West, & Tein, 1996).

Measurement model invariance was tested following the methodrecommended by Byrne (2010). In the initial step, an unconstrainedbaseline model was compared with a fully constrained model with cor-responding Time 1 and Time 2 factor loadings, latent variable variances,and synchronous latent correlations constrained to equality. A chi-squaredifference test and goodness-of fit statistics determined whether theunconstrained and fully constrained models were significantly different.All invariant parameters were held constant in subsequent models,whereas the noninvariant parameters were freely estimated (Byrne, 2010).

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

264 Biggs, Brough, and Barbour

Page 11

Model Estimation

The adequacy of model-fit was determined according to the (a)chi-square statistic, along with its corresponding degrees of freedom andp value; (b) pattern of standardized residuals; (c) goodness-of-fit statis-tics; (d) squared multiple correlations of the indicators (measurementmodel) and endogenous latent factors (structural model); and (e) modi-fication indices (Byrne, 2010; Kline, 2011). For the goodness-of-fit sta-tistics, cutoffs recommended in previous research were observed (Byrne,2001; Hu & Bentler, 1999): values exceeding .95 for the comparative fitindex (CFI; Bentler, 1990) and the Tucker-Lewis index (TLI; Bentler &Bonett, 1980), and values below .05 for the standardized root mean squareresidual (SRMR; Bentler, 1990) and root mean square error of approxi-mation (RMSEA; Browne & Cudeck, 1993; Steiger, 1990). Finally, formodel comparisons, the Akaike information criterion (AIC) was used tocompare two competing models, with a smaller AIC indicating superiormodel fit (Byrne, 2001).

Structural Model Determining Direct and Indirect Effects

Two structural models were evaluated to test the research hypotheses.First, a structural model containing direct effects from exposure to aspectsof a natural disaster to job characteristics (demands, control, and support)and outcomes (job satisfaction, work engagement, psychological strain,and turnover intentions) was tested (pathways from job demands, jobcontrol, supervisor support, and work culture support to the outcomeswere constrained to zero in this model). Second, an indirect effects modelwas constructed with all indirect and direct pathways freely estimatedfrom exposure to aspects of a natural disaster to job characteristics(demands, control, and support) to the outcomes. This indirect model is anintervening variable model in which a portion of the effect of theindependent variables (IV) on the dependent variables (DV) is expectedto occur via mediating variables (MV) (Hayes, 2009). In both structuralmodels, the Time 1 baseline measures of each of the Time 2 variables(except for exposure to aspects of a natural disaster) were included ascontrol variables. Each indirect effect was formally tested by calculatingthe product of a significant Path A (IV to MV) and significant Path B (MVto DV; MacKinnon, Warsi, & Dwyer, 1995), and obtaining bias-correctedbootstrapped confidence intervals.

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

265Exposure to Extraorganizational Stressors

Page 12

RESULTS

Differences Between Completers and Noncompleters

ANOVAs were conducted to test for differences between completers(completed both Time 1 and Time 2 surveys) and noncompleters (completedonly the Time 1 survey) for the research variables. There were no significantdifferences at Time 1 for work culture support, psychological strain, jobdemands, supervisor support, work engagement, and intrinsic job satisfac-tion. Significant differences occurred for turnover intentions, F(1, 3433) �10. 39, p � .001, �p

2 � .00, and job control, F(1, 3403) � 13.33, p � .001,�p

2 � .00. Job control was higher for completers (M � 3.52, SD � 0.86)compared with noncompleters (M � 3.41, SD � 0.87). Conversely, turnoverintentions were higher for noncompleters (M � 2.51, SD � 1.20) comparedwith completers (M � 2.38, SD � 1.14).

Descriptives and Bivariate Relationships

The scale descriptives, internal consistency (Cronbach’s alpha), andcorrelations are presented in Table 1. Several small, but statistically signif-icant, correlations were observed between natural disaster exposure, jobcharacteristics, and outcomes measured at Time 2. Involvement in majorincidents/communication was positively associated with demands, work cul-ture support, intrinsic job satisfaction, and work engagement, and was neg-atively associated with turnover intentions at Time 2. Deployment to front-line operations was negatively associated with work culture support andintrinsic job satisfaction, and positively associated with turnover intentions atTime 2. Volunteering was negatively associated with job control at Time 2only. Finally, personal loss or property damage was negatively associatedwith work culture support and intrinsic job satisfaction, and positivelyassociated with turnover intentions at Time 2. All of the variables weresignificantly related to one another in the expected direction, with theexception of job demands, which was not significantly associated with all ofthe research variables. Nonetheless, the correlations are comparable withpreviously published research results (e.g., Brough & Biggs, 2013).

CFA and Invariance of Measurement Model

The CFA results indicated that the hypothesized measurement modelproduced an acceptable fit to the data, �2(1064) � 2219.29, p � .001,

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

266 Biggs, Brough, and Barbour

Page 13

SRMR � .03, TLI � .98, CFI � .98, parsimony-adjusted comparative fitindex (PCFI) � .79, RMSEA � .03. Squared multiple correlations for theindicators of the latent factors ranged from .66 to .81 for supervisor support,.44 to .69 for work culture support, .72 to .80 for turnover intentions, .62 to.75 for intrinsic job satisfaction, .74 to .79 for psychological strain, .47 to .91for work engagement, .67 to .91 for job control, and .63 to .73 for jobdemands. Therefore, in most cases, the latent factors explained at least 60%of the variance in the indicators. Inspection of the standardized residualcovariances indicated that the majority were below the recommended cutoffof 2.58 (Byrne, 2010).

The longitudinal invariance of the factor loadings, latent variances, andlatent covariances was tested by comparing an unconstrained model (Step 1,Table 2) with a fully constrained model (Step 2, Table 2). A chi-squaredifference test indicated that there was a significant difference between thetwo models, indicating that parameters were not equivalent over time,��2(52) � 70.73, p � .043. Factor loadings, latent variances, and latentcovariances were progressively constrained to determine the source of thevariance. Further comparisons determined that the factor loadings (Step 3,Table 2) and latent variances (Step 4, Table 2) were invariant. However, thecovariances between work culture support and supervisor support, and be-tween work culture support and turnover intentions, were not equivalent overtime. We note that despite the significant chi-squared difference test, thegoodness of fit statistics were similar for the unconstrained and fully con-strained model. Based on this, and the finding that all factor loadings, latentvariances, and (most) latent covariances were equivalent over time, weconsider an acceptable level of factorial invariance was achieved (Byrne,2010). The factor loadings were constrained to equality in the subsequentstructural models.

Associations Between Natural Disaster Exposure andWork and Health Outcomes

Exposure to aspects of a natural disaster was hypothesized to ad-versely affect perceptions of job demands, job resources, and work-related outcomes. The direct effects model produced acceptable fit sta-tistics, �2(1184) � 3439.40, p � .001, SRMR � .07, TLI � .96, CFI �.96, PCFI � .86, RMSEA � .03. Pathways from job demands andresources to the outcomes were constrained to zero, and baseline (Time 1)values of the Time 2 research variables were controlled for in the analysis.The significant paths are presented in Figure 1 and show that involvementin major operations and communication was significantly related to work

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

267Exposure to Extraorganizational Stressors

Page 14

culture support, intrinsic job satisfaction, work engagement, and turnoverintentions. Personal or property damage was also significantly associatedwith work culture support, intrinsic job satisfaction, and turnover inten-tions. Deployment to frontline operations and volunteering had no sig-nificant impact on any of the research variables. Exposure to aspects ofthe natural disaster was not significantly related to job demands, jobcontrol, supervisor support, or psychological strain.

We also tested indirect relationships in which exposure to a naturaldisaster was hypothesized to adversely impact perceptions of job demandsand job resources, which, in turn, would have a negative influence onwork-related outcomes (Hypothesis 2). As there were no significanteffects for exposure to aspects of the natural disaster on job demands, jobcontrol, and supervisor support, we tested a revised model in which onlythe indirect effects via perceived work culture support were included.Furthermore, we removed the two aspects of exposure to a major disaster(volunteering and deployment to major operations) that had no significantassociation with any of the psychosocial work characteristics and out-comes. The revised indirect effects model produced acceptable fit statis-tics, �2(1094) � 2484.03, p � .001, SRMR � .04, TLI � .97, CFI � .98,PCFI � .87, RMSEA � .03. The significant pathways for the revisedindirect effects model are depicted in Figure 2.

The revised indirect effects model was analyzed within 10,000 boot-strap samples to ascertain the magnitude and significance of the indirect

Table 1. Bivariate Associations of the Research Variables (N � 1,623)

M SD 1 2 3 4 5 6 7 8

1. Communications –2. Frontline .09*** –3. Volunteer .03 .09*** –4. Damage .01 .09*** .15*** –5. Demands T1 3.90 0.69 .08** �.00 �.02 .03 (.88)6. Control T1 3.53 0.86 .01 �.03 �.04 �.02 .04 (.93)7. Supervisor support T1 2.62 0.95 �.04 �.03 �.03 �.01 �.01 .21*** (.92)8. Work culture support T1 2.85 0.82 .08*** �.08** .01 �.03 �.03 .28*** .35*** (.81)9. Intrinsic satisfaction T1 4.30 1.07 �.00 �.08** �.05* �.03 �.04 .46*** .49*** .61***

10. Engagement T1 3.91 1.40 .04 �.01 .01 .00 .07** .39*** .36*** .45***

11. Turnover intentions T1 2.38 1.14 .02 .05 .05 .06* .08** �.26*** �.28*** �.41***

12. Psychological strain T1 1.00 0.46 .02 .04 .02 .04 .15*** �.27*** �.28*** �.29***

13. Demands T2 3.84 0.69 .06* .02 .00 .02 .61*** .02 �.02 �.0114. Control T2 3.58 0.86 .03 �.03 �.06* �.04 .01 .61*** .17*** .27***

15. Supervisor support T2 2.63 0.94 �.04 �.04 .01 �.01 .00 .16*** .52*** .29***

16. Work culture support T2 2.92 0.84 .12*** �.08** �.02 �.08*** .00 .26*** .26*** .66***

17. Intrinsic satisfaction T2 4.34 1.11 .06* �.05* �.04 �.07** �.01 .36*** .36*** .51***

18. Engagement T2 3.93 1.41 .09*** .01 .02 �.03 .05* .30*** .28*** .42***

19. Turnover intentions T2 2.47 1.18 �.05* .05* .04 .09*** .03 �.26*** �.23*** �.37***

20. Psychological strain T2 1.00 0.46 �.01 �.01 �.01 .05 .09*** �.19*** �.18*** �.21***

Note. Communications � involvement in major incidents/communication; Frontline � de-ployment to frontline operations; Volunteer � volunteering; Damage � personal loss orproperty damage; T1 � Time 1; T2 � Time 2.* p � .05. ** p � .01. *** p � .001.

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

268 Biggs, Brough, and Barbour

Page 15

relationships. The unstandardized indirect effect, bootstrapped standarderrors, and bias-corrected confidence intervals are presented in Table 3.All of the indirect effects were significant, demonstrating that a propor-tion of the total effect of exposure to the natural disaster on the outcomevariables occurred indirectly via perceived work culture support. Thisrevised model accounted for 28.82% of the variance of supervisor sup-port, 43.50% of the variance of job control, 45.42% of the variance ofdemands, 53.89% of the variance of work culture support, 32.67% of thevariance of strain, 75.53% of the variance of intrinsic job satisfaction,58.44% of the variance of turnover intentions, and 61.05% of the varianceof work engagement.

Overall, the results indicated that involvement in communications andmajor operations, and personal property damage or loss, were associatedwith perceived work culture support, which, in turn, was associated withintrinsic job satisfaction, work engagement, turnover intentions, andpsychological strain. Job control and supervisor support were also sig-nificantly associated with intrinsic job satisfaction, work engagement,turnover intentions, and psychological strain. Job demands were posi-tively associated with both work engagement and psychological strain.Finally, the strongest predictors of the Time 2 work-related characteristicsand outcomes were the baseline measures of each of the variables at Time1, demonstrating their stability over time.

9 10 11 12 13 14 15 16 17 18 19 20

(.86).53*** (.92)

�.49*** �.44*** (.89)�.42*** �.39*** .39*** (.90)�.03 .06* .04 .13*** (.89)

.37*** .29*** �.22*** �.24*** .08** (.94)

.36*** .24*** �.21*** �.22*** .02 .27*** (.92)

.51*** .38*** �.38*** �.26*** �.02 .33*** .31*** (.84)

.68*** .42*** �.40*** �.32*** �.01 .49*** .52*** .65*** (.88)

.41*** .68*** �.36*** �.29*** .12*** .40*** .37*** .49*** .56*** (.93)�.41*** �.35*** .69*** .31*** .02 �.30*** �.27*** �.48*** �.53*** �.45*** (.89)�.28*** �.22*** .26*** .50*** .11*** �.29*** �.28*** �.31*** �.40*** �.38*** .38*** (.91)

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

269Exposure to Extraorganizational Stressors

Page 16

Tab

le2.

Lon

gitu

dina

lIn

vari

ance

Tes

ting

ofth

eM

easu

rem

ent

Mod

el(N

�1,

623)

Step

Mod

elC

ompa

rativ

em

odel

�2

df�

�2

�df

pC

FI�

CFI

SRM

RR

MSE

AA

IC

1U

ncon

stra

ined

mod

el(M

odel

A)

—22

19.2

910

64—

——

.98

—.0

3.0

329

51.2

92

Fully

cons

trai

ned

mod

elM

odel

A22

90.0

211

1670

.73

52.0

43.9

80

.03

.03

2918

.02

3C

onst

rain

edfa

ctor

load

ings

(Mod

elB

)M

odel

A22

41.5

210

8022

.23

16.1

36.9

80

.03

.03

2941

.52

4C

onst

rain

edla

tent

vari

ance

s(M

odel

C)

Mod

elB

2251

.08

1088

9.56

8.2

97.9

80

.03

.03

2935

.08

5C

onst

rain

edla

tent

cova

rian

ces

for

wor

kcu

lture

supp

ort

Mod

elC

2265

.06

1095

13.9

87

.052

.98

0.0

3.0

329

35.0

6

6C

onst

rain

edla

tent

cova

rian

cebe

twee

nw

ork

cultu

resu

ppor

tan

dsu

perv

isor

supp

ort

Mod

elC

2255

.23

1089

4.15

1.0

42.9

80

.03

.03

2937

.23

7C

onst

rain

edla

tent

cova

rian

cebe

twee

nw

ork

cultu

resu

ppor

tan

dtu

rnov

erin

tent

ions

Mod

elC

2256

.53

1089

5.45

1.0

20.9

80

.03

.03

2938

.53

8C

onst

rain

edla

tent

cova

rian

ces

for

rem

aini

ngw

ork

cultu

resu

ppor

tco

vari

ance

s(M

odel

D)

Mod

elC

2253

.70

1093

2.62

5.7

58.9

80

.03

.03

2927

.70

9C

onst

rain

edla

tent

cova

rian

ces

for

all

rem

aini

ngva

riab

les

Mod

elD

2279

.06

1114

25.3

621

.232

.98

0.0

3.0

329

11.0

6

Not

e.df

�de

gree

sof

free

dom

;CFI

�co

mpa

rativ

efit

inde

x;SR

MR

�st

anda

rdiz

edro

otm

ean-

squa

rere

sidu

al;R

MSE

A�

root

mea

n-sq

uare

erro

rof

appr

oxim

atio

n;A

IC�

Aka

ike

info

rmat

ion

crite

rion

.

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

270 Biggs, Brough, and Barbour

Page 17

Figure 1. Final direct effects model with standardized beta coefficients for significant paths only.Pathways from job demands, job control, supervisor support, and work culture support tooutcomes were constrained to equal zero and were thus excluded in this figure. Significant pathsbetween Time 1 and Time 2 variables were omitted to simplify the model: Time 1 work culturesupport to Time 2 work culture support (� � .74, p � .001); Time 1 supervisor support to Time2 supervisor support (� � .54, p � .001); Time 1 demands to Time 2 demands (� � .68, p �.001); Time 1 control to Time 2 control (� � .66, p � .001); Time 1 intrinsic job satisfactionand Time 2 intrinsic job satisfaction (� � .72, p � .001); Time 1 engagement to Time 2engagement (� � .70, p � .001); Time 1 turnover intentions to Time 2 turnover intentions (� �.71, p � .001); Time 1 psychological strain to Time 2 psychological strain (� � .51, p � .001).

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

271Exposure to Extraorganizational Stressors

Page 18

Figure 2. Final revised indirect effects model with standardized beta coefficients for significantpaths only. Significant paths between Time 1 and Time 2 variables were omitted to simplify themodel: Time 1 work culture support to Time 2 work culture support (� � .73, p � .001); Time1 supervisor support to Time 2 supervisor support (� � .54, p � .001); Time 1 demands toTime 2 demands (� � .67, p � .001); Time 1 control to Time 2 control (� � .66, p � .001);Time 1 intrinsic job satisfaction and Time 2 intrinsic job satisfaction (� � .34, p � .001); Time1 engagement to Time 2 engagement (� � .55, p � .001); Time 1 turnover intentions to Time2 turnover intentions (� � .61, p � .001); Time 1 psychological strain to Time 2 psychologicalstrain (� � .42, p � .001).

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

272 Biggs, Brough, and Barbour

Page 19

DISCUSSION

Impact of Exposure to Natural Disasters on Individual andOrganizational Outcomes

Although previous research has predominantly focused on individualpsychological health outcomes, this study evaluated the effect of exposureto a natural disaster on specific work-related outcomes. Hypothesis 1predicted that exposure to a natural disaster would be associated withintensified psychological strain and turnover intentions, as well as de-creased job satisfaction and work engagement. In support of this hypoth-esis, experiencing personal property damage or loss was associated withlower intrinsic job satisfaction and higher turnover intentions. In contrastto the hypothesized negative associations, involvement in communica-tions and major operations was associated with higher levels of intrinsicjob satisfaction and work engagement, and lower turnover intentions.Finally, volunteering and deployment to frontline operations were notsignificantly related to work-related outcomes. These findings demon-strate that different aspects of exposure to a natural disaster have uniqueassociations with organizational outcomes, thereby highlighting the im-portance of assessing different aspects of exposure as opposed to mea-suring exposure as a singular variable.

In addition to the hypothesized direct effects, we found support forindirect effects of exposure to aspects of a natural disaster on work-related outcomes, via job resources (i.e., work culture support; Hypoth-esis 2). Experiencing personal property damage or loss was associatedwith more negative perceptions of work culture support, whereas involve-ment in communications and major operations were associated with morepositive perceptions of work culture support. Contrary to expectations,exposure to aspects of a natural disaster was not associated with job

Table 3. Indirect Effects Through Work Culture Support (N � 1,623)

B SE 95% BC CI p

CommunicationsPsychological strain T2 �0.02 0.01 [�0.03, �0.01] �.001Intrinsic job satisfaction T2 0.10 0.02 [0.06, 0.14] �.001Turnover intentions T2 �0.07 0.02 [�0.11, �0.04] �.001Work engagement T2 0.07 0.02 [0.04, 0.11] �.001

DamagePsychological strain T2 0.01 0.01 [0.01, 0.03] �.001Intrinsic job satisfaction T2 �0.09 0.03 [�0.14, �0.04] �.001Turnover intentions T2 0.06 0.02 [0.03, 0.11] �.001Work engagement T2 �0.06 0.02 [�0.11, �0.03] �.001

Note. BC CI � bias corrected confidence interval; T2 � Time 2.

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

273Exposure to Extraorganizational Stressors

Page 20

demands, job control, or supervisor support. As a result, a revised indirecteffects model was tested, which demonstrated a significant indirect effectbetween exposure to some aspects of a natural disaster (i.e., personalproperty damage or loss and involvement in communications and majoroperations) and work-related outcomes via work culture support. There-fore, Hypothesis 2 was partially supported.

Indirect Effect of Work Culture Support

Employees who were personally affected by the natural disasterreported lower perceived work culture support after the disaster, which isan important work-related resource predicting job satisfaction, workengagement, psychological strain, and turnover intentions. Thus, it ap-pears these employees experienced cumulative resource loss both withinand beyond work boundaries. This pattern of findings reflects a resourceloss spiral described by Hobfoll (1989, 2001), in which an initial loss ofresources diminishes one’s capacity to cope with stress, increasing thelikelihood of further resource loss, strain, and withdrawal.

Further exploration of the data indicated that individuals who werenot affected personally by the natural disaster reported a significantincrease in work culture support over time, whereas those who werepersonally affected perceived significantly lower work culture supportafter the natural disaster. This suggests that organizationally supportiveresponses were not perceived as effectively by employees personallyaffected by the natural disaster. There are several plausible explanationsfor this finding. First, employees personally affected by the naturaldisaster may have withdrawn from work in an attempt to cope with theirpersonal circumstances, diminishing their opportunity to receive supportfrom organizational sources. Second, the timing and type of supportavailable may not have been beneficial to those who experienced personalloss from the natural disaster. Alexander and Wells (1991), for example,cautioned that organizationally supportive responses after a disastershould be timely, readily available, and unobtrusive. Third, as the lossoccurred outside the work domain, individuals may have been morereluctant to seek support from work-related sources. Stephens et al.(1997), for instance, demonstrated that negative attitudes toward theexpression of emotions at work strengthened the positive associationbetween traumatic stressors and PTSD.

Conversely, individuals involved in major operations and communi-cations reported positive associations with perceived work culture supportand work-related outcomes. This unexpected positive association may

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

274 Biggs, Brough, and Barbour

Page 21

have occurred if employees involved in the disaster relief effort appraisedtheir additional job demands as challenging, rather than threatening,especially if they felt they had the resources to cope with the extrademands (Lepine, Podsakoff, & Lepine, 2005). These employees mayhave benefited from additional supportive responses provided by theorganization. In addition, participation in the disaster relief effort mayhave increased the perceived meaningfulness of their work tasks ifemployees felt that they were positively contributing to the lives of otherswithin flood-affected communities (May, Gilson, & Harter, 2004). Sim-ilarly, Hochwarter et al. (2008) demonstrated that job satisfaction wasgreater for employees reporting high levels of both hurricane-inducedjob-related stressors and perceived resources.

Overall, our findings concur with COR theory (Hobfoll, 1989), whichemphasizes the critical role of job resources. Sufficient resource reservesare integral during exposure to acute extraorganizational stressors, asresources attenuate stressor–strain associations, allow behavior regulationto meet unpredictable and emotionally challenging work demands, andenable individuals to engage in proactive coping (Hobfoll, 1989;Hochwarter et al., 2008). The importance of work culture support inparticular is unsurprising, as the value of this resource has been demon-strated in previous research. For example, Byron and Peterson (2002)reported that employees whose companies provided targeted supportiveresponses after the September 11, 2001, disaster in the United States weremore satisfied with their jobs, demonstrating that “employees may expecttheir companies to ‘step up to the plate,’ even when a stressor is outsidethe domain of the organization” (p. 906). Similarly, Alexander and Wells(1991) demonstrated that police officers’ reports of positive organiza-tional morale and support mitigated the effect of exposure to a disaster onpsychological outcomes, concluding that “powerful antidotes to the del-eterious effects of unpleasant tasks are organizational and managerial innature” (p. 553).

The importance of work culture support relative to job characteristicsin the research model has broader theoretical implications. Studies onsociopolitical support constructs, such as work culture support, havedemonstrated their association with personal feelings of value, meaning-fulness, self-determination, and task competence (Seibert, Wang, &Courtright, 2011). Examining the interplay between organizational andindividual constructs is an important addition to research that has largelyfocused on job characteristics. For instance, evaluating whether workculture support builds meaningfulness or is most effective for individualswho already possess a high level of meaningfulness could be investigatedfurther.

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

275Exposure to Extraorganizational Stressors

Page 22

Relevance of Work-Related Demands and Resources

Job demands and job resources accounted for a larger proportion ofvariance in the outcomes compared with exposure to the flood. In particular,control over how and when to perform job tasks and appropriate levels ofsupervisor support were associated with more positive attitudes toward theworkplace, higher work engagement and job satisfaction, and lower psycho-logical strain and turnover intentions. Previous research also demonstratedthe importance of job resources for both normal work performance, and whenextreme uncertainty occurs because of traumatic events or acute extraorga-nizational stressors (e.g., Freedy et al., 1992; Hobfoll, 1989; Hochwarter etal., 2008).

These findings also support previous research within high-stress occu-pations, demonstrating that chronic exposure to routine work stressors pro-vokes more intense and enduring effects on work-related outcomes comparedwith acute stressors (e.g., Alexander & Wells, 1991; Brough, 2004, 2005;Brough & Biggs, 2010). Despite the perception that traumatic events areespecially stressful, because of the presumption that the intense arousalassociated with the trauma overcompensates for its relatively short duration,research consistently fails to support this notion (Beehr, 1998). This researchsuccessfully extends previous studies and makes a significant contribution toscholarly discussions by evaluating the effects of different aspects of expo-sure to a natural disaster on both individual and organizational outcomes,while also accounting for job demands and job resources.

Practical Implications

Law enforcement organizations typically have well-established policiesand practices in place to ensure the provision of support to employeesexposed to traumatic events. However, major disasters are becoming increas-ingly prevalent throughout the world, and all organizations, not just thoseinvolving emergency services, need to be adequately prepared for theirunique impact on employee well-being and organizational processes (James,2011a). Accordingly, “improved understanding of how organizations canprepare for and respond to disaster and terrorism is clearly needed to enhancetheir and their employees’ safety and success” (James, 2011a, p. 933).Concerned employers may see value in identifying employees most vulner-able to the adverse effects of extraorganizational stressors, to ensure thatsupportive responses are appropriate and available for these individuals(Byron & Peterson, 2002). The finding that exposure to different aspects ofa natural disaster uniquely influenced outcomes has important practical

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

276 Biggs, Brough, and Barbour

Page 23

implications for the administration of organizational stress managementduring disasters. For example, individuals who experience personal damageand loss particularly require targeted, supportive responses in the short term,to minimize their adverse work-related outcomes.

Although employees involved in assisting in disaster-relief efforts mayinitially express positive perceptions of their work and well-being, theseemployees also require access to adequate job resources to deal with inten-sified work pressures, and long-term monitoring and access to supportservices to prevent latent effects from arising. Recognition of the extra effortsexerted by employees to meet intensified work demands is especially impor-tant, as effort invested without recognition is likely to transform into disen-gagement over a longer period of time (George, 2011).

Work culture support, supervisor support, and job control, were associ-ated with the outcomes to a greater extent than natural disaster exposure.According to COR theory, employees require resources they can invest toadequately meet demands, and their ability to invest resources “rests in largepart on the collective pool of resources available within that organizationalecology, and individuals’ and groups’ ability to access those resources”(Hobfoll, 2011, p. 118). As workplace resources become diminished duringdisasters, a focus on ensuring employees have access to adequate re-sources is especially important. The results of this study particularlyemphasize the importance of the provision of appropriate levels of workculture support. Therefore, it is recommended that the provision of asupportive organizational culture, in addition to targeted supportive or-ganizational responses to disaster, will most effectively minimize adverseoutcomes. Although conducted with a sample of law enforcement officers,the implications of the research are also applicable to a broader range oforganizational settings.

Research Limitations

In a recent special issue on the subject of organizational disastermanagement, James (2011a, p. 935) argued that “organizational scienceneeds to attend much more to disasters and organizations.” The practicalimplications of this research are especially beneficial in light of thisrecognized need. The ability to control for predisaster baseline values ofthe research variables is extremely rare, given that disasters are generallyunforeseen, and calls for more research controlling for pre- and postde-signs were successfully addressed in this study (Bacharach & Bamberger,2007; Byron & Peterson, 2002). Nonetheless, there are some importantlimitations of this study.

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

277Exposure to Extraorganizational Stressors

Page 24

First, although we controlled for baseline measures, the direct andindirect effects evaluated in this research were cross-sectional, and truecausation cannot be established. Second, the response rates for the twosurvey distributions were modest, although consistent with other organi-zational self-report surveys (e.g., Brough, 2004). Low response rates areparticularly problematic when respondents differ from the population onvariables of interest (Schalm & Kelloway, 2001). In this particularinstance, it was not possible to determine whether respondents wererepresentative of the general organizational population in terms of theirlevel of exposure to aspects of the natural disaster—the main variable ofinterest in this research. However, we found few differences with com-parisons on demographic characteristics and other research variables.Nonetheless, it is plausible that a nonresponse bias influenced the resultsif the most severely affected employees were unable to participate in thepostdisaster survey, or if this sample consisted of respondents mostimpacted by the disaster. Third, despite conceptualizing exposure toaspects of a natural disaster as a stressor, it is probable that involvementin the disaster relief effort was not a uniformly stressful experience for allrespondents. Stronger effects may have been attained if individual ap-praisals of the intensity of the stressful experience were also assessed.Finally, we acknowledge that obtaining longer-term postdisaster datawould be of value and may support or influence the results reported here.For instance, Morren et al. (2007) reported that the effects of disasterexposure upon sick-leave absences were not fully apparent until someyears postdisaster, demonstrating the latency of certain effects.

CONCLUSIONS

By employing a longitudinal design, we demonstrated that exposureto a natural disaster was associated with work-related outcomes indirectlyvia perceived work culture support, after controlling for baseline mea-sures of the outcomes. By considering varying forms of disaster exposure,this research also provided evidence that within-disaster factors uniquelyimpact outcomes and identified subgroups that would benefit from tar-geted supportive organizational responses. Although previous researchhas largely focused on adverse outcomes of disasters, the inclusion ofboth positive work-related outcomes (work engagement) and negativeoutcomes (turnover intentions) demonstrated that assisting individualswithin disaster affected communities may enhance one’s engagement withtheir job. Finally, the importance of job resources, such as job control andwork culture support, is especially apparent in the aftermath of a disaster.

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

278 Biggs, Brough, and Barbour

Page 25

REFERENCES

Alexander, D. A., & Wells, A. (1991). Reactions of police officers to body-handling after amajor disaster: A before-and-after comparison. The British Journal of Psychiatry, 159,547–555. doi:10.1192/bjp.159.4.547

Anderson, J. C., & Gerbing, D. W. (1988). Structural equation modeling in practice: A reviewand recommended two-step approach. Psychological Bulletin, 103, 411–423. doi:10.1037/0033-2909.103.3.411

Bacharach, S. B., & Bamberger, P. A. (2007). 9/11 and New York City Firefighters’ post hocunit support and control climates: A context theory of the consequences of involvementin traumatic work-related events. Academy of Management Journal, 50, 849–868. doi:10.5465/AMJ.2007.26279180

Beehr, T. A. (1998). An organizational psychology meta-model of occupational stress. In C. L.Cooper (Ed.), Theories of organizational stress (pp. 6–27). New York, NY: OxfordUniversity Press.

Beehr, T. A., Jex, S. M., Stacy, B. A., & Murray, M. A. (2000). Work stressors and coworkersupport as predictors of individual strain and job performance. Journal of OrganizationalBehavior, 21, 391– 405. doi:10.1002/(SICI)1099-1379(200006)21:4�391::AID-JOB153.0.CO;2-9

Bentler, P. M. (1990). Comparative fit indices in structural models. Psychological Bulletin, 107,238–246. doi:10.1037/0033-2909.107.2.238

Bentler, P. M., & Bonett, D. G. (1980). Significance tests and goodness of fit in the analysis ofcovariance structures. Psychological Bulletin, 88, 588–606. doi:10.1037/0033-2909.88.3.588

Biggs, A., Brough, P., & Barbour, J. P. (2014a). Enhancing work-related attitudes and workengagement: A quasi-experimental study of the impact of a leadership developmentintervention. International Journal of Stress Management, 21, 43–68. doi:10.1037/a0034508

Biggs, A., Brough, P., & Barbour, J. P. (2014b). Strategic alignment with organizationalpriorities and work engagement: A multi-wave analysis. Journal of OrganizationalBehavior, 35, 301–317. doi:10.1002/job.1866

Biggs, A., Brough, P., & Barbour, J. P. (in press). The longitudinal prediction of supportive jobresources and work engagement: Testing alternative models of causality. Work & Stress.

Brough, P. (2004). Comparing the influence of traumatic and organisational stressors upon thepsychological health of police, fire, and ambulance officers. International Journal ofStress Management, 11, 227–244. doi:10.1037/1072-5245.11.3.227

Brough, P. (2005). A comparative investigation of the predictors of work-related psychologicalwell-being within police, fire and ambulance workers. New Zealand Journal of Psychol-ogy, 34, 127–134.

Brough, P., & Biggs, A. (2010). Occupational stress in police and prison staff. In J. Brown &E. Campbell (Eds.), The Cambridge handbook of forensic psychology (pp. 707–717).Cambridge, UK: Cambridge University Press. doi:10.1017/CBO9780511730290.086

Brough, P., & Biggs, A. (2013). Job demands job control interaction effects: Do occupation-specific job demands increase their occurrence? Stress and Health. Advance onlinepublication. doi:10.1002/smi.2537

Brough, P., & Frame, R. (2004). Predicting police job satisfaction and turnover intentions: Therole of social support and police organisational variables. New Zealand Journal ofPsychology, 33, 8–16.

Brough, P., O’Driscoll, M., & Biggs, A. (2009). Parental leave and work-family balance amongemployed parents following childbirth: An exploratory investigation in Australia and NewZealand. Kotuitui: New Zealand Journal of Social Sciences Online, 4, 71–87. doi:10.1080/1177083X.2009.9522445

Brough, P., O’Driscoll, M., Kalliath, T., Cooper, C. L., & Poelmans, S. (2009). Workplacepsychological health: Current research and practice. Cheltenham, UK: Edward Elgar.doi:10.4337/9781781952672

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

279Exposure to Extraorganizational Stressors

Page 26

Brough, P., Timms, C., Siu, O. L., Kalliath, T., O’Driscoll, M., Sit, C., . . . Lu, C. Q. (2013).Longitudinal application of the job demands-resources model to psychological strain andengagement in cross-national samples. Human Relations, 66, 1311–1335. doi:10.1177/0018726712472915

Browne, M. W., & Cudeck, R. (1993). Alternative ways of assessing model fit. In K. A. Bollen& J. S. Long (Eds.), Testing structural equation models (pp. 136–162). Newbury Park,CA: Sage.

Byrne, B. M. (2001). Structural equation modeling with AMOS: Basic concepts, applications,and programming. Mahwah, NJ: Erlbaum.

Byrne, B. M. (2010). Structural equation modeling with AMOS: Basic concepts, applications,and programming (2nd ed.). New York, NY: Routledge.

Byron, K., & Peterson, S. (2002). The impact of a large-scale traumatic event on individual andorganizational outcomes: Exploring employee and company reactions to September 11,2001. Journal of Organizational Behavior, 23, 895–910. doi:10.1002/job.176

Caplan, R. D., Cobb, S., French, J. R. P., Jr., Van Harrison, R., & Pinneau, S. R., Jr. (1980).Job demands and worker health: Main effects and occupational differences. Ann Arbor,MI: Survey Research Center, Institute for Social Research, University of Michigan.

Demerouti, E., Bakker, A. B., Nachreiner, F., & Schaufeli, W. B. (2001). The job demands-resources model of burnout. Journal of Applied Psychology, 86, 499–512. doi:10.1037/0021-9010.86.3.499

Demerouti, E., Shimazu, A., Bakker, A. B., Shimada, K., & Kawakami, N. (2013). Work-selfbalance: A longitudinal study on the effects of job demands and resources on personalfunctioning in Japanese working parents. Work & Stress, 27, 223–243. doi:10.1080/02678373.2013.812353

Freedy, J. R., Shaw, D. L., Jarrell, M. P., & Masters, C. R. (1992). Towards an understandingof the psychological impact of natural disasters: An application of the conservation ofresources model. Journal of Traumatic Stress, 5, 441–454. doi:10.1002/jts.2490050308

George, J. M. (2011). The wider context, costs, and benefits of work engagement. EuropeanJournal of Work and Organizational Psychology, 20, 53–59. doi:10.1080/1359432X.2010.509924

Goldberg, D. P. (1972). The detection of psychiatric illness by questionnaire. Oxford, UK:Oxford University Press.

Gracia, N. J. (2007). Police leadership, culture and well-being: Implementing positive work-place change. Unpublished honours thesis, Griffith University, Brisbane, Australia.

Hayes, A. F. (2009). Beyond Baron and Kenny: Statistical mediation analysis in the new millen-nium. Communication Monographs, 76, 408–420. doi:10.1080/03637750903310360

Hobfoll, S. E. (1989). Conservation of resources. A new attempt at conceptualising stress.American Psychologist, 44, 513–524. doi:10.1037/0003-066X.44.3.513

Hobfoll, S. E. (2001). The influence of culture, community, and the nested-self in the stressprocess: Advancing conservation of resources theory. Applied Psychology: An Interna-tional Review, 50, 337–421. doi:10.1111/1464-0597.00062

Hobfoll, S. E. (2011). Conservation of resources caravans and engaged settings. Journal ofOccupational and Organizational Psychology, 84, 116–122. doi:10.1111/j.2044-8325.2010.02016.x

Hochwarter, W. A., Laird, M. D., & Brouer, R. L. (2008). Board up the windows: Theinteractive effects of hurricane-induced job stress and perceived resources on workoutcomes. Journal of Management, 34, 263–289. doi:10.1177/0149206307309264

Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis:Conventional criteria versus new alternatives. Structural Equation Modeling, 6, 1–55.doi:10.1080/10705519909540118

James, K. (2011a). Introduction to the special issue: Terrorism, disaster, and organizationalscience. Journal of Organizational Behavior, 32, 933–937. doi:10.1002/job.758

James, K. (2011b). The organizational science of disaster/terrorism prevention and response:Theory-building toward the future of the field. Journal of Organizational Behavior, 32,1013–1032. doi:10.1002/job.782

Kastenmüller, A., Greitemeyer, T., Aydin, N., Tattersall, A. J., Peus, C., Bussmann, P., . . .Fischer, P. (2011). Terrorism threat and networking: Evidence that terrorism salience

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

280 Biggs, Brough, and Barbour

Page 27

decreases occupational networking. Journal of Organizational Behavior, 32, 961–977.doi:10.1002/job.745

Kleber, R. J., & van der Velden, P. G. (2009). Acute stress at work. In C. L. Cooper, J. C.Quick, & M. J. Schabracq (Eds.), International handbook of work and health psychology(3rd ed., pp. 269–291). Oxford, UK: Wiley-Blackwell.

Kline, R. B. (2011). Principles and practice of structural equation modeling (3rd ed.). NewYork, NY: Guilford Press.

Lepine, J. A., Podsakoff, N. P., & Lepine, M. A. (2005). A meta-analytic test of the challengestressor–hindrance stressor framework: An explanation for inconsistent relationshipsamong stressors and performance. Academy of Management Journal, 48, 764–775.doi:10.5465/AMJ.2005.18803921

Little, T. D., Cunningham, W. A., Shahar, G., & Widaman, K. F. (2002). To parcel or not toparcel: Exploring the question, weighing the merits. Structural Equation Modeling, 9,151–173. doi:10.1207/S15328007SEM0902_1

Little, T. D., Rhemtulla, M., Gibson, K., & Schoemann, A. M. (2013). Why the items versusparcels controversy needn’t be one. Psychological Methods, 18, 285–300. doi:10.1037/a0033266

MacKinnon, D. P., Warsi, G., & Dwyer, J. H. (1995). A simulation study of mediated effectmeasures. Multivariate Behavioral Research, 30, 41– 62. doi:10.1207/s15327906mbr3001_3

May, D. R., Gilson, R. L., & Harter, L. M. (2004). The psychological conditions of meaning-fulness, safety and availability and the engagement of the human spirit at work. Journalof Occupational and Organizational Psychology, 77, 11–37. doi:10.1348/096317904322915892

Morren, M., Dirkzwager, A. J. E., Kessels, F. J. M., & Yzermans, C. J. (2007). The influenceof a disaster on the health of rescue workers: A longitudinal study. Canadian MedicalAssociation Journal, 176, 1279–1283. doi:10.1503/cmaj.060626

O’Driscoll, M., & Brough, P. (2010). Work organization and health. In S. Leka & J. Houdmont(Eds.), Occupational health psychology (pp. 57–87). Chichester, UK: Blackwell.

Pitts, S. C., West, S. G., & Tein, J. (1996). Longitudinal measurement models in evaluationresearch: Examining stability and change. Evaluation and Program Planning, 19, 333–350. doi:10.1016/S0149-7189(96)00027-4

Queensland Floods Commission of Inquiry. (2012). Final report. Retrieved from http://www.floodcommission.qld.gov.au/__data/assets/pdf_file/0007/11698/QFCI-Final-Report-March-2012.pdf

Schalm, R. L., & Kelloway, E. K. (2001). The relationship between response rate and effect sizein occupational health psychology research. Journal of Occupational Health Psychology,6, 160–163. doi:10.1037/1076-8998.6.2.160

Schaufeli, W. B., Bakker, A. B., & Salanova, M. (2006). The measurement of work engagementwith a short questionnaire: A cross-national study. Educational and Psychological Mea-surement, 66, 701–716. doi:10.1177/0013164405282471

Seibert, S. E., Wang, G., & Courtright, S. H. (2011). Antecedents and consequences ofpsychological and team empowerment in organizations: A meta-analytic review. Journalof Applied Psychology, 96, 981–1003. doi:10.1037/a0022676

Steiger, J. H. (1990). Structural model evaluation and modification: An interval estimation approach.Multivariate Behavioral Research, 25, 173–180. doi:10.1207/s15327906mbr2502_4

Stephens, C., Long, N., & Miller, I. (1997). The impact of trauma and social support onposttraumatic stress disorder: A study of New Zealand police officers. Journal of CriminalJustice, 25, 303–314. doi:10.1016/S0047-2352(97)00015-9

Waldman, D. A., Carmeli, A., & Halevi, M. Y. (2011). Beyond the red tape: How victims ofterrorism perceive and react to organizational responses to their suffering. Journal ofOrganizational Behavior, 32, 938–954. doi:10.1002/job.710

Wall, T. D., Jackson, P. R., & Mullarkey, S. (1995). Further evidence on some new measuresof job control, cognitive demand and production responsibility. Journal of OrganizationalBehavior, 16, 431–455. doi:10.1002/job.4030160505

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

281Exposure to Extraorganizational Stressors

Page 28

Warr, P., Cook, J., & Wall, T. (1979). Scales for the measurement of some work attitudes andaspects of psychological well-being. Journal of Occupational Psychology, 52, 129–148.doi:10.1111/j.2044-8325.1979.tb00448.x

Received September 21, 2012Revision received April 17, 2014

Accepted April 24, 2014 �

Members of Underrepresented Groups:Reviewers for Journal Manuscripts Wanted

If you are interested in reviewing manuscripts for APA journals, the APAPublications and Communications Board would like to invite your participation.Manuscript reviewers are vital to the publications process. As a reviewer, youwould gain valuable experience in publishing. The P&C Board is particularlyinterested in encouraging members of underrepresented groups to participatemore in this process.

If you are interested in reviewing manuscripts, please write APA Journals [email protected]. Please note the following important points:

• To be selected as a reviewer, you must have published articles in peer-reviewed journals. The experience of publishing provides a reviewer with thebasis for preparing a thorough, objective review.

• To be selected, it is critical to be a regular reader of the five to six empiricaljournals that are most central to the area or journal for which you would liketo review. Current knowledge of recently published research provides areviewer with the knowledge base to evaluate a new submission within thecontext of existing research.

• To select the appropriate reviewers for each manuscript, the editor needsdetailed information. Please include with your letter your vita. In the letter,please identify which APA journal(s) you are interested in, and describe yourarea of expertise. Be as specific as possible. For example, “social psychol-ogy” is not sufficient—you would need to specify “social cognition” or“attitude change” as well.

• Reviewing a manuscript takes time (1–4 hours per manuscript reviewed). Ifyou are selected to review a manuscript, be prepared to invest the necessarytime to evaluate the manuscript thoroughly.

APA now has an online video course that provides guidance in reviewingmanuscripts. To learn more about the course and to access the video, visithttp://www.apa.org/pubs/authors/review-manuscript-ce-video.aspx.

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

282 Biggs, Brough, and Barbour