The Relationship Between Stress, Strain and Social Capital

The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1363-951X.htm
Stress, strain and

The relationship between stress, social capital
strain and social capital
Martin Gächter
University of Innsbruck, Innsbruck, Austria, and 515
Davd A. Savage and Benno Torgler
Queensland University of Technology, Brisbane, Australia Received 22 April 2010
Revised 30 November 2010
Accepted 31 May 2011
Abstract
Purpose – The purpose of this study is to examine the relationship of social capital with the negative
externalities associated with stress, or the psychological and physiological strains experienced by
police officers.
Design/methodology/approach – Using data collected in 1999 from a survey of Baltimore Police
officers designed to examine questions about the relationship between police stress and domestic
violence in police families and using multivariate regression analysis, the paper focuses on five
different proxies for stress and strain, and two proxies for social capital and conducting several
robustness checks.
Findings – Results show that an increase in social capital is significantly correlated to a decrease in
the level of strain, in the psychological, physical, burnout and health areas.
Research limitations/implications – While this study examines the social capital/strain
relationship with US officers, more research is needed, as these findings may not extrapolate well
into other national settings. It may also be interesting to further explore sub-cultures within
departments. Additionally, the data may be dated and, as major changes and events have occurred
since the survey, a newer study of officers would be needed to observe whether these changes have
had significant impact.
Practical implications – From a policy perspective, the findings suggest that stress reduction
programs should actively engage employees to build stronger social networks.
Originality/value – This study comprehensively examines the ability of social capital at negating
the impacts of strains, and significantly reduces the impact of major trauma events. This paper adds to
the literature as there are few multivariate analyses of the social capital/strain relationship.
Keywords Social capital, Trust, Stress, Gender, Burnout, Police, United States of America
Paper type Research paper
Introduction
The two concepts of social capital and stress have received much attention across a
diverse range of research fields from anthropology, economics, and sociology to
psychology and the medical sciences. However, very little work has been done to link
these two seemingly disparate concepts together, and to show how the relationship
between the two may work. Social capital has been the catch phrase for a large volume
of research, investigating the enigmatic links between the individual and their access
to resources via personal networks. The concept has resisted definitive definition, even Policing: An International Journal of
though many prominent researchers across several fields have attempted to do so (see, Police Strategies & Management
Vol. 34 No. 3, 2011
e.g. Bourdieu, 1985; Coleman, 1988; Fukuyama, 2003; Grootaert, 2001; Knack and pp. 515-540
Keefer, 1997; Loury, 1977; Manski, 2000; Portes, 1998; Putnam, 1993). Bourdieu (1985, q Emerald Group Publishing Limited
1363-951X
p. 248) defined social capital as: DOI 10.1108/13639511111157546
PIJPSM [. . .] the aggregate of the actual or potential resources which are linked to possession of a
durable network of more or less institutionalized relationships of mutual acquaintance or
34,3 recognition.
A decade and much research later, Portes (1998, p. 6) declared that the consensus
position is simply that:
516 [. . .] social capital stands for the ability of actors to secure benefits by virtue of membership in
social networks or other social structures.
The benefits of social capital that are best understood and most commonly cited
usually include only the physical or economic type resources, such as access to capital
or individuals with influence, but the less-tangible benefits such as social support may
be of equal import.
Like social capital, stress and its effects has become so inexorably linked to that of
modern society that the topic is also researched and discussed across a diverse range of
fields. The economic costs or the negative externalities generated by stress are
considerable, as it not only includes the cost of administering mental and physical
support for sufferers but also the lost work hours. International Labor Organization
(ILO) estimates that in the USA one in ten workers are diagnosed with depression, with
a treatment cost of between US$30-$44 billion, resulting in approximately 200 million
lost working days each year (Gabriel and Liimatainen, 2000). Public service workers,
like police officers or nurses, are recognized as suffering from high levels of stress
through performing work that is both physically and emotionally draining (Kopel and
Friedman, 1999; Schwartz and Schwartz, 1981; Stotland, 1991). Numerous research
studies have demonstrated that the high levels of stress in these professions can lead to
detrimental health consequences, or strains (see, e.g. Gächter et al., 2010). These
consequences can include mental and physical illnesses; aggressive or violent
behavior; alcohol abuse and decreased work performance (McCarty et al., 2007; Morash
and Haarr, 1995; Swatt et al., 2007).
Although literature has explored a large set of factors that may determine stress, the
examination of whether social capital has an effect on stress or strains is still an
underdeveloped topic. Specifically, we want to investigate the impact of the non-tangible
aspects of social capital on stress and its manifestations of strain. We therefore work
with an extensive survey of police officers from the Baltimore Police Department in
Maryland, USA (Gershon, 1999, 2000). The extensive survey covered many job related
factors (both positive and negative), as well as personal, organizational and social
questions and is representative of the demographic characteristics of that department.
Within this work we utilize the definitions of stress and strain as set out in O’Driscoll and
Dewe (2001) such that: stressors are characteristics of the work environment that cause
strain, and strains are the labels for the resulting physical or psychological impacts such
as burnout or ill-health. Thus, individuals who are exposed to high levels of stress are
likely to exhibit or manifest detrimental health issues as strains (which can include
burnout, somatization, etc.). We contend that greater levels of social capital alleviate
work related strains, such that as levels of social capital increase so too does access to
social support, which in turn can reduce many of the negative aspects of stress. If, as
indicated by Coleman (1988, p. 304), social capital works by “facilitating the achievement
of goals that could not be achieved in its absence or could be achieved only at higher
cost” and the resource which is being facilitated is social support and the informational
exchange of coping strategies, then higher levels of social capital should result in lower Stress, strain and
stress levels and measurably lower strain outcomes. This idea is well supported in the social capital
literature such that “the short and long term effects of stress depend in large part upon
the individual’s available coping strategies and their effective use” (Loo, 1984, p. 10).
Searching for improvements for law enforcers can generate large societal benefits.
Utilizing the benefits of social capital may provide a good alternative or additional
program in situations where traditional stress reducing instruments fail or where the 517
necessary information to design and enforce suitable instruments and directives cannot
be effectively used.
Theoretical background
A generalized definition of stress covers conditions of a physical, biological or
psychological nature that strain an organism beyond its power to adapt (Cannon,
1929a, b, 1935; Lovallo, 2005; Selye, 1936). Stress has been explained as a physical
process with an almost mechanical or automatic response from the human body[1].
Cannon suggested such an automatic response existed for psychosocial threats
(Cannon, 1929a, b, 1935). Selye’s (1936) General Adaption Syndrome (GAS) expanded
on this concept, stating that the human body has an innate drive to maintain a
biological steady state or equilibrium known as homeostasis. Stress, infections and
excessive work demands disrupt this equilibrium and trigger a natural response from
the body, returning it to homeostasis. Everly and Lating (2002, p. 40) describe the GAS
as a tri-phase phenomenon. The first phase refers to the alarm phase, representing a
generalized somatic shock. The second phase is the stage of resistance, in which there
is a dramatic reduction in most alarm stage processes and the body fights to
re-establish and maintain homeostasis. The first two stages are constantly repeated
throughout an individual’s life; however, should the body remain in the second stage
for protracted periods of time, then the body enters stage three, the exhaustion stage.
Prolonged exposure to this stage manifests itself as disease and dysfunction of bodily
organs and – when applied to the entire body – may cause death. Research into the
third stage has demonstrated the links between stress and physiological and
psychological disease/illness (see, e.g. Everly and Benson, 1989; Lachman, 1972;
Sternbach, 1966; Schwartz, 1977, 1979). These models indicate that it is the inability to
adapt to these stresses and/or extended durations of stress that creates ill-effect, and
the greater the impact stress has on an individual the greater the probability of harm.
Police officers are routinely exposed to a vast array of stresses that are dissimilar to
most common work environments, but resemble those experienced by emergency service
workers (ambulance, E/R and fire) as well as those in the armed forces (see, e.g. Liberman
et al., 2002). However, some of the more common stresses are driven by the
organizational structure, social interactions, and job requirements observable in many
working environments. These common stresses can include shift work, excessive
overtime, heavy workload, discrimination and harassment, poor working conditions, and
strong interactions with the public. The most commonly reported stresses amongst
police officers around the world are the everyday operational stresses (see, e.g. Brooks
and Piquero, 1998; Brown and Campbell, 1990; Crank and Caldero, 1991; Kroes et al.,
1974; Liberman et al., 2002). However, we contend that these stresses are ubiquitous and
as such are experienced by all members of the police force to form the environmental
conditions within which they work. From a theoretical and empirical perspective, it is
PIJPSM important to analyze data where individuals have a similar job profile, as many of the
34,3 potential stress factors are common across a large group of individuals. Remaining
differences within the homogenous environment can then be controlled as good as
possible in a multivariate analysis. For example, it has been proposed that female
officers may face additional stress factors than those experienced by their male
colleagues (He et al., 2005; Morash and Haarr, 1995; O’Farrell, 1980; Stotland, 1991) and
518 that gender varies the way in which work stress is experienced. First, men and women
can experience different stressors or varying levels of the same stressor from the same
single event. Second, the reaction of men and women to the event can vary greatly so that
each gender experiences a different level of stress from the event. Finally, the
stressor-strain relationship observed for males and females may differ or gender may act
as a moderator such that it affects one but not the other (Desmarais and Alksnis, 2005).
Many theories have been proposed to explain the variations in stress effects between
males and females, some of which include: gender socialization and gender roles,
tokenism, family status and children, workloads, social support, or self-reported
variations (see Gächter et al., 2011). He et al. (2005) investigated the gender and racial
variations of stress within the police force, claiming three initial findings:
(1) female officers have higher levels of stress than males;
(2) white males have higher stress levels than black males; and
(3) there are no significant stress differences between white and black female
officers.
Similarly, we also control for ethnic groups to deal with the issue that multiple
subcultures exist within a department.
Thus, the advantage of focusing on a particular profession such as police officers
within a regional department is the chance of improving the ceteris paribus
assumption, holding important potential factors constant[2]. The remaining stresses
can be classified by the frequency by which they occur as well as the intensity of the
impact on the officer (Brown et al., 1999) and it is these high impact traumatic
exposures inherent in police work that are of interest. Police officers can encounter,
witness or hear about fellow officers’ involvement in extreme situations such as
physical or even life threatening danger and the exposure to disturbing events in
general (Gershon et al., 2009). It is the involvement in events such as shootings,
attachment to victims, or the attendance of a gruesome crime scene that have been
identified as triggers for mental disorders such as Post Traumatic Stress Disorder
(Stephens et al., 1997).
The current Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition,
Text Revision (DSM-IV-TR) from the American Psychiatric Association (APA, 2000)
states that an individual has been exposed to a traumatic event in which both of the
following were present:
.
The person experienced actual or threatened death or serious injury, or other
threat to one’s physical integrity; or witnessing an event that involves death,
injury, or a threat to the physical integrity of another person; or learning about
unexpected or violent death, serious harm, or threat of death or injury
experienced by a family member or other close associate (Criterion A1).
. The person’s response involved intense fear, helplessness, or horror (Criterion A2).
Traditional attempts to measure stress or traumatic events, have utilized self-reported Stress, strain and
surveys such as: the Impact of Events Scale (IES) or the revised version (IES-R), social capital
encompass a 15 (22) question set with a ranking scale (see Horowitz et al., 1979; Weiss
and Marmar, 1997). Across a number of the most common standardized measures of
trauma, Criterion A1 is the sole or primary focus[3] and most scales do not assess
Criterion A2 at all (Norris and Hamblen, 2004). Thus, the use of high impact traumatic
events utilized in this paper are in line with the measures of traumatic events generally 519
accepted in research and as defined by the APA criteria. We are not creating a new
assessment scale but are utilizing the trauma questions and scales provided by officers
in the survey, to illustrate the relational effects that social capital has on physiological
and psychological stress indicators or health.
As seen from Table I, stressful incidents can have a broad range of impact
responses across the different events. Attending a police funeral and being the subject
of an IID investigation clearly rate as being some of the highest impact events. On the
other hand, it is surprising to observe that chemical spills, violent crime scenes and
hostage events are viewed as having little or no impact. These events can affect officers
in several ways, either physically, psychologically or both (Gershon, 2000). Some of the
noted physical strains associated with police stress include: hypertension, stroke,
ulcers, high blood pressure, or sexual dysfunction (Bartollas and Hahn, 1999; Berkman
and Syme, 1979; Kroes, 1985; Mitchell and Bray, 1990; Peak, 1993, Stratton, 1984,
Violanti et al., 1983). The psychological strains associated with police stress can
include: depression, PTSD, burnout, suicide and alcoholism (Hawkins, 2001; Kawachi
et al., 1996; Kopel and Friedman, 1999; Schaufeli and Enzmann, 1998; Schwartz and
Schwartz, 1981; Stephens et al., 1997). Thus, officers who suffer from high levels of
stress or trauma, should exhibit higher incidents of physiological and psychological
strain, allowing for demographic variances (see, e.g. Gächter et al., 2010). However, this
simple relationship does not allow for the mitigating effects of social capital on
reducing stress and promoting positive health outcomes.
It has been argued that a person may be more or less insulated against the effects of
stressors depending on whether the individual possesses social supports. It has also
been suggested that low levels of social capital exacerbate these problems, as lack of
social capital can indicate a predisposition for depression (Brown and Harris, 1978;
Caplan, 1974). More recent studies have shown that social capital in the form of social
support buffers individuals against both chronic and acute forms of stress (Cohen and
Not at all A little Very much N/A

Frequency and impact of incident (1) (2) (3) (0) Totals
Making a violent arrest 220 554 210 105 1,089

Shooting someone 98 92 90 813 1,093
Subject of IID investigation 86 258 370 378 1,092
Respond to chemical spill 267 197 49 581 1,094
Respond to bloody crime scene 378 445 170 100 1,093
Personally know victim 121 323 180 469 1,093
Involved in hostage situation 249 311 86 447 1,093
Attending police funeral 41 262 602 186 1,091 Table I.
Experience needle stick injury 76 198 325 494 1,093 Descriptive statistics
PIJPSM Willis, 1985; Prince et al., 1997; Whitley and McKenzie, 2005). However, it has also been
34,3 reported that gender may play a role in the effectiveness of social support, such that
females are reported to have a higher use of social support strategies then males
(Biggam et al., 1997). This is related to the literature on social environment that states
that supportive, non-conflictual social relations at work are able to reduce stress, strain
outcomes and enhance health, meeting basic human needs such as approval, affiliation,
520 and a sense of belonging (Repetti, 1993). More specifically, this would mean that
individuals who confront difficult or traumatic job situations by themselves will
experience higher levels of stress or other damaging consequences than those who are
able to turn to friends or colleagues for advice, comfort, or material aid (Cullen et al.,
1985). Social support in this context is the end product of an individual’s social capital,
i.e. social support is the result of expenditure of social capital and the higher the level of
an individual’s social capital the greater the available social support. Conversely, an
individual who has a low level of social capital will be unable to access assistance or
social support from colleagues. One of the causes of low social capital may come from
being an outsider, whether it is from gender, race or simply the individual’s inability to
fit into with the behavioral expectations of the group.
The social support being offered through the social network of officers consists of
two parts: first, as informational exchange of coping strategies; and second, as support
from fellow officers (e.g. an understanding shoulder to lean on). Coping strategies are
methods of reducing or manage stress, being able to move an individual from the
higher phases of the GAS to the lower ones becomes important. Coping has been
defined as cognitive and behavioral efforts to manage environmental or internal
demands, and the conflicts between them, which tax or exceed a person’s resources
(see, e.g. Cohen and Lazarus, 1979; Lazarus and Folkman, 1984). Put more simply,
coping creates ways for individuals to deal with or reduce the effects of stress. Coping
strategies can be either adaptive or maladaptive (Girdano et al., 2001) such that
adaptive strategies reduce stress whilst promoting long-term health (see Figure 1),
whereas maladaptive strategies (alcohol/drug abuse, smoking, etc.) may reduce
short-term stress effects but have negative long-term health outcomes (Everly, 1979).
Furthermore, maladaptive coping strategies have been shown to result in job burnout
and quitting behaviors (Burke and Deszca, 1986; Malloy and Mays, 1984). When coping
strategies are successful, the detrimental effects of the third stage (exhaustion) of the
GAS model are reduced or eliminated and homeostasis is re-established, but if
unsuccessful then the chances of “disease and dysfunction” are increased (Everly and
Lating, 2002).
Research has identified several of the coping mechanisms utilized by police officers
to alleviate stress as being either adaptive or maladaptive. These can result in either
positive or negative outcomes such as: social and spiritual support systems, alcohol
and substance abuse, and violence (Gershon et al., 2009; Haarr and Morash, 1999; Swatt
Figure 1.
et al., 2007). Waters and Ussery (2007) list of proposed individual coping strategies Stress, strain and
included: development of a dependable support system; improved communication social capital
skills and means of venting feelings appropriately, as well as lifestyle factors like diet,
exercise and vacations. It is obvious from these strategies that a high level of social
capital would be needed for them to be successful, as support, venting and
communication would be all but impossible without it. According to Dasgupta (1999),
social capital can lead to more efficient transactions by giving agents access to more 521
information, enabling them to coordinate activities for mutual benefit, and, through
frequent transactions with the same person, reducing therefore the likelihood of
opportunistic behavior.
Social capital can be viewed as being akin to financial capital, as it can be generated
and held in account for the future and used to purchase social objects such as support
or assistance. Social capital can be generated through a shared sense of identity (or
sense of belonging/solidarity) reciprocity and norms of cooperation that induces a
sense of obligation to help others, along with a confidence that such assistance will be
returned (Putnam, 1993). It is through the generation of trust and cooperation that
individuals are able to engage in informational exchange, which include coping
strategies or load sharing with colleagues. Game theory and experimental findings
have shown that a high level of social capital enables co-operation between actors and
facilitates superior social outcomes (Boix and Posner, 1998).
Coleman (1988) stated that for any systematic treatment of social capital three
things must be clearly distinguishable:
(1) the possessor of social capital or those making claims;
(2) the sources of social capital, or those agreeing to these demands; and
(3) the resource being demanded/supplied.
In this case the stressed officers are seeking social support or coping mechanisms (the
resource) from other experienced officers. However, access to the resources granted
through social capital can be stymied or limited, not only through the amount of social
capital available to an individual, but also due to other social or demographic factors
involved in the generation of social capital. These factors can include: gender, race,
education, age or rank but layered on top of these restrictions are the social norms and
behavioral requirements of the police force. As such, it is important to use these
demographic variables as controls as to get at the true nature of the relationship
between stress and social capital. Portes (1998) concluded that “at the individual level
[. . .] social ties can bring about greater control over wayward behavior and provide
privileged access to resources; they can also restrict individual freedoms and bar
outsiders from gaining access to the resources through particular preferences.” Thus,
two of the negative consequences of social capital include the exclusion of outsiders
(see, e.g. Waldinger, 1995) and the restriction of individual freedoms through control of
behavior (see, e.g. Boissevian, 1974; Simmel, 1964). Table II shows a predominately
white male working environment within the police force, where outsiders may include
other subcultures such as: non-whites, women, or those not meeting behavioral
expectation of fellow officers.
How can we measure social capital? We are interested not only in an analytical
concept but also in an empirical one. Grootaert (2001, pp. 10-11) points out that there
are three major views on social capital: First, Putnam (1993) interprets social capital as
PIJPSM
Variable Count Percent n Mean s2 Min Max
34,3
Gender
Male 943 85.73 1,100
Female 157 14.27
Ethnic group
522 African-American 355 32.51 1,092
Caucasian 696 63.74
Hispanic 14 1.28
Other 27 2.47
Current rank
Officer trainee 91 8.27 1,100
Officer 601 54.64
Agent 62 5.64
Detective 144 13.09
Sergeant 143 13.00
Lieutenant or above 59 5.36
Marital status
Married 658 59.87 1,099
Live-in partner 88 8.01
Divorced/separated 135 12.28
Single 213 19.38
Widowed 5 0.45
Age 1,081 36.04 9.09 20 66
Experience 1,078 11.52 9.28 0 44
Children 1,090 1.18 1.16 0 7
Psychological strain 1,064 10.57 3.02 7 28
Physical strain 1,086 6.61 1.84 5 20
Overall strain 1,060 17.18 4.36 12 48
Burnout 1,092 7.91 2.56 3 15
Health 1,104 1.18 1.35 0 9
Social capital 1,075 7.19 1.60 2 10
Table II. Home index 1,078 6.60 1.41 2 9
Descriptive statistics Trauma index 1,077 11.98 5.79 0 27
a social network, as networks of civic engagement facilitating coordination and

cooperation. Second, Coleman (1988, p. 598) defined social capital as: “a variety of
different entities” that consists of social structure aspects, that also facilitate certain
actions, taking into account not only horizontal (co-worker) but also vertical social
relationships (officer rankings). The third concept considers the social and political
environment that enforces norms and shapes social structures. In this case, we have the
ability to hold such environmental factors constant as we observe police officers within
the same working environment. Social capital is therefore used to describe aspects of
social networks, relationships and trust (Coleman, 1988; Fukuyama, 2003; Portes, 1998;
Woolcock and Narayan, 2000). Putnam’s (1983) five principles include: a local/civic
identity, a sense of belonging, solidarity, and/or equality with other members of the
community, and reciprocity and norms of cooperation inducing a sense of obligation to
help others, along with a confidence that such assistance will be returned (Putnam,
1993). Similarly, Paldam (2000, p. 630), describes three families of social capital
concepts: trust (cognitive social capital), cooperation (collective action) and networks.
He points out that these conceptual families come together because “most people build
trust in and networks to others and come to cooperate with them” (Paldam, 2000, p. Stress, strain and
629). Paldam’s view is in line with our rationale for working with the following two
proxies for social capital, namely whether “there is a good and effective cooperation
social capital
between units” and whether one “can trust his/her work partner”. The trust variable
that we use can be classified according to Uslaner (2002) as particularized (or personal)
trust, a proxy that relies strongly upon experiences. Particularized trust is only related
to a specific group such as co-workers, family members, or to specific institutions. 523
Trust is then often connected with the element of reciprocity or interactions depending
upon specific individual or group characteristics. This notion is essential for our
analysis as we are exploring the work environment and its implication on individuals’
stress level. Good effective managerial behavior is crucial to the formation of social
capital in a workplace, such that a well-organized workplace fosters an environment of
trust between all members of staff (Hodson, 2005). Thus, one could stress that social
capital within any workplace is important but the special nature of police work, similar
to the military, makes trust, reciprocity and cooperation between colleagues even more
vital (Torgler, 2003). Trust leads to the idea of “having one’s back” such that support
and assistance will be freely given even in the most dangerous situations by your
fellows. However, this concept is not a given in either policing or the military, but
through the creation of social capital individuals “know” that others will have their
back as they have built up high levels of trust. Furthermore, individuals with lower
levels of social capital may not experience the same level of trust in colleagues and as
such would be concerned about others having their back in a dangerous situation.
Research has shown that higher levels of social support decrease strain outcomes for
police officers (Morash et al., 2006, Morris et al., 1999). However, many previous studies
have only utilized a narrow selection of environmental and demographic factors and
utilized a single stress variable within the analysis. In this paper we have therefore
examined several aspects of stress and strain outcomes as well as an extended set of
control and environmental factors in our multivariate analysis.
Data and methods

The data for our analysis are taken from the Study to Help Identify, Evaluate and Limit
Department Stress (SHIELDS), which was conducted by Gershon (1999, 2000). Study
participants were recruited from the Baltimore Police Department in Baltimore which
provides law enforcement services to about 700,000 inhabitants in Maryland. The
study aimed to examine questions about the relationship between police stress and
domestic violence in police families. The sampling strategy (Gershon, 1999, p. 34)
involved:
.
obtaining the number of sworn employees at each precinct at each shift;
.
attending one or two roll calls for each shift at each precinct to obtain a
convenience sample of volunteers; and
.
distributing questionnaires to all officers volunteering to complete the
questionnaire.
The response rate[4] in the original study was very high, amounting to 68 percent (see
Gershon, 1999 for more details). From approximately 1,200 questionnaires distributed
1,104 were returned (more than 92 percent). The five-page questionnaire covers
questions in four main areas:
PIJPSM (1) symptoms of stress and likely stressors;
34,3 (2) perceived (current) stress;
(3) coping strategies; and
(4) health outcomes.
Thus, the very high response rate, the excellent sampling strategies and the
524 anonymous nature of the study makes it very interesting to analyze such a dataset.
Moreover, the sample closely resembles the demographic characteristics of the police
department in 1996. At that time, the department had 3,061 sworn employees,
including 2,636 males (86 percent) and 425 females (14 percent). Thus, the sample
covers roughly a third of the whole study population.
Table II presents an overview of the data set. Almost 86 percent of the employees
are male. Regarding the ethnic group, a majority is Caucasian (64 percent), followed by
African-American (33 percent) and Hispanic (1 percent). Approximately 26 percent
attended college, while just about 4 percent hold a graduate degree. The predominant
rank was officer (55 percent), followed by detective and sergeant (13 percent each). The
majority of employees were either married or had a live-in partner (68 percent), while
19 percent declared themselves as singles. The mean age was 36 years, ranging from
20 to 66. On average, people have been working in the department for 11.5 years
(lasting from 0 to 44) and have 1.18 children living at home (varying between 0 and 7).
For the purpose of this study, we have constructed several indices to measure
different aspects of stress and strain. Moreover, to better isolate the impact of social
capital on stress we control for factors such as demographic characteristics (age,
gender, ethnic group, number of children, marital status), as well as experience and
rank within the department. There is a large body of work that has suggested that
gender and ethnicity have either a confounding or moderating effect on stress. In
Gächter et al. (2011), this premise is tested and shown to have validity, for this purpose
we utilize both variables as controls for a clearer picture. Additionally, it is naı̈ve to
assume that home life has no impact upon work stresses, how an individual perceives
the job is not completely isolated from life outside work and their life-satisfaction. It
has been well documented that social capital is not isolated to the work environs,
officers who are married or have domestic partners are be able utilize this additional
capital and reduce work stresses. Therefore, we have added control factors for
marriage and children, as well as a measure of social capital in the domestic
environment. Many previous studies report that rank and experience is relevant (for an
overview see Brown and Campbell, 1990) and in the reverse direction age. It is
generally believed that as individuals age they become better able to mitigate stress
effects (Lennings, 1997) through either acclimatization or a quasi-natural for of stress
inoculation training (see Meichenbaum, 2007). However, age is correlated with both
rank and experience and thus we control for all three to separate out the effects of each.
To check robustness of the results we also conduct a sensitivity analysis, extending
the baseline specification with a trauma index. This index measures whether police
officers have experienced certain potentially dangerous or traumatic events in the line
of duty and how much they were emotionally affected by them. Additionally, we
include an index that measures police officers’ stability and support in the home. For
simplicity and comparability we will use the same independent variables for all the five
strain indices used as dependent variables (see Table II for descriptive statistics).
Dependent variables Stress, strain and
To measure different types, aspects and outcomes of stress, as well as being able to social capital
distinguish between certain effects and their specific influences on strain we construct
five different indices of stress. Using a large set of dependent variables also offers a
good robustness test for the relationship between social capital and stress. Following
Kurtz (2008, p. 224), we develop indices of psychological and physical strain as well as
an index that combines these two factors. In the first index (referred to as psychological 525
strain), participants were asked if they experienced the following signs of
psychological strain in the past 6 months: restlessness, feeling hopeless, panic
attacks, irritability, withdrawal, depression, and emotional depletion. A four-point
Likert scale (Likert, 1932) with possible answers ranging from never (1) to always (4)
was used for each question. These items were then used to create a summative scale
that ranged from 7 to 28, with higher levels indicating a higher level of (psychological)
strain. The measure showed a satisfactory level of internal consistency (Cronbach’s
a ¼ 0:83). The physical strain index (physical strain) uses five questions assessing
whether respondents had experienced nausea, trouble getting breath, a lump in the
throat, pains or pounding in the chest, and faintness or dizziness in the six months
prior to the survey. As the construction of the index is similar as explained above, the
summative scale ranged from 5 to 20, with higher levels indicating a higher level of
(physical) strain (a ¼ 0:72). Our third strain indicator (overall strain) combines the
psychological and physical components and, therefore, gives an overall indicator of
perceived strain ranging from 12 to 48 (a ¼ 0:86).
In addition to these three strain indices, we construct indices considering burnout
symptoms and health outcomes. Our burnout index (burnout) follows the approach of
Kurtz (2008, p. 225), taking into account the three questions related to burnout,
specifically: feeling like an automatic pilot most times, feeling burned out from the job,
and feeling like being at the end of the rope. The possible answers range from strongly
disagree (1) to strongly agree (5) resulting in an index from 3 to 15 (a ¼ 0:73). Our
index of health outcomes (health), as opposed to the index of physical strain, considers
chronic health outcomes, including migraines, diabetes, chronic low back pain, high
blood pressure, liver disease, foot problems, heart disease, reproductive problems and
chronic insomnia. Possible answers of these questions were yes (1) or no (0). Thus, the
index includes nine questions ranging from 0 to 9 (a ¼ 0:56) with increasing levels
indicating higher levels of burden or negative health outcomes, respectively[5].
Correlation coefficients of the five indices explained above are reported in Table III.
Not surprisingly, as all indices measure various aspects of strain, respectively, they
show positive correlations. While the correlation between the psychological and the
physical index with the overall strain index is not surprising, since the latter is the sum
of the former two aspects of strain. The remaining correlations (varying between 0.35
Psychological strain Physical strain Overall strain Burnout Health
Strain 1 1.0000
Strain 2 0.5873 1.0000
Strain 3 0.9403 0.8277 1.0000 Table III.
Burn 0.5853 0.3761 0.5640 1.0000 Cross-correlation table of
Health 0.4373 0.5301 0.5261 0.3457 1.0000 dependent variables
PIJPSM and 0.59) show that these measures reflect different dimensions of stress and strain,
34,3 albeit the correlation between them is considerably positive.
By measuring stress and various aspects of strain by means of five different indices, we
are confident to cover a wide range of stress aspects as well as strain. The following
section explains our explanatory variables while focusing on our measure of social capital
at work. Moreover, it covers our choice of control variables such as demographic variables
526 and specific characteristics of the current position within the department.
Explanatory and control variables

To address our main research question, we construct as mentioned in the previous
section a narrow index measuring social capital at work (referred to as social capital)
by focusing on two specific questions in the survey, namely whether there is good and
effective cooperation between units and trust in work partners. Possible answers range
from strongly agree (1) to strongly disagree (5). To facilitate a more intuitive
interpretation of our results we have reversed the index, so as to range from 2 to 10
with higher levels indicating a higher level of social capital. Although the level of
internal consistency was at the lower bound of acceptability (a ¼ 0:53) we included it
in our following regressions as such moderate level of Cronbach’s alpha could also be
due to the low number of items included in the index. Moreover, the low a also
indicates that there is lower redundancy in our index of social capital, as the index is
measuring different dimensions of social capital[6]. Additionally, we add the number of
years working for the department to control for experience (referred to as exp), age and
ranking (rank) as explanatory variables. We use all these three factors to isolate their
effects even though they are correlated with each other[7]. As further control variables
we include the number of children (ranging from 0 to 7, referred to as child), as well as
dummies for the ethnic group (1 if Caucasian, referred to as Caucasian) and the marital
status (1 if married or live-in partner, referred to as marital status). To consider
possible differences between genders, we simply constructed a gender dummy with
value 1 if female and 0 otherwise (female).
For robustness purposes, we extended our model by including indices for “trauma” and
“stability at home” in our empirical analysis. Following Swatt et al. (2007), trauma was
measured using a nine-item negative work-related events scale. More detailed, participants
were asked whether they have experienced certain potentially dangerous or traumatic
events in the line of duty and how much it emotionally affected them. In total we included
nine incidents such as a violent arrest, shooting someone, being the subject of an IID
investigation, responding to a call related to a chemical spill, responding to a bloody crime
scene, personally knowing the victim, being involved in a hostage situation, attending a
police funeral and experiencing a needle stick injury or other exposure to blood and body
fluids. For each event officers were asked if they ever experienced this event, and if so, how
much it affected them. Possible answers ranged from “not experienced” (0), “not at all” (1),
“a little” (2) to “very much” (3). Thus, we assume that experiencing an event, although
without affecting the officer emotionally, was more stressful than not experiencing the
event at all. The resulting summative scale ranged from 0 to 27 with higher levels
indicating more individual trauma (a ¼ 0:79).
We construct an additional index measuring social capital or “stability at home”
(referred to as home) we consider questions about reliability on support from the
family, friends etc. and talking about problems with the spouse, relative or friend
(Cullen et al., 1985; He et al., 2002, Howard et al., 2004). In constructing the index, we Stress, strain and
had to recode the question about reliability on the family (“I feel that I can rely on social capital
support from my family, friends etc.”), as the answers originally ranged from strongly
agree (1) to strongly disagree (5). However, the second question (“I talk with my spouse,
relative or friend about problems”) could be answered with never (1) to always (4).
Therefore, we reverse the measure of the first question by putting the numbers upside
down from strongly disagree (1) to strongly agree (5). Subsequently, we construct an 527
index ranging from 2 to 9 with a moderate level of internal consistency (a ¼ 0:53).
Empirical results
Baseline model
Our findings in the baseline model are presented in Table IV. In all regressions we use
standard errors robust to heteroskedasticity of unknown form. Remarkably, in all five
models, the measure of social capital has the expected negative sign, being highly
statistically significant at the 1 percent level in all five cases (see Models 1 to 5). The
estimated regression coefficients indicate that with each additional one unit increase in
social capital strain decreases on average between 0.246 and 0.782 points. Interestingly,
social capital affects psychological and physical strain in quite a similar way, as the
standardized beta coefficients for the first three equations vary between 20.214 and
20.287[8]. Thus, a one standard deviation increase of social capital reduces strain by more
than 0.2 standard deviations. The same applies to our measurement of burnout
(standardized a ¼ 20:287) whereas the effects on health outcomes is slightly smaller
(a ¼ 20:142) but still highly significant. Remarkably, the magnitude of our standardized
beta coefficient of social capital is quite high as compared to other explanatory variables in
our estimation, which shows the relative importance of social capital.
Looking at the control variables we observe that strain levels are negatively
correlated with increasing age holding ranking and experience constant, while our
measure of experience (number of years worked in the department) has ceteris paribus
a positive sign. In almost all the cases both coefficients are statistically significant. On
the other hand, the ranking position is not statistically significant in most of the
regressions. For our burnout index there is a negative relationship observable that is
statically significant at the 1 percent level. The dummy variable for ethnic
identification is also statistically significant, indicating that Caucasian employees
experience higher strain levels, particularly in psychological terms, while there is no
statistically significant difference between races for our health measure. Furthermore,
while our gender dummy variable is not statistically significant in our measures for
psychological strain (Models 1 and 4), the coefficient turns out to be highly statistically
significant in all physical aspects of strain (Models 2, 3 and 5). Thus, as compared to
men, women report suffering from higher levels of physical strain, while there is no
significant difference between genders in terms of perceived levels of psychological
strain and its aspects. Overall, the number of children, marital status and the current
rank do not seem to have a reliable influence on our measurements, although a higher
rank within the department seems to reduce the liability for burnout.
Extensions of the model

To check the reliability of these results, we test the robustness by including the indices
“trauma” and “stability at home” (see Models 6-10 in Table IV). The variable “trauma”,
34,3
528
models
Table IV.
PIJPSM
Baseline and extension

Psychological Physical Overall Psychological Physical Overall
strain strain strain Burnout Health strain strain strain Burnout Health
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)
Social capital
Regression 20.519 * * * 20.246 * * * 20.782 * * * 20.460 * * * 20.120 * * * 20.441 * * * 20.212 * * * 20.667 * * * 20.392 * * * 20.093 * *
Robust standard
errors 27.390 25.588 27.612 28.909 24.041 26.516 24.956 26.858 27.566 23.208
Beta coefficients 20.275 20.214 20.287 20.287 20.142 20.232 20.184 20.244 20.245 20.110
Child
Regression 0.046 0.044 0.075 20.005 20.023 20.025 0.014 20.027 20.069 20.051
Robust standard
errors 0.565 0.873 0.634 20.066 20.646 20.321 0.278 20.246 21.003 21.463
Rank
Regression 20.053 20.024 20.083 20.184 * * * 0.034 20.144 * 20.067 20.215 * 20.271 * * * 20.009
Robust standard
errors 20.642 20.472 20.682 22.728 0.906 21.844 21.292 21.850 24.164 20.224
Exp.
Regression 0.077 * * * 0.049 * * * 0.127 * * * 0.060 * * 0.040 * * * 0.026 0.024 * 0.051 * 0.015 0.020 * *
Robust standard
errors 3.881 3.720 4.286 3.076 4.525 1.373 1.925 1.870 0.818 2.237
Age
Regression 20.066 * * * 20.028 * * 20.093 * * * 20.049 * * * 20.007 20.052 * * * 20.022 * 20.074 * * * 20.037 * * 20.001
Robust standard
errors 23.569 22.252 23.424 22.748 20.830 22.941 21.865 22.885 22.226 20.131
Female
Regression 0.228 0.759 * * * 0.954 * * 20.182 0.507 * * * 0.414 0.864 * * * 1.230 * * * 20.016 0.579 * * *
Robust standard
errors 0.788 3.806 2.208 20.820 3.955 1.554 4.455 3.090 20.082 4.693
(continued)
Psychological Physical Overall Psychological Physical Overall
strain strain strain Burnout Health strain strain strain Burnout Health
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)
Caucasian
Regression 0.773 * * * 0.122 0.919 * * * 0.008 0.037 0.699 * * * 0.092 0.803 * * * 20.08 20.012
Robust standard
errors 3.777 0.946 3.041 0.048 0.418 3.553 0.726 2.764 20.521 20.142
Marital status
Regression 20.059 0.04 20.021 0.055 20.061 0.043 0.072 0.115 0.134 20.042
Robust standard
errors 20.272 0.301 20.067 0.299 20.632 0.203 0.532 0.366 0.760 20.433
Trauma
Regression 0.157 * * * 0.077 * * * 0.233 * * * 0.136 * * * 0.066 * * *
Robust standard
errors 8.607 6.823 8.938 8.923 8.150
Beta coefficients 0.295 0.239 0.305 0.302 0.276
Home
Regression 20.437 * * * 20.214 * * * 20.657 * * * 20.383 * * * 20.125 * * *
Robust standard
errors 25.604 24.545 25.813 27.150 23.880
Constant 15.406 * * * 8.660 * * * 24.199 * * * 12.841 * * * 1.714 * * * 16.241 * * * 9.117 * * * 25.506 * * * 13.614 * * * 1.747 * * *
20.203 18.414 22.389 19.405 5.351 18.052 15.943 19.393 19.370 4.538
R-squared 0.112 0.097 0.127 0.097 0.113 0.22 0.169 0.243 0.212 0.188
F 13.681 * * * 12.831 * * * 16.313 * * * 12.619 * * * 14.416 * * * 22.749 * * * 17.872 * * * 26.485 * * * 24.187 * * * 26.186 * * *
n 991 1,009 987 1,019 1,024 972 990 969 995 997
Notes: Significance levels: *0.05 , p , 0:10, * *0:01 , p , 0:05, * * *p , 0:01
social capital
Stress, strain and
529
Table IV.
PIJPSM as explained above, allows controlling for experiencing extreme situations, while such a
34,3 potential stressor is not found in many other job profiles. Not surprisingly, we observe a
strong relationship between trauma and strain. The trauma index influences the strain
level positively, as more trauma leads to a higher level of (perceived) stress and thus,
higher levels of strain. This relationship holds for all our five specifications. On the
contrary, as expected, stability at home reduces stress at work, being highly statistically
530 significant in all five regressions. The impact is quite strong, as shown by means of the
standardized beta coefficients. Moreover, the results of this extension once again confirm
the baseline model. The index for social capital is still highly statistically significant,
while the magnitude of the coefficients does not change drastically. Remarkably, the
magnitude of the standardized beta coefficient for trauma is comparable to the influence
of our social capital variable. Thus, even under high trauma levels strain levels do not
increase if there is a certain degree of social capital within the police unit. It seems that
social capital consisting of trust between working partners and effective cooperation
between units is able to negate a considerable level of trauma within a job. Considering
the relative magnitude of the coefficients (by comparing standardized betas) it is obvious
that social capital in general, particularly at work, plays a major role in reducing
perceived stress levels and negative strain effects.
Further robustness tests

Taking into account the moderate internal consistency of our measure of social capital
we conducted further robustness tests by splitting up the social capital variable into its
two single parts, namely the question about good and effective cooperation between
units (cooperation) and trust in work partners (trust). For reasons of simplicity, just the
coefficients for the single measures of social capital are shown, while we use the same
set of control variables reported in the previous specifications in Table IV (including
the indices “trauma” and “stability at home”). As expected, the results reported in
Table V are very robust and do not change. Both single factors are still highly
statistically significant in almost all specifications reporting comparable quantitative
Psychological strain Physical strain Overall strain Burnout Health

(11) (12) (13) (14) (15)
Cooperation
Regression 0.475 * * 0.190 * * 0.673 * * 0.325 * * 0.067
Robust standard errors 5.054 3.111 4.907 3.975 1.581
Trust
Regression 0.399 * * 0.240 * * 0.660 * * 0.475 * * 0.124 *
Robust standard errors 3.345 3.170 3.851 4.986 2.469
Other control factors Yes Yes Yes Yes Yes
R-squared 0.22 0.169 0.243 0.213 0.188
Table V. F 20.891 * * 16.237 * * 24.140 * * 22.175 * * 23.826 * *
Robustness test: splitting n 972 990 969 995 997
up to single factors of
social capital Note: Significance levels: *0.01 , p , 0.05, * *p , 0.01
effects between trust and cooperation, with slightly lower coefficients than in former Stress, strain and
regressions as they are just measuring one part of the original social capital index. social capital
Thus, although the index of social capital exhibits only a moderate scale of internal
consistency, the estimates of the influence of social capital on strain are confirmed by
these regressions including the separated variables.
Furthermore, we also took into account a possible endogeneity issue regarding our
measure of social capital. More precisely, as various stress measures are investigated, 531
questions about causality between strain and, e.g. aspects of work environment,
camaraderie, unfairness, coping mechanisms, etc. necessarily rises. Surprisingly, very
few previous studies raised the question about possible endogeneity issues in this
context. Thus, in a further robustness test, we approached this issue by applying an
instrumental two-stage-least square setting where the index of social capital is
assumed to be endogenous. We used personal characteristics and personal perceptions
of the environment as excluded instruments[9]. Both the validity as well as the
relevance of our excluded instruments is confirmed by a number of tests to assess the
reliability and efficiency of the IV estimations[10]. The results of our 2SLS estimation
taking into account the endogeneity of our social capital index confirm the results
derived from our former models[11]. The index for social capital reduces strain
significantly for all five measurements which confirm the importance and significance
of social capital and interpersonal skills at work for reducing strain, even when
controlling for endogeneity of the social capital measure.
Discussion and conclusion

The aim of this paper was to investigate the relational effect that social capital has on
stress and strain by utilizing a large set of strain indices from amongst police officers
and within their physically and emotionally stressful work environment. While many
police stressors are comparable to other work environments (e.g. shift work, excessive
overtime, heavy workload, poor working conditions and strong interactions with the
public), it is the highly stressful and traumatic events of police work that make this
work environment different. Unlike other work environments, officers can also
encounter, witness or hear about fellow officers’ involvement in extreme situations
such as physical or even life threatening danger and general exposure to disturbing
events. Police officers form an essential part for a well-functioning society and thus the
necessity to find effective methods to reduce the negative aspects created by working
in this highly stressful job. We illustrate in this paper that social capital within a work
environment is a breeding ground for social stability among workers and from this
trust flows information and social support. In other words, the negative aspects of
stress are lowered through the generation of trust and cooperation established between
co-workers and units, through the passing on of coping strategies and support of
colleagues. Social cohesion reduces transaction costs, allows for better access to
information and enables a better coordination of activities. Thus, social capital is the
conduit through which police officers can draw on resources in their personal and
professional lives which should help them to deal with stressful situations.
The results provide strong empirical support that social capital has a highly
significant and potent effect on reducing all of the factors of stress/strain under
investigation, even after controlling for many of the usual confounding demographic
factors. However, some of these control factors have thrown up some interesting
PIJPSM results that demand further attention such as race, gender and rank. In line with much
34,3 of the literature, gender has affected stress levels but in surprising ways, as we observe
that females suffer significantly more from the physical and health aspects of stress
but not from burnout or the psychological effects. This is in line with the idea that
females do indeed have better capabilities to alleviate the mental side of stress through
communication and load sharing with female colleagues, whereas males cannot or will
532 not do the same. Additionally, we observed that Caucasians are affected significantly
more by the psychological effects of stress, but do not display significant differences in
the physical strain, burnout or health outcomes. The findings are consistently strong
and robust in both the baseline and extension models. This provides additional support
to the idea that not only does gender and race affect how stress impacts individuals but
also determines the type of stress, be it physical or psychological in nature. This
differentiation of stress effects requires much more research that could ultimately lead
to tailored programs with greater probability of success if both race and gender of the
individual is taken into account. Furthermore, we find an interesting inverse
relationship between age and experience of individuals and the effects of stress. As
expected, with increasing age officers become better at divesting themselves of large
amounts of stress, excluding health where age is expected to have a much more direct
impact on health regardless of stress. However, the more years of experience an officer
has the greater the negative stress outcome, which would indicate that stress is
cumulative over time and the more years spent in the force the greater the effect. This
effect effectively disappears once we control for traumatic events and home life. It is
clear from the results that social capital in this work environment is by far the largest
and most significant factor in reducing the negative effects of stress, closely followed
by social capital at home. Just as significant is the explanatory power of the traumatic
events index across all the aspects of stress explored in this study.
Social capital has therefore shown to be extremely effective in negating the impacts
of the majority of the strains, and significantly reduces the impact of the major trauma
events (shootings, hostages and funerals, etc). This finding would indicate that police
management and police officers themselves would be greatly benefited through the
implementation of social programs that enhance social capital, in our case measured by
trust, cooperation and communication between units. Further benefits derived from
greater levels of social capital can be observed in staff retention rates, where higher
levels of social capital can be seen to have a positive effect on lowering staff turnover
rates (Gächter et al., 2009). However we do concede that the age of the data does
slightly reduce the explanatory power for the current work environment and consists
of only one American police department. It may also be interesting to explore police
environments in different countries to check whether the extrapolation of the results is
possible. Brown and Campbell (1990) point to the fact that there are divergent
traditions between countries and results from the USA cannot be extrapolated to other
countries such as the UK. Moreover, the source of stress may be driven by the nature of
the organization itself. However, it is also useful to test whether the obtained results
also hold in other environments that are comparable to the police one (e.g., military).
Nevertheless, additional studies of highly stressed work employees and
environments in other areas would contribute to a better understanding of the
relationship between stress, strain and social capital and may improve the quality of
relief programs and greatly reduce the costs and its externalities accumulated through
strained employees. Currently, the predominant stress reduction programs are Stress, strain and
counseling services, utilized in the hope that this will stem the flood of stress related social capital
retirements and burnouts. This hope has been labeled occasionally as too simplistic
given the very complex relationships between stressful incidents, individual
demographic variables and organizational structure (Dick, 2000). In addition, social
capital might be a good alternative instrument in situations where common stress
reducing instruments fail or where the necessary information to design and enforce 533
suitable instruments and directives cannot effectively be used.
Notes
1. For example, when an individual’s core temperature rises, the body evokes a sweating
response to shed the excess heat.
2. For example, environmental factors are better controlled or isolated compared to the case
where individuals within a survey have heterogeneous job profiles and are acting in different
environments (noisy stress comparison).
3. See, e.g. the Stressful Life Events Screening Questionnaire (Goodman et al., 1998); Trauma
History Questionnaire (Green, 1996); Traumatic Life Events Questionnaire (Kubany et al.,
2000); Traumatic Stress Schedule (Norris, 1990); Brief Trauma Questionnaire (Schnurr et al.,
1999); Traumatic Events Questionnaire (Vrana and Lauterbach, 1994); Life Stressor
Checklist – Revised (Wolfe et al., 1996).
4. The response rate was calculated by the number returned by each precinct compared with
the average number of sworn employees at each precinct on the day of the survey.
5. At this point it seems important to mention the slightly differing number of observations
depending on various variables and indices (see Table II) ranging from 1,060 to 1,104. The
reason for this is some missing observations in the data, as some participants did not
respond to all questions. However, as the missing observations amount to 44 cases in the
worst case (index overall strain, not even 4 percent of the data,) this should not be a major
problem in our analysis. Moreover, preliminary analyses indicate that excluded cases did not
significantly differ from the others on key demographic variables such as gender, age, rank,
or race.
6. Thus, it is also important to check the results by splitting up the index of social capital to
examine the effects of the single parts of the index for all the dependent variables (see
Table V).
7. However, our results indicate that there is enough remaining variation on each of the
variables when the other two variables are held constant. In addition, when assuming
multicollinearity issues, although it would be difficult to get distinct coefficient estimates for
them, it would only affect the coefficient estimates for those variables that are collinear and
not the coefficient estimates of our main independent variable, namely the social capital
index. Nevertheless, we have run estimations with these single factors independently
without observing major changes in the reported results.
8. Standardized coefficients convert all the variables into standard deviations inducing the
same metric which allows us to compare them across different variables.
9. More precisely, we assume that personal characteristics and personal perceptions of the
environment have a significant impact on the personal perception of social capital at work,
namely whether the individual is well integrated into the department or not. In other words,
even if there is a considerable degree of social capital in a department, certain individuals
who have difficulties with interpersonal relationships in general should report a lower
degree of social capital in that specific department (as they are not able to participate in this
PIJPSM social process), although other more socialized employees may experience high trust and
good cooperation, respectively.
34,3
10. Further information about the estimation procedure and the tests conducted is available on
request.
11. For the purposes of brevity and space we have not included the results for the 2SLS
regressions in the paper, however the results are available on request.
534
References
American Psychiatric Association (APA) (2000), Diagnostic and Statistical Manual of Mental
Disorders, Fourth Edition, Text Revision (DSM-IV-TR), APA, Arlington, VA.
Bartollas, C. and Hahn, L.D. (1999), Policing in America, Allyn & Bacon, Needham Heights, CA.
Berkman, L. and Syme, S. (1979), “Social networks, host resistance and mortality: a nine year
follow-up study of Alameda county residents”, American Journal of Epidemiology, Vol. 109
No. 2, pp. 186-204.
Biggam, F.H., Power, K.G. and MacDonald, R.R. (1997), “Coping with the occupational stressors
of police work: a study of Scottish police officers”, Stress Medicine, Vol. 13, pp. 109-15.
Boissevian, J. (1974), Friends of Friends, Networks, Manipulators and Coalitions, Basil Blackwell,
Oxford.
Boix, C. and Posner, D. (1998), “Social capital: explaining its origins and effects on government
performance”, British Journal of Political Science, Vol. 28 No. 4, pp. 686-95.
Bourdieu, P. (1985), “The forms of capital”, in Richardson, J.G. (Ed.), Handbook of Theory and
Research for the Sociology of Education, Greenwood, New York, NY, pp. 241-58.
Brooks, L.W. and Piquero, N.L. (1998), “Police stress: does department size matter?”, Policing:
An International Journal of Police Strategies and Management, Vol. 21 No. 4, pp. 600-17.
Brown, G.W. and Harris, T.O. (1978), Social Origins of Depression, Tavistock, London.
Brown, J.M. and Campbell, E.A. (1990), “Sources of occupational stress in the police”, Work and
Stress, Vol. 4 No. 4, pp. 305-18.
Brown, J., Fielding, J. and Grover, J. (1999), “Distinguishing traumatic, vicarious and routine
operational stressor exposure and attendant adverse consequences in a sample of police
officers”, Work and Stress, Vol. 13 No. 4, pp. 312-25.
Burke, R.J. and Deszca, E. (1986), “Correlates of psychological burnout phases among police
officers”, Human Relations, Vol. 39 No. 6, pp. 487-502.
Cannon, W.B. (1929a), Bodily Changes in Pain, Hunger, Fear and Rage, D. Appleton and Co., New
York, NY.
Cannon, W.B. (1929b), “Organization for physiological homeostasis”, Physiological Review, Vol. 9,
pp. 399-431.
Cannon, W.B. (1935), “Stresses and strains of homoeostasis”, American Journal of Medical
Science, Vol. 189 No. 1.
Caplan, G. (1974), Support Systems and Community Mental Health, Behavioral Publications, New
York, NY.
Cohen, F. and Lazarus, R.S. (1979), “Coping with the stress of illness”, in Stone, G., Cohen, F. and
Alder, N. (Eds), Health Psychology, Jossey-Bass, San Francisco, CA.
Cohen, S. and Willis, T.A. (1985), “Stress, social support and the buffering hypothesis”,
Psychological Bulletin, Vol. 98 No. 2, pp. 310-57.
Coleman, J.S. (1988), “Social capital in the creation of human capital”, American Journal of Stress, strain and
Sociology, Vol. 94, pp. 95-120.
social capital
Crank, J.P. and Caldero, M. (1991), “The production of occupational stress in medium-sized police
agencies: a survey of line officers in eight municipal departments”, Journal of Criminal
Justice, Vol. 19 No. 4, pp. 339-49.
Cullen, F.T., Lemming, T., Link, B.G. and Wozniak, J.F. (1985), “The impact of social supports on
police stress”, Criminology, Vol. 23 No. 3, pp. 503-22. 535
Dasgupta, P. (1999), “Economic progress and the idea of social capital”, in Dasgupta, P. and
Serageldin, I. (Eds), Social Capital: A Multifaceted Perspective, The World Bank,
Washington, DC.
Desmarais, S. and Alksnis, C. (2005), “Gender issues”, in Barling, J., Kelloway, E.K. and
Frone, M.R. (Eds), Handbook of Work Stress, Sage Publications, Thousand Oaks, CA.
Dick, P. (2000), “The social construction of the meaning of acute stressors: a qualitative study of
the personal accounts of police officers using a stress counseling service”, Work and
Stress, Vol. 14 No. 3, pp. 226-44.
Everly, G.S. (1979), Strategies for Coping with Stress: An Assessment Scale, Office of Health
Promotion, Department of Health and Human Services, Washington, DC.
Everly, G.S. and Benson, H. (1989), “Disorders of arousal and the relaxation response:
speculations on the nature and treatment of stress-related diseases”, International Journal
of Psychosomatics, Vol. 36 No. 1, pp. 15-21.
Everly, G.S. and Lating, J.M. (2002), A Clinical Guide to the Treatment of the Human Stress
Response, Kluwer, New York, NY.
Fukuyama, F. (2003), “Social capital and civil society”, in Ostrom, E. and Ahn, T.K. (Eds),
Foundations of Social Capital, Edward Elgar Publishing, Cheltenham.
Gabriel, P. and Liimatainen, M.R. (2000), Mental Health in the Workplace, International Labor
Office, Geneva.
Gächter, M., Savage, D.A. and Torgler, B. (2009), “Retaining the thin blue line: what shapes
workers’ willingness not to quit the current work environment?”, School of Economics and
Finance Discussion Papers and Working Papers Series, No 253, School of Economics and
Finance, Brisbane.
Gächter, M., Savage, D.A. and Torgler, B. (2010), “The role of social capital in reducing negative
health outcomes among police officers”, International Journal of Social Inquiry, Vol. 3 No. 1,
pp. 141-61.
Gächter, M., Savage, D.A. and Torgler, B. (2011), “Gender variations of physiological and
psychological stress amongst police officers”, Gender Issues, Vol. 28 Nos 1/2, pp. 66-93.
Gershon, R. (1999), Project SHIELDS, National Institute of Justice, Washington, DC.
Gershon, R. (2000), Police Stress and Domestic Violence in Police Families in Baltimore, Maryland,
1997-1999, Inter-university Consortium for Political and Social Research, Ann Arbor, MI.
Gershon, R., Barocas, B., Canton, A.N., Li, X. and Vlahov, D. (2009), “Mental, physical, and
behavioral outcomes associated with perceived work stress in police officers”, Criminal
Justice and Behavior, Vol. 36 No. 3, pp. 275-89.
Girdano, D.A., Dusek, D.E. and Everly, G.S. (2001), Controlling Stress and Tension, 6th ed., Allyn
& Bacon, Boston, MA.
Goodman, L.A., Corcoran, C., Turner, K., Yuan, N. and Green, B.L. (1998), “Assessing traumatic
event exposure: general issues and preliminary findings for the Stressful Life Events
Screening Questionnaire”, Journal of Traumatic Stress, Vol. 11 No. 2, pp. 521-42.
PIJPSM Green, B.L. (1996), “Psychometric review of Trauma History Questionnaire (self-report)”,
in Stamm, B.H. and Varra, E.M. (Eds), Measurement of Stress, Trauma and Adaptation,
34,3 Sidran, Lutherville, MD.
Grootaert, C. (2001), “Social capital: the missing link?”, in Dekker, P. and Uslaner, E. (Eds), Social
Capital and Participation in Everyday Life, Routledge, London.
Haarr, R.N. and Morash, M. (1999), “Gender, race, and strategies of coping with occupational
536 stress in policing”, Justice Quarterly, Vol. 16 No. 2, pp. 304-32.
Hawkins, H.C. (2001), “Police officer burnout: a partial replication of Maslach’s Burnout
Inventory”, Police Quarterly, Vol. 4 No. 3, pp. 343-60.
He, N., Zhao, J. and Archibold, C.A. (2002), “Gender and police stress: the convergent and
divergent impact of work environment, work-family conflict, and stress coping
mechanisms of female and male police officers”, Policing: An International Journal of
Police Strategies and Management, Vol. 25 No. 4, pp. 687-708.
He, N., Zhao, J. and Ren, L. (2005), “Do race and gender matter in police stress? A preliminary
assessment of the interactive effects”, Journal of Criminal Justice, Vol. 33, pp. 535-47.
Hodson, R. (2005), “Management behavior as social capital: a systematic analysis of
organizational ethnographies”, British Journal of Industrial Relations, Vol. 43 No. 1,
pp. 41-65.
Horowitz, M.J., Wilner, N.R. and Alvarez, W. (1979), “Impact of event scale. a measure of
subjective stress”, Psychosomatic Medicine, Vol. 41 No. 3, pp. 209-18.
Howard, G.W., Howard, H.D. and Boles, J.S. (2004), “Inter-domain work-family, family-work
conflict and police work satisfaction”, Policing: An International Journal of Police
Strategies and Management, Vol. 27 No. 3, pp. 380-95.
Kawachi, I., Colditz, G.A. and Ascherio, A. (1996), “A prospective study of social networks in
relation to total mortality and cardiovascular disease incidence in men”, Journal of
Epidemiology and Community Health, Vol. 50 No. 3, pp. 245-51.
Knack, S. and Keefer, P. (1997), “Does social capital have an economic payoff? A cross-country
investigation”, Quarterly Journal of Economics, Vol. 112 No. 4, pp. 1251-88.
Kopel, H. and Friedman, M. (1999), “Effects of exposure to violence in South African police”,
in Violanti, J. and Paton, D. (Eds), Police Trauma: Psychological Aftermath of Civilian
Combat, Charles C. Thomas, Springfield, IL.
Kroes, W.H. (1985), Society’s Victim, the Police: An Analysis of Job Stress in Policing, 2nd ed.,
Charles C. Thomas, Springfield, IL.
Kroes, W.M., Hurrell, J.J. and Margolis, B. (1974), “Job stress in police administration”, Journal of
Police Science and Administration, Vol. 2 No. 4, pp. 386-7.
Kubany, E.S., Haynes, S.N., Leisen, M.B., Owens, J.A., Kaplan, A.S., Watson, S.B. and Burns, K.
(2000), “Development and preliminary validation of a brief broad-spectrum measure of
trauma exposure: the Traumatic Life Events Questionnaire”, Psychological Assessment,
Vol. 12 No. 2, pp. 210-24.
Kurtz, D.L. (2008), “Controlled burn: the gendering of stress and burnout in modern policing”,
Feminist Criminology, Vol. 3 No. 3, pp. 216-38.
Lachman, S. (1972), Psychosomatic Disorders: A Behavioristic Interpretation, John Wiley & Sons,
New York, NY.
Lazarus, R.S. and Folkman, S. (1984), Stress, Appraisal and Coping, Springer, New York, NY.
Lennings, C. (1997), “Police and occupationally related violence: a review”, Policing: An International
Journal of Police Strategies and Management, Vol. 20 No. 3, pp. 555-66.
Liberman, A.M., Best, S.R., Metzler, T.J., Fagan, J.A., Weiss, D.S. and Marmar, C.R. (2002), “Routine Stress, strain and
occupational stress and psychological distress in police”, Policing: An International Journal
of Police Strategies and Management, Vol. 25 No. 2, pp. 421-39. social capital
Likert, R. (1932), “A technique for the measurement of attitudes”, Archives of Psychology, Vol. 140,
pp. 1-55.
Loo, R. (1984), “Occupational stress in the law enforcement profession”, Canada’s Mental Health,
Vol. 31, pp. 10-13. 537
Loury, G.C. (1977), “A dynamic theory of racial income differences”, in Wallace, P.A. and
La Mond, A.M. (Eds), Women, Minorities, and Employment Discrimination, Heath,
Lexington, MA, pp. 153-86.
Lovallo, W.R. (2005), Stress and Health: Biological and Psychological Interaction, 2nd ed., Sage
Publications, Thousand Oaks, CA.
McCarty, W.P., Zhao, J.S. and Garland, B.E. (2007), “Occupational stress and burnout between male
and female police officers: are there any gender differences?”, Policing: An International
Journal of Police Strategies and Management, Vol. 30 No. 4, pp. 672-91.
Malloy, T.E. and Mays, L. (1984), “The police stress hypothesis: a critical evaluation”, Criminal
Justice and Behavior, Vol. 11 No. 2, pp. 197-224.
Manski, C.F. (2000), “Association economics: analysis of social interactions”, The Journal of
Economic Perspectives, Vol. 14 No. 3, pp. 115-36.
Meichenbaum, D. (2007), “Stress inoculation training: a preventative and treatment approach”, in
Lehrer, M., Woodfolk, R.L. and Slime, W.S. (Eds), Principles and Practices of Stress
Management, Guilford Press, New York, NY.
Mitchell, J. and Bray, G. (1990), Emergency Services Stress: Guidelines for Reserving the Health
and Careers of Emergency Service Personnel, Prentice Hall, Englewood Cliffs, NJ.
Morash, M. and Haarr, R.N. (1995), “Gender, workplace problems, and stress in policing”, Justice
Quarterly, Vol. 12 No. 1, pp. 113-40.
Morash, M., Haarr, R. and Kwak, D.H. (2006), “Multilevel influences on police stress”, Journal of
Contemporary Criminal Justice, Vol. 22 No. 1, pp. 26-43.
Morris, A., Shinn, M. and Du Mont, K. (1999), “Contextual factors affecting the organizational
commitment of diverse police officers: levels of analysis perspective”, American Journal of
Community Psychology, Vol. 27 No. 1, pp. 75-105.
Norris, F.H. (1990), “Screening for traumatic stress: a scale for use in the general population”,
Journal of Applied Science, Vol. 20, pp. 1704-18.
Norris, F.H. and Hamblen, J.L. (2004), “Standardized self-report measures of civilian trauma and
PTSD”, in Wilson, J.P. and Keane, T.M. (Eds), Assessing Psychological Trauma and PTSD,
2nd ed., Guilford Press, New York, NY.
O’Driscoll, M.P. and Dewe, P. (2001), “Mediators and moderators of stressor-strain linkages”,
in Perrewe, P.L. and Ganster, D.C. (Eds), Research in Occupational Stress and Wellbeing:
Exploring Theoretical Mechanisms and Perspectives, Elsevier, New York, NY.
O’Farrell, B. (1980), Women and Traditional Blue-collar Jobs: A Case Study of Local 1,
US Department of Labor, Employment, and Training Administration, Washington, DC.
Paldam, M. (2000), “Social capital: one or many? Definition and measurement”, Journal of
Economic Surveys, Vol. 14 No. 5, pp. 629-53.
Peak, K. (1993), Policing America: Methods, Issues, Challenges, Prentice Hall, Englewood Cliffs,
NJ.
PIJPSM Portes, A. (1998), “Social capital: its origins and applications in contemporary sociology”, Annual
Review of Sociology, Vol. 24, pp. 1-24.
34,3
Prince, M., Harwood, R.H., Blizard, R.A., Thomas, A. and Mann, A.H. (1997), “Impairment,
disability and handicap as risk factors for depression in old age: the Gospel Oak Project
V”, Psychological Medicine, Vol. 27, pp. 311-21.
Putnam, L.L. (1983), “The interpretive perspective: an alternative to functionalism”, in Putnam, L.L.
538 and Pacanowsky, M.E. (Eds), Communication and Organization, Sage, Beverly Hills, CA,
pp. 31-54.
Putnam, R. (1993), Making Democracy Work: Civic Traditions in Modern Italy, Princeton
University Press, Princeton, NJ.
Repetti, R.L. (1993), “The effects of workload and the social environment at work on health”,
in Goldberger, L. and Breznitz, S. (Eds), Handbook of Stress: Theoretical and Clinical
Aspects, Free Press, New York, NY.
Schaufeli, W.B. and Enzmann, D. (1998), The Burnout Companion to Study and Practice: A Critical
Analysis, Taylor & Francis, London.
Schnurr, P.P., Vieilhauer, M.J., Weathers, F. and Findler, M. (1999), The Brief Trauma
Questionnaire, National Center for Post-Traumatic Stress Disorder, White River Junction,
VT.
Schwartz, G.E. (1977), “Psychosomatic disorders and biofeedback: a psychobiological model of
disregulation”, in Maser, J.D. and Seligman, M.E.P. (Eds), Psychopathology: Experimental
Models, Freeman, San Francisco, CA.
Schwartz, G.E. (1979), “The brain as a health care system”, in Stone, G.C. and Cohen, F. (Eds),
Health Psychology – A Handbook, Jossey-Bass, San Francisco, CA, pp. 549-71.
Schwartz, J. and Schwartz, C. (1981), “The personal prolems for the police officer: a plea for
action”, in Territo, L. and Vetter, H. (Eds), Stress and Police Personnel, Allyn & Bacon,
Boston, MA.
Selye, H. (1936), “A syndrome produced by diverse noxious agents”, Nature, Vol. 138 No. 32.
Simmel, G. (1964), “The metropolis and mental life”, in Wolff, K.H. (Ed.), The Sociology of Georg
Simmel, Free Press, New York, NY, trans. by Wolff, K.H. (originally published in 1902).
Stephens, C., Long, N. and Miller, I. (1997), “The impact of trauma and social support on
post-traumatic stress disorder in New Zealand police officers”, Journal of Criminal Justice,
Vol. 25, pp. 303-14.
Sternbach, R. (1966), Principles of Psychophysiology, Academic Press, New York, NY.
Stotland, E. (1991), “The effects of police work and professional relationships on health”, Journal
of Criminal Justice, Vol. 19 No. 4, pp. 371-9.
Stratton, J.G. (1984), Police Passages, Glennon, Manhattan Beach, CA.
Swatt, M.L., Gibson, C.L. and Piquero, N.L. (2007), “Exploring the utility of general strain theory
in explaining problematic alcohol consumption by police officers”, Journal of Criminal
Justice, Vol. 35 No. 6, pp. 596-611.
Torgler, B. (2003), “Why do people go to war?”, Defense and Peace Economics, Vol. 14 No. 4,
pp. 261-80.
Uslaner, E.M. (2002), The Moral Foundation of Trust, Cambridge University Press, Cambridge.
Violanti, J.M., Marshall, J.R. and Howe, B. (1983), “Police occupational demands, psychological
distress and the coping function of alcohol”, Journal of Occupational Medicine, Vol. 25
No. 6, pp. 455-8.
Vrana, S. and Lauterbach, D. (1994), “Prevalence of traumatic events and post-traumatic Stress, strain and
psychological symptoms in a non-clinical sample of college students”, Journal of
Traumatic Stress, Vol. 7 No. 2, pp. 289-302. social capital
Waldinger, R. (1995), “The ‘other side’ of embeddedness: a case study of the interplay between
economy and ethnicity”, Ethnic and Racial Studies, Vol. 18 No. 3, pp. 555-80.
Waters, J.A. and Ussery, W. (2007), “Police stress: history, contributing factors, symptoms, and
interventions”, Policing: An International Journal of Police Strategies and Management, 539
Vol. 30 No. 2, pp. 169-88.
Weiss, D.S. and Marmar, C.R. (1997), “The Impact of Event Scale-revised”, in Wilson, J.P. and
Keane, T.M. (Eds), Assessing Psychological Trauma and PTSD: A Practitioner’s
Handbook, Guilford, New York, NY, pp. 399-411.
Whitley, R. and McKenzie, K. (2005), “Social capital and psychiatry: review of the literature”,
Harvard Review of Psychiatry, Vol. 13 No. 2, pp. 71-84.
Wolfe, J.W., Kimerling, R., Brown, P.J., Chrestman, K.R. and Levin, K. (1996), “Psychometric
review of the Life Stressors Checklist-Revised”, in Hudnall Stamm, B, (Ed.), Measurement
of Stress, Trauma, and Adaptation, Sidran Press, Lutherville, MD, pp. 198-201.
Woolcock, M. and Narayan, D. (2000), “Social capital: implications for development theory,
research and policy”, The World Bank Research Observer, Vol. 15 No. 2, pp. 225-49.
Further reading
Adler, P.S. and Kwon, S.W. (2002), “Social capital: prospects for a new concept”, Academy of
Management Review, Vol. 27 No. 1, pp. 17-40.
Alesina, A. and La Ferrara, E. (2002), “Who trusts others?”, Journal of Public Economics, Vol. 85
No. 2, pp. 207-34.
Baum, C.F. (2006), An Introduction to Modern Econometrics Using Stata, Stata Press, College
Station, TX.
Bourne, L.E. and Yaroush, R.A. (2003), Stress and Cognition: A Cognitive Psychological
Perspective, National Aeronautics and Space Administration (NASA), Washington, DC.
Cannon-Bowers, J.A. and Salas, E. (1998), “Individual and team decision making under stress:
theoretical underpinnings”, in Cannon-Bowers, J.A. and Salas, E. (Eds), Making Decisions
under Stress, American Psychological Association, Washington, DC.
Derogatis, L. and Savitz, K. (1999), “The SCL-90-R, brief symptom inventory and matching
clinical rating scales”, in Maruish, M. (Ed.), The Use of Psychological Testing for
Treatment, Planning and Outcomes Assessment, Lawrence Erlbaum Associates, Mahwah,
NJ.
Gambetta, D. (1988), Trust: Making and Breaking Cooperative Relations, Blackwell, Oxford.
Hardin, R. (1993), “The street-level epistemology of trust”, Politics and Society, Vol. 21 No. 4,
pp. 505-31.
La Porta, R., Lopez-de-Silanes, F., Shleifer, A. and Vishny, R. (1999), “The quality of government”,
Journal of Law, Economics, and Organization, Vol. 15 No. 1, pp. 222-79.
Leana, C.R. and Van Buren, H.J. (1999), “Organizational social capital and employment
practices”, Academy of Management Review, Vol. 24 No. 3, pp. 538-55.
Schaltegger, C.A. and Torgler, B. (2007), “Government accountability and fiscal discipline: a panel
analysis using Swiss data”, Journal of Public Economics, Vol. 91 Nos 1-2, pp. 117-40.
PIJPSM Stock, J.H. and Yogo, M. (2005), “Testing for weak instruments in linear iv regression”,
in Andrews, D.W.K. and Stock, J.H. (Eds), Identification and Inference for Econometric
34,3 Models: Essays in Honor of Thomas Rothenberg, Cambridge University Press, Cambridge.
Torgler, B. (2007), Tax Compliance and Tax Morale: A Theoretical and Empirical Analysis,
Edward Elgar, Cheltenham.
540 Corresponding author

David A. Savage can be contacted at: david.savage@qut.edu.au
To purchase reprints of this article please e-mail: reprints@emeraldinsight.com

Or visit our web site for further details: www.emeraldinsight.com/reprints

The Relationship Between Stress, Strain and Social Capital

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Relationship Between Stress, Strain and Social Capital

Uploaded by

Copyright:

Available Formats

The current issue and full text archive of this journal is available at

Stress, strain and

Not at all A little Very much N/A

Making a violent arrest 220 554 210 105 1,089

a social network, as networks of civic engagement facilitating coordination and

Data and methods

Psychological strain Physical strain Overall strain Burnout Health

Explanatory and control variables

Extensions of the model

Baseline and extension

Further robustness tests

Psychological strain Physical strain Overall strain Burnout Health

Discussion and conclusion

540 Corresponding author

To purchase reprints of this article please e-mail: reprints@emeraldinsight.com

You might also like