8/18/2019 Social Support’s Relationship to Correctional Staff Job Stress,Job Involvement, Job Satisfaction, And Organizationalc…
1/11
Please cite this article in press as: Lambert, E. G., et al. Social support’s relationship to correctional staff
job stress, job involvement, job satisfaction, and organizational commitment. The Social Science Journal (2015),
http://dx.doi.org/10.1016/j.soscij.2015.10.001
ARTICLE IN PRESSG Model
SOCSCI-1290; No.of Pages11
The Social Science Journal xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
The Social Science Journal
j ourna l homepage: www.e lsev ier .com/locate /sosc i j
Social support’s relationship to correctional staff job stress,
job involvement, job satisfaction, and organizational
commitment
Eric G. Lambert a,∗, Kevin I. Minor b,1, James B. Wells b,2, Nancy L. Hogan c,3
a Department of Legal Studies, TheUniversity of Mississippi, 202 Odom Hall, Mississippi, MS 38677, USAb School of Justice Studies, Eastern Kentucky University, Stratton 467, 521 Lancaster Avenue, Richmond, KY 40475, USAc School of Criminal Justice, Ferris State University, 525 BishopHall, Big Rapids, MI 49307, USA
a r t i c l e i n f o
Article history:
Received 18 December 2014
Received in revised form 18 April 2015
Accepted 5 October 2015
Available online xxx
Keywords:
Correctional staff
Social support
Job stress
Job satisfaction Job involvement
Organizational commitment
a b s t r a c t
The literature suggests that social support, in general, is linked to positive outcomes among
correctional staff, but the different types of social support may differ in their effects. Using
survey data from staff working at a privately-owned,maximum security prison for juveniles
sentenced as adults, this study analyzes three intra-organizational supports, administra-
tive, supervisory, and coworker, andone extra-organizational form, family/friends support,
as antecedents of job stress, job involvement, job satisfaction, and organizational com-
mitment. Personal variables serve as controls. Administrative and supervisory support are
inversely related to job stress, while job involvement is affected positively by supervi-
sory support and negatively by family/friends support. All three intra-organizational forms
of support are significant antecedents of job satisfaction and organizational commitment
however family/friends support is not. These findings hold implications for improving job
outcomes among correctional staff and for future research.
© 2015 Western Social Science Association. Published by Elsevier Inc. All rights reserved.
Prisons are expensive to administer and operate.
Henrichson and Delaney (2012) report that over 35 billion
dollars is spent on prisons each year in the United States
to house 1.4 million adult inmates. Operating prisons is
labor intensive, with staff being the largest expenditure,
often accounting for over 75% of a prison’s annual budget
(Camp & Lambert, 2005). Additionally, staff are a valu-
able resource, as they are responsible for the multitude of tasks and duties necessary for the operation of a humane,
∗ Corresponding author. Tel.: +1 662 915 2672.
E-mail addresses: [email protected] (E.G. Lambert),
[email protected] (K.I. Minor), [email protected] (J.B. Wells),
[email protected] (N.L. Hogan).1 Tel.: +1 859 622 2240.2 Tel.: +1 859 622 1158.3 Tel.: +1 231 591 2664.
secure, and safe prison. In a sense, staff are a prison’s heart
and soul. In an era of reduced government budgets, inves-
tigating the factors that may affect correctional staff is
imperative.
A growing body of research focuses on prison staff,
particularly in terms of job stress, job involvement, job
satisfaction, and organizational commitment. Stressed,
uninvolved, dissatisfied, and uncommitted staff can bedetrimental to a prison’s operations. On the other hand,
having relatively unstressed, involved, satisfied, and com-
mitted staff is a desired outcome. To reach this outcome,
researchers and administrators need information on the
factors that help shape the job stress, job involvement,
job satisfaction, and organizational commitment of prison
staff.
Working in a prison differs from working in other
types of organizations. The prison work environment is
unique because it involves confining individuals against
http://dx.doi.org/10.1016/j.soscij.2015.10.001
0362-3319/© 2015 Western Social Science Association. Published by Elsevier Inc. All rights reserved.
http://localhost/var/www/apps/conversion/tmp/scratch_4/dx.doi.org/10.1016/j.soscij.2015.10.001http://localhost/var/www/apps/conversion/tmp/scratch_4/dx.doi.org/10.1016/j.soscij.2015.10.001http://www.sciencedirect.com/science/journal/03623319http://www.elsevier.com/locate/soscijmailto:[email protected]:[email protected]:[email protected]:[email protected]://localhost/var/www/apps/conversion/tmp/scratch_4/dx.doi.org/10.1016/j.soscij.2015.10.001http://localhost/var/www/apps/conversion/tmp/scratch_4/dx.doi.org/10.1016/j.soscij.2015.10.001mailto:[email protected]:[email protected]:[email protected]:[email protected]://www.elsevier.com/locate/soscijhttp://www.sciencedirect.com/science/journal/03623319http://localhost/var/www/apps/conversion/tmp/scratch_4/dx.doi.org/10.1016/j.soscij.2015.10.001http://localhost/var/www/apps/conversion/tmp/scratch_4/dx.doi.org/10.1016/j.soscij.2015.10.001
8/18/2019 Social Support’s Relationship to Correctional Staff Job Stress,Job Involvement, Job Satisfaction, And Organizationalc…
2/11
Please cite this article in press as: Lambert, E. G., et al. Social support’s relationship to correctional staff
job stress, job involvement, job satisfaction, and organizational commitment. The Social Science Journal (2015),
http://dx.doi.org/10.1016/j.soscij.2015.10.001
ARTICLE IN PRESSG Model
SOCSCI-1290; No.of Pages11
2 E.G. Lambert et al. / The Social Science Journal xxx (2015) xxx–xxx
their will—individuals who have been convicted of violat-
ingcriminallaw. As Armstrong and Griffin (2004) point out,
“Few other organizations are charged with the central task
of supervising and securing an unwilling and potentially
violent population” (p. 577). As the work environment in
prisons varies from that found in most other organizations,
the factors which influence job stress, job involvement, job
satisfaction, and organizational commitment may alsovary
compared to other organizations. Research needs to iden-
tify factors that mitigate job stress and contribute to the
prison staff’s job involvement, job satisfaction, and organi-
zational commitment.
The literature suggests that social support is important
for staff to deal with the unique strains and challenges
encountered in a prison work environment (Keinan &
Maslach-Pines, 2007; Whitehead & Lindquist, 1986). Social
support should help staff deal with strains that would
otherwise increase job stress and decrease job involve-
ment, job satisfaction, and organizational commitment.
There are different types of social support, and the liter-
ature is limited and unclear about what types of social
support are linked with job stress, job involvement, job
satisfaction,and organizationalcommitment amongprison
staff. This study’s objective is to explore the nature of the
relationship between administrative support, supervisory
support, coworker support, and family and friends support
with job stress, job involvement, job satisfaction, and orga-
nizational commitment. Expanding knowledge of these
relationships provides a framework for the future devel-
opment of research in this area and provides correctional
administrators with information on how different forms of
social support affect prison staff.
1. Literature review
1.1. Job stress
Accordingto Mattesonand Ivancevich (1987), “thereare
literallyhundredsof definitions for stress to be found in the
research and professional literature. Virtually all of them
can be placed into one of two categories, however: stress
can be defined as either a stimulus or a response” (p. 10).
Stressors arenegativestimuli that cause strainfor a person,
which ultimately can result in stress (Cullen, Link, Wolfe,
& Frank, 1985). Job stress is psychological strain leading to
leading to job-related hardness, tension, anxiety, frustra-
tion, and worry arising from work (Misis, Kim, Cheeseman,
Hogan, & Lambert, 2013). Job stress can be harmful over
time, leading to increased mental withdrawal from the job,
reduced interactions with clients and coworkers, increased
conflict with family and friends, absenteeism, substance
abuse, turnover, burnout, health/medical problems, and
even premature death (Cheek, 1984; Cheek & Miller, 1983;
Lambert, Edwards, Camp, & Saylor, 2005a; Matteson &
Ivancevich, 1987; Mitchell, MacKenzie, Styve, & Gover,
2000; Slate & Vogel, 1997; Whitehead, 1989; Whitehead
& Lindquist, 1986; Woodruff, 1993). Job stress is harm-
ful to both staff and the prison organization, and its
antecedents need to be studied. Social support may be one
such antecedent.
1.2. Job involvement
Job involvement is the level of psychological identifica-
tion with a job (Kanungo, 1982a,b). As Elloy, Everett, and
Flynn (1992) note, it refers to a cognitive bond people can
form with their jobs. Paullay, Alliger, and Stone-Romero
(1994) point out that job involvement is a psychological
state wherein an individual “is cognitively preoccupied
with, engaged in, and concerned with one’s present job”
(p. 224).DeCarufeland Schaan (1990) notethat“an individ-
ual with a high degree of job involvement would place the
job at the center of his/her life’s interests. The well-known
phrase ‘I live, eat, and breathe my job’ would describe
someone whose job involvement is very high” (p. 86). Per-
sons with low job involvement focus on interests other
than their work (Hogan, Lambert, & Griffin, 2013). Elloy
et al. (1992) contend that job involvement is a measure
worklife’s quality.ChenandChiu(2009)point outthatpeo-
ple with “high job involvement are more independent and
self-confident–they not only conduct their work in accor-
dance with the job duties required by the company but are
also more likely to do their work in accordance with the
employees’ perception of their own performance” (p. 478).
Pfeffer (1994) contends that job involvement translates to
organizational effectiveness in the long run. Additionally,
job involvement is linked with reduced turnover intent
among jail staff (Lambert & Paoline, 2010). There is a need
to examine possible antecedents of prison staff job involve-
ment, and social support may be an important antecedent.
1.3. Job satisfaction
Locke (1976) defines job satisfaction as “a pleasurable
or positive emotional state resulting from the appraisal
of one’s job or job experiences” (p. 1300). Weiss (2002)
contends job satisfaction is the summative positive and
negativeemotions arising fromthe job. Simply, job satisfac-
tion is “the extent to which people liketheir jobs” (Spector,
1996, p. 214). In all these definitions, job satisfaction is
an affective/emotional response by an employee concern-
ing his/her particular job and whether the employee likes
the job. Job satisfaction is a salient and powerful work-
place concept. Low levels have been found to be associated
with absenteeism, turnover, and job burnout among cor-
rectional staff (Byrd, Cochran, Silverman, & Blount, 2000;
Whitehead & Lindquist, 1986; Wright, 1993). Conversely,
high levels are related to improved work performance,
organizationalinnovation,greatercompliance withorgani-
zational rules, greater support for rehabilitationof inmates,
and greater life satisfaction among correctional staff (Fox,
1982; Lambert, 2004; Lambert & Hogan, 2010; Lambert
et al., 2009; Lambert, Hogan, Paoline, & Baker, 2005b;
Robinson, Porporino, & Simourd, 1992). Given its signif-
icance, it is important to explore how different forms of
social support may be associated with job satisfaction.
1.4. Organizational commitment
Organizational commitment is the bond between an
employee and the employing organization (Mowday,
Porter, & Steers, 1982). Continuance and affective
http://localhost/var/www/apps/conversion/tmp/scratch_4/dx.doi.org/10.1016/j.soscij.2015.10.001http://localhost/var/www/apps/conversion/tmp/scratch_4/dx.doi.org/10.1016/j.soscij.2015.10.001
8/18/2019 Social Support’s Relationship to Correctional Staff Job Stress,Job Involvement, Job Satisfaction, And Organizationalc…
3/11
Please cite this article in press as: Lambert, E. G., et al. Social support’s relationship to correctional staff
job stress, job involvement, job satisfaction, and organizational commitment. The Social Science Journal (2015),
http://dx.doi.org/10.1016/j.soscij.2015.10.001
ARTICLE IN PRESSG Model
SOCSCI-1290; No.of Pages11
E.G. Lambert et al. / The Social Science Journal xxx (2015)xxx–xxx 3
commitment are two major forms of commitment, and
they differ in how the bond is formed (Hogan et al., 2013).
Continuance commitment refers to a bond that occurs due
to investments made in the employing organization, such
as social relationships, pension, salary and benefits, and
nontransferable job skills (Allen & Meyer, 1990). These
investmentsbond theperson to theorganization andresult
in a desire to stay with the organization (Garland, Hogan,
Kelley, Kim, & Lambert, 2013). With affective commitment,
a voluntary bond is formed by positive work experiences
(Allen & Meyer, 1990). The concept encompasses the ele-
ments of loyalty to theorganization, identification with the
organization, such as pride in the organization and inter-
nalization of organizational goals, and involvement in the
organization, such as personal effort made for the sake
of the organization (Hogan et al., 2013; Mowday et al.,
1982). With continuance commitment, a worker bonds
with the organization because they must do so. With affec-
tive commitment, a worker bonds with the organization
because they choose to do so (Meyer & Allen, 1991). Of the
two, affective commitment tends to have far more salient
outcomes for both employees and organizations. Further,
continuance commitment has been found to be associ-
ated with some negative consequences, as burned out
employees may feel trapped in the job and view quitting
as an attractive but untenable option. As such, the current
study focused on affective commitment, a concept gener-
ally viewed as a highly desirable outcome (Lambert, Kim,
Kelley, & Hogan, 2013). Among correctional staff, affective
commitment has been inversely linked with absenteeism
and turnover and positively associated with job perfor-
mance, life satisfaction, and organizational citizenship
behaviors, which is going beyond what is expected at work
(Camp, 1994; Culliver, Sigler, & McNeely, 1991; Lambert
et al., 2005a, 2013; Lambert, Hogan, & Griffin, 2008).
1.5. Social support
Social support refers to a network of connections with
other human beings that can provide assistance, support,
andhelpforaperson(Cohen, Underwood, & Gottlieb, 2000;
Harvey, 2014; Lambert, Altheimer, & Hogan, 2010). A lack
of social support can lead employees to feel isolated and
alone at work, exacerbatingthe effects of workplace stress-
ors (Ileffe & Steed, 2000). In addition, social support can
provide resources for people to deal more effectively with
stressors,ultimately reducing job stress (Neveu, 2007). Fur-
thermore, social support is a valuable resource because it
provides psychological support, assistance, feedback, and
motivation for employees (Lambert et al., 2010). Most peo-
pleare social creatures, and social support can be a positive
element for employees (Cohen et al., 2000), but its benefits
extend to both employees and employers. Social support
systems can allow innovation to occur, which can result
in quicker solutions to workplace issues, allowing work to
become more productive and enjoyable. Conversely, a lack
of socialsupportcan be seen as a form of resource depletion
for prison workers, detractingfrom positive outcomes from
work. Social support can be an excellent way to boost self-
confidence and self-esteem (Lambert et al., 2010). Social
support can increase the likelihood that a staff member
will bond with thejob. In addition,the positive experiences
with support can result in greater satisfaction from the job.
Based on the social exchangetheory, affectivecommitment
arises due to positive work experiences that allow individ-
uals to see the organization in a favorable light (Colquitt
et al., 2013).
There is a gap in the literature in terms of which
forms of social support are potential antecedents of prison
staff’s job stress, job involvement, job satisfaction, and
organizational commitment. Social support comes in dif-
ferent forms (Lee & Ashford, 1996; Whitehead & Lindquist,
1986). The social support’s major forms are administra-
tive support, supervisory support, coworker support, and
family andfriends support. Support fromcoworkers, super-
visors, and management represents intra-organizational
social support systems, while support from family and
friends represents an extra-organizational support system
(Lambert et al., 2010; Ray & Miller, 1994).
Administrative support refers to the staff’s percep-
tion that they are supported by the prison’s manage-
ment/administration (Garland & McCarty, 2006; Griffin,
2006). Administrative support can help staff deal with
stressors and being stressed from work. A lack of admin-
istrative support can make work difficult and straining.
Cheek and Miller (1983) contend that a lack of adminis-
trative social support can lead to staff feeling pressured
by both management and the inmates, resulting in being
between the proverbial rock and a hard place. They term
this situation a “double-bind” of stress for correctional
staff. Administrative support sends a message to staff that
they are valued and respected by organization. In addition,
administrative support can help staff be more successful
in their jobs, possibility resulting in greater identification
with the job and satisfaction. Administrative support, or
a lack thereof, for most staff will represent the organiza-
tion. Eisenberger, Huntington, Hutchison, and Sowa (1986)
report that employees “personify the organization, viewing
actions by agents of the organization as actions of theorga-
nization itself” (p. 504). In a sense, administrative support
reflects the level of commitment an organization has for
its employees. If staff perceive that theadministration sup-
ports them, it is easier for them to form a cognitive bond
with the organization. These postulations have a basis in
the literature. Administrative support is inversely associ-
ated with correctional staff job stress (Armstrong & Griffin,
2004; Auerbach, Quick, & Pegg, 2003; Griffin, 2006). Like-
wise, administrative support is positively associated with
job satisfaction and organizational commitment (Lambert
& Hogan, 2009).
Supervisory support focuses on staff perceptions of the
level that their supervisor supports them (Cullen et al.,
1985). Supervisory support can help remove stressors
or reduce their negative effects, which should translate
to lower stress from the job (Brough & Williams, 2007).
Supervisors can help staff see challenges in a more positive
light. Lee and Ashford (1996) note that “with the right kind
of supervisory support, workers may come to perceive
ambiguous role expectations as opportunities to carry
out their own initiatives (potential gains) rather than as
restrictions on their actions (certain losses)” (p. 131). On
the other hand, a lack of supervisory support can be a
http://localhost/var/www/apps/conversion/tmp/scratch_4/dx.doi.org/10.1016/j.soscij.2015.10.001http://localhost/var/www/apps/conversion/tmp/scratch_4/dx.doi.org/10.1016/j.soscij.2015.10.001
8/18/2019 Social Support’s Relationship to Correctional Staff Job Stress,Job Involvement, Job Satisfaction, And Organizationalc…
4/11
Please cite this article in press as: Lambert, E. G., et al. Social support’s relationship to correctional staff
job stress, job involvement, job satisfaction, and organizational commitment. The Social Science Journal (2015),
http://dx.doi.org/10.1016/j.soscij.2015.10.001
ARTICLE IN PRESSG Model
SOCSCI-1290; No.of Pages11
4 E.G. Lambert et al. / The Social Science Journal xxx (2015) xxx–xxx
straining and frustrating experience, which over time can
lead to job stress (Lambert & Hogan, 2009). Supervisory
support is important for employees to succeed in their
jobs. It is a resource which can help staff do their jobs in
an effective manner. Being productive tends to provide
people with a sense of pride and accomplishment. In the
end, prison staff are probably more likely to feel that the
job is meeting some of their needs and wants, and, as
such, will view the job in a more favorable light, increasing
the level of job involvement and satisfaction. Supervisors
are also seen as the prison organization’s representatives
(Lambert, 2004). Supervisory support sends a message
that staff are valued; thus, supervisory support likely
will increase favorable views of the organization, and, in
turn, strengthen the bond to it. On the contrary, a lack of
supervisory support can be a type of resource withdrawal
that leads to staff to question the organization, and, in
the long run, drive a wedge between staff and the prison
organization. Supervisory support is associated with lower
job stress and increased job satisfaction and organizational
commitment of correctional staff (Cullen et al., 1985;
Grossi, Keil, & Vito, 1996; Lambert, 2004; Lambert &
Hogan, 2009; Van Voorhis, Cullen, Link, & Wolfe, 1991).
Coworker support refers to the perceptions of support
from peers at work (Liaw, Chi, & Chuang, 2010). Coworker
support can help deal with stressors and facilitate man-
agement of minor problems quickly before they get out
of hand (Bakker & Demerouti Schaufeli, 2005). A lack of
peer support can be a trying experience in itself, raising
the degree of frustration encountered at work. Coworker
support should help reduce job stress among prison staff.
Support from work peers is also a resource for many peo-
ple (Lambert et al., 2010). It allows a chance to discuss
work issues and, at times, to “vent.” Cherniss (1980) points
out that “when one can discuss work experiences with
colleagues, those experiences often become more inter-
esting and meaningful” (p. 120). It provides a place for
advice and social companionship, as well as a support sys-
tem (Whitehead & Lindquist, 1986). In the end, coworker
support can make thejob enrichingand enjoyable, increas-
ing the level of job involvement and job satisfaction for
staff. A lack of support for coworkers could lead to less
identification with the job, as well as having unmet needs
from the job. Positive experiences from coworker support
may spillover and allow staff to seethe prison organization
in a more positive light, promoting greater attachment to
the organization. Having positive relations with coworkers
is associated with increased satisfaction and lowered job
stress among jail staff (Paoline, Lambert, & Hogan, 2006),
and lower job stress among Australian prison staff (Dollard
& Winefield, 1998).
Family and friends social support deals with the per-
ception by staff that their non-work significant others
care and help them with work matters (Lambert et al.,
2010). Family and friends are the primary social support
for people outside the workplace (Adams, King, & King,
1996). This form of support can help individuals deal
with work problems (Kurtz, 2008). Working in a prison
can be a trying experience (Armstrong & Griffin, 2004).
Having support at home may help reduce stress from
work. This form of support may also provide a safe haven
for staff to escape the trying work experiences, even if it is
a temporary respite, and, in turn, may help alleviate stress
from the job. In addition, having support from family and
friends for working in a prison may give a person a sense
of pride, allowing them to have a greater bond with the job
and the organization. In addition, having positive social
support from family and friends concerning the job may
result in positive feelings about their lives and their jobs,
spilling over to raise job satisfaction levels (Kwok, Cheng,
& Wong, 2014). Finally, family and friends support can
allow prison staff to lead a more balanced life, which, in
the end could result in general positive feelings that help
raise the levels of job involvement, job satisfaction, and
affective commitment (Cordes & Dougherty, 1993).
1.6. Research focus
While limited research exists on the effects of the dif-
ferent forms of social support on correctional staff, gaps
remain to be addressed. First, there has not been sufficient
research to date to uphold derive firm conclusions. Sec-ond, there has been far more research on the effects of
supervisory support than there has been for administra-
tion, coworker,and friends and family support. There needs
to be more research on all four forms of social support
among prison staff in order to understand the role social
support plays. Third, no published study has examined the
relationships of the four forms of support with job stress,
job involvement, job satisfaction, and organizational com-
mitment among prison staff in the same study. As a result,
theknowledge about theeffects of socialsupport in limited
and fragmentary. For example, one form of support may
be more important in shaping a particular outcome area
among prison staff than the other forms of support. Thisstudy’s purpose is to address these gaps in knowledge.
2. Method
2.1. Participants
All the available staff at a private Midwestern prison for
juvenile offenders were provided a survey packet, which
contained a cover letter, survey, and return envelope. The
cover letter explained the study’s purpose, that participa-
tion was voluntary, how to participate in the study, how to
returnthe surveys,and stressedthatall responseswould be
anonymous and would be kept confidential. At the time of the survey, there were 220 employees assigned to work at
the prison, but due to leave, such as sick leave and admin-
istrative leave, only 200 were available to be provided the
survey packet. A total of 160 usable surveys were returned,
which represents a response rate of 80% of those given the
packet and 72.7% of those who worked at the prison1. The
1 The survey measured a wide array of workplace perceptions, views,
attitudes, intentions, and behaviors of employees. Because of the wealth
of information from the survey, other studies have been conducted using
differentparts of the survey. The full citations of thesestudiesare available
upon request. None of these previous studies examined the impact of the
fourforms of social support on jobstress,job involvement,job satisfaction,and organizational commitment together.
http://localhost/var/www/apps/conversion/tmp/scratch_4/dx.doi.org/10.1016/j.soscij.2015.10.001http://localhost/var/www/apps/conversion/tmp/scratch_4/dx.doi.org/10.1016/j.soscij.2015.10.001
8/18/2019 Social Support’s Relationship to Correctional Staff Job Stress,Job Involvement, Job Satisfaction, And Organizationalc…
5/11
Please cite this article in press as: Lambert, E. G., et al. Social support’s relationship to correctional staff
job stress, job involvement, job satisfaction, and organizational commitment. The Social Science Journal (2015),
http://dx.doi.org/10.1016/j.soscij.2015.10.001
ARTICLE IN PRESSG Model
SOCSCI-1290; No.of Pages11
E.G. Lambert et al. / The Social Science Journal xxx (2015)xxx–xxx 5
prison housed approximately 450 juvenile offenders under
the age of 20 who were juveniles sentenced in adult court.
The prison was a high security closed facility.
Employees came from all positions except administra-
tion. Demographic characteristic of the respondents are
listed in Table 1. Institutional records indicate that, at
the time of the survey, the demographic breakdown of
the prison’s staff was approximately 81% White and 61%
male; the staff’s median age was about 35, the median
tenure was about 20.64 months, and about two-thirds of
theemployees held a custody position.As such, therespon-
dents appear to be demographically representative of the
staff at the private correctional facility.
2.2. Variables
2.2.1. Dependent variables
The dependent variables are additive indexes mea-
suring job stress, job involvement, job satisfaction, and
organizational commitment. All the items were answered
using a five-point Likert response scale ranging from
strongly disagree (coded 1) to strongly agree (coded 5).
The responses to the items were summed together to form
indexes, and Cronbach’s alpha values were calculated and
are listed in Table 1. The job stress index comprises five
items from Crank, Regoli, Hewitt, and Culbertson (1995).
For example, two job stress items are “I am usually under a
lot of pressure when I am at work” and “When I’m at work
I often feel tense or uptight.” The job involvement index
comprises three items adapted from Kanungo (1982a,b).
For example, two job involvement items are “I live, eat,
and breathe my job” and “The most important things that
happento me inmy life occur at work.” Five items, adapted
from Brayfield and Rothe (1951), are used to measure job
satisfaction. For example, two job satisfaction items are “I
like my job better than the average worker does” and “Most
days I am enthusiastic about my job.” The affective orga-
nizational commitment index comprises six items from
Mowdayet al.(1982). For example,two commitment items
are “I feel very little loyalty to this prison” (reverse coded)
and “I find that my values and the prison’s values are very
similar.”
2.2.2. Independent variables
Fourtypes of support are measured:administrative sup-
port, supervisory support, coworker support, and family
and friends support. The support items were answered
using a Likert scale ranging from strongly disagree (coded
1) to strongly agree (coded 5), and the responses to the
items were summed together to form indexes. See Table 1
for Cronbach’s alpha values. The administrative support
index comprises two items. For example, one item is “For
the most part, management at this prison supports its
workers.”The index for supervisory support comprisesfour
items. For example, three supervisory support items are
“Supervisors at this prison are supportive of employees,”
“At this facility, supervisors often criticize employees over
minor things” (reverse coded), and “My supervisor looks
out for my personal welfare.” The coworker support index
comprises six items. For example, three of the items are
“I am able to discuss problems with my coworkers,” “My
coworkers provide me support in solving personal prob-
lems”, and “The people I work with are friendly.” An index
for support for family and friends comprises two items. For
example, one item is “When my job gets me down, I know
that I can turn to family and friends for support.” Finally,
the personal characteristics of gender, age, tenure, posi-
tion, educational level, race, and supervisory status were
included in the study more as control than explanatory
variables. See Table1 for how these personalcharacteristics
are coded.
3. Results
The descriptive statistics for the variables in this study
are presented in Table 1. There appears to be significant
variationin both the dependent and independent variables.
None of the variables are constants. The median and mean
are similar to one another for the variables, suggesting
that the variables were normally distributed. Cronbach’s
alpha, a measure of internal reliability, was above .70 for all
the indexes, which indicates acceptable internal reliability
(Gronlund, 1981). The items used to measure job stress, job
involvement, job satisfaction, and organizational commit-
mentwere enteredintoan exploratory factor analysisusing
generalized least squares method with an equamax rota-
tion. The items for each latent concept loaded upon same
factorwithfactor loading scoresof .41or higher, andmostly
.50 or higher. For example, thethreejob involvement items
loaded on a single factor with factor loading scores of .61,
.84, and .83. The four factors explained approximately 64%
of the variance of the items. Similarly the 14 social sup-
port items were entered into an exploratory factor analysis
using generalized least squares with an equamax rotation.
Four factors explained for about 55% of the variance. The
items for each latent concept load upon same factor with
factor loading scores of .41 or higher, and mostly .50 or
higher. For example, the two administrative support items
loaded on one factor, with .67 and .61 factor loading scores.
The Pearson’s correlation coefficient results are pre-
sented in Table 2. Tenure, administrative support, super-
visory support, and coworker support have statistically
significant correlations with job stress. Increases in tenure
are associated with higher levels of stress from the job.
Conversely, increases in administrative, supervisory, and
coworker support are related to lower reported stress from
the job. Gender, age, position, educational level, race, and
family and friends support each have non-significant cor-
relations with job stress.
Position, supervisory status, administrative support,
supervisory support, and coworker support have sig-
nificant correlations with job involvement. Correctional
officers report, on average, lower levels of involvement
with thejob as compared to non-custody staff. Supervisors,
in general, report higher job involvement than their non-
supervisory counterparts. Administrative, supervisory, and
coworker support have positive correlations, which means
an increase in each type of social support is linked with
a rise in job involvement. Gender, age, tenure, educa-
tional level, race, and family and friends support have
non-significant correlations.
http://localhost/var/www/apps/conversion/tmp/scratch_4/dx.doi.org/10.1016/j.soscij.2015.10.001http://localhost/var/www/apps/conversion/tmp/scratch_4/dx.doi.org/10.1016/j.soscij.2015.10.001
8/18/2019 Social Support’s Relationship to Correctional Staff Job Stress,Job Involvement, Job Satisfaction, And Organizationalc…
6/11
Please cite this article in press as: Lambert, E. G., et al. Social support’s relationship to correctional staff
job stress, job involvement, job satisfaction, and organizational commitment. The Social Science Journal (2015),
http://dx.doi.org/10.1016/j.soscij.2015.10.001
ARTICLE IN PRESSG Model
SOCSCI-1290; No.of Pages11
6 E.G. Lambert et al. / The Social Science Journal xxx (2015) xxx–xxx
Table 1
Descriptive statistics for study variables.
Variable Description Md Min Max Mean SD
Gender 41% women (coded 0) 1 0 1 .59 .49
59% men (coded 1)
Age Measured in continuous years 33 19 68 35.77 10.82
Tenure Tenure at the prison in months 17 1 53 20.64 13.84
Position 38% non-CO (coded 0) 1 0 1 .62 .49
62% CO (coded1)
Educ level 53% have no college degree (coded 0) 0 0 1 .47 .50
47% have a college degree (coded 1)
Race 21% Nonwhite (coded 0) 1 0 1 .79 .50
79% White (coded 1)
Supervisor 79% non-supervisors (coded 0) 0 0 1 .21 .41
21% supervisors (coded1)
Admin support 2 item additive index, ˛= .84 6 2 10 5.31 2.21
Super support 4 item additive index, ˛= .80 11 4 20 10.92 3.67
Cowrk support 6 item additive index, ˛= .81 20 7 27 19.52 4.53
Family support 2 item additive index, ˛= .70 8 2 10 7.12 1.72
Job stress 5 item additive index, ˛= .82 14 6 25 14.29 4.51
Job involve 3 item additive index, ˛= .81 6 3 15 5.90 2.51
Job sat 5 item additive index, ˛= .92 19 5 25 18.34 4.67
Org commit 6 item additive index, ˛= .88 19 10 26 18.14 4.00
Note. Md stands for median value, Min stands for minimum value, Max stands for maximum value, SD stands for standard deviation, CO stands for
correctional officer, Educ stands for educational, Supervisor for supervisory status, Admin stands for administrative, Super stands for supervisory, Cowrkstands for coworker, Family stands for family and friends, involve stands for involvement, sat stands for satisfaction, Org commit stands for organizational
commitment, and stands for Cronbach’s alpha, a measure of internal reliability. For Educ level, having a college degree includes those who had earned
an Associate’s/Vocational degree, a Bachelor’s degree, and Graduate/Professional degree. The data for this table are from a survey of 160 staff members at
private Midwestern prison for juvenile offenders tried as adults.
Table 2
Pearson correlation matrix for study variables.
Variables 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.
1. Gender 1.00
2. Age −.17* 1.00
3. Tenure .00 .07 1.00
4. Position .28** −.53** −.20** 1.00
5. Educ lev −.13 .14 .05 −.29** 1.00
6. Race .08 −.05 .06 −.02 .05 1.007. Supervisor .12 .16* .39** −.47** .16* .04 1.00
8. Adm sup .02 .29** −.09 −.32** .09 .11 .29** 1.00
9. Sup −.05 .21** −.05 −.30** .15 .06 .27** .76** 1.00
10. Co sup .06 .13 −.13 −.26** .05 .11 .02 .49** .42** 1.00
11. Fam sup .06 −.05 −.04 −.01 −.08 −.14 .03 .20* .13 .35** 1.00
12. Job str −.12 −.02 .20* .07 .01 .04 −.10 −.49** −.51** −.31** −.11 1.00
13. Job inv .03 −.01 .12 −.27** .13 .05 .26** .37** .43** .21** −.10 −.34** 1.00
14. Job sat .07 .18* −.07 −.30** .18* .01 .28** .57** .57** .40** .09 −.70** .41** 1.00
15. Org com .05 .17* −.04 −.24** .12 .00 .23** .66** .68** .46** .16* −.53** .46** .75**
Note. See Table 1 for a description of the variables and how they were coded. Educ lev stands for educational level, Supervisor for supervisory status, Adm
sup stands for administrative support, Sup for supervisory support, Co sup for coworker support, Fam sup for family and friends support, Job str for job
stress, Job inv for job involvement, Job sat for job satisfaction, and Org com for organizational commitment. The data for this table are from a survey of 160
staff members at private Midwestern prison for juvenile offenders tried as adults.* p≤ .05.
** p≤ .01.
http://localhost/var/www/apps/conversion/tmp/scratch_4/dx.doi.org/10.1016/j.soscij.2015.10.001http://localhost/var/www/apps/conversion/tmp/scratch_4/dx.doi.org/10.1016/j.soscij.2015.10.001
8/18/2019 Social Support’s Relationship to Correctional Staff Job Stress,Job Involvement, Job Satisfaction, And Organizationalc…
7/11
Please cite this article in press as: Lambert, E. G., et al. Social support’s relationship to correctional staff
job stress, job involvement, job satisfaction, and organizational commitment. The Social Science Journal (2015),
http://dx.doi.org/10.1016/j.soscij.2015.10.001
ARTICLE IN PRESSG Model
SOCSCI-1290; No.of Pages11
E.G. Lambert et al. / The Social Science Journal xxx (2015)xxx–xxx 7
Table 3
Multivariate ordinary least squares regression results.
Variables Job stress Job involvement Job satisfaction Org commitment
B ˇ B ˇ B ˇ B ˇ
Gender −1.00 −0.11 0.41 0.08 .96 .10 .43 .05
Age 0.05 0.11 −0.06 −.26** −.01 −.03 −.01 −.02
Tenure 0.04 0.14 0.02 0.12 −.03 −.08 .01 .02
Position 0.14 0.02 −1.24 −.24
*
−.86 −.09 .00 .00Educ level 0.57 0.06 0.08 0.02 .77 .08 .11 .01
Race 1.13 0.1 −0.37 −0.06 −.87 −.08 −.89 −.09
Supervisor −0.11 −0.01 0.07 0.01 1.09 .10 .16 .02
Admin support −0.48 −.24* 0.17 0.15 .47 .22* .52 .28**
Super support −0.41 −.34** 0.21 .30** .37 .29** .44 .40**
Cowrk support −0.07 −0.07 0.04 0.07 .16 .16* .16 .18*
Family support 0.11 0.04 −0.31 −.21** −.17 −.06 −.07 −.03
R-Squared .35** .31** .42** .54**
Note. See Table 1 f or a description of thevariablesand howthey were coded.B stands for the unstandardized regression coefficient,ˇ for the standardized
regression coefficient, Educ for educational,Supervisorfor supervisory status, Admin for administrative,Super for supervisory, Cowrk for coworker, Family
for family and friends, and Org for organizational. The data for this table is from a survey of 160 staff members at private Midwestern prison for juvenile
offenders tried as adults.* p≤ .05.
** p≤ .01.
Age, position, educational level, supervisory status,
administrativesupport,supervisory support, and coworker
support have significant correlations with job satisfaction.
Older workers tend to report higher job satisfaction than
younger workers. Correctional officers tend to report lower
levels of job satisfaction compared to non-custody staff.
More highly educated and supervisory staff report higher
levels of job satisfaction compared to their less educated
and non-supervisory coworkers. Administrative support,
supervisory support, and coworker support have positive
correlations with job satisfaction, meaning an increase in
each type of social support is linked with a rise in job satis-
faction. Allthe other variables, includingfamily and friends
support, have non-significant correlations with job satis-
faction.
Age, position, supervisory status, administrative sup-
port, supervisory support, coworker support, and family
and friends support have positive correlations with
commitment. Correctional officers, in general, report
lower commitment than their non-custody counterparts.
Supervisors on average have higher levels of affectivecom-
mitment than non-supervisory staff. Increases in age and
the four forms of support are associated with greater levels
of commitment to the organization.
Along with the personal characteristics, the social sup-
port variables were entered as independent variables into
Ordinary Least Squares (OLS) regressionequations with job
stress, job involvement, job satisfaction, and organizational
commitment as the dependent variables. The results for
the four OLS regression equations are presented in Table 3.
While not reported, the tolerance and variance inflation
factor statistics indicate no problem of multicollinearity
among the independent variables. High multicollinearity
occurs when an independent variable shares a very large
part of its variance with the other independent variables in
the regression equation, and this impedes the estimation
of the effects of the affected independent variables on the
dependent variable (Berry, 1993).
For the job stress equation, the R-Squared statistic is
.35, which means that the independent variables explained
about 35% of the variance observed in the job stress vari-
able. Administrative support and supervisory support have
significant relationships with job stress, and direction of
the relationship is negative for both. In other words, an
increase in either form of support is associated with lower
reported stress from the job. The regression coefficients
in the columns in Table 3 estimates the magnitude of the effect of an independent variable on the dependent
variable. For the job stress equation, supervisory sup-
port has the greatest effect, followed by administrative
support.
About 31% of the variance in the job involvement
variable is accounted for by the independent variables.
Age, position, supervisory support, and family and friends
support each have significant relationships with job
involvement. Correctional officers generally report lower
job involvement. Increases in age and support from fam-
ily and friends are linked to lower reported involvement
in the job, while greater supervisory support is linked
to higher involvement. Among the significant variables,
supervisory support has the largest sized effect, followed
closely by age, position, and family and friends sup-
port.
Approximately 42% of the variance in the job sat-
isfaction index is accounted for in the OLS regression
equation.Administrative support, supervisory support,and
coworker support have significant associations with the
dependent variable, and the direction of the relation-
ship for each is positive. Among the significant variables,
supervisory support has the largest effect, followed by
administrative support and then coworker support.
Approximately 54% of the observedvariancein the orga-
nizational commitment variable is accounted for in the
multivariateequation. Administrativesupport, supervisory
support, and coworker support have a significant posi-
tive relationship with commitment to the organization.
http://localhost/var/www/apps/conversion/tmp/scratch_4/dx.doi.org/10.1016/j.soscij.2015.10.001http://localhost/var/www/apps/conversion/tmp/scratch_4/dx.doi.org/10.1016/j.soscij.2015.10.001
8/18/2019 Social Support’s Relationship to Correctional Staff Job Stress,Job Involvement, Job Satisfaction, And Organizationalc…
8/11
Please cite this article in press as: Lambert, E. G., et al. Social support’s relationship to correctional staff
job stress, job involvement, job satisfaction, and organizational commitment. The Social Science Journal (2015),
http://dx.doi.org/10.1016/j.soscij.2015.10.001
ARTICLE IN PRESSG Model
SOCSCI-1290; No.of Pages11
8 E.G. Lambert et al. / The Social Science Journal xxx (2015) xxx–xxx
Similar to the job satisfaction results, supervisory support
has the largest sized effect, followed by administrativesup-
port, and then coworker support.
4. Discussion and conclusion
As suggested by limited past research (Keinan &
Maslach-Pines, 2007; Lambert & Hogan, 2009; Whitehead
& Lindquist, 1986), data from this study indicate that
perceived social support is an important antecedent of
such favorable outcomes among correctional staff; how-
ever, the extent to which different types of social support
are related to the outcomes of job stress, job involvement,
job satisfaction, and organizational commitment–variables
known to be integral to agency operations and effective-
nessin corrections–was unknown until this study (Lambert
et al., 2005a, 2008; Pfeffer, 1994; Wright, 1993). Correc-
tional administrators seeking to improve these outcomes
among staff can exercise more direct control over some
forms of support, such as administrative and supervisory,
than other forms, such as coworkers and family/friends.
The existing literature is unclear about how administra-
tive, coworker, and family/friends support relates to job
outcomes; most past research has concentrated on super-
visory support. Finally, previous research on social support
hasbeen fragmentary, in that no singlestudy hasexamined
the construct in relationship to job stress, job involve-
ment, job satisfaction, and organizational commitment. In
short,knowledgeis very limited about whichtypes of social
support influence which job outcomes among correctional
staff.
Findings from the present study shed light on these
issues. First, social support is far more important in shap-
ingthe four job outcomes than thepersonal characteristics.
This is consistent with what research on correctional
staff has generally established about the importance of
work environment factors relative to personal factors with
regards to job attitudes and behaviors (Matz, Wells, Minor,
& Angel, 2013; Minor, Wells, Lambert, & Keller, 2014). Only
two of the personal variables studied here are significantly
related to a single job outcome, job involvement. Older
employees and those in correctional officer positions dis-
played lower job involvement. A plausible interpretation
of the age effect is that older correctional staff may be
more likely than younger ones to have developed capac-
ity to keep work in perspective relative to other priorities,
to have devised routines and strategies for getting work
done efficiently, and also to have developed alternative
interests and avenues of psychological identification and
for deriving life satisfaction, such as activities of their chil-
dren, hobbies, and so forth. By contrast, younger staff who
are newer to the world of full-time work may experience
their jobs, at least initially, as more engulfing. They may
alsobe relatively more dependent than their oldercounter-
parts on their jobs for money due to having had less time
to establish savings, and some might see engrossment in
work as a pathway to loftier career ambitions. Compared
with other jobs in prison, correctional officer work is usu-
ally more regimented with clearly demarcated times for
starting and ending shifts; thus, correctional officers could
be more likely to see their work as an occupation instead
of a profession, and this could help account for the lower
job involvement among correctional officers found in this
study. These, of course, are an untested postulations.
Job involvement’s other two significant correlates are
supervisory support and family support. Staff members
who feel their supervisor supports them are likely to iden-
tify more closely with their jobs and are less likely to feel
alienated than those who feel unsupported; these staff are
probably more likely to experience motivation to become
immersed in their work. Research indicates that a moti-
vating work climate, which is something that a supportive
supervisor can inspire, promotes job involvement (Brown
& Leigh, 1996).
By contrast, the data from this study show an inverse
relationship between support from family/friends and job
involvement. People who experience less support from
their family and friends may gravitate more closely toward
identifying with their jobs. Additionally, Lambert (2008)
indicates that greater levels of job involvement are linked
to a greater probability of work-induced family conflict
among correctional staff. That is, staff who feel overloaded
by the job role experience greater work-on-family conflict
(Lambert, Minor, Wells, Lambert, & Hogan, 2015). Social
support from familyand friends could function as a double-
edged sword with respect to job involvement. On the one
hand, such support could help the employee maintain a
proper perspective on work, causing the employee to bal-
ance work with other life domains and refrain from an
unhealthy obsession with his/her job. On the other hand,
relations with family and friends could lead the employee
to become overly distracted from work responsibilities by
outside interests and commitments (Hogan et al., 2013).
Further research is needed to improve understanding of
the relationship between family/friends support and job
involvement.
Consistent with past research (Brough & Williams,
2007; Cheek & Miller, 1983; Cullen et al., 1985; Griffin,
2006; Keinan & Maslach-Pines, 2007), correctional staff
in the present study who feel more supported by their
administrationand supervisors report lower job stress than
those whoperceive less support from these sources. As line
staff members’ bureaucratic superiors, administrators and
supervisors wield considerable power and influence, and
feeling that these persons are unsupportive can make the
job more stressful for staff. A lack of support from adminis-
trators and supervisors is likely to make staff feel devalued
and vulnerable, to exacerbate workplace stressors, and to
deprivestaff of a valuableresourcefor coping withstressors
(Ileffe & Steed, 2000; Lambert et al., 2010; Neveu, 2007).
Similarly, this study’s findings show that higher lev-
els of social support from administrators, supervisors, and
coworkers were significant antecedents of both job satis-
faction and organizational commitment. These results are
consistent with exchange theory principles (Colquitt et al.,
2013) as well as with findings from past studies (Lambert &
Hogan, 2009; Paoline et al., 2006). Feeling supported on the
job by bureaucraticsuperiors and peers can lead employees
to reciprocate by liking their jobs and developing loyalty
and solidarity with the organization. By contrast, perceiv-
ing these groups as non-supportive probably renders the
job unsatisfying and impairs bonding to the organization.
http://localhost/var/www/apps/conversion/tmp/scratch_4/dx.doi.org/10.1016/j.soscij.2015.10.001http://localhost/var/www/apps/conversion/tmp/scratch_4/dx.doi.org/10.1016/j.soscij.2015.10.001
8/18/2019 Social Support’s Relationship to Correctional Staff Job Stress,Job Involvement, Job Satisfaction, And Organizationalc…
9/11
Please cite this article in press as: Lambert, E. G., et al. Social support’s relationship to correctional staff
job stress, job involvement, job satisfaction, and organizational commitment. The Social Science Journal (2015),
http://dx.doi.org/10.1016/j.soscij.2015.10.001
ARTICLE IN PRESSG Model
SOCSCI-1290; No.of Pages11
E.G. Lambert et al. / The Social Science Journal xxx (2015)xxx–xxx 9
Thefindings of thepresentstudyare consistent with the
conclusion that administrative, supervisory, and coworker
support operate in tandem, with supervisory support
exerting the strongest and most consistent effects across
the four outcomes studied, followed by administrative
support, which is related to all outcomes except job
involvement, and finally support of coworkers, which is
related to job satisfaction and organizational commitment
only. As an extra-organizational source of social support
(Lambert et al., 2010; Ray & Miller, 1994), family/friends
support is not significantly related with the outcomes of
job stress, job satisfaction, or organizational commitment;
it is only related to job involvement. This pattern of find-
ings underscores the importance of differentiating forms of
social support and examining these forms simultaneously,
as not all types relate to the same job outcomes.
Moreover, it seems that intra-organizational sources
of support can be differentiated according to the posi-
tion of the parties providing, or not providing, support
to a staff member. Such differentiation can be conceptu-
alized as having vertical and lateral dimensions. On the
vertical dimension is support coming from those in the
organization positioned above the staff member, includ-
ing supervisors and higher level administrators, as well
as those positioned below the staff member. The lateral
dimension involves support from peer workers. Because
parties positioned above a staff member can be a pressing
source of job demands, such as daily expectations, mon-
itoring and evaluation of performance, and so forth, and
also a resource for coping with job strains, it makes sense
that low levels of perceived support from administrators
and supervisors might promote more job stress than a lack
of lateral support from coworkers. Similarly, support com-
ing from immediately above an employee in the vertical
chain in terms of supervisory support would logically have
a greater likelihood of leading to high job involvement
than support coming from higher, further removed levels
in terms of administrative support, or support from lateral
levels in terms of coworker support. Also, the present find-
ings suggest that both the vertical and lateral dimensions
of intra-organizational support are important for fostering
a liking for the job and commitment to the organization
among correctional staff. Although the supervisor variable
is notsignificantly related to theoutcomes examinedin this
study is indicated in Table 3, f uture research should attend
to how perceived support from subordinates is related to
these and other job outcomes.
The current study has limitations. This was a single
study of staff at a private prison. The vast majority of pris-
onsin theU.S. arepublic facilities (Lambert & Hogan,2009).
Results could differ among staff at a public prison. More-
over, the private prison held juveniles sentenced as adults,
which is a uniqueinmatepopulation notfoundat many pri-
vate or even public prisons. Dealing with juvenile offenders
is a differentfocusthan dealing with adult offenders. Treat-
ment was a major focus at this private prison as state law
requires educational services be provided to all inmates
under 18 years of age. The unique inmate population
and treatment efforts could have affected the associations
between the different forms of social support and the out-
come areas of job stress, job involvement, job satisfaction,
and organizational commitment. Research at other prisons
and types of correctional facilities, such as jails, public pris-
ons, adult facilities, and so forth, is needed to determine
whether the findings can be replicated. With replication,
the effects of different forms of social support on prison
staff can be understood more fully. Additionally, it remains
to be determined if effects are contextual and situational,
varying across different types of correctional facilities. The
current study uses a cross-sectional design of collectingthe
data at one point in time, which precludes demonstrat-
ing causal relationships. Longitudinal designs are needed
to demonstrate the casual process of the effects of dif-
ferent forms of social support on prison staff job stress,
job involvement, job satisfaction, and organizational com-
mitment. Future researchers should consider using more
in-depth measures of job involvement, administrative sup-
port, and family support, which were measured in this
study with either 2 or 3 items. Future research should
examine theeffectsof other forms of socialsupport, such as
subordinate support as mentioned above. Another type of
social support not examined here was community support
(Gerstein, Topp, & Correll, 1987).
Research is needed to determine the best ways to
build social support among staff. For example, future
studies could examine whether teamwork and integra-
tion help build social support, or whether social support
plays a role in other outcome areas, such as work per-
formance or turnover/turnover intent. Additional studies
are needed to generate a more complete understanding of
the antecedents and effects of social support among prison
staff.
Two decades ago, Cullen (1994) urged that the concept
of social support be construed as an organizing concept
for the field of criminology. He saw social support as a
theme common to various theories of criminal behavior
and viewed it as a means of helping unify the field. Cullen’s
argument is that a lack of social support is criminogenic.
The present study builds on a limited body of past research
to demonstrate that social support in its various forms is
integral to understanding not only criminal action, but also
pertinent job outcomes among people working in prisons.
The findings presented here suggest that low levels of sup-
port are also associated with undesirable outcomes among
correctional staff. There is a kind of vicious cycle here,
in that staff who feel unsupported and experience high
job stress, low job involvement, low job satisfaction, and
low organizational commitment seem unlikely to provide
much support to the offenders in their charge. In this man-
ner, lowsupport could beget lowsupport, which, following
Cullen’s logic, might decrease the chances of successful
reentry of ex-offenders into the community.
The results of this study imply that correctional poli-
cymakers and leaders can promote better job outcomes
among their staff and improved effectiveness of their orga-
nizations by directly providingadministrative and supervi-
sory support to staff and also by making efforts to encour-
age peer coworkers to support one another. Such support
can be instrumental/goal oriented or expressive/affective
in nature, delivered by individuals or groups, and delivered
through either formal or informal channels (Cullen, 1994).
Likewise, the findings of this study imply giving the
http://localhost/var/www/apps/conversion/tmp/scratch_4/dx.doi.org/10.1016/j.soscij.2015.10.001http://localhost/var/www/apps/conversion/tmp/scratch_4/dx.doi.org/10.1016/j.soscij.2015.10.001
8/18/2019 Social Support’s Relationship to Correctional Staff Job Stress,Job Involvement, Job Satisfaction, And Organizationalc…
10/11
http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0250http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0250http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0250http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0250http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0250http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0250http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0250http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0250http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0250http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0250http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0250http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0250http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0250http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0250http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0250http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0250http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0245http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0245http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0245http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0245http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0245http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0245http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0245http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0245http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0245http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0245http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0245http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0245http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0245http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0245http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0245http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0245http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0245http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0245http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0245http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0245http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0245http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0245http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0245http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0245http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0245http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0245http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0245http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0240http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0240http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0240http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0240http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0240http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0240http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0240http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0240http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0240http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0240http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0240http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0240http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0240http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0240http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0240http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0240http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0240http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0240http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0240http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0240http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0240http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0235http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0235http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0235http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0235http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0235http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0235http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0235http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0235http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0235http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0235http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0235http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0235http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0235http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0235http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0235http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0235http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0235http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0235http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0235http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0235http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0235http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0235http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0235http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0235http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0235http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0230http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0230http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0230http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0230http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0230http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0230http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0230http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0230http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0230http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0230http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0230http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0230http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0230http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0230http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0230http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0230http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0230http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0230http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0225http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0225http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0225http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0225http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0225http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0225http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0225http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0225http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0225http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0225http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0225http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0225http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0225http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0225http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0225http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0225http://link.springer.com/article/10.1007%2Fs10902-014-9522-7http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0215http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0215http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0215http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0215http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0215http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0215http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0215http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0215http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0215http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0215http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0215http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0215http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0215http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0215http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0215http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0215http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0215http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0215http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0215http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0210http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0210http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0210http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0210http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0210http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0210http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0210http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0210http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0210http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0210http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0210http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0210http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0210http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0210http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0210http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0210http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0210http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0210http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0210http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0210http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0210http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0205http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0205http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0205http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0205http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0205http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0205http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0205http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0205http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0205http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0205http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0200http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0200http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0200http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0200http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0200http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0200http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0200http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0200http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0200http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0200http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0200http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0200http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0200http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0200http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0200http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0200http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0195http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0195http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0195http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0195http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0195http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0195http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0195http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0195http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0195http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0195http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0195http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0195http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0195http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0195http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0195http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0195http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0195http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0195http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0195http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0195http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0195http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0195http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0195http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0195http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0190http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0190http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0190http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0190http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0190http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0190http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0190http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0190http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0190http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0190http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0190http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0190http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0190http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0190http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0190http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0190http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0190http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0190http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0190http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0190http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0190http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0190http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0190http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0190http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0190http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0190http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0185http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0185http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0185http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0185http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0185http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0185http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0185http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0185http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0185http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0185http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0185http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0185http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0185http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0185http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0185http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0185http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0180http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0180http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0180http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0180http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0180http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0180http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0180http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0180http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0180http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0180http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0180http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0180http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0180http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0180http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0180http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0180http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0180http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0180http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0180http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0180http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0175http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0175http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0175http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0175http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0175http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0175http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0175http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0175http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0175http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0175http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0175http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0175http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0175http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0175http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0175http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0175http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0175http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0175http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0175http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0175http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0170http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0170http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0170http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0170http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0170http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0170http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0170http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0170http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0170http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0170http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0165http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0165http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0165http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0165http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0165http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0165http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0165http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0165http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0165http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0165http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0165http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0165http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0165http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0165http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0165http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0165http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0165http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0165http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0165http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0165http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0165http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0165http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0165http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0165http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0160http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0160http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0160http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0160http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0160http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0160http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0160http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0160http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0160http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0160http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0160http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0160http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0160http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0160http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0160http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0160http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0160http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0160http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0160http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0160http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0160http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0160http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0155http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0155http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0155http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0155http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0155http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0155http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0155http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0155http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0155http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0155http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0155http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0155http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0155http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0155http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0155http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0155http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0155http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0155http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0155http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0155http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0155http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0155http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0155http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0155http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0155http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0155http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0155http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0155http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0155http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0150http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0150http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0150http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0150http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0150http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0150http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0150http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0150http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0150http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0150http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0150http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0150http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0150http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0150http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0150http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0150http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0150http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0150http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0150http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0150http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0150http://refhub.elsevier.com/S0362-3319(15)00085-3/sbref0150http://refhub.elsevier.com/S0362-
Top Related