Full Terms & Conditions of access and use can be found athttps://www.tandfonline.com/action/journalInformation?journalCode=wssr20

Journal of Social Service Research

ISSN: 0148-8376 (Print) 1540-7314 (Online) Journal homepage: https://www.tandfonline.com/loi/wssr20

Compassion Satisfaction Among Social WorkPractitioners: The Role of Work–Life Balance

Junghee Bae, Porter F. Jennings, Christi P. Hardeman, Eunhye Kim, MeganLee, Tenesha Littleton & Sherinah Saasa

To cite this article: Junghee Bae, Porter F. Jennings, Christi P. Hardeman, Eunhye Kim, MeganLee, Tenesha Littleton & Sherinah Saasa (2019): Compassion Satisfaction Among SocialWork Practitioners: The Role of Work–Life Balance, Journal of Social Service Research, DOI:10.1080/01488376.2019.1566195

To link to this article: https://doi.org/10.1080/01488376.2019.1566195

Published online: 27 Feb 2019.

Submit your article to this journal

Article views: 123

View Crossmark data

Compassion Satisfaction Among Social Work Practitioners: The Role ofWork–Life Balance

Junghee Baea, Porter F. Jenningsa, Christi P. Hardemana, Eunhye Kima, Megan Leea, Tenesha Littletona ,and Sherinah Saasab

aSchool of Social Work, University of Georgia, Athens, Georgia, USA; bSchool of Social Work, Brigham Young University, Provo,Utah, USA

ABSTRACTPrevious research has established that social work practitioners are especially vulnerable towork-related psychological distress and burnout due to the high-stress nature of the profes-sion, yet less research has focused on examining factors are associated with social workerretention. Emerging research on compassion satisfaction suggests that this factor could miti-gate professional burnout, yet there is a gap in research focusing explicitly on examiningcompassion satisfaction among social workers. To address this gap in knowledge, this quan-titative study collected survey data on practicing social workers who were alumni from alarge southeastern university (n¼ 120) to examine individual and organizational factors asso-ciated with compassion satisfaction. Multiple regression analyses revealed that higher levelsof emotional intelligence, perceived work autonomy, and perceived work–life balance wereassociated with an increase in compassion satisfaction among experienced, licensed socialwork practitioners. Findings have implications for how social work employers can promotecompassion satisfaction through cultivating emotional intelligence among practitioners,allowing social workers autonomy in decision making, and providing supportive work envi-ronments. Recommended directions for future research include longitudinal studies withlarge sample sizes that expound research on compassion satisfaction by examining the roleof additional factors, such as client population, job role characteristics, and super-visor support.

KEYWORDSSocial work practitioners;compassion satisfaction;work–life balance; workautonomy; emotionalintelligence

Introduction

Social work is a profession committed to address-ing some of the most challenging social problemsin the world, including oppression, discrimin-ation, poverty, and mental health issues.

When addressing these complex issues, socialwork practitioners often experience stressful workenvironments where they are met with highneeds and limited resources. These factorsincrease their risk of experiencing job-relatedpsychological distress and burnout (Adams,Boscarino, & Figley, 2006; Bride, 2007; Sanchez-Moreno, de La Fuente Roldan, Gallardo-Peralta& de Roda, 2015). Burnout is broadly defined as“a psychological syndrome of emotional exhaus-tion, depersonalization, and reduced personalaccomplishment that can occur among individu-als who work with other people in some

capacity” (Zalaquett & Wood, 1997, p. 192), andcan manifest in symptoms such as “chronic phys-ical and emotional fatigue, the depletion of theempathic resources and of compassion, boredom,cynicism, diminution of enthusiasm, temporarydistress and depression” (Diaconescu, 2015,p. 57). These adverse responses can also lead pro-fessionals to experience feelings of professionalfailure that can be expressed in negative attitudestoward the people they serve. Therefore, under-standing potential protective factors that preventburnout is necessary not only for the well-beingof social work practitioners, but also for the pro-tection of clients they serve.

Although much of the existing research insocial work has focused on risk factors thatincrease burnout (such as work overload, insuffi-cient resources, traumatic stress, compassion

CONTACT Junghee Bae jb36946@uga.edu School of Social Work, The University of Georgia, 279 Williams Street, Athens, GA 30602, USA.� 2019 Taylor & Francis Group, LLC

JOURNAL OF SOCIAL SERVICE RESEARCHhttps://doi.org/10.1080/01488376.2019.1566195

fatigue, and negative relationship with coworkersand managers; Butler, Carello, & Maguin, 2017;Kim, 2011; Lloyd, King, & Chenoweth, 2002;McFadden, 2018; Simon, Pryce, Roff, &Klemmack, 2006), a few studies have used astrengths-based lens to focus on examining pro-tective factors that mitigate burnout. Such studieshave identified a negative relationship betweenburnout and compassion satisfaction amongsocial work practitioners (Cummings, Singer,Hisaka, & Benuto, 2018, Stamm, 2002), and havesuggested that compassion satisfaction may miti-gate the undesired effects of burnout in the socialwork profession (Conrad & Kellar-Guenther,2006; Cummings et al., 2018). However, despitethe empirical evidence that increased rates ofcompassion satisfaction are associated with lowerrates of burnout (Butler et al., 2017; Rossi et al.,2012; Wagaman, Geiger, Shockley & Segal, 2015),there is a limited amount of research that hasexamined compassion satisfaction as a potentialprotective factor associated with burnout reduc-tion among social work practitioners specifically.Research examining factors that promote com-passion satisfaction has important implications,as this knowledge can be used to protect socialwork practitioners against burnout and improvethe well-being of both social work practitionersand the clients they serve. To respond to theabove identified gap in knowledge, the aim ofthis study is to examine factors that can increasecompassion satisfaction among social workpractitioners.

Compassion satisfaction

Compassion satisfaction is defined as “the level ofsatisfaction helping professionals find in their joband the degree to which they feel successful intheir job” (Conrad & Kellar-Guenther, 2006,p. 1074), and presents as “motivation, stamina,interest, and a sense of accomplishment in aidingclients” (Wagaman et al., 2015, p. 203). This con-cept encompasses the rewards social workersreceive from observing improvements in clientfunctioning and growth, as the practitioner oftenshares in experiencing the clients’ feelings ofempowerment, energy, and exhilaration (Rossiet al., 2012). When social workers experience

compassion satisfaction, they also often experi-ence increased positive perceptions of their work,which can result in enhanced professional prac-tice (Wagaman et al., 2015).

Factors associated with compassion satisfaction

Studies have identified several primary factorsassociated with the concept of compassion satis-faction, including emotional intelligence, workautonomy, and work–life balance (Athey et al.,2016; Ingram, 2013; Saragih, 2011; Skaalvik &Skaalvik, 2014).

Emotional intelligence

Emotional intelligence refers to the ability to con-trol individual emotions, the ability to influenceothers’ emotions, and the utilization of emotionsin problem-solving contexts (Balakrishnan &Saklofske, 2015). It has been identified as a “keypredictor of adaptive coping and interpersonalsuccess in the workplace,” and is thought to“enhance life satisfaction and psychological andphysical health” (Kinman & Grant, 2011, p. 263).Emotional intelligence has been established ascore component of social work practice, yet littleresearch has examined the specific relationshipbetween emotional intelligence and compassionsatisfaction among social workers (Ingram, 2013).

Work autonomy

Work autonomy refers to the amount of controlthat individuals believe they have in their job tomake work-related decisions and includes issuessuch as the perception of “workload, control,reward, community, fairness, and values”(Whitebird, Asche, Thompson, Rossom &Heinrich, 2013). Research on helping professio-nals (such as nurses and teachers) concludes thatwork autonomy has been known to increase jobsatisfaction (Athey et al., 2016; Skaalvik &Skaalvik, 2014). Additionally, research has foundhigh levels of work autonomy to be associatedwith factors that positively impact job perform-ance, including increased motivation and satisfac-tion, more control over job stress, and reducedwork and family conflict (Saragih, 2011).However, the effects of work autonomy on

2 J. BAE ET AL.

compassion satisfaction among social work prac-titioners have not yet been comprehen-sively explored.

Work–life balance

Work–life balance is an additional factor that hasbeen associated with job satisfaction (Orkibi &Brandt, 2015). As professions experience theimpact of demographic shifts that includeyounger generations of employers within a soci-ety of changing gender and familial norms, con-cerns regarding manageable lifestyle factors thatallow for a balance between work and personallife have been placed at high priority in choosinga profession (Keeton, Fenner, Johnson &Hayward, 2007). However, although recent stud-ies have shown that work–life balance plays acritical role in increasing general job satisfaction(Athey et al., 2016; Skaalvik & Skaalvik, 2014),there has yet to be extensive research conductedon the impact of work–life balance among socialwork practitioners specifically.

Summary of the literature and identified gaps

In summary, review of the literature reveals thatthere is empirical evidence to suggest that com-passion satisfaction could mitigate burnoutamong social workers (Cummings et al., 2018;Stamm, 2002), but there is a lack of researchcomprehensively examining factors that can pro-mote compassion satisfaction in this specific pro-fession (Butler et al., 2017; Rossi et al., 2012;Wagaman et al., 2015). Furthermore, while emo-tional intelligence, work autonomy, and work–lifebalance have been associated with increased com-passion satisfaction among general employees,there is a lack of studies examining the associ-ation of these factors among social work practi-tioners. Based on these identified gaps in theliterature, the purpose of this study is to exploreindividual and organizational factors that areassociated with compassion satisfaction amongsocial work practitioners. Specifically, this studyexamines the relationship between emotionalintelligence, work autonomy, work–life balance,and compassion satisfaction to generate findings

that can enhance social work practice and guidefuture research in this area.

Theoretical foundations

To more thoroughly understand the factors thatare related to compassion satisfaction, this studyapplied two prominent theories from the litera-ture to identify factors that promote compassionsatisfaction: the job-demands resource (JD-R)model and the conservation of resources(COR) theory

The job-demands resource model

The JD-R model (Demerouti, Bakker, Nachreiner& Schaufeli, 2001) has been used to examine theinfluence of internal and external resources onemployee wellbeing (Demerouti & Bakker, 2011).This model proposes that when job demands(e.g., physical workload, time pressure) are high,and job resources (e.g., rewards, supervisor sup-port) are low, employees respectively experienceexhaustion and disengagement that result inburnout. However, while the JD-R model pro-vides a framework for understanding burnout asa consequence of an imbalance between jobdemands and resources (Bakker & Demerouti,2007), it fails to highlight the role of internalresources (e.g., strong coping skills, social sup-port), which research has shown to be correlatedwith the burnout process (Alvaro et al., 2010).Therefore, as suggested by Ojedokun andIdemudia (2014), this study builds upon the JD-Rmodel by incorporating COR theory’s inclusionof personal and social resources in order to morethoroughly examine compassion satisfaction.

The conservation of resources theory

The COR theory (Hobfoll, 1989) is an integratedstress model positing that individuals seek toobtain and conserve resources, and the loss of orthreat to such resources in the workplace resultsin stress that may lead to job dissatisfaction.However, while the COR theory does examinethe personal and social resources not included inthe JD-R model, it only examines factors in rela-tion to burnout, and not the inverse relationsresulting in compassion satisfaction.

JOURNAL OF SOCIAL SERVICE RESEARCH 3

Integrated resource model on compassionsatisfaction

Due to the above-identified gaps in these existingtheories, this study proposes a novel integratedresource model on compassion satisfaction. Thismodel suggests that emotional intelligence (e.g.,individual resource), work autonomy (e.g., organ-izational resource), and work–life balance (e.g.,individual and organizational resources) are allfactors that may influence compassion satisfac-tion. Figure 1 presents this integrated conceptualmodel to enhance understanding of the combinedinfluence of these variables on compassionsatisfaction.

Research questions and hypotheses

Informed by the above-discussed theoreticalfoundations and the integrated resource modelon compassion satisfaction, the following researchquestion was generated to guide this study: Whatfactors determine compassion satisfaction amongsocial workers? Based on the review of existingresearch, the following hypotheses emerged:

1. Emotional intelligence will be positively associ-ated with compassion satisfaction.

2. Work autonomy will be positively associatedwith compassion satisfaction.

3. Work–life balance dimensions will be associ-ated with compassion satisfaction as follows:work interference with personal life and per-sonal life interference with work will be nega-tively associated with compassion satisfaction,and work/personal life enhancement will bepositively associated with compassionsatisfaction.

Methods

Sample

This study utilized a convenience sampling strat-egy consisting of social work graduates from alarge southeastern university who graduatedbetween 1961 and 2016, and who registered theiremail on the alumni networking database. All reg-istered alumni were emailed an invitation to com-plete the survey voluntarily, and a preliminaryqualifying question allowed only currently practic-ing social work alumni to complete the survey. Atotal of 120 participants responded (N¼ 120). 91%(n¼ 109) of participants were female with anaverage of 36.5 years of age (SD¼ 12.6;range¼ 23–75). 86.7% (n¼ 104) of the partici-pants identified as white, 8.3% (n¼ 10) identifiedas black, 5% (n¼ 6) identified as “other.”Approximately 77% (n¼ 92) of the participantsreported current active social work licensure sta-tus, and 23% (n¼ 28) did not report active licen-sure status. 65% (n¼ 78) of participants reportedengagement in direct practice with clients at thetime of the survey, while 35% (n¼ 42) of partici-pants engaged in indirect practice.

Instruments

The survey questionnaire was created using exist-ing validated instruments designed to examinethe following factors: compassion satisfaction,emotional intelligence, work autonomy, work–lifebalance, and socioeconomic characteristics.

Compassion satisfaction

One of the three concepts measured in theProfessional Quality of Life Scale (ProQOL; Stamm,2010) is compassion satisfaction. This concept was

Emotional Intelligence

Work Autonomy

Work-life Balance

Compassion Satisfaction

Figure 1. Integrated resource model on compassion satisfaction.

4 J. BAE ET AL.

assessed using participant ratings in response tostatements in the ProQOL related to key constructsof compassion satisfaction using a five-point Likertscale ranging from 1 (never) to 5 (very often). Anexample of a statement on this scale is: I get satis-faction from being able to [help] people. Higherscores indicate that a person finds greater satisfac-tion in his or her ability to be an effective caregiverin his or her professional role. The Cronbach’salpha for our sample was .92, which was compar-able to that of .88 found by Stamm.

Emotional intelligence

Emotional intelligence was measured using theBrief Emotional Intelligence Scale (BEIS-10;Davies, Lane, Devonport, & Scott, 2010). Theshortened scale was adapted from the original 33-item Schutte Emotional Intelligence Scale (SEIS;Schutte et al., 1998) and Salovey and Mayer’s(1990) Emotional Intelligence Framework. TheBEIS-10 includes 10 statements (e.g., I know whymy emotions change) and uses a five-point Likertscale ranging from 1 (strongly agree) to 5(strongly disagree). Responses were reverse codedsuch that high scores on all items indicatedhigher emotional intelligence (a¼ .68).

Work autonomy

Work autonomy was measured using Breaugh’s(1999) Work Autonomy Scale. This scale consistsof nine items on a seven-point Likert scale rang-ing from 1 (strongly agree) to 7 (strongly dis-agree). The scale measures three dimensions ofautonomy: methods autonomy (e.g., I am allowedto decide how to go about getting my job done),scheduling autonomy (e.g., I have control over thescheduling of my work), and criteria autonomy(e.g., My job allows me to modify the normal waywe are evaluated so that I can emphasize someaspects of my job and play down others). Higherscores reflected greater levels of work autonomy.The Cronbach’s alpha was .91, which was com-parable to the reliability reported by Breaugh.

Work–life balance

Hayman’s (2005) Work–life Balance Scale wasused to measure work–life balance. This scale

consists of three dimensions: work interferencewith personal life (WIPL; e.g., personal life suffersbecause of work), personal life interference withwork (PLIW; e.g., personal life drains me ofenergy for work), and work/personal life enhance-ment (WPLE; e.g., personal life gives me energyfor my job). The scale consists of fifteen itemsand is measured on a five-point Likert scale rang-ing from 1 (not at all) to 5 (almost all of thetime; Agha, Azmi & Irfan, 2017). The WIPL andPLIW sub-scale items with higher means indicatelower levels of work–life balance. However,higher means on the WPLE sub-scale indicatehigher levels of perceived WLB. The Cronbach’salpha coefficients in the sample for WIPL (.79),PLIW (.84), and WPLE (.80) were comparable tothose reported by Hayman.

Sociodemographic variables

Several sociodemographic factors were included ascontrol variables in analyses: gender (male¼ 0,female¼ 1), race (White¼ 0, Black¼ 1, otherraces¼ 2), marital status (single¼ 0, married¼ 1,other =2), length of professional social workpractice (in years), and currently a licensed socialworker (yes¼ 0, no¼ 1). Also, practice role indi-cating whether participants worked directly withclients or in indirect services (such as manage-ment, supervision, planning, etc.) was coded as abinary measure (direct practice¼ 0, indir-ect practice¼ 1).

Procedures

Data were collected from January 9, 2017 toFebruary 5, 2017 through an online survey usingQualtrics (2016). A link to the survey was sent to2,100 potential participants who completed bach-elor’s degree in social work (i.e., BSW) or/andmaster’s degree in social work (i.e., MSW) at alarge southeastern university between 1961 and2016 via email addresses obtained from an offi-cial school alumni list. Following standard proto-col for online survey administration suggested byDillman (2000), an initial pre-notice invitationemail followed by another email a week laterwith the cover letter/invitation, and link toparticipate was used to collect the data. An

JOURNAL OF SOCIAL SERVICE RESEARCH 5

invitation to participate along with the surveylink was also posted on social media (Twitter andFacebook) at the time the invitation email wassent out. To decrease respondent burden andincrease the likelihood of participation, the surveyinstrument was designed to not be too long,approximately 15 minutes to complete (Dillman,Smyth, & Christian, 2014).

All analyses were conducted using IBM SPSS,Version 22. Univariate, bivariate and multivariateanalyses were performed on the data. One-wayanalyses of variance (ANOVA) and Pearsoncorrelations were used to examine the bivariaterelationship between the dependent variable andeach independent variable. Multiple regressionanalysis was used to analyze the relationshipbetween the continuous dependent variable,compassion satisfaction, and the independentvariables (i.e., emotional intelligence, workautonomy, work–life balance) while controllingfor sociodemographic characteristics. A poweranalysis using G�Power 3.1 software suggested asample size of 95 in order to detect medium-sized effects given our modeling strategy (linearmodeling approach with 11 predictors, detectionof effect sizes .30 or greater), which was achievedin our sample size (Faul, Erdfelder, Buchner &Lang, 2009).

Results

Descriptive analysis

A brief summary of descriptive statistics is pre-sented in Table 1. Overall, participants reportedlevels of compassion satisfaction at a mean scoreof 38 (SD¼ 12.2). Respondents in this studyscored a mean of 37.1 (SD¼ 12.4) on emotionalintelligence and 5.1 (SD¼ 1.3) on work auto-nomy. Regarding the sub-scales of work–lifebalance, they reported an average work interfer-ence with personal life of 2.5 (SD¼ 0.7), personallife interference with work of 1.8 (SD¼ 0.7) andwork/personal life enhancement of 3.3 (SD¼ 0.8).

Bivariate analysis

The results of bivariate analysis between compas-sion satisfaction and each independent variableare shown in Table 2. The findings showed that

at the bivariate level, emotional intelligence, workautonomy, work/personal life enhancement, yearsof practice experience, and active social worklicensure status were all factors that positivelycorrelated with compassion satisfaction (p <. 05).Work interference with personal life and personallife interference with work were negatively corre-lated with compassion satisfaction (p < .05).

Multivariate analyses

Multivariate models of compassion satisfactionare presented in Table 3. In the first model whichincluded emotional intelligence as an independ-ent variable, emotional intelligence had a

Table 1. Sample demographic characteristics and distributionof study variables.

n ( %) M SD Range

Compassion Satisfaction 119 38.0 12.2 0–50Emotional Intelligence 118 37.1 12.4 0–50Work Autonomy 100 5.1 1.3 1–7Work Interference with Personal Life 108 2.5 0.7 1.29–5Personal Life Interference with Work 107 1.8 0.7 1–4.25Work/Personal Life Enhancement 108 3.3 0.8 1.75–5Age 119 36.5 12.6 23–75Years of Practice 118 10.9 11.5 0–50Gender Male 8 (6.7)

Female 109 (90.8)Race White 104 (86.7)

Black 10 (8.3)Other Races 6 (5.0)

Marital Status Single 44 (36.7)Married 60 (50.0)Other 16 (13.3)

Currently Licensed Yes 92 (76.7)No 28 (23.3)

Practice Role Direct Service 78 (65.0)Indirect Service 42 (35.0)

Table 2. Bivariate analysis between compassion satisfactionand independent variable.Variables Pearson’s r M t or (F)

Emotional Intelligence 0.808��Work Autonomy 0.335��Work Interference with Personal Life �0.254��Personal Life Interference with Work �0.491��Work/Personal Life Enhancement 0.493��Years of Practice 0.226�Gender Male 38.50 0.051

Female 38.28Race White 37.84 (0.779)

Black 35.80Other Races 43.50

Marital status Single 35.74 (2.024)Married 40.17Other 35.63

Currently Licensed Yes 39.95 2.475�No 31.50

Practice role Direct Service 37.05 �1.100Indirect Service 39.62

�p < .05,��p < .01.

6 J. BAE ET AL.

significant, positive association with compassionsatisfaction (B¼ 0.729, p < .001) after controllingfor sociodemographic variables. In the secondmodel (which included work autonomy as anindependent variable) individuals who hadhigher work autonomy experienced a significantlyhigher level of compassion satisfaction (B¼ 1.442,p < .01) when controlling for sociodemographicvariables. In the third model (which includedthree sub-scales of work–life balance) individualswho had higher personal life interference withwork had a significantly lower level of compas-sion satisfaction (B¼�3.024, p < .01). However,individuals who had higher work/personal lifeenhancement reported a higher level of compas-sion satisfaction (B¼ 2.953, p < .001). In theintegrated model which included all threeindependent variables in this study, work auto-nomy (B¼ 1.153, p < .01) and work/personal lifeenhancement (B¼ 2.534, p <.01) were positivelyassociated with compassion satisfaction. Also,there was a negative association between personalinterference with work and compassion satisfac-tion (B¼�3.043, p <.01). Emotional intelligenceand work interference with personal life were notsignificant factors in the integrated model.

Discussion

This study explored whether emotional intelli-gence, work autonomy, and work–life balancewere factors that contributed to compassionsatisfaction among social work practitioners.

Results concluded that emotional intelligence,work autonomy, and work–life balance were allfactors significantly associated with compassionsatisfaction when controlling for gender, race,marital status, license status, practice role, andyears of practice. These findings have importantimplications for combatting burnout amongsocial workers.

Regarding emotional intelligence, resultsshowed that social workers who had higher emo-tional intelligence experienced more compassionsatisfaction. This finding provides empirical evi-dence supporting a previous conceptual study byIngram (2013), which argued that emotionalintelligence among social workers contributes tothe core process of social work practice and thedevelopment of positive interaction between prac-titioners and service users (Ingram, 2013).Additionally, findings are consistent with existingresearch by Grant, Kinman, and Alexander(2014), who concluded that that developing emo-tional intelligence among social work studentsdecreases their psychological distress (Grantet al., 2014). Having a greater insight into others’and one’s own emotions during the helping pro-cess may improve a social worker’s ability tomanage the stress that is associated with workingwith distressed clients thereby increasing compas-sion satisfaction. Consistent with Ingram’s find-ings, results from this study suggest that at theindividual level, social workers should be encour-aged to cultivate emotional intelligence as notonly a relevant practice skill, but also as a

Table 3. Multivariate analyses of compassion satisfaction.EmotionalIntelligence

Model

WorkAutonomyModel

Work–lifeBalance Model

IntegratedModel

B SE B SE B SE B SE

Gender (Female) 3.829 2.946 2.061 2.151 3.185 2.000 1.047 1.949Race Black 0.399 2.572 4.943 1.995 3.844 1.828 4.732 1.695

Other races 3.457 3.307 4.695 0.046 2.095 2.079 2.574 1.934Marital Status Single �0.783 1.650 �0.288 1.277 �0.006 1.079 0.407 1.056Other �1.411 2.147 �0.659 1.609 �1.753 1.484 �0.964 1.382Currently Licensed (Yes) 3.128 1.877 0.603 1.46 �0.082 1.301 �0.272 1.206Practice Role 0.904 1.491 �0.127 1.164 0.979 0.985 0.507 0.981Years of Practice 0.099 0.069 0.034 0.051 0.053 0.045 0.02 0.043Emotional Intelligence 0.729 0.061 0.114 0.134Work Autonomy 1.442 0.449 1.153 0.38Work Interference with Personal Life 1.608 0.879 1.477 0.84Personal Life Interference with Work �3.024 0.876 �3.043 0.848Work/Personal Life Enhancement 2.953 0.719 2.534 0.716�p < .05,��p < .01,���p < .001.

JOURNAL OF SOCIAL SERVICE RESEARCH 7

protective factor that may increase professionallongevity. Such a goal can be achieved throughthe increased inclusion of emotional intelligencecontent in social work undergraduate and gradu-ate education, continuing education, and work-place training.

Regarding work autonomy, the results foundthat social workers who had a higher level of per-ceived work autonomy reported higher levels ofcompassion satisfaction. Findings are consistentwith previous studies identifying a positive rela-tionship between work autonomy and job satis-faction among other professions (Athey et al.,2016; Skaalvik & Skaalvik, 2014), and provideempirical evidence expanding this conclusion tosocial work practitioners specifically. Priorresearch examining the relationship betweensocial workers’ perceptions of their jobs andburnout found that social workers often experi-ence ambiguous job roles and responsibilities,which can contribute to burnout (Lloyd et al.,2002). The finding that work autonomy wasrelated to increased compassion satisfactionimplies that when social workers perceive moreindependence and freedom to control tasks intheir job role, they experienced a greater realiza-tion of fulfillment and efficacy in helping others.This finding suggests that employers should con-sider how to allow social workers more inde-pendence in decision making and more controlover their work.

Regarding work–life balance, the results dem-onstrated that work/personal life enhancementwas associated with higher levels of compassionsatisfaction while personal life interference withwork was associated with lower levels of compas-sion satisfaction. This lends support to existingliterature on work–life balance, which found thatemployees’ perception of achieving work–life bal-ance is associated with job satisfaction and men-tal wellbeing (Jang, Park & Zippay, 2011; Orkibi& Brandt, 2015), and provides evidence for socialworkers. The perception of work–life balancemay be particularly salient for social workers whooften work in human service organizations thatare underfunded, understaffed, and highly bur-eaucratic (Lloyd et al., 2002). This finding sug-gests that employers should provide supportivework environments that minimize conflicts

between personal life and work demands.Work–life balance programs and policies thatpromote self-care, flexible work scheduling, andfamily friendly leave policies might be consideredto increase employee wellbeing (Jang et al., 2011).

These results have implications for how theexecutive directors of social service organizationsshould support their employees. They shouldpromote emotional intelligence of employees,encourage work autonomy of employees, andsecure better work–life balance through trans-formational leadership that empowers employees(Fisher, 2009). In contrast to transactional leader-ship based on rewards and punishment in returnfor doing a job, transformational leadershipinspires employees, intellectually stimulates them,and considers them individually (Bass, 1999;Fisher, 2009). Based on this perspective, a leaderis defined as “anyone who takes responsibility forfinding the potential in people and processes, andwho has the courage to develop that potential”(Brown, 2018, p. 4). This transformational leader-ship style is congruent with social work principlesthat value the empowerment of individuals andthe interaction of individuals in a system (Fisher,2009) and has been known to increase compas-sion satisfaction among employees (Tafvelin,Hyv€onen, & Westerberg, 2014). Rather than lean-ing on leadership styles based on compliance andcontrol that uses fear to motivate employees(Brown, 2018), transformational leadership thatinspires emotional intelligence and work auton-omy of employees is needed for social work lead-ers to increase compassion satisfaction.

Limitations

While this study adds to the scant literatureexamining factors that contribute to compassionsatisfaction among social workers, there are sev-eral noteworthy limitations. First, this study uti-lized a cross-sectional design, and thus causationcannot be assumed from the results (Rubin,2013). Second, the sample size was limited, andadditionally was largely homogenous in terms ofrace and gender in comparison to the generalpopulation, which thus limits generalizability.Lastly, this study did not examine client popula-tion, job role characteristics (e.g., workload,

8 J. BAE ET AL.

time-consuming paperwork), or other organiza-tional factors (e.g., social support from supervisorand coworkers) that could also contribute tocompassion satisfaction among social workers.

Conclusion

Despite these limitations, this study contributesto a richer understanding of what factors contrib-ute to compassion satisfaction among socialworkers that have been discussed little so far.This study found that emotional intelligence,work-autonomy, and work–life balance playimportant roles in increasing the compassion sat-isfaction of social workers. Examining compas-sion satisfaction shifts the narrative to astrengths-based perspective focused on whatkeeps social workers in the profession, and thisknowledge can be used to increase retention.Furthermore, these findings suggest that trans-formational leadership that inspires employees isneeded. Future research may contribute to thefurther understanding of compassion satisfactionamong social workers by conducting a longitu-dinal study with larger sample size and examin-ing other factors such as client population, jobrole characteristics, and support from supervisorsand coworkers.

Acknowledgments

The authors would like to thank Dr. Llewellyn Cornelius,Dr. Tiffany Washington, and Dr. Orion Mowbray for theirinvaluable feedback on this manuscript.

ORCID

Tenesha Littleton http://orcid.org/0000-0002-0486-5402

References

Adams, R. E., Boscarino, J. A., & Figley, C. R. (2006).Compassion fatigue and psychological distress amongsocial workers: A validation study. American Journal ofOrthopsychiatry, 76(1), 103–108. doi:10.1037/0002-9432.76.1.103

Agha, K., Azmi, F. T., & Irfan, A. (2017). Work-life balanceand job satisfaction: An empirical study focusing onhigher education teachers in Oman. International Journalof Social Science and Humanity, 7(3), 164–171. doi:10.18178/ijssh.2017.V7.813

Alvaro, C., Lyons, R. F., Warner, G., Hobfoll, S. E.,Martens, P. J., Labonte, R., & Brown, R. E. (2010).Conservation of resources theory and research use inhealth systems. Implementation Science, 5(79), 1–20. doi:10.1186/1748-5908-5-79

Athey, E. K., Leslie, M. S., Briggs, L. A., Park, J., Falk, N. L.,Pericak, A., … Greene, J. (2016). How important areautonomy and work setting to nurse practitioners’ jobsatisfaction?. Journal of the American Association ofNurse Practitioners, 28(6), 320–326. doi:10.1002/2327-6924.12292

Bakker, A. B., & Demerouti, E. (2007). The job demands-resources model: State of the art. Journal of ManagerialPsychology, 22(3), 309–328. doi:10.1108/02683940710733115

Balakrishnan, A., & Saklofske, D. (2015). Be mindful howyou measure: A psychometric investigation of the BriefEmotional Intelligence Scale. Personality and IndividualDifferences, 87, 293–297. doi:10.1016/j.paid.2015.08.030

Bass, B. M. (1999). Two decades of research and develop-ment in transformational leadership. European Journal ofWork and Organizational Psychology, 8(1), 9–32. doi:10.1080/135943299398410

Breaugh, J. (1999). Further investigation of the work auton-omy scales: Two studies. Journal of Business andPsychology, 13(3), 357–373. doi:10.1023/A:1022926416628

Bride, B. E. (2007). Prevalence of secondary traumatic stressamong social workers. Social Work, 52(1), 63–70. doi:10.1093/sw/52.1.63

Brown, B. (2018). Dare to lead: Brave work. Tough conversa-tions. Whole hearts. New York: Random House.

Butler, L. D., Carello, J., & Maguin, E. (2017). Trauma, stress,and self-care in clinical training: Predictors of burnout,decline in health status, secondary traumatic stress symp-toms, and compassion satisfaction. Psychological Trauma,9(4), 416–424. doi:10.1037/tra0000187

Conrad, D., & Kellar-Guenther, Y. (2006). Compassionfatigue, burnout, and compassion satisfaction amongColorado child protection workers. Child Abuse & Neglect,30(10), 1071–1080. doi:10.1016/j.chiabu.2006.03.009

Cummings, C., Singer, J., Hisaka, R., & Benuto, L. T.(2018). Compassion satisfaction to combat work-relatedburnout, vicarious trauma, and secondary traumaticstress. Journal of Interpersonal Violence. Retrieved fromhttps://doi.org/10.1177/0886260518799502

Davies, K. A., Lane, A. M., Devonport, T. J., & Scott, J. A.(2010). Validity and reliability of a brief emotional intelli-gence scale (BEIS-10). Journal of Individual Differences,31(4), 198–208. doi:10.1027/1614-0001/a000028

Demerouti, E., & Bakker, A. B. (2011). The job demands-resources model: Challenges for future research. SAJournal of Industrial Psychology, 37(2), 1–9. doi:10.4102/sajip.v37i2.974

Demerouti, E., Bakker, A. B., Nachreiner, F., & Schaufeli,W. B. (2001). The job demands-resources model of burn-out. Journal of Applied Psychology, 86(3), 499–512. doi:101037//0021-9010863499

JOURNAL OF SOCIAL SERVICE RESEARCH 9

Diaconescu, M. (2015). Burnout, secondary trauma andcompassion fatigue in social work. Revista De AsistentaSociala (Social Work Review), 14(3), 57–63.

Dillman, D. (2000). Mail and internet surveys: The tailoreddesign method. New York: Wiley.

Dillman, D. A., Smyth, J. D., & Christian, L. M. (2014).Internet, phone, mail, and mixed-mode surveys: The tail-ored design method (4th ed.). Hoboken, NJ: Wiley.

Faul, F., Erdfelder, E., Buchner, A., & Lang, A.-G. (2009).Statistical power analyses using G�Power 3.1: Tests forcorrelation and regression analyses. Behavior ResearchMethods, 41(4), 1149–1160. doi:10.3758/BRM.41.4.1149

Fisher, E. A. (2009). Motivation and leadership in socialwork management: A review of theories and related stud-ies. Administration in Social Work, 33(4), 347–367. doi:10.1080/03643100902769160

Grant, L., Kinman, G., & Alexander, K. (2014). What’s allthis talk about emotion? Developing emotional intelli-gence in social work students. Social Work Education,33(7), 874–889. doi:10.1080/02615479.2014.891012

Hayman, J. (2005). Psychometric assessment of an instru-ment designed to measure work life balance. Researchand Practice in Human Resource Management, 13(1),85–91.

Hobfoll, S. E. (1989). Conservation of resources: A newattempt at conceptualizing stress. American Psychologist,44(3), 513–524. doi:10.1037/0003-066X.44.3.513

Ingram, R. (2013). Locating emotional intelligence in theheart of social work practice. The British Journal of SocialWork, 43(5), 987–1004. https://doi.org/10.1093/bjsw/bcs029 doi:10.1093/bjsw/bcs029

Jang, S. J., Park, R., & Zippay, A. (2011). The interactioneffects of scheduling control and work–life balance pro-grams on job satisfaction and mental health.International Journal of Social Welfare, 20(2), 135–143.doi:10.1111/j.1468-2397.2010.00739.x

Keeton, K., Fenner, D., Johnson, T., & Hayward, R. (2007).Predictors of physician career satisfaction work-life bal-ance, and burnout. Obstetrics & Gynecology, 109(4),949–955. doi:10.1097/01.AOG.0000258299.45979.37

Kim, H. (2011). Job conditions, unmet expectations, andburnout in public child welfare workers: How differentfrom other social workers?. Children and Youth ServicesReview, 33(2), 358–367. doi:10.1016/j.childyouth.2010.10.001

Kinman, G., & Grant, L. (2011). Exploring stress resiliencein trainee social workers: The role of emotional andsocial competencies. British Journal of Social Work, 40(4),261–275. doi:10.1093/bjsw/bcq088

Lloyd, C., King, R., & Chenoweth, L. (2002). Social work,stress, and burnout: A review. Journal of Mental Health,11(3), 255–265. doi:10.1080/09638230020023642

McFadden, P. (2018). Two sides of one coin? Relationshipsbuild resilience or contribute to burnout in child protec-tion social work: Shared perspectives from Leavers and

Stayers in Northern Ireland. International Social Work.Retrieved from https://doi.org/10.1177/0020872818788393

Orkibi, H., & Brandt, Y. I. (2015). How positivity links withjob satisfaction: Preliminary findings on the mediatingrole of work-life balance. Europe’s Journal of Psychology,11(3), 406–418. doi:10.5964/ejop.v11i3.869

Ojedokun, O., & Idemudia, E. (2014). Burnout among para-military personnel in Nigeria: A perspective from conser-vation of resources theory. South African Journal ofPsychology, 44(1), 106–117. doi:10.1177/0081246313516256

Qualtrics. (2016). [Computer software] Retrieved fromhttp://www.qualtrics.com

Rossi, A., Cetrano, G., Pertile, R., Rabbi, L., Donisi, V.,Grigoletti, L., … Amaddeo, F. (2012). Burnout, compas-sion fatigue, and compassion satisfaction among staff incommunity-based mental health services. PsychiatryResearch, 200(2-33), 933–938. doi:10.1016/j.psychres.2012.07.029

Rubin, A. (2013). Statistics for evidence-based practice andevaluation (3rd ed). Belmont, CA: Brooks/Cole, CengageLearning.

Salovey, P., & Mayer, J. D. (1990). Emotional intelligence.Imagination, Cognition and Personality, 9(3), 185–211.doi:10.2190/DUGG-P24E-52WK-6CDG

Sanchez-Moreno, E., de La Fuente Roldan, I., Gallardo-Peralta, L. P., & de Roda, A. B. (2015). Burnout, informalsocial support and psychological distress among socialworkers. British Journal of Social Work, 45(8), 2366–2386.doi:10.1093/bjsw/bcu084

Saragih, S. (2011). The effects of job autonomy on workoutcomes: Self efficacy as an intervening variable.International Research Journal of Business Studies, 4(3),203–215. doi:10.21632/irjbs.4.3.203-215

Schutte, N. S., Malouff, J. M., Hall, L. E., Haggerty, D. J.,Cooper, J. T., Golden, C. J., & Dornheim, L. (1998).Development and validation of a measure of emotionalintelligence. Personality and Individual Differences, 25(2),167–177. doi:10.1016/S0191-8869(98)00001-4

Simon, C. E., Pryce, J. G., Roff, L. L., & Klemmack, D.(2006). Secondary traumatic stress and oncology socialwork. Journal of Psychosocial Oncology, 23(4), 1–14. doi:10.1300/J077v23n04_01

Skaalvik, E. M., & Skaalvik, S. (2014). Teacher self-efficacyand perceived autonomy: Relations with teacher engage-ment, job satisfaction, and emotional exhaustion.Psychological Reports, 114(1), 68–77. doi:10.2466/14.02.PR0.114k14w0

Stamm, B. H. (2002). Measuring compassion satisfaction aswell as fatigue: Developmental history of the compassionsatisfaction and fatigue test. In C. R. Figley (Ed.).Treating compassion fatigue (pp. 107–119). New York:Brunner-Routledge.

Stamm, B. H. (2010). The concise ProQOL manual (2nded.). Retrieved from https://owl.purdue.edu/owl/research_

10 J. BAE ET AL.

and_citation/apa_style/apa_formatting_and_style_guide/reference_list_electronic_sources.html

Tafvelin, S., Hyv€onen, U., & Westerberg, K. (2014).Transformational leadership in the social work context:The importance of leader continuity and co-worker sup-port. The British Journal of Social Work, 44(4), 886–904.doi:10.1093/bjsw/bcs174

Wagaman, M. A., Geiger, J. M., Shockley, C., & Segal, E. A.(2015). The role of empathy in burnout, compassion

satisfaction, and secondary traumatic stress among socialworkers. Social Work, 60(3), 201–209. doi:10.1093/sw/swv014

Whitebird, R., Asche, S., Thompson, G., Rossom, R., &Heinrich, R. (2013). Stress, burnout, compassion fatigue,and mental health in hospice workers in Minnesota.Journal of Palliative Medicine, 16(12), 1534–1539 doi:10.1089/jpm2013.0202.

Zalaquett, C. P. & Wood, R. J. (Eds.). (1997). A book ofresources. Lanham, MD: The Scarecrow Press.

JOURNAL OF SOCIAL SERVICE RESEARCH 11