Statistical Social Work Study Paper/Post Week 5 BSHS

article.pdf

Home >Mathematics homework help >Numerical analysis homework help >Statistical Social Work Study Paper/Post Week 5 BSHS

Differences between Children’s Social Workers and Adults’ Social Workers on Sense of Burnout, Work Conditions and Organisational Social Support

Liat Hamama*

Liat Hamama is a social worker and a lecturer in the Bob Shapell School of Social Work at Tel Aviv University. She received her PhD at Tel Aviv University. Her research interests include

physical health, emotional distress, coping resources and well-being among children/ adolescence, their families and among health care professionals.

*Correspondence to Liat Hamama, Ph.D., Bob Shapell School of Social Work, Tel-Aviv University, 69978, Israel. E-mail: [email protected]

Abstract

This research study aimed to examine the experience of burnout among 232 Israeli social

workers (126 who were directly treating children and adolescents and 106 who were

directly treating adults). Burnout was investigated in relation to social workers’ demo-

graphic characteristics, extrinsic and intrinsic work conditions, and social support at the

workplace by colleagues, their direct supervisor and the head of their agency. Social

workers of children did not report a higher experience of burnout than social workers

of adults in Israel. Both groups indicated, on average, a moderate intensity of burnout.

However, significant differences emerged between the two groups on perceived work

conditions and on support from the agency head. That is, children’s social workers per-

ceived better extrinsic work conditions than workers of adults, who reported better

intrinsic working conditions. In addition, children’s social workers reported higher

support from their agency heads than did workers serving adults. Burnout was signifi-

cantly negatively correlated with age, professional experience, intrinsic and extrinsic

work conditions, and social support from colleagues within the organisation and from

the agency head. Professional experience and support from colleagues and from the

agency head contributed significantly to explaining the variance in burnout intensity.

Moreover, intrinsic work conditions mediated between professional experience and

experience of burnout. Various explanations for these findings are discussed.

Keywords: Mental health, practitioners, social support

Accepted: August 2012

# The Author 2011. Published by Oxford University Press on behalf of

British Journal of Social Work (2012) 42, 1333–1353 doi:10.1093/bjsw/bcr135 Advance Access publication September 28, 2011

Introduction

Burnout and stress among social workers are well documented in Israel (e.g. Abu-Bader, 1999, 2000; Bargal and Guterman, 1996, 1997; Katchalnick et al., 1991; Stav et al., 1987) and in other countries (e.g. Acker, 1999; Evans et al., 2006; Kim and Stonere, 2008; Söderfeldt et al., 1995; Um and Harri- son, 1998). Daily, social workers face the busy and complex world of human behaviour in social contexts—a world in which relationships break down, emotions run high and personal needs go unmet (Howe, 2004). Moreover, the constant demand for one-sided giving on an emotional level—by social workers to the users of their services or agencies—can lead to emotional depletion and eventual burnout (Lloyd et al., 2002; Maslach et al., 2001; Pines and Aronson, 1988).

Despite this, there is little research on social workers’ conditions and their absenteeism, burnout and turnover. Moreover, the lack of Israeli national statistics on burnout and turnover intentions in the social work profession precludes obtaining a detailed picture of the current situation or gaining an overview of developments in recent years. High turnover requires new personnel recruitment; incurs administrative, selection, induc- tion and training costs along with loss of knowledge and experience; renders adverse effects on staff morale; and decreases user satisfaction with services (Bargal and Guterman, 1996). In addition, little research has compared burnout among social workers serving children versus those serving adults (age twenty-one and up). To enhance insight into organisational factors and their possible relations to burnout, the current study presented a comparative analysis of children’s social workers and adults’ social workers in Israel, under the assumption that the former may be at higher risk for burnout due to the unique challenges and characteristics of thera- peutic work with children.

Social workers treating adults versus social workers treating children and adolescents

Social workers serving adults in need aim to improve their social functioning by the provision of practical and psychological help (Abu-Bader, 1999; Kim and Stonere, 2008; Rabin et al., 2000). The accepted view held by these social workers is that many of their users’ difficulties are linked in diverse ways to their social, economic and political status in society (Lloyd et al., 2002). According to Abu-Bader (1999, 2000), those who apply for aid from social welfare services usually exhibit the most severe economic and social pro- blems. Unlike adult users, children and adolescents rarely apply directly for aid from social workers. Yet, children and adolescents constitute more than 50 per cent of referrals to social welfare services in the USA

1334 Liat Hamama

(Kazdin, 2000). Similar trends were also found in Israel (e.g. Ronen and Hamama, in press) and in the UK (e.g. Kazi, 2000; Waterhouse, 2008). Epi- demiological studies indicate that from 17 to 22 per cent of children under age eighteen suffer from developmental, emotional or behavioural pro- blems, most of which are long lasting and predictive of problems in adult- hood (Kazdin, 2000). Children’s young age, naivety and vulnerabilities, along with the need to work concurrently with a complex mix of other sur- rounding systems, such as parents, family, schools, health workers and col- leagues, demand the maintenance of particularly high ethical and professional standards with respect to working with vulnerable groups (Lonergan et al., 2004). Furthermore, due to children’s developmental needs, children’s social workers need to summon all their creativity, intellec- tual ability, emotions and compassion (Pearlman and Saakvitne, 1995).

Several studies in the UK have noted the particular demands made upon child-care social workers in social services settings, which may lead them to experience more stress, less job satisfaction, limited support and more diffi- culties in coping (Bennett et al., 1993; Coffey et al., 2004; Lecroy and Rank, 1987). The child-care workers were ‘least able to distance themselves emotionally from the work they were engaged in’ (Bennett et al., 1993, p. 37). Yet, other studies in the UK have produced different and contradictory findings. For example, Thompson et al.’s (1996) study indicated that working with particular user groups was not a significant variable in social workers’ perceptions of stress; instead, the organisational context itself generated the most significant stress. Collings and Murray (1996) also concluded that stress was not significantly linked to working with a particular user group, while Balloch et al.’s (1998) extensive research survey suggested that, contrary to expectations, statutory social workers working with older people experi- enced higher levels of stress than those undertaking child-care work.

A comparative analysis of two groups of Israeli social workers—working with children and working with adults—was conducted to highlight the similarities and differences regarding their sense of burnout.

The concept of burnout

The term ‘burnout’ refers to a phenomenon associated with care-giving and service occupations (Maslach et al., 2001) that has been observed among human service workers dealing with emotionally demanding individuals. Burnout was first defined by Freudenberger (1980) as ‘a state of fatigue or frustration brought about by devotion to a cause, way of life or relation- ship that failed to produce the expected reward’ (Freudenberger, 1980, p. 13), and it was later defined by Maslach et al. (2001) as a ‘psychological syndrome in response to chronic interpersonal stressors on the job’ (Maslach et al., 2001, p. 399). In investigating 218 Israeli social workers, Abu-Bader (1998) found a strong correlation (r¼– 0.40; p,0.001)

Social Workers’ Burnout 1335

between burnout and work satisfaction, which refers to ‘the amount of overall positive affect (or feelings) that individuals have toward their jobs’ (Arnold and Feldman, 1986, p. 86). That is, social workers reported less burnout when they felt satisfied with their work.

The majority of burnout studies have been based on Maslach and Jack- son’s (1986) conceptualisation of burnout as having three components: emotional exhaustion (feelings of being overextended and depleted of emotional and physical resources), depersonalisation or cynicism (negative or excessively detached responses to various aspects of the job), and dimin- ished personal accomplishment (feelings of incompetence and a lack of achievement at work).

Bargal and Guterman (1996) and Maslach et al. (2001) asserted that the factors underlying burnout include workers’ personal characteristics and personality traits, environmental factors and various organisational and social conditions. To determine whether different factors may reduce or eliminate a sense of burnout among social workers involved with children and social workers involved with adults, the present study investigated per- sonal demographic characteristics, environmental work conditions and social support at work.

Demographic characteristics associated with burnout

Research has linked five personal factors with burnout: age, gender, pro- fessional experience, family status and education level (Bargal and Guter- man, 1996; Maslach, 1982).

Age

Research in Israel (e.g. Amrani-Cohen, 1999; Bar-Zaselvisky, 2003) showed that young employees, particularly below age thirty, are more prone to burnout than older ones. These researchers suggested that older workers are more stable, mature, confident and financially secure, with more familial supports, self-identity and life experience than younger workers. Further- more, young workers tend to develop unrealistically high expectations of the workplace. However, Bargal and Guterman’s (1996) study of 591 Israeli social workers found the opposite—older workers reported more depersonalisation (a burnout component) than younger workers.

Gender

Research showed that burnout was more prevalent among female than male social workers in Israel (Bargal and Guterman, 1996; Cohen, 1990;

1336 Liat Hamama

Etzion and Pines, 1986). Maslach et al. (2001) attributed this finding to female-typed occupations that tend to be rich in emotional and interperso- nal stressors, such as negative user feedback.

Professional experience

Most research has focused on professional experience in terms of age, because the two correlate highly (Navaro, 1999). Bar-Zaselvisky (2003) and Navaro (1999) found that Israeli workers with high professional experi- ence demonstrate less burnout because they are exposed to more challen- ging roles and rewards. In contrast, Bargal and Guterman (1996) found that age correlated with burnout. That is, older workers reported more depersonalisation (a burnout component) than younger workers.

Family status

In a series of studies conducted in Israel (Bargal and Guterman, 1996; Cohen, 1990) and in the USA (Siefert et al., 1991), single female social workers reported higher rates of burnout than married ones. This finding was linked to greater emotional support supplied by the married worker’s family and to the married worker’s shared financial burden. Data are una- vailable regarding males.

Level of education

Kahana-Friedman’s (2000) study of 295 Israeli social workers showed that more highly educated workers reported stronger burnout levels. Other researchers linked education with burnout if workers feel overqualified or ‘sub-challenged’ and thus become bored and frustrated (Lauderdale, 1982; Pines and Aronson, 1988).

The present study examined whether these demographic variables were differentially linked to burnout intensity among Israeli social workers serving children compared with those serving adults.

Work conditions associated with burnout

The current study examined three dimensions of social workers’ work environment that may induce burnout (Maslach et al., 2001; Pines, 1982): physical, psychological and social. The physical dimension comprises extrinsic work conditions like temperature, light, noise, crowdedness, work- place structure and the agency’s space and flexibility for changing those physical characteristics. All these extrinsic conditions were found to

Social Workers’ Burnout 1337

correlate positively with burnout (Abu-Bader, 2000; Pines, 1982; Vinokur- Kaplan, 1991). That is, social workers reported less burnout when their physical work conditions were comfortable and flexible regarding tempera- ture, light, noise and crowdedness.

The psychological dimension comprises intrinsic emotional and cognitive work conditions. Emotionally, low sense of meaning at work and low per- ceived potential for personal growth, job challenge or self-fulfilment may lead to frustration and burnout. For example, ‘job challenge’, defined by Jayaratne and Chess (1984) as an emotional aspect referring to skill utilisation and interest level, was the most significant contributor to explaining burnout rates among 591 Israeli social workers in welfare agencies (Bargal and Guter- man, 1996). Cognitive work conditions were found to increase stress and thus may lead to burnout (Dollard et al., 2000; Kary, 1990; Pines, 1982). Such con- ditions included job autonomy—workers’ degree of control over their own immediate scheduling and tasks (Bargal and Guterman, 1996; Liu et al., 2005), role ambiguity—uncertainty about the job’s scope and others’ expec- tations (Bargal and Guterman, 1996; Kahn et al., 1964) and role conflict— when expectations are in conflict at work (Kahn et al., 1964).

The social dimension, presented here through the social support variable, relates to the workplace social environment, including users, colleagues, the direct supervisor and the head of the agency (Bargal and Guterman, 1996; Deeter-Schmelz and Ramsey, 1997; Maslach et al., 2001). An environment that provides social support can reduce and even prevent burnout (Houkes et al., 2003; Kim and Stonere, 2008; Peterson et al., 2008; Pines, 1982). According to Karasek and Theorell (1990) and Pines et al. (1981), the social support system at work has six functions: listening and attentiveness, professional support, professional challenge, emotional support, mental challenge and shared worldview. Abu-Bader (1999) found that worker relationships with colleagues and quality of supervision significantly pre- dicted burnout among 218 Israeli social workers. Maslach and Goldberg (1998) suggested that colleagues’ support might provide emotional comfort, new insights, and personal rewards and recognition, thereby diminishing burnout.

In the UK, Huxley et al. (2005) found that support from supervisors was valuable for revealing symptoms of distress to workers. An effective super- visor provides a front line social worker with essential guidelines, pro- fessional skills and knowledge related to services while also communicating with social workers about opinions, feelings and decision making with mutual trust and respect (Peterson et al., 2008).

The present study hypotheses

This study first investigated the extent to which Israeli social workers experience work burnout and possible differences between workers

1338 Liat Hamama

serving the two populations. Second, this study examined the contribution of demographic variables, work conditions and workplace social support to workers’ development of burnout. The following hypotheses were tested:

(1) Social workers who treat children will report a higher experience of

burnout than social workers who treat adults.

(2) Social workers who report better work conditions (extrinsic and/or

intrinsic) will report less burnout.

(3) Social workers who report higher organisational social support by col-

leagues, the direct supervisor and the agency head will report lower burnout.

(4) Greater burnout will be reported by social workers who are younger, are

unmarried, have less professional experience and have higher education

levels (compared to older, married, more experienced and less educated workers, respectively).

Method Participants

The sample consisted of 232 social workers comprising 126 social workers serving children and 106 social workers serving adults. Information was col- lected on participants’ age, gender, family status, education, professional experience and full-time or part-time position.

As seen in Table 1, 75 per cent of participants were younger than forty (M ¼ 35.54, SD¼7.92); the vast majority (93 per cent) were women, 63 per cent were married and 58 per cent held a first academic degree. Regarding pro- fessional experience, 42 per cent had up to five years’ experience, and 33 per cent had over eleven years’ experience (M ¼ 9.30, SD ¼ 7.38).

Table 1 Study participants’ characteristics (N¼232)

Characteristics Values n %

Age (in years) 30 – 20 80 34.0 40 – 31 94 41.0 50 – 41 42 18.0 51+ 16 7.0

Gender Male 16 7.0 Female 216 93.0

Family status Unmarried (single, divorced, widowed) 80 35.0 Married 152 65.0

Education Bachelor’s degree 134 58.0 Master’s degree 98 42.0

Seniority in the social work profession (in years)

,1 11 5.0 1 – 5 86 37.0 6 – 10 55 25.0 11 – 15 35 15.0 16+ 43 18.0

Social Workers’ Burnout 1339

Preliminary analyses of socio-demographic variables (i.e. age, gender, family status, education, professional experience) revealed no significant differences between children’s and adults’ workers.

Social workers were employed in one of two work settings: primary- practice or secondary-practice settings. Fifty-eight per cent (n¼134) of par- ticipants worked in a primary-practice social welfare agency within a local municipal department. These local agencies oversee non-profit public organ- isations sponsored by the Israeli national government (Office of Welfare and Social Services) and local authorities. Such agencies’ primary function is to provide social and human services to low-income families and users of all ages (e.g. at-risk teenage girls, elderly people). The other 42 per cent of par- ticipants (n ¼ 98) worked in secondary-practice settings where the agency’s major function is other than social work practice and social workers support that primary function (Kane, 1984). These included settings providing proba- tion services, afterschool child-care facilities, home-based child-care ser- vices, residential facilities and hospitals. Mental health social workers did not participate in this study because it was undertaken at a time of uncer- tainty in Israeli mental health services owing to a move towards privatisation. No significant differences were found in participants’ socio-demographic variables in relation to their work setting.

Measures

The study included three questionnaires: burnout, work conditions and workplace social support.

Burnout questionnaire

The Hebrew translation (Stav, 1982) of Maslach and Jackson’s (1981) Maslach Burnout Inventory (MBI) comprised three components: emotion- al depletion (e.g. ‘I feel emotionally exhausted because of my work’); depersonalisation (e.g. ‘I feel as though I relate to some of my clients as objects’); and diminution of personal competence (e.g. ‘I can easily under- stand how my clients feel in the context of different situations’). Partici- pants rated the twenty-two-item inventory responses on a five-point Likert scale ranging from Not at all (1) to With very great intensity (5). Higher scores indicated higher burnout intensity.

Work conditions questionnaire

Gorlick’s (1998) twenty-one-item questionnaire was developed, following consultation with agency heads, to examine occupational stress and

1340 Liat Hamama

burnout among social workers in health and family treatment agencies. In the present study, work environment items were categorised as intrinsic, extrinsic or neither by ten social worker agency heads. Agreement (by more than six heads) emerged regarding four extrinsic aspects of work environments (privacy, relaxed atmosphere in the room, accessibility to tel- ephone and noise) and nine items comprising six intrinsic aspects (ability to succeed in interventions, extent of challenge, interest, self-fulfilment, meaning and logic at work). Participants rated these thirteen items along a five-point Likert-type scale ranging from To a great extent (1) to To a low extent or not at all (5), where higher mean scores for the two aspects indicated higher extrinsic/intrinsic work conditions.

Social support questionnaire

Kershner-Cohen’s (1998) ten-item support questionnaire in Hebrew was used based on House’s (1981) four types of workplace social support: instru- mental (e.g. ‘To what extent are they willing to temporarily replace you in your job, when necessary?’), appraisal (e.g. ‘To what extent do they voice appraisal as to your ability to cope with difficulties that arise at work?’), emotional (e.g. ‘To what extent do you receive encouragement when you encounter difficulties?’) and informative (e.g. ‘To what extent do you receive advice that can help you cope with difficulties at work?’). Partici- pants completed the same ten-item scale three times, for three sources of support in the work agency (colleagues, the direct supervisor and the agency head) along a five-point Likert scale from Low level of support (1) to High level of support (5). A mean score was calculated for each source of support; higher scores indicated higher perceived workplace social support.

Procedure

The current sample recruited social workers in various settings from differ- ent areas of Israel. A snowball method over two time periods was used because the Israeli Association of Social Workers provides no information on child and/or adult social workers. Moreover, other lists from employing agencies were not available to the researcher. In autumn 2008, thirty social workers treating only children and adolescents were located by the author through professional practice contacts (i.e. the author is a child’s social worker therapist in a health agency). In autumn 2009, fifteen social workers treating only adults were located by the author through pro- fessional contacts. Social workers who concurrently served both groups were excluded. All forty-five social workers agreed to participate in the study when approached, and all obtained agreement for their participation

Social Workers’ Burnout 1341

from their agency and different supervisors. Ethical approval was obtained from the Ministry of Social Welfare to conduct the research.

Six questionnaire sets were then mailed to each social worker, including one set for the worker to complete independently and five other sets to dis- tribute to colleagues either serving children and adolescents or else serving adults. These colleagues were from within their own work setting or from parallel settings in local or other areas of the country. Questionnaire sets included stamped, addressed envelopes for mailing the completed question- naires. From the first period, 180 questionnaire sets were distributed in total and 136 were returned, but only 126 were complete—a return rate of 75 per cent. From the second period, 140 questionnaire sets were added and 112 were returned, but only 106 were complete—a return rate of 80 per cent (sixteen questionnaires were excluded because of missing data). Meticulous examination of participants’ demographic details revealed that no partici- pant completed the questionnaire set twice. In sum, 232 social workers par- ticipated in the study: 126 children’s social workers and 106 social workers of adults.

Data analysis

Data analysis utilised the SPSS method version 15 – 0 for Windows. First, descriptive data were calculated to obtain the measures’ psychometric properties. Second, correlations were calculated. Last, regression analysis (95 per cent confidence limits) examined the contribution of each variable to burnout.

Results Descriptive information on measures’ scores

Table 2 presents means, standard deviations, ranges and reliabilities for the study variables. As seen in the table, participants reported moderate

Table 2 Psychometric characteristics of study variables (N ¼ 232)

Variable M SD Minimum Maximum Reliability a

Burnout intensity 2.10 0.47 1.14 3.76 0.87 Work conditions

A. Extrinsic 3.44 0.95 1.00 5.00 0.83 B. Intrinsic 3.90 0.58 2.22 5.00 0.89

Social support A. Colleague support 4.05 0.72 1.40 5.00 0.92 B. Supervisor support 3.83 1.03 0.30 5.00 0.94 C. Agency head support 3.06 1.13 0.80 5.00 0.96

1342 Liat Hamama

burnout intensity (M ¼ 2.10, range ¼ 1.14 – 3.76), good work conditions (extrinsic: M ¼ 3.44, range ¼ 1.00 – 5.00; intrinsic: M ¼ 3.90, range ¼ 2.22 – 5.00), and high support from colleagues (M ¼ 4.05, range ¼ 1.40 – 5.00) and the direct supervisor (M ¼ 3.83, range ¼ 0.30 – 5.00) but only moderate support from the agency head (M ¼ 3.06, range ¼ 0.80 – 5.00).

To examine differences between the two social worker groups (children’s or adults’) in sense of burnout, work conditions and workplace social support, univariate one-way MANOVAs were conducted. Significant differences emerged between groups (F (6, 225) ¼ 7.88, p , 0.001, Eta2 ¼ 0.17). Table 3 presents the group means and standard deviations for the univariate one-way MANOVAs. As seen in the table, no significant differences emerged between children’s and adults’ workers regarding burnout or workplace social support. Only for work conditions did signifi- cant differences emerge: children’s workers rated higher extrinsic work conditions, whereas workers of adults rated higher intrinsic work conditions.

Correlations between burnout intensity and demographic characteristics

The links between burnout and demographic characteristics (age, sex, family status, education and professional experience) were examined by Pearson correlations and one-way ANOVAs. Pearson correlations were conducted to examine links between burnout and workers’ age and pro- fessional experience. Burnout correlated negatively with age (r ¼ – 0.29, p , 0.001) and with professional experience (r ¼ – 0.25, p , 0.001). Thus, older social workers and those with more professional experience reported less burnout. One-way ANOVAs conducted for gender, family status and education revealed no significant differences in burnout levels between males and females, married and unmarried workers or between workers with a first versus second academic degree (p . 0.05).

Table 3 Means, standard deviations and F scores of the dependent variables among children’s and adults’ social workers (N ¼ 232)

Variables

Serving children

Serving adults

F (1,230) Eta2

95% confidence of the difference

M SD M SD Lower Upper

Burnout intensity 2.14 0.44 2.05 0.49 1.97 0.00 20.35 0.21 Extrinsic work conditions 3.70 0.98 3.14 0.83 21.02** 0.08 0.32 0.79 Intrinsic work conditions 3.81 0.58 4.02 0.54 8.95* 0.04 20.36 20.07 Colleague support 4.02 0.71 4.10 0.75 0.88 0.00 20.28 0.10 Supervisor support 3.88 0.95 3.78 1.13 0.49 0.00 20.17 0.36 Agency head support 2.93 1.10 3.23 1.15 4.24 0.01 20.60 20.013

* p , 0.01; ** p , 0.001.

Social Workers’ Burnout 1343

Correlations between burnout intensity and study variables

Table 4 presents Pearson correlation coefficients for the study variables. As seen in the table, social workers’ burnout was significantly lower when extrinsic (physical) work conditions were better and this negative corre- lation was even stronger between burnout and intrinsic (psychological) work conditions. Workers’ burnout was also significantly lower when they received better support from colleagues and this negative correlation was stronger between burnout and agency head support.

To examine differences between the groups (children’s or adults’) regarding burnout’s correlations with work conditions and social support, we conducted Z Fisher’s analysis. Differences emerged only between burnout and support perceived from the agency head (r ¼ 2.85, p , 0.001). Among children’s social workers, a high negative correlation emerged (r ¼ – 0.43, p , 0.001), whereas social workers of adults showed a lower, non-significant correlation (r ¼ – 0.08, p . 0.05). Thus, less support from the agency head was linked more strongly to higher burnout among children’s workers than among workers of adults.

Regression analysis

To examine the joint contribution of all the variables to the explained var- iance of burnout, hierarchical regression analysis was performed with burnout intensity as the dependent variable and the predictors entered in four steps. The first step assessed the contributions of the demographic characteristics; professional experience, education and group (children’s or adults’) were entered. Professional experience but not age was entered in this step due to these two characteristics’ high intercorrelation (r ¼ 0.76, p,0.001). Education was entered in the first step due to the inter- action found between education and agency head support. The support

Table 4 Pearson correlations between study variables (N ¼ 232)

Work conditions Support from

Extrinsic Intrinsic Colleagues Supervisor Agency head

Work conditions: Extrinsic – – Intrinsic 0.25** Support from: Colleagues 0.08 0.18* – – – Supervisor 0.13 0.12 0.47** 0.27** Agency head 0.13 0.38** 0.20* Burnout intensity – 0.18** – 0.55** – 0.18* – 0.11 – 0.27**

* p, 0.01; ** p, 0.001.

1344 Liat Hamama

variable (colleagues/supervisor/head) was entered in the second step. Work conditions (extrinsic and intrinsic) were entered in the third step. In the fourth step, interactions between the different predictors were entered. The first three steps used forced entry, where each block of vari- ables was entered as a single step. In the fourth step, interaction variables entered according to the significance of their contribution (p , 0.05).

Table 5 presents the hierarchical regression coefficients, explaining 35 per cent of the overall variance in social workers’ burnout. In Step 1, edu- cation, professional experience and group (children’s or adults’) contribu- ted 7 per cent to the explained variance. A significant contribution with negative b coefficient emerged only for professional experience. Thus, greater professional experience predicted lower burnout. In Step 2, the support variables added another 8 per cent to the explained variance. A sig- nificant contribution with negative b coefficient emerged for support by colleagues and by the agency head. Thus, more support from one’s col- leagues and from the agency head predicted workers’ lower burnout. In Step 3, work conditions added a further 18 per cent to the explanation of variance. Only the intrinsic index was significant, with a negative b coeffi- cient. Thus, higher intrinsic work conditions predicted lower burnout.

Adding extrinsic work conditions (i.e. privacy, relaxed atmosphere in the room, accessibility to telephone and noise) and intrinsic work conditions (i.e. ability to succeed in interventions, extent of challenge, interest, self-

Table 5 Hierarchical regression coefficients (b) to explain differences in burnout (N ¼ 232)

First step Second step Third step Fourth step

Predictors t b t b t b t b

1. Education 21.13 20.07 21.59 20.10 21.15 20.06 21.25 20.07 Experience 23.48*** 20.23 23.26*** 20.20 21.84 20.10 22.07 20.12 Group 20.84 20.06 23.26 20.02 0.43 0.03 0.57 0.03

2. Colleague support 22.16* 20.15 21.56 20.10 21.37 20.09 Supervisor support

0.17 0.01 0.14 0.01 0.25 0.02

Agency head support

23.26*** 20.21 20.91 20.06 21.11 20.07

3. Extrinsic work conditions

20.21 20.01 20.27 20.02

Intrinsic work conditions

27.39*** 20.48 27.18*** 20.46

4. Education × agency head support

22.18* 20.12

Group × agency head support

1.97* 0.11

R2 0.07** 0.15*** 0.33*** 0.35*** DR2 0.07** 0.08*** 0.18*** 0.02*

* p, 0.05; ** p, 0.01; *** p, 0.001.

Social Workers’ Burnout 1345

fulfilment, meaning and logic at work) caused the significant b coefficient found in the previous steps to become non-significant; therefore, intrinsic work conditions comprised a mediator between professional experience and agency head support to the sense of burnout. To examine this mediation, Sobel analysis was performed. Intrinsic work conditions were found to mediate between professional experience and experience of burnout (z ¼ 3.65, p , 0.001), indicating that more professional experi- ence was associated with higher intrinsic work conditions, and that higher intrinsic work conditions were associated with less burnout. In addition, intrinsic work conditions were found to mediate between agency head support and the experience of burnout (z ¼ 5.09, p , 0.001), indicating that stronger support from the agency head was associated with higher intrinsic work conditions and that higher intrinsic work conditions were associated with less burnout.

In Step 4, two interactions contributed significantly to the explained var- iance of burnout, adding 2 per cent: Agency Head Support × Education and Agency Head Support × Group. To explain the first interaction, sep- arate correlations were calculated for participants with Bachelor’s and Master’s degrees. Significant correlations emerged for both: Bachelor’s: r ¼ – 0.22, p , 0.05, Master’s: r ¼ – 0.36, p , 0.001. Thus, lower support from the agency head was associated with higher burnout among social workers with a Master’s degree.

To explain the interaction for Agency Head Support × Group (chil- dren’s or adults’), separate correlations were calculated for the children’s and adults’ social workers. A significant correlation emerged only for chil- dren’s social workers (r ¼ – 0.43, p , 0.001), but not for adults’ (r ¼ – 0.08, p . 0.05). Thus, lower support from the agency head was associated with higher burnout among children’s social workers.

Discussion

The study investigated two groups of Israeli social workers—those treating children and those treating adults—under the assumption that children’s and adolescents’ unique characteristics demand particular professional resources and specific work conditions that might promote burnout more than in adults’ workers. Moreover, demographic characteristics, work con- ditions and perceived workplace social support were examined vis-à-vis burnout intensity experienced by both groups.

This study has some limitations, including its solely quantitative design, which should be complemented in the future by qualitative methods that give voice to the unique perspectives on burnout among children’s or adults’ social workers. Furthermore, the study’s non-random snowball recruitment method limits the findings’ generalisability because the original participants likely referred workers with similar personal characteristics to

1346 Liat Hamama

their own. Moreover, the study’s self-report methodology makes the assumption that participants’ responses accurately reflect their feelings (Heppner et al., 1992). Thus, future research would do well to collect comp- lementary objective data on the actual work environment, to confirm current outcomes. And, finally, future studies should also examine the burnout subscales (emotional depletion, depersonalisation, diminution of personal competence) and their correlates to the independent variables (i.e. extrinsic and intrinsic work conditions, support from colleagues, the direct supervisor and the agency head).

In the present study, in contrast to the first study hypothesis, Israeli chil- dren’s social workers did not report a higher experience of burnout than their peers serving adults. Perhaps the lack of higher burnout levels among children’s workers can be partly explained by Reagh’s (1994) quali- tative findings where child welfare workers remained at work despite stress- ful working conditions because they felt ‘called’ to child welfare and believed they could improve their users’ lives. Perhaps these workers con- sistently failed to meet the criteria for burnout because of their feelings of personal accomplishment related to their work helping children in crisis. Such an explanation highlights the importance of employees’ appraisals of situations in the work context (e.g. appraisals of challenge and control or of overload and stress at work) as determinants of outcomes such as burnout or job satisfaction (Ben-Zur and Michael, 2007; Lazarus, 1999).

The moderate rather than high burnout intensity can perhaps be explained using Hobfoll’s (1989) conservation of resources model (COR). This model assumes that people consciously and unconsciously utilise the resources at their disposal and compensate for threatened resources by uti- lising others. Thus, one may speculate that children’s social workers may effectively use social support as an environmental resource when sensing that their personal psychological resources are threatened by the emotional overload that users elicit.

Regarding the second hypothesis, both extrinsic work conditions (i.e. privacy, relaxed atmosphere in the room, accessibility to telephone and noise) and intrinsic work conditions (i.e. ability to succeed in interventions, extent of challenge, interest, self-fulfilment, meaning and logic at work) were correlated negatively with workers’ burnout, but the intrinsic com- ponent showed a higher association. This corroborated previous Israeli research outcomes (Bargal and Guterman, 1996; Katchalnick et al., 1991) whereby Israeli social workers who worked directly with users attributed the intrinsic component of their work with the greatest impact on their decision to remain at work.

Children’s social workers perceived better extrinsic work conditions than workers of adults, who reported higher intrinsic working conditions. This outcome may stem from the need for special physical settings when treating children (e.g. workplace structure, noise) to facilitate helping processes such as game therapy or drawing (Ronen, 2003).

Social Workers’ Burnout 1347

Indeed, this study’s most prominent finding was the mediating role high- lighted for intrinsic work conditions. That is, younger social workers or social workers who reported less support from their agency head experi- enced a stronger sense of burnout when their intrinsic work conditions were lower. In previous studies in the USA, younger social workers reported lower personal accomplishments, more psychological strain and more depersonalisation than older workers (Guterman and Bargal, 1996; Jayaratne et al., 1991, 1995). In the current study, younger participants who reported less favourable intrinsic work conditions perhaps referred to a low sense of meaning, low autonomy and low perceived potential for personal growth and self-fulfilment in their work. This may contribute to the lower psychological resilience and job mastery that may lead to the experience of burnout (Amrani-Cohen, 1999).

Regarding the third hypothesis, the current results corroborated previous research that reported workplace social support’s significant contribution to reducing burnout among human service professionals (Bargal and Guter- man, 1996; El-Bassel et al., 1998; Ezer, 2003; Houkes et al., 2003). Nonethe- less, in both groups of social workers in the present study, burnout was linked negatively only to colleagues’ and to the agency head’s support, not to the supervisor’s. That is, the current workers who perceived stronger support from their colleagues and from the agency head experienced less burnout. Colleagues generally provide workers with emotional and infor- mative support, technical assistance and a sense of partnership in a similar reality (Ezer, 2003; Jenkins and Elliott, 2004), which may tend to reduce difficulties with users and encourage employees to remain at work. The agency head often provides practical, instrumental assistance to workers, such as helping set priorities and make decisions about tasks, as well as allocating resources to benefit workers’ extrinsic and intrinsic work conditions such as job-related training programmes (Abu-Bader, 2000; Tham, 2007). Therefore, workers who experience the agency head as supportive and as ‘valuing’ their work may manifest a lower experience of burnout (Huxley et al., 2005).

The fact that support from the agency head led to less burnout for the more educated social workers (Master’s degree) than for workers with only a Bachelor’s degree conflicted with previous research on 295 social workers in Israeli welfare agencies (Kahana-Friedman, 2000), which demonstrated stronger burnout among more highly educated workers. Kahana-Friedman suggested that higher burnout was related to higher expectations from work alongside high personal ambitions, in light of their education. One may conjecture that the current finding—where support from the agency head was found important in reducing highly edu- cated workers’ burnout—may indicate the agency head’s awareness of employees’ high stress levels and possibly even the agency head’s initiation of changes to improve work conditions, which may encourage employee longevity. However, such speculation must be followed up by further

1348 Liat Hamama

quantitative and qualitative study to uncover the links between education and burnout.

An additional related finding was that children’s social workers reported higher support from the agency head than did workers serving adults. This finding may stem from the unique characteristics of working with children. Children’s social workers need open and sup- portive communication with the agency head to access special resources conducive to complex work with young users. For example, children’s social workers must creatively incorporate play materials to motivate and treat children (Pearlman and Saakvitne, 1995; Rae and Fournier, 1999; Ronen, 2003).

Regarding the correlation between demographic characteristics and experience of burnout, as predicted in the fourth hypothesis, the current results showed the following. Older workers were found to be less prone to burnout than younger ones. This finding strengthened prior research showing young employees, particularly below thirty, to be more prone to burnout (e.g. Bar-Zaselvisky, 2003; Lauderdale, 1982; Maslach, 1982; McNeelly, 1992). Regarding professional experience, the current results revealed that social workers with more experience tended to report less burnout, thus substantiating prior research on Israeli workers (Bar-Zaselvisky, 2003; Navaro, 1999). Possibly, workers who are older and more experienced tend to be more resilient and perhaps less preoccu- pied with work on account of family commitments, thus enhancing their sense of strength and security and helping them develop realistic expec- tations of the workplace. Moreover, they may be exposed to more challen- ging roles and rewards (Navaro, 1999).

This study’s findings may contribute to existing theoretical knowledge on burnout and stress, and may also assist in developing practical organis- ational strategies to reduce workers’ experience of burnout or increase immunity to it. This study offered the opportunity to examine theoretical differences in burnout between social workers who directly treat children and youth and those who treat adults, related to the psychological, physical and social dimensions of their workplace. Future research needs to develop and validate these findings.

From a practicable point of view, the results of this study may contribute to some reappraisal of policies for social work agencies. Such organisations may wish to work on strengthening the opportunities available for social support at all levels of their organisations: support by colleagues, the direct supervisor and the agency head. In particular, they may do well to enhance intrinsic rewards such as providing workers with an improved sense of autonomy, an increased sense of control, a greater variety of work and tasks, a stronger sense of meaning in work and greater pro- fessional fulfilment.

Social Workers’ Burnout 1349

Acknowledgement

The author would like to thank Dee B. Ankonina for her editorial assistance.

References

Abu-Bader, S. H. (1998) ‘Predictors of Work Satisfaction between Arab and Jewish

Social Workers in Israel’, unpublished doctoral dissertation, Salt Lake City, Univer-

sity of Utah.

Abu-Bader, S. H. (1999) ‘Predictors of burnout among social workers in Israel’, Journal

of Comparative Social Welfare, 15(1), pp. 52 – 61.

Abu-Bader, S. H. (2000) ‘Work satisfaction, burnout, and turnover among social workers

in Israel: A causal diagram’, International Journal of Social Welfare, 9, pp. 191 – 200.

Acker, G. (1999) ‘The impact of clients’ mental illness on social workers’ job satisfaction

and burnout’, Health and Social Work, 24, pp. 112 – 19.

Amrani-Cohen, I. (1999) ‘Resilience among social workers: A cross cultural study of

Americans and Israelis’, unpublished doctoral dissertation, Boston College.

Arnold, J. and Feldman, D. (1986) Organizational Behavior, New York, McGraw-Hill.

Balloch, S., Pahl, J. and McLean, J. (1998) ‘Working in the social services: Job satisfac-

tion, stress and violence’, British Journal of Social Work, 28, pp. 329 – 50.

Bar-Zaselvisky, T. (2003) ‘The link between seniority and dimensions of job satisfaction:

A meta-analysis’, unpublished Master’s thesis, Tel Aviv University.

Bargal, D. and Guterman, N. (1996) ‘Perceptions of job satisfaction, service effectiveness

and burnout among social workers in Israel’, Society and Welfare, 16(4), pp. 541 – 65

(in Hebrew).

Bargal, D. and Guterman, N. (1997) ‘Career outcomes among medical vs. family service

social workers in Israel’, Social Work in Health Care, 25(1/2), pp. 223 – 41.

Ben-Zur, H. and Michael, K. (2007) ‘Burnout, social support, and coping at work among

social workers, psychologists, and nurses: The role of challenge/control appraisals’,

Social Work in Health Care, 45(4), pp. 63 – 85.

Bennett, P., Evans, R. and Tattersall, A. (1993) ‘Stress and coping in social workers: A

preliminary investigation’, British Journal of Social Work, 23, pp. 31 – 44.

Coffey, M., Dugdill, L. and Tattersall, A. (2004) ‘Research note: Stress in social services:

Mental well-being, constraints and job satisfaction’, British Journal of Social Work,

34(5), pp. 735 – 47.

Cohen, E. (1990) ‘The burnout phenomenon among social workers during their first two

and five years in professional work and its link to the worker’s sex and social support’,

unpublished Master’s thesis, Tel Aviv University (in Hebrew).

Collings, J. and Murray, P. (1996) ‘Predictors of stress amongst social workers: An

empirical study’, British Journal of Social Work, 26, pp. 375 – 87.

Deeter-Schmelz, D. R. and Ramsey, R. P. (1997) ‘Considering sources and types of social

support: A psychometric evaluation of the House and Wells (1978) instrument’,

Journal of Personal Selling and Sales Management, 17(1), pp. 49 – 61.

Dollard, M. F., Winefield, H. R., Winefield, A. M. and De Jonge, J. (2000) ‘Psychological

job strain and productivity in human service workers: A test of the

1350 Liat Hamama

demand-control-support model’, Journal of Occupational and Organizational Psy-

chology, 73(4), pp. 501 – 10.

El-Bassel, N., Guterman, N., Bargal, D. and Kuo-Hsien, S. (1998) ‘Main and buffering

effects of emotional support on job and health related strains: A national survey of

Israeli social workers’, Employee Assistance Quarterly, 13(3), pp. 1 – 18.

Etzion, D. and Pines, A. (1986) ‘Sex and culture factor explaining coping and burnout

among human service professionals: A social psychological perspective’, Journal of

Cross-Culture Psychology, 17, pp. 191 – 209.

Evans, S., Huxley, P., Gately, C., Webber, M., Mears, A., Pajak, S., Medina, J., Kendall,

T. and Katona, C. (2006) ‘Mental health, burnout and job satisfaction among mental

heath social workers in England and Wales’, British Journal of Psychiatry, 188, pp.

75 – 80.

Ezer, S. (2003) ‘Burnout and organizational commitment among teachers at MIFTAN’,

unpublished Master’s thesis, Bar Ilan University, Ramat Gan, Israel (in Hebrew).

Freudenberger, H. J. (1980) Burnout: The High Cost of High Achievement, Garden City,

NY, Doubleday.

Gorlick, Y. (1998) ‘Occupational stress and burnout among social workers in health and

family treatment agencies’, unpublished doctoral thesis, Hebrew University, Jerusa-

lem (in Hebrew).

Guterman, N. and Bargal, D. (1996) ‘Social workers’ perceptions of their power and

service outcomes’, Administration in Social Work, 20(3), pp. 1 – 20.

Heppner, P. P., Kivlighan, D. M. and Wampold, B. E. (1992) Research Design in Coun-

seling, Pacific Grove, CA, Brooks/Cole.

Hobfoll, S. E. (1989) ‘Conservation of resources: A new attempt at conceptualizing

stress’, American Psychologist, 44, pp. 513 – 25.

Houkes, I., Janssen, P. P. M., de Jonge, J. and Bakker, A. B. (2003) ‘Specific determinants

of intrinsic work motivation, emotional exhaustion and turnover intention: A multi-

sample longitudinal study’, Journal of Occupational and Organizational Psychology,

76(4), pp. 427 – 50.

House, J. S. (1981) Work Stress and Social Support, Reading, MA, Addison-Wesley.

Howe, D. (2004) ‘Relating theory to practice’, in M. Davies (ed.), The Blackwell Compa-

nion to Social Work, 2nd edn, Oxford, Blackwell, pp. 82 – 7.

Huxley, P., Evans, S., Gately, C., Webber, M., Mears, A., Pajak, S., Kendall, T., Medina,

J. and Katona, C. (2005) ‘Stress and pressures in mental health social work: The

worker speaks’, British of Journal of Social Work, 35, pp. 1063 – 79.

Jayaratne, S. and Chess, W. A. (1984) ‘Job satisfaction, burnout, and turnover: A national

study’, Social Work, 29, pp. 448 – 53.

Jayaratne, S., Davis-Sacks, M. L. and Chess, W. A. (1991) ‘Private practice may be good

for your health and well-being’, Social Work, 36(3), pp. 224 – 9.

Jayaratne, S., Vinokur-Kaplan, D. and Chess, W. A. (1995) ‘The importance of personal

control: A comparison of social workers in private practice and public agency set-

tings’, Journal of Applied Social Sciences, 19(1), pp. 47 – 59.

Jenkins, R. and Elliott, P. (2004) ‘Stressors, burnout and social support: Nurses in acute

mental health settings’, Journal of Advanced Nursing, 48(6), pp. 622 – 31.

Kahana-Friedman, G. (2000) ‘Stressors and their influence on burnout among social

workers in the welfare services: Self-efficacy and colleague support as moderating

variables’, unpublished Master’s thesis, Tel Aviv University (in Hebrew).

Kahn, R., Wolfe, D., Quinn, R. P., Snoek, J. O. and Rosenthal, R. A. (1964) Organiz-

ational Stress: Studies in Role Conflict and Ambiguity, New York, Wiley.

Social Workers’ Burnout 1351

Kane, R. (1984) ‘Social work as a health profession’, in D. Mechanic (ed.), Handbook of

Health, Health Care and the Health Professions, New York, Free Press, pp. 495 – 522.

Karasek, R. and Theorell, T. (1990) Healthy Work, New York, Basic Books.

Kary, H. (1990) ‘The influence of the job on burnout among workers in the welfare

service agencies’, unpublished Master’s thesis, University of Haifa, Israel (in

Hebrew).

Katchalnick, K., Aviram, U. and Katan, J. (1991) ‘Graduates of social work schools in the

first stages of their career: Realization of occupational values, satisfaction and com-

mitment toward the career’, Society and Welfare, 12(1), pp. 56 – 71 (in Hebrew).

Kazdin, A. E. (2000) Psychotherapy for Children and Adolescents: Directions for

Research and Practice, New York, Oxford University Press.

Kazi, M. A. F. (2000) ‘Evaluation of social work practice in England’, Journal of Social

Work Research and Evaluation, 1, pp. 101 – 9.

Kershner-Cohen, M. (1998) ‘The link between workers’ stress and the quality of service:

The moderating influence of social support’, unpublished Master’s thesis, Tel Aviv

University (in Hebrew).

Kim, H. and Stonere, M. (2008) ‘Burnout and turnover intention among social workers:

Effects of role stress, job autonomy and social support’, Administration in Social

Work, 32(3), pp. 5 – 25.

Lauderdale, M. L. (1982) Burnout: Strategies for Personal and Organizational Life:

Speculations on Evolving Paradigms, Austin, TX, Learning Concepts.

Lazarus, R. S. (1999) Stress and Emotion: A New Synthesis, London, Free Association

Books.

Lecroy, C. and Rank, M. (1987) ‘Factors associated with burnout in the social services:

An exploratory study’, Journal of Social Services Research, 11, pp. 23 – 39.

Liu, C., Spector, P. E. and Jex, S. M. (2005) ‘The relation of job control with job strains: A

comparison of multiple data sources’, Journal of Occupational and Organizational

Psychology, 78, pp. 325 – 36.

Lloyd, C., King, R. and Chenoweth, L. (2002) ‘Social work, stress and burnout: A review’,

Journal of Mental Health, 11(3), pp. 255 – 65.

Lonergan, B. A., O’Halloran, M. S. and Crane, S. C. (2004) ‘The development of the

trauma therapist: A qualitative study of the child therapist’s perspectives and experi-

ences’, Brief Treatment and Crisis Intervention, 4(4), pp. 353 – 66.

Maslach, C. (1982) Burnout: The Cost of Caring, Englewood Cliffs, NJ, Prentice Hall.

Maslach, C. and Goldberg, J. (1998) ‘Prevention of burnout: New perspectives’, Applied

and Preventive Psychology, 7, pp. 63 – 74.

Maslach, C. and Jackson, S. E. (1981) ‘The measurement of experienced burnout’,

Journal of Occupational Behavior, 2, pp. 99 – 113.

Maslach, C. and Jackson, S. E. (1986) The Maslach Burnout Inventory Manual, 2nd edn,

Palo Alto, CA, Consulting Psychologists Press.

Maslach, C., Schaufeli, W. B. and Leiter, M. P. (2001) ‘Job burnout’, Annual Review of

Psychology, 52, pp. 397 – 422.

McNeelly, R. (1992) ‘Job satisfaction in the public social services: Perspective on struc-

ture, situational factors, gender and ethnicity’, in Y. Hasenfeld (ed.), Human Services

as Complex Organizations, Newbury Park, CA, Sage, pp. 224 – 56.

Navaro, H. (1999) ‘The link between job overload and work satisfaction: Direct and mod-

erated effects of self-efficacy, seniority, and teamwork in different health professions’,

unpublished Master’s thesis, Tel Aviv University (in Hebrew).

1352 Liat Hamama

Pearlman, L. A. and Saakvitne, K. W. (1995) ‘Treating therapists with vicarious trauma-

tization and secondary stress disorders’, in C. R. Figley (ed.), Compassion Fatigue,

New York, Brunner/Mazel, pp. 150 – 77.

Peterson, U., Bergström, G., Samuelsson, M., Asberg, M. and Nygren, A. (2008) ‘Reflect-

ing peer-support groups in the prevention of stress and burnout: Randomized con-

trolled trial’, Journal of Advanced Nursing, 63(5), pp. 506 – 16.

Pines, A. (1982) ‘Changing organizations: Is a work environment without burnout an

impossible goal?’, in W. S. Paine (ed.), Job Stress and Burnout, Beverly Hills, Sage,

pp. 189 – 211.

Pines, A. and Aronson, E. (1988) Career Burnout: Causes and Cures, New York, Free

Press.

Pines, A., Aronson, E. and Kafry, D. (1981) Burnout: From Tedium to Personal Growth,

New York, Free Press.

Rabin, S., Saffer, M., Weisberg, E., Kornitzer-Enav, T., Peled, I. and Ribak, J. (2000) ‘A

multifaceted mental health training program in reducing burnout among occupational

social workers’, Israel Journal of Psychiatry and Related Sciences, 37(1), pp. 12 – 19.

Rae, W. A. and Fournier, C. J. (1999) ‘Ethical and legal issues in the treatment of chil-

dren and families’, in S. W. Russ and T. H. Ollendick (eds), Handbook of Psy-

chotherapies with Children and Families, New York, Kluwer Academic/Plenum,

pp. 67 – 83.

Reagh, R. (1994) ‘Public child welfare professionals: Those who stay’, Journal of Soci-

ology and Social Welfare, 21, pp. 69 – 78.

Ronen, T. (2003) Cognitive Constructivist Psychotherapy with Children and Adolescents,

New York, Kluwer/Plenum.

Ronen, T. and Hamama, L. (in press) ‘Social work with children and adolescence in

Israel’, in M. Hovav, Y. Katan and E. Lawental (eds), Social Work in Israel,

Tel-Aviv, Probook.

Siefert, K., Jayaratne, S. and Chess, W. A. (1991) ‘Job satisfaction, burnout and turnover

in health care social workers’, Health and Social Work, 16, pp. 193 – 202.

Söderfeldt, M., Söderfeldt, B. and Warg, L. E. (1995) ‘Burnout in social work’, Social

Work, 40, pp. 638 – 46.

Stav, A. (1982) ‘The link between rehabilitation treatment in situations of loss: Bereave-

ment and disability and the professional burnout phenomenon among rehabilitation

workers’, unpublished Master’s thesis, University of Haifa, Israel (in Hebrew).

Stav, A., Florian, V. and Zernitsky-Shurka, E. (1987) ‘Burnout among social workers

working with physically disabled persons and bereaved families’, Journal of Social

Service Research, 10(1), pp. 81 – 94.

Tham, P. (2007) ‘Why are they leaving? Factors affecting intention to leave among social

workers in child welfare’, British Journal of Social Work, 37, pp. 1225 – 46.

Thompson, N., Stradling, S., Murphy, M. and O’Neill, P. (1996) ‘Stress and organizational

culture’, British Journal of Social Work, 26, pp. 647 – 65.

Um, M. Y. and Harrison, D. F. (1998) ‘Role stressors, burnout, mediators, and job satis-

faction: A stress-strain-outcome model and an empirical test’, Social Work Research,

22(2), pp. 100 – 15.

Vinokur-Kaplan, D. (1991) ‘Job satisfaction among social workers in public and volun-

tary child welfare agencies’, Child Welfare, 70(1), pp. 81 – 91.

Waterhouse, L. (2008) ‘Child abuse’, in M. Davies (ed.), The Blackwell Companion to

Social Work, 3rd edn, Oxford, Blackwell, pp. 18 – 26.

Social Workers’ Burnout 1353

may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express

written permission. However, users may print, download, or email articles for individual use.