Emotional Intelligence, Work Satisfaction, and Affective Commitment: An Occupational Health Study of Social Workers

Introduction

Social work, a profession which is expected to provide practical and psychological help to people in need, improve their social functioning (Lloyd et al., 2002). This societal expectation is widely accepted and makes people perceive that their clients’ concerns and problems should be solved by social workers (Jones & Novak, 1993). For social workers, working under this expectation could generate a lot of difficulties, making them to have various issues in personal and working aspects. Previous studies have found that social workers have high levels of general anxiety and poor mental well-being (Aiello & Tesi, 2017; Bennett et al., 1993; Bradley & Sutherland, 1995). In working aspect, social worker is a specific professional group particularly exposed to a high risk of developing stress and burnout (e.g., Bamber, 2006; Wu & Pooler, 2014).

In the literature of burnout, the main ideas of tremendous studies have aimed to find out factors to eliminate burnout, but with the evolution of positive psychology (Seligman & Csikszentmihalyi, 2000) scholars have suggested that work engagement can be the opposite of burnout out (see Maslach et al., 2001; Schaufeli et al., 2002). Recent research has proved that work engagement and burnout can coexist instead of being antipode (see Kuok et al., 2022; Kuok & Taormina, 2017). Work engagement has been found to be related to employees’ outcomes at work as well, e.g., work engagement negatively related to turnover intention and positively related to career success (Kuok, 2020; Kuok & Taormina, 2017). By adopting the idea of positive psychology in work setting, it becomes critical to understand employees’ both dark side and bright side of work, especially, studying positive work-related health outcomes in the social work context, which was relatively rare (Aiello & Tesi, 2017). This is a uniqueness of this study: extending the latest work engagement theory (i.e., burnout and work engagement are coexistent, work engagement not being the opposite of burnout) to the field of professional health, social workers in this study.

In Lloyd et al.’s (2002) review of social workers’ stress and burnout, it is suggested that previous studies have mainly focused on two main directions. The first one was about making comparison in stress level between social workers and other occupations (Bradley & Sutherland, 1995; Rushton, 1987). The second direction of studying social workers’ burnout was aimed to found out the possible organizational factors (e.g., job satisfaction and organizational commitment) that related to their stress (Balloch et al., 1998; Collings & Murray, 1996; Glisson & Hemmelgarn, 1998; Martin & Schinke, 1998). Moreover, in the recent literature of human service management, there has been relatively more research studying social workers’ retention in terms of job satisfaction, commitment, burnout, and work engagement as outcomes of social worker about their work (Brown et al., 2019; Geisler et al., 2019).

While there have been either very limited studies focusing on studying the relationship between personal factors and burnout, it was not found associated with burnout, and client-related factors were rarely mentioned in studies (Collings & Murray, 1996; Lloyd et al., 2002). In particular, the demands on social services have been increasing dramatically, while social workers have experienced a loss of available personal and organizational resources and control in their work tasks. (Aiello & Tesi, 2017). Thus, this study addressed another research gap, which is to investigate the antecedents (not either organizational or personal factor, but both) of social workers burnout and work engagement.

It becomes very important to study social workers’ both positive and negative conditions, not limited to organizational factors, but also the personal and client-related factors. Thus, emotional intelligence may be the key element as a mix of personal and client-related factors. Since the job nature of social workers has been expected to use up a lot of personal and organizational recourses due to dealing with clients’ difficulties, it is understandable that emotions could be generated though the interaction with clients. Therefore, social workers’ ability to understand other’s emotions and controlling one’s emotions become key to affecting their working condition (Salovey & Mayer, 1990). In addition, scholars have suggested that the study of emotional intelligence, burnout, and work engagement is very limited in the literature of social workers (Geng et al., 2011) and relevant health professionals (Nstas & Frca, 2015). Therefore, this study was conducted to understand how personal resources (i.e., emotional intelligence) could influence social workers’ perception at work (i.e., burnout and work engagement).

In addition, the social exchange theory assumes that people participate in exchange behavior because they think their reward would justify their cost (Porter et al., 1974), i.e., the exchange of employees’ effort and loyalty for the organization’s provision of material and socio-emotional benefits and vice versa. Previous studies have shown that work satisfaction and organizational commitment (organizational variables) are the consequences of being treated fairly by the organization (Kuok & Taormina, 2015). By adapting the social exchange theory to this study, work satisfaction and organizational commitment instead of being counted as the outcomes can be considered as the antecedents that may have an impact on social workers’ perception at work. That is, if they gain satisfaction at work and commitment from the organization, they would perceive less burnout and more engagement in return.

Thus, this research aimed to (1) to investigate the impact of emotional intelligence on burnout and work engagement respectively and (2) to understand the additional effect of work satisfaction and commitment on burnout and work engagement respectively.

Maslach and Jackson (1981) described burnout as a state of physical, emotional, and mental exhaustion, with three components, namely emotional exhaustion, depersonalization, and reduced personal accomplishment. Their study initially addressed human service providers, same as social workers, who work with people in highly demanding working conditions. Emotional exhaustion is a central component of burnout, a feeling of excessive emotional stress while being drained by contacting with demanding workload or other people, specifically to a large extent of tiredness, somatic symptoms, decreased emotional resources, and a sense of nothing left to give to work. Depersonalization is an essential factor for detecting burnout; the characteristic is about callous feelings without involving any personal reaction towards the acceptance of others’ interest or care service. Reduced personal accomplishment is a subjectively self-evaluation of work performance and ability – one perceives that one cannot perform as good as one originally could think (Maslach et al., 1996).

In a review study of burnout, it was mentioned that work engagement could be the fourth component of burnout, a positive facet of burnout (Maslach et al., 2001). This concept was quickly adopted by several scholars, and the Utrecht work engagement measurement was created based on the concept of work engagement as the antipode of burnout (Schaufeli et al., 2002). However, recent research has provided evidence to suggest this concept of work engagement as the antipode of burnout may not be correct, and proved that work engagement and burnout can be coexistent (Kuok & Taormina, 2017).

Kahn (1990) first conceptualized engagement at work as the “harnessing of organizational members’ selves to their work roles; in engagement, people employ and express themselves physically, cognitively, and emotionally during role performances” (p. 694). That is, people bring their personal selves into their work. However, this remains as a concept until Rothbard (2001), who was inspired by Kahn (1990), developed a similar idea by defining engagement as a two-dimensional construct that includes “attention” (i.e., the cognition and the time one spends thinking about one's role at work) and “absorption” (i.e., the intensity of one’s focus on a role at work). These two dimensions were mainly focused on the cognitive aspect of Kahn’s (1990) conceptualization of work engagement. In addition, Saks (2006) adopted the conceptualization of work engagement as role related and created a new idea. That is, an employee is psychologically present in a particular organization role. For Saks (2006), “the two most dominant roles for most organizational members are their work role and their role as a member of an organization” (p. 604). Therefore, he suggested that work engagement could be distinguished from organizational engagement. Kuok and Taormina (2017) also adopted Kahn’s (1990) concept of work engagement, and created the operational definition of cognitive, emotional, and physical work engagement.

Cognitive Work Engagement is defined as “the intentional and actively focused awareness of one’s tasks, objectives, or organizational activities that is characterized by willingly calling one’s attention to and having positive thoughts about one’s work, with the purpose of improving one’s effectiveness at those tasks, objectives, or activities” (Kuok & Taormina, 2017, p. 266).

Emotional Work Engagement is defined as “the willing attachment to tasks, objectives, or organizational activities that is characterized by having positive feelings, such as pride, enthusiasm, and excitement, about actively executing and completing those tasks, objectives, or activities” (Kuok & Taormina, 2017, p. 266).

Physical Work Engagement is defined as “the bodily involvement in tasks, objectives, or organizational activities by intentionally and voluntarily utilizing one’s energy and effort to execute and complete those tasks, objectives, or activities” (Kuok & Taormina, 2017, p. 267).

Emotional Intelligence

Emotional intelligence explicates the cognitive and emotional mechanisms that process emotional information, and provides a unified framework to study the role of emotional abilities in social functioning (Mayer & Salovey, 1997). The term of emotional intelligence consists of appraisal and expression of emotion, regulation of emotion, and utilization of emotions in solving problems (Salovey & Mayer, 1990). Higher emotional intelligence is one of the resilient factors for preventing burnout in social service workers. For example, research has found that of emotional intelligence was negatively related with their level of burnout for marriage and family counselors burnout (Gutierrez & Mullen, 2016) and social workers (Stone, 2016). Similarly, research about burnout among similar professionals has shown a negative correlation between burnout and emotional intelligence among doctors (Swami et al., 2013) and teachers (Mérida-López & Extremera, 2017).

Mayer and Salovey (1997) divided emotional intelligence into four facets: perceiving others’ emotion, using emotion to solve problems, understanding one’s own emotion, and managing one’s own emotion, which gives more emphasis to the cognitive components of emotional intelligence and conceptualizes emotional intelligence in terms of potential for intellectual and emotional growth (Brackett et al., 2006). Moreover, Ciarrochi et al. (2001) also suggested emotional intelligence has four components, namely perception of self and others’ emotions (i.e., the ability to understand one’s own and others’ emotions), skills at managing self and others’ emotions (i.e., the ability to control one’s own and others’ emotions). In this research, “self-control” through managing one’s own emotion and “understanding others” through perceiving emotions, which are both abilities emphasized by the psychological professions and are effective for personal intervention, were studied.

Research has found that understanding others and self-control can be effective factors in reducing the possibility of burnout among helping professionals. For example, research has found that understanding others is negatively correlated with devaluing the personal accomplishment, and self-control is negatively correlated with depersonalization and personal accomplishment among nurses (Görgens-Ekermans & Brand, 2012). Thus, H1: Social workers’ emotional intelligence of self-control is negatively correlated to their (a) emotional exhaustion, (b) depersonalization, and (c) reduced personal accomplishment.

Similarly, a research about burnout and emotional intelligence among teachers found that the negative predictor for depersonalization of burnout was intrapersonal competence of emotional intelligence, and interpersonal competence was the positive predictor of personal accomplishment (Pishghadam & Sahebjam, 2012). Along these lines, research about counselors has found that global trait emotional intelligence can effectively predict exhaustion negatively (Gutierrez & Mullen, 2016). Thus, H2: Social workers’ emotional intelligence of understanding others is negatively correlated to their (a) emotional exhaustion, (b) depersonalization, and (c) reduced personal accomplishment.

In the emotional intelligence literature, research on both emotional intelligence and work engagement in the field of helping professionals (e.g., social workers) is rarely found. However, in similar professionals, namely teachers, some researchers actually have tried to study emotional intelligence as a mediator between work satisfaction and work engagement (Mérida-López et al., 2019) and as an antecedent of work satisfaction (Extremera et al., 2018). In addition, both above mentioned studies suggest suggested that there is a positive relationship between emotional intelligence and work engagement. Thus, H3: Social workers’ emotional intelligence of self-control is positively correlated to their (a) cognitive, (b) emotional, and (c) physical work engagement.

H4: Social workers’ emotional intelligence of understanding others is positively correlated to their (a) cognitive, (b) emotional, and (c) physical work engagement.

Work Satisfaction

Job satisfaction is the “pleasurable emotional state resulting from the appraisal of one’s job as achieving or facilitating one’s job values” (Locke, 1969, p. 316). Apart from the single global dimension of job satisfaction, Smith et al. (1969) identified that there were five facets of job satisfaction, i.e., pay, promotion, coworkers, supervisors, and the work itself. In this research, only employees’ satisfaction with the work itself was selected. The main reason to keep job satisfaction at work only was that this research was aimed to study social workers across organizations. It was more meaningful to study their satisfaction at work than other aspects, as the organizations may vary in their organizational structure while the nature of social workers was similar to each other, thus, work satisfaction referred to social workers’ affective orientation toward their job.

In addition, Abu-Bader’s (2003) study of social workers found that there was a negative relationship between work satisfaction and burnout. Thus, H5: Social workers’ work satisfaction is negatively correlated to their (a) emotional exhaustion, (b) depersonalization, and (c) reduced personal accomplishment.

Besides, a recent study has confirmed that work engagement was positively related to work satisfaction among social workers in Sweden (Geisler et al., 2019). Thus, H6: Social workers’ work satisfaction is positively correlated to their (a) cognitive, (b) emotional, and (c) physical work engagement.

Affective Organizational Commitment

Allen and Meyer (1990) developed the theory and the measurement of organizational commitment. There are three facets of organizational commitment, namely affective commitment – the emotional attachment to the organization, i.e., the employees want to stay in the organization - continuance commitment – motivation to remain with an organization because they have to stay - and normative commitment – a sense of moral obligation to remain with an organization because they ought to. In this research, it is mainly addressed how social workers want to stay in the organization (affective organizational commitment). It was similar to Jaskyte and Lee’s (2009) study of social workers’ organizational commitment, that is, an affective state based on the perception of concordance between one’s and the organization’s values. In addition, Kuok (2017) stated that employees’ affective commitment was average in a study of employees’ dissonance across industries (including social work) in a casino dominant economy, suggesting whether an employee wants to stay in one’s organization is very important to resolve the conflict of one's perception at work.

Moreover, a recent study on social workers’ burnout has found that there is a negative relationship between burnout and organizational commitment (Brown et al., 2019). Thus, H7: Social workers’ affective commitment is negatively correlated to their (a) emotional exhaustion, (b) depersonalization, and (c) reduced personal accomplishment.

On the other hand, work engagement was confirmed to have a positive relationship with organizational commitment (Geisler et al., 2019). Thus, H8: Social workers’ affective commitment is positively correlated to their (a) cognitive, (b) emotional, and (c) physical work engagement.

Method

Respondents

The sample consisted of 199 social workers (62 males, 137 females), with an average age is 30.85 (SD = 7.40) and an age range from 22 to 61 years old. Concerning their working experience, the mean was 60.88 months, ranging from 0 to 370 months. As for the major of qualifications, 138 were graduated in social work, and 61 were graduated in psychology.

Measures

Burnout

This varaible was measured by Maslach and Jackson’s (1981) Burnout Inventory. It is a 22-item scale with three facets, namely 9 items for emotional exhaustion, 5 items for depersonalization, and 8 items for reduced personal accomplishment. A sample item of emotional exhaustion is “I feel emotionally drain from my work” (original alpha = .86); an item of depersonalization is “I feel I treat some recipients as if they were impersonal ‘object’” (original alpha = .81); an item of personal accomplishment is “I can easily understand how my recipients feel about things” (original alpha = .81). The respondents were asked to what extend they agreed or disagreed about the statements describing them, in a 5-point Likert scale (from 1 = strongly disagree to 5 = strongly agree).

Emotional Intelligence

The Self-Rated Emotional Intelligence Scale (SREIS; Brackett et al., 2006) was adopted to measure this variable in the current research. The 9-item perceiving emotions subscale in the original scale was adopted to measure the domain of understanding others (e.g., “I can tell how people are feeling by listening to the tone of their voice” (original alpha = .84). The 9-item managing emotions subscale in the original scale was adopted to measure the domain of self-control (e.g., “I have difficulty managing my emotions” (original alpha = .84). The respondents were asked to what extend they agreed or disagreed about the statements describing them, in a 5-point Likert scale (from 1 = strongly disagree to 5 = strongly agree).

Work Satisfaction

This variable was measured by using the updated version of the Smith et al.’s (1969) Job Descriptive Index by Roznowski’s (1989). Items (e.g. “Gives a sense of accomplishment” (original alpha = .87) for work satisfaction were used. The respondents were asked whether they agreed or disagreed about the statements describing their work. Following Roznowski’s (1989) scoring, a negative response (‘Yes’ to a negative or ‘No’ to a positive item) was scored 1; a ‘Not sure’ was scored 2; and a positive response (‘Yes’ to a positive or ‘No’ to a negative item) was scored 3.

Affective Organizational Commitment

This variable was measured through the affective commitment subsacle of Allen and Meyer’s (1990) Organziational Commitment Scale. A sample item of affective commitment was “I enjoy discussing my organization with people outside it” (original alpha = .86). The respondents were asked to what extent they agreed or disagreed about the statements described their organization, in a 5-point Likert scale (from 1 = strongly disagree = to 5 = strongly agree).

Procedure

Since all the original instruments were in English but the target population were Chinese, a back-to-back translation was adopted for all the items in order to ensure the participants understood all the questions. A formal letter with the purpose and the organiser of the study was sent to seven non-profit social service organizations in Macao, asking their permission for the researchers to contact their social workers to participate in this study. As a result, five provided the permission, and the social workers belonging to these entities were approached individually, informing them about the purpose of the study, asking their consent and providing them with the questionnaires to fill in; they were also informed that they could drop out at any time while they were filling in the questionnaire. Moreover, in order to motivate the respondents to participate in this study, non-monetary incentives were applied, but only ensuring anonymity and confidentiality (see Burns et al., 2014).

Ethical Considerations

Ethical approval was obtained from the university research ethics committee before the study was conducted, and ethical guidelines of the American Psychological Association were followed. Informed consent of the participants was requested both verbally and for completing the questionnaire, which stated the purpose of the survey and provided the researcher’s contact information. Potential respondents were informed that their participation was voluntary, and they could stop responding at any time. They were also told that no personal information was being asked, and their responses would remain confidential. They were also informed that the data would be used only for academic purposes and only in aggregated statistical form.

Results

Test for Mulitcollinearity

Multicollinearity is a statistical phenomenon in which predictor variables in a multiple regression model are highly correlated, which could lead to inaccuracy in statistical analyses. This was assessed by a “tolerance” (1- R²) test for each independent variable. According to Hair et al. (1998, pp. 191-193), a tolerance value of less than .10 is problematic. This test used all the independent variables (for the planned regressions), and regressed each one on all the other independent variables (excluding demographics, because many of them are naturally correlated, e.g., age and marital status). Tolerance values for the independent variables ranged from .76 to .88, all above the .10 cutoff, indicating that multicollinearity was not a concern.

Test for Common Method Bias

Common method bias is a statistical phenomenon in which statistical relationships may be based on the measurement method but not on the measure of the construct. This was assessed by factor-analyzing all the variables in this study together, and using the “maximum-likelihood” approach with a forced, one-factor solution (see Harman, 1960). If a ratio of the resulting chi-square value over the degrees of freedom is less than 2.00:1, it indicates common-method bias (i.e., a single factor). For this study, the ratio was 6.65:1, suggesting that common-method bias was not a concern.

Intercorrelations

Means, standard deviations, and intercorrelations were computed for all variables to test the hypothesised relationships between emotional intelligence, work satisfaction, affective organizational commitment as well as burnout and work engagement.

For emotional intelligence, there was no correlation between emotional intelligence of controlling self and emotional exhaustion, but negative correlations were found between emotional intelligence of controlling self and depersonalization (r = -.17, p < .05), and reduced personal accomplishment (r = -.41, p < .001) respectively. These supported H1b and c, but not H1a. Also, there was no correlation between emotional intelligence of understanding others and emotional exhaustion and depersonalization respectively, but negative correlation was found between emotional intelligence of understanding others and reduced personal accomplishment (r = -.36, p < .001) respectively. These supported H2c, but not H2a and b.

Moreover, emotional intelligence of controlling self was found to have positive correlations to cognitive work engagement (r = .19), emotional work engagement (r = .29), and physical work engagement (r = .27) with all ps < .001. These supported H3a to c. On the other hand, no correlation was found between emotional intelligence of understanding others and the three facets of work engagements, which did not support H4a to c.

For work satisfaction, it was found to have negative correlations to emotional exhaustion (r = -.48), depersonalization (r = -.46), and reduced personal accomplishment (r = -.27) with all ps < .001. In addition, it was found to have positive correlations to cognitive work engagement (r = .19, p < .01), emotional work engagement (r = 03, p < .001), and physical work engagement (r = .35, p < .001). These supported H5a to c as well as H6a to c.

For affective commitment, it was found to have negative correlations to emotional exhaustion (r = -.33), depersonalization (r = -.27), and reduced personal accomplishment (r = -.28) with all ps < .001. In addition, it was found to have positive correlations to cognitive work engagement (r = .27), emotional work engagement (r = 0.46), and physical work engagement (r = .50) with all ps < .001. These supported H7a to c as well as H8a to c. The results for these correlations were shown in Table 1. As well as referring to the summary of the hypotheses in Table 2.

In order to further investigate the effects of emotional intelligence, as well as the additional effects of work satisfaction and affective commitment, on social workers’ burnout and work engagement, six linear stepwise regressions were run, the first three were using the three factors of burnout as criterion measures in which the predictors were emotional intelligence of controlling self and understanding others in the first block, then putting work satisfaction and affective commitment in the second block to check the additional effects, including demographics as control variables. Then, the procedure of the first three linear regressions for the rest three by replacing the three facets of work engagement as criterion measures was repeated.

For emotional exhaustion, age and work satisfaction proved to be very good positive predictors. The strongest predictor was “work satisfaction”, showing that β was -.460 and ΔR² was .21, with a significant p < .001. The second predictor was “age”, showing that β was -.194 and ΔR² was .05, with a significant p < .001. These variables combined and formed as powerful predictors (R² = .26, F = 35.13, p < .001).

For depersonalization, emotional intelligence of controlling self and work satisfaction proved to be very good positive predictors. The strongest predictor was “work satisfaction”, showing that β was -.439 and ΔR² was .17, with a significant p < .001. The second predictor was “emotional intelligence of controlling self”, showing that β was -.132 and ΔR² was .04, with a significant p < .05. These variables combined and formed as powerful predictors (R² = .21, F = 20.51, p< .001).

For reduced personal accomplishment, emotional intelligence of controlling self and understanding others, work satisfaction and affective commitment proved to be very good positive predictors. The strongest predictor was “emotional intelligence of controlling self”, showing that β was -.303 and ΔR² was .16, with a significant p < .001. The second predictor was “emotional intelligence of understanding others”, that showed that β was -.224 and ΔR² was .05, with a significant p < .005. The third one was “work satisfaction”, that showed that β was -.167 and ΔR² was .05, with a significant p < .05. The last one was “affective commitment”, that showed that β was -.146 and ΔR² was .02,with a significant p < .05. These variables combined and formed as powerful predictors (R² = .28, F = 19.89, p < .001). These results are shown in Table 3.

For cognitive work engagement, emotional intelligence of controlling self and affective commitment proved to be very good positive predictors. The strongest predictor was “affective commitment”, that showed that β was .254 and ΔR² was .06, with a significant p < .001. The second predictor was “emotional intelligence of controlling self”, that showed that β was .163 and ΔR² was .03, with a significant p < .05. These variables combined and formed as powerful predictors (R² =.09, F = 10.90, p < .001).

For emotional work engagement, emotional intelligence of controlling self, work satisfaction, and affective commitment proved to be very good positive predictors. The strongest predictor was “affective commitment”, that showed that β was .298 and ΔR² was .16, with a significant p < .001. Another predictor was “emotional intelligence of controlling self”, that showed that β was .240 and ΔR² was .08, with a significant level p < .001. The third one was “work satisfaction”, that showed that β was .266 and ΔR² was .05, with a significant p < .001. These variables combined and formed as powerful predictors (R² = .31, F = 30.40, p < .001).

For physical work engagement, emotional intelligence of controlling self and affective commitment proved to be very good positive predictors. The strongest predictor was “affective commitment”, that showed that β was .473 and ΔR² was .22, with a significant level p < .001. The second predictor was “emotional intelligence of controlling self”, that showed that β was .218 and ΔR² was .07 with a significant p < .001. These variables combined and formed as powerful predictors (R² = .29, F = 40.88, p < .001). These results were shown in Table 4.

Discussion

Social Workers’ Burnout and Work Engagement

Social workers’ burnout, in terms of emotional exhaustion, depersonalization, and reduced personal accomplishment, was slightly below average. While their cognitive, emotional, and physical work engagements were all slightly above average, suggesting that social workers were work engaged and fewer problems in burnout. Although the overall results of social workers’ burnout seemed to suggest that they have no problem in regulating emotions, while breaking down in each item of emotional exhaustion subscale, there are two items, namely “I feel emotionally drained from my work” and “I feel used up at the end of the workday” that were rated very high, suggesting that they have a specific emotional exhaustion rather than a general one. That is, social workers felt that their emotions are used up by their work. In addition, this study also provides a theoretical contribution, since the pattern between work engagement (by using the newly work engagement measure) and burnout is similar to the original study (Kuok & Taormina, 2017) and the study of teachers in inclusive education (Kuok et al., 2022) further confirmed that the concept of coexistence between work engagement and burnout can be applied to another helping professional – social workers. Therefore, it is very critical for the social service management to understand the possible factors that decrease burnout and increase work engagement among social workers.

Personal and Organizational Factors Influence Social Workers’ Burnout

The impact of emotional intelligence on burnout varied depending on the three factors of burnout: emotional intelligence of controlling self was a negative predictor of depersonalization and reduced personal accomplishment, while emotional intelligence of understanding others was a negative predictor of reduced personal accomplishment only. Moreover, the organizational variable work satisfaction was a powerful negative predictor of all three factors of burnout and another organizational variable, whereas affective commitment was a negative predictor of reduced personal accomplishment.

Personal and Organizational Factors Influence Social Workers’ Work Engagement

Emotional intelligence of controlling self was a positive predictor of all three factors of work engagement. Moreover, the organizational variable affective commitment was a powerful positive predictor of all three factors of work engagement and another organizational variable; work satisfaction was a positive predictor of emotional work engagement only.

All these findings revealed that both personal and situational factor emotional intelligence and organizational variables work satisfaction and affective commitment can influence burnout and work engagement with different intensity respectively. Emotional intelligence of controlling self can affect all three facets of work engagement and burnout (except emotional exhaustion). Work satisfaction mainly minimizes burnout but affective commitment can boost work engagement. It provided insights for the management in social service organizations of how to decrease burnout and increase work engagement in practice. Interestingly, such a discrepancy of the predictors between work engagement and burnout in practical findings evidenced and provided theoretical contribution that the distinctiveness and coexistence between the work engagement and burnout theories are correct.

Conclusion and Recommendation

Social workers with high emotional intelligence of controlling self are more likely to work engaged cognitively, emotionally, and physically. On the other hand, they are less likely to treat clients as objects, and to feel less accomplished about how one’s job can be performed. In addition, work satisfaction is a powerful predictor and has an additional effect on all three factors of burnout, as well as affective commitment is a powerful predictor and has an additional effect on all three faces of work engagement.

Therefore, it is important for the management of social service to find ways to help social workers to improve their emotional intelligence of controlling self, by adopting Dolev and Leshem’s (2017) suggestions on developing teachers’ emotional intelligence, increasing one’s self-awareness in terms of growing awareness of thoughts, emotions, and moods – those brought from clients and social expectation – and how they impact upon their work and well-being, as well as increasing self-assertiveness via daily practice by setting realistic expectations, and confidence in approaching clients.

Moreover, in order to decrease burnout and increase work engagement among social workers, it is critical for them to increase work satisfaction and affective commitment. Both Kuok (2017) and Kuok and Taormina (2015) found that the organizations create a sense of “justice” for distributing work that will result in increasing their employees’ work satisfaction and affective commitment, and by sharing and executing the ethical standards of conduct to guide employees’ decisions and behaviors their employees would be less likely to experience uncertainty when they encountered problems as well as to provide their employees pay, rewards, and allocation (their out-comes) under a distributive justice system, i.e., based on their effort and performance (their inputs) in non-profit making companies (e.g., social service organizations). So their work satisfaction and affective commitment will raise and, in return, their burnout will decrease and their work engagement will increase.