The Consequences of Emotional Labor for Public Sector Workers and the Mitigating Role of Self-Efficacy

Abstract

Emotional labor that requires workers to suppress their truly felt emotions and create a fake emotional display has negative consequences for workers including psychological distress and lowered job satisfaction. This type of emotional labor, called surface acting, is often necessary in public service work. In an effort to identify ways to reduce the harmful effects of emotional labor for workers, this research further specifies the relationship between emotional labor and worker well-being using data from a large sample of public service workers working in a variety of occupations (n = 1,395). The analyses test a mediator and moderators of the relationship between emotional labor and worker well-being. The findings suggest that surface acting emotional labor is harmful to workers because it increases feelings of self-estrangement. In addition, a worker’s sense of self-efficacy in emotional labor performance is shown to reduce the negative effects of surface acting.

Keywords

public administration generally research methods–quantitative personnel/human resource management training and development

In her 1983 study of flight attendants and bill collectors, Arlie Hochschild drew attention to the importance of considering the emotional aspects of working. In particular, Hochschild (1983) developed the concept of emotional labor as the efforts made by interactive service workers to manage their emotions and create desired emotional states in themselves and in customers. Workers engage in emotional labor to meet expectations of service in customer interactions, to display deference to superiors, to make favorable impressions on others, or to maintain workplace norms of emotional expression (e.g., Erickson & Ritter, 2001; Hochschild, 1979, 1983; Pugliesi, 1999; Wharton, 2009). Emotional labor is often a necessary part of service work; however, the constant management of emotions can have harmful consequences for the worker. More specifically, emotional labor that involves masking one’s inner feeling to display an appropriate emotion can make workers feel inauthentic and estranged from their own, personal emotions, and is associated with psychological distress, burnout, and lowered job satisfaction (e.g., Brotheridge & Grandey, 2002; Erickson & Ritter, 2001; Glasø & Einarsen, 2008; Glomb & Tews, 2004; Guy, Newman, & Mastracci, 2008; Hochschild, 1983; Sloan, 2008).

One factor that may mitigate the negative impact of emotional labor is a worker’s feelings of effectiveness in the emotional components of the work role. Emotional self-efficacy has beneficial direct effects on workers including increased job satisfaction and decreased emotional stress (Hsieh & Guy, 2009; Guy et al., 2008). Self-efficacy may also moderate the relationship between emotional labor and worker well-being (Heuven, Bakker, Schaufeli, & Huisman, 2006; Pugh, Groth, & Hennig-Thurau, 2011). While the suppression of inner feelings in the effort to display appropriate emotions may cause workers to feel inauthentic, such feelings and their consequences may be lessened if workers feel that their emotion management efforts are successful. That is, feelings of competence in emotional labor may offset the costs of fake emotional display to worker well-being (Pugh et al., 2011; Thoits, 2004).

Identifying moderators of the negative relationship between emotional labor and worker well-being, such as worker self-efficacy, will provide insight into lessening the burden of emotional labor for workers. Because most public service jobs require either face-to-face or voice-to-voice contact with the public, emotional labor is essential to public service work (Guy & Newman 2004; Guy, Newman, & Mastrtacci, 2008; Mastracci, Newman, & Guy, 2006; Meier, Mastracci, & Wilson, 2006; Steinberg & Figart, 1999). Assessments of how public service workers are affected by emotional labor can enable agency and human resource managers to gain a better understanding of how consequences of emotional labor performance may be alleviated. However, researchers have just recently begun to examine the experiences of emotional labor among public sector workers (e.g., Guy et al., 2008; Guy & Newman 2004; Hsieh, Jin, & Guy, 2012; Meier et al., 2006). Much of the work that has been done has focused on workers in a small number of occupations (Hsieh et al., 2012). The current study examines the consequences of emotional labor and analyzes the role of self-efficacy in reducing its harmful effects in a random sample of 1,395 public sector workers from a wide variety of occupations.

Consequences of Emotional Labor

Emotional labor is the management of emotion that is performed for a wage (Hochschild, 1983). For example, front line service workers are encouraged to deliver “service with a smile” or individualized attention as part of the consumer’s experience (Erickson, 1995; Hochschild, 1983; Leidner, 1993). Some organizations may actually specify the exact emotions that are to be displayed by workers in interactions with customers or clients. For example, Leidner (1993) observed that at “Hamburger University,” McDonald’s employees are instructed to “be enthusiastic and smile” as they greet customers, and at various restaurants waitresses are required to smile, defer, and flirt when interacting with diners (Hall, 1993). In his study of nursing home organizations, Lopez (2006) observed requirements of workers to inhibit all personal emotions and accept abuse from patients. Although it may not be explicitly recognized as a part of the work role, the performance of emotional labor is often an essential part of work in the public sector (Guy et al., 2008; Hsieh & Guy, 2009; Mastracci et al., 2006; Steinberg & Figart, 1999). Consider the efficient and friendly service of a post office clerk, the analytical calm of a 911 call taker, or the tough exterior presented by a corrections officer in the face of danger (e.g., Guy et al., 2008).

Two ways that workers may accomplish the emotional labor necessary in their work role are by deep acting and surface acting (Hochschild, 1983). In deep acting, workers regulate their internal feelings to be consistent with the external emotional display that is necessary in a given situation. Hochschild (1983) gave the deep acting example of a flight attendant who imagined a disrespectful passenger as a fearful child in order to actually feel the benevolence she was expected to display. Thus, with deep acting, workers actually feel the emotions that they need to express to others. In contrast, surface acting involves masking one’s internal emotional state and presenting an, often contrasting, expression. When surface acting, workers essentially “fake” their emotional displays. For instance, an administrative worker at the Department of Motor Vehicles may need to present a pleasant demeanor despite her feelings of anger when dealing with a rude customer.

The effects of emotional labor on worker well-being depend on the type of emotional labor performed. According to Hochschild’s (1983) argument, surface acting that involves the constant management of truly felt emotions may cause the worker to feel dissonance between her true self and the person she presents to others. The suppression of one’s true feelings often requires the presentation of a certain self to others that is inconsistent with the self that the individual may wish to present, and this process distances a worker from her true feelings or true self (Brotheridge & Grandey, 2002; Erickson, 1995; Hochschild, 1983). This feeling of self-estrangement, then, is distressing to workers and is associated with work-related consequences (Erickson & Wharton, 1997; Hochschild, 1983; Sloan, 2008). For example, Guy et al. (2008) found that “false face acting,” a type of emotional labor that involves faking a worker’s true feelings, was linked with increased burnout and lowered job satisfaction (also see Hsieh et al., 2012). In addition, Brotheridge and Grandey (2002) revealed a heightened sense of depersonalization among workers who engage in surface acting, and Glomb and Tews (2004) linked surface acting to emotional exhaustion.

In contrast, because deep acting involves emotional display that is consistent with internally felt emotion, workers do not experience the same feelings of dissonance as they do with surface acting. Instead, with successful deep acting, the worker feels satisfied with his or her ability to provide personal service (Hochschild, 1983). Indeed, Brotheridge and Grandey (2002) found that deep acting was associated with feelings of effectiveness and accomplishment at work while surface acting predicted depersonalization. In addition, becoming emotionally involved in work, as opposed to simply faking emotional displays, is associated with increased job satisfaction and pride in work (Guy et al., 2008; Jin & Guy, 2009; Hsieh et al., 2012; Lopez, 2006; Morris & Feldman, 1997; Wharton, 2009).

Thus, the extant literature suggests that the effects of emotional labor on workers depend on the type of the emotional labor performed and the context in which workers perform it (Grandey, 2003; Guy et al., 2008; Hsieh et al., 2012; Jin & Guy, 2009; Lopez, 2006). More specifically, workers who engage in deep acting may reap rewards from their emotional engagement in their work while workers who regularly surface act experience dissonance, self-estrangement, and reduced job satisfaction (Hülsheger, Lang, & Maier, 2010; Hülsheger & Schewe, 2011). Although it is associated with negative consequences, many workers may need to rely on surface acting to meet the affective requirements of their jobs (Guy et al., 2008; Hochschild, 1983; Hsieh, & Guy, 2009; Lopez, 2006; Wharton, 2009). Furthermore, by stipulating emotional display rules and limiting self-expression, many workplaces do not provide the opportunity for workers to become emotionally engaged in their work, leaving workers little choice other than to fake their emotional displays (Lopez, 2006).

Self-Efficacy as a Moderator

Given that workers may be unable to avoid surface acting in their work roles, this research draws on the findings of previous studies to examine how the consequences of this emotional labor may be lessened. Specifically, a key factor that may mitigate the impact of emotional labor on workers is a worker’s sense of self-efficacy, or the worker’s feeling of effectiveness in her work role (Bandura, 1996; Heuven et al., 2006; Pugh et al., 2011; Thoits, 2004). Empirical research lends support to the positive effects of self-efficacy on workers (Guy et al., 2008; Hsieh et al., 2012; Pugh et al., 2011), and Thoits (2004) suggests that a worker’s sense of self-efficacy in her emotional labor performance may help prevent negative consequences of emotional labor. Research has shown that failure at emotion management (i.e., being a poor emotion manager) has negative consequences for an individual’s self-esteem (MacRae, 1998; Taylor, 2000). On the other hand, if a worker feels effective at managing emotions on the job, the effective performance of emotion labor at work may be a positive experience for the worker. Just as workers who engage in successful deep acting and intense emotional labor report feelings of accomplishment, satisfaction, and pride in work instead of feelings of self-estrangement, workers may also experience feelings of competence in regard to their surface acting performance (e.g., that they provided good service, made a customer happy, prevented conflicts in the workplace, or pleased a boss). Thus, although workers may need to fake their emotional displays at work, feelings of self-efficacy may enable workers to view this emotional labor as a skill that can be effectively used at work rather than an act that makes them feel self-estranged (Heuven et al., 2006; Pugh et al., 2011; Thoits, 2004).

Hypotheses

The current article examines three known consequences of emotional labor: Self-estrangement, decreased job satisfaction, and increased psychological distress (e.g., Erickson & Ritter, 2001; Glasø & Einarsen, 2008; Guy et al., 2008; Hsieh et al., 2012; Morris & Feldman, 1997; Pugliesi, 1999; Wharton, 2009; Zapf & Holz, 2006).¹ The analyses focus specifically on surface acting as it is the type of emotional labor associated with negative consequences. Feelings of self-estrangement are expected to mediate the relationship between surface acting and its negative consequences (Heuven et al., 2006; Sloan, 2008). The false emotions expressed when surface acting cause workers to feel inauthentic and estranged from their true selves, which in turn is associated with the negative outcomes of psychological distress and lowered job satisfaction (Erickson & Ritter, 2001; Erickson &Wharton, 1997; Hochschild, 1983; Hülsheger & Schewe, 2011). Self-efficacy will serve a moderator of the relationship between surface acting and feelings of self-estrangement.

Hypothesis 1: Surface acting will have negative consequences for public service workers.

Hypothesis 1a: Surface acting will be negatively associated with job satisfaction.

Hypothesis 1b: Surface acting will be positively associated with psychological distress.

Hypothesis 2: Feelings of self-estrangement will mediate the relationships between surface acting and its negative consequences.

Hypothesis 3: The relationship between surface acting and self-estrangement will depend on the worker’s self-efficacy. Self-efficacy will moderate the relationship between surface acting and self-estrangement.

Research Design

Data

The hypotheses are tested using data from a 2004 survey of Tennessee state employees.² The author designed and distributed a 10-page anonymous mail survey to a simple random sample of 2,500 “career service” workers who had worked in their position with the state for at least 6 months. Career Service jobs are all positions within the state in which the workers do not serve the governor directly.³ The population sampled included over 37,000 individuals working in over 1,400 positions for the state. Dillman’s (2000) Tailored Design Method was followed for the survey implementation and up to three follow-up contacts with potential respondents were made in an effort to maximize participation. The response rate for the survey was 62% (n = 1,550).⁴ The analyses presented here include 1,395 respondents with full data on all measures. A review of the literature indicates that this project is the first large-scale survey to collect data from a random sample of public service workers in a wide variety of occupations (see Hsieh et al., 2012; Hülsheger & Schewe, 2011). Nevertheless, the limited generalizability of cross-sectional data from one state must be acknowledged. The implications of cross-sectional data for the interpretation of the analyses are discussed in greater detail below.

Measures

Surface acting

Surface acting emotional labor is measured by a scale of summed responses to five statements about managing emotions at work. A sample item is “I spend time at work acting “artificially” friendly to others.” Each item was scored on a scale from 1 (strongly disagree) to 5 (strongly agree). The scale was averaged to produce a range of 1 to 5, and a higher score on this measure indicates greater emotion management effort. Adapted from Pugliesi (1999), the items used in this scale measure the act of covering up one’s own feelings—similar to the concepts of “employee-focused emotional labor” or “false face acting” that have been linked with negative consequences in previous research (Brotheridge & Grandey, 2002; Guy et al., 2008; Hochschild, 1983; Hsieh et al., 2012; Pugliesi, 1999; Pugliesi & Shook, 1997). The wording of the survey items used in each scale and Cronbach’s alphas for the measures are reported in Table 1.

Table 1.

Constructs and Reliability of Key Measures.

Variable	Survey item	Cronbach’s alpha
Surface acting	At work I have to be nice to people no matter how they treat me.	0.72
	I express my emotions freely when I am at work (reversed).
	I am unable to express my true feelings to the people I work with.
	I spend time at work acting “artificially” friendly to others.
Self-estrangement	The way I act at work is very different from the way I act at home.	0.81
	I feel most like myself at work (reversed).
	I feel that I cannot express my true self when I am at work.
	When I am at work, I am unsure of what my “real” feelings are.
	I basically have to become a different person when I am at work.
Job satisfaction	Given the chance, I would change my present type of work for something different (reversed).	0.84
	I am proud of the type of work I do.
	My work gives me a chance to do the things I enjoy.
	In general, I am satisfied with my job.
Relational self-efficacy	My coworkers rely on me to help them handle their problems with other workers.	0.74
	I enjoy helping coworkers when they have personal difficulties.
	My coworkers come to me for help with their problems.
	I often help coworkers deal with their emotional problems.
Service self-efficacy	I am proud of my ability to provide good customer or client service.
	I am skilled at dealing with customers and/or clients.
Job complexity	There is a lot of variety in the kinds of things I do at work.	.66
	I keep learning new things in my job.
	My job requires that I do the same things over and over (reversed).
Worker autonomy	My job allows me to make a lot of decisions on my own.	.64
	I have flexibility in my work to decide when to do different tasks.
	My supervisor decides how my work tasks should be done (reversed).

Self-estrangement

The self-estrangement measure was modeled after a scale used by Erickson and Ritter (2001) and is comprised of the averaged sum of responses to five statements about feelings of self-estrangement at work. “I feel that I cannot express my true self when I am at work” is a sample item. Each of the five items was rated on a scale from 1 (strongly disagree) to 5 (strongly agree), and a factor analysis revealed that all items load on a single factor. Responses were coded so that a high score on this measure indicates a greater experience of self-estrangement.⁵

Job satisfaction

Job satisfaction is measured by a summed scale of responses to the following five statements: “Given the chance, I would change my present type of work for something different (reversed),” “I am proud of the type of work I do,” “My work gives me a chance to do the things I enjoy,” “In general, I am satisfied with my job,” and “I am respected for the work that I do” (Cronbach’s alpha = .84). Responses to each statement ranged from 1 (strongly disagree) to 5 (strongly agree) such that a high score on this measure indicates a high level of job satisfaction. This scale provides a global assessment of job satisfaction and is similar to measures utilized in previous research (e.g., Glasø & Einarsen, 2008; Pugliesi, 1999).

Psychological distress

Psychological distress is measured with the mean of respondents’ scores on a series of questions about emotions. Respondents were asked to indicate on how many days in the past week they experienced each of the following symptoms of anxiety and depression: felt that he or she could not shake the blues, sad, lonely, happy (reversed), calm (reversed), fearful, worried, anxious, and restless. These symptoms of distress have been combined to form a distress index in previous research (Mirowsky & Ross, 2003; Thoits, 2002). The distress scores range from 0 to 7, with higher scores indicating more frequent psychological distress (Cronbach’s alpha = 0.77).

Self-efficacy

A principal components analysis of items used to assess worker self-efficacy revealed two factors: Relational self-efficacy and service self-efficacy. Relational self-efficacy is measured by the averaged sum of responses to four statements. A sample item is “My coworkers rely on me to help them handle their problems with other workers.” Service self-efficacy is measured by the average of responses to the following two statements: “I am proud of my ability to provide good customer or client service” and “I am skilled at dealing with customers and/or clients.” The responses for all items ranged from 1 (strongly disagree) to 5 (strongly agree), with a high score on the measures indicating a high level of self-efficacy.

In addition, a measure of perceived influence over others is included as another dimension of self-efficacy. Research suggests that emotional labor that shapes the emotions of others can be empowering for workers (Bolton & Boyd, 2003; Jin & Guy, 2009; Tolich, 1993). Although the data do not contain a specific measure a worker’s ability to influence the emotions of others, a general measure of perceived influence may assess a worker’s confidence in affecting others. Respondents were asked how much they disagree or agree with the following statement: “I have a lot of influence over others in the workplace.” Responses ranged from 1 (strongly disagree) to 5 (strongly agree), and a higher score indicates greater perceived influence over others.

Control variables

Several control variables are included in the analyses in order to account for demographic and work-related influences on emotional labor, job satisfaction, and psychological distress. As a measure of income and status within the workplace, each respondent’s pay grade was obtained from the state’s job classification database and coded onto the respondent’s self-reported job title. Pay grade values range from 4 to 41, with a higher pay grade indicating greater income. Because working in a supervisory position may influence relations with others in the workplace and affect job satisfaction, a dummy variable for working in a managerial position (1 = yes, 0 = no) is included. In addition, an ordinal measure of job tenure is included in the analyses. Respondents indicated how long they have worked in their current position using the following categories: less than 1 year; 1 to 2 years; 3 to 4 years; 5 to 10 years; 11 to 15 years; 16 to 20 years and more than 21 years. To control for any differences in emotional labor requirements by occupation, dummy variables were created for the following occupational groups: administrative, office or clerical; biological or physical sciences; health care; information systems or telecommunications; public safety; service or maintenance; skilled crafts or trades; and social, human, or employment services.

Also, the extent to which workers interact with different groups of people is likely to influence emotional labor, so controls for the amount of time the worker spends with customers coworkers, and supervisors coded separately on a scale of 1 (none of my time) to 5 (almost all of my time) are included in the analyses.

Two job characteristics often considered in analyses of worker well-being and emotional labor are also included in the analyses: Job complexity and control over work (worker autonomy). Both variables are negatively correlated with the performance of surface acting and are positively correlated with job satisfaction (Erickson & Ritter, 2001; Erickson & Wharton, 1997; Kohn & Schooler, 1973; Mirowsky & Ross, 2007; Pugliesi, 1999; Wharton, 1993).

Finally, all models include demographic controls that have been shown to influence emotional labor, worker well-being, and job satisfaction: gender (0 = women, 1 = men), race (0 = non-White, 1 = White), age (in years), education (measured ordinally); marital status (0 = not married, 1 = married); and the presence of minor children in the home (0 = no, 1 = yes; e.g., Bolger, DeLongis, Kessler, & Schilling, 1989; Erickson & Ritter, 2001; Hsieh & Guy, 2009; Meier et al., 2006; Pugliesi, 1999). The demographic composition of the sample and its distribution on key variables is presented in Table 2.

Table 2.

Sample Characteristics and Distribution on Key Variables (n = 1,395).

Variable	% of sample	Mean	Standard deviation	Range
Gender
Women	58.8
Men	41.2
Race
White	74.4
Non-White	25.6
African American	20.9
Other	4.7
Married	62.4
Minor children in the home	32.6
Education
Less than high school	1.9
High school degree or equivalent	20.4
Some college	22.6
Associate degree/junior college	8.5
4-year college degree	24.8
Some graduate school	9.5
Graduate/professional degree	12
Occupational Group
Administrative, office or clerical	24.2
Biological or physical sciences	4.6
Health care	11.2
Information systems or telecommunications	5.8
Public safety	13.1
Service or maintenance	7.3
Skilled crafts or trades	7.6
Social, human or employment services	26.2
Managerial position	35.2
Age		47.33	11.01	20-80
Tenure in job		4.06	1.92	1-7
Pay grade		20.76	6.72	4-41
Autonomy		10.91	2.37	3-15
Complexity		10.24	2.42	3-15
Time spent with clients/customers		2.94	1.38	1-5
Time spent with coworker		2.88	1.17	1-5
Time spent with supervisors		2.40	0.94	1-5
Surface acting		3.24	0.63	1-5
Self-estrangement		4.87	1.31	2-10
Job satisfaction		3.64	0.85	1-5
Psychological distress		1.54	1.42	0-7
Service self-efficacy		4.13	0.75	1-5
Relational self-efficacy		3.23	0.77	1-5
Perceived influence over others		3.28	0.99	1-5

Plan of Analysis

A series of Ordinary Least Squares Regression models is used to test the hypotheses presented above. First, as a preliminary analysis to assess differences in emotional labor performance by occupation and other work characteristics, surface acting is regressed on the job characteristic, demographic and control variables. Next, the consequences of surface acting, job satisfaction and psychological distress are regressed on surface acting and the control variables. To examine the mediating role of self-estrangement in the relationship between surface acting and its consequences, self-estrangement is first regressed on surface acting and the control variables and then self-estrangement is added to the regression of job satisfaction and psychological distress on surface acting. If self-estrangement serves a mediating function in the relationship between surface acting and its consequences, surface acting will be positively associated with feelings of self-estrangement, net of the control variables, and self-estrangement will explain the relationship between surface acting and psychological distress and job satisfaction. Finally, to test for the moderating effect of self-efficacy, the multiplicative interactions between the self-efficacy measures and surface acting are added to a regression of self-estrangement on surface acting and the control variables.

It should be noted that because the data are cross-sectional, the causal direction of the relationships analyzed cannot be determined. Theoretically, it is expected that the presentation of fake emotions in surface acting will cause workers to feel inauthentic and estranged from their selves and experience subsequent costs to their well-being (Erickson & Ritter, 2001; Erickson & Wharton, 1997; Grandey, 2000; Hochschild, 1983). However, it is possible that job attitudes and worker well-being also influence the performance of emotional labor (Guy et al., 2008; Grandey, 2003). Although previous research and theory on the relationship between emotion management, self-estrangement and their effects on workers suggest that emotional labor and self-estrangement precede work-related consequences (Erickson &Wharton, 1997; Heuven et al., 2006; Hsieh et al., 2012; Morris & Feldman, 1997; Pugliesi, 1999; Wharton, 2009; Wolpin, Burke, & Greenglass, 1991), most of the empirical research on the consequences of emotional labor has used cross-sectional data. A recent meta-analysis of the past 30 years of emotional labor research revealed just one longitudinal survey study (Hülsheger & Schewe, 2011). In a two-wave survey study of teachers, Hülsheger et al. (2010) revealed that surface acting led to psychological strain but did not find support for the reverse pathway. Nevertheless, while the current research makes the assumption that surface acting precedes worker well-being, panel data are ultimately necessary to substantiate the direction of these relationships.

Findings

Predictors of Surface Acting

The first column of Table 3 presents the regression of surface acting on the demographic, occupational, and job characteristic variables. Overall, the model shows that job characteristics are more important predictors of surface acting than occupational group or demographic characteristics. Consistent with previous research, worker autonomy and job complexity are both negatively associated with surface acting (e.g., Bulan, Erickson, & Wharton, 1997; Erickson & Ritter, 2001). In addition, both the amount of time spent with clients or customers and the amount of time spent with supervisors are associated with increased surface acting. Thus, surface acting is important in customer or client interactions and also when working with supervisors. Conversely, the amount of time a worker spends with coworkers is negatively associated with surface acting. This suggests that workers may feel more comfortable and casual around coworkers. Coworkers may also serve as an outlet for emotional expression that workers need to inhibit in other contexts (Lively, 2000). The model also reveals that managers engage in more surface acting than other workers. The role of a manager to negotiate between the needs of employees and those of the public may place managers in a position that requires extensive surface acting (Bolton & Houlihan, 2010).

Table 3.

OLS Regression of Surface Acting, Job Satisfaction, Psychological Distress, and Self-Estrangement (n = 1,395).

	Surface acting	Job satisfaction	Psychological distress	Self-estrangement
Surface acting	—	−.181*** (.030)	.447*** (.061)	.671*** (.052)
Autonomy	−.052*** (.008)	.104*** (.009)	−.059** (.017)	−.097*** (.015)
Complexity	−.029*** (.007)	.143*** (.008)	−.081*** (.016)	−.081*** (.014)
Pay grade	.000 (.004)	.003 (.004)	.002 (.008)	−.004 (.007)
Time spent with clients/customers	.078*** (.013)	.020 (.015)	.020 (.030)	−.014 (.026)
Time spent with coworkers	−.049** (.015)	.011 (.017)	.005 (.035)	.005 (.030)
Time spent with supervisors	.074*** (.019)	.030 (.021)	−.006 (.043)	−.022 (.037)
Occupational Group^b
Administrative, office or clerical	.024 (.051)	.060 (.057)	−.040 (.116)	.214* (.099)
Biological or physical sciences	−.058 (.085)	.262** (.096)	−.068 (.194)	−.127 (.164)
Health care	.002 (.059)	.267*** (.067)	−.095 (.135)	.006 (.115)
Information systems or telecommunications	−.026 (.080)	.167 (.091)	−.218 (.183)	.241 (.155)
Public safety	.025 (.062)	.150* (.071)	−.113 (.143)	.327** (.121)
Service or maintenance	.138 (.082)	.033 (.093)	−.062 (.189)	.179 (.160)
Skilled crafts or trades	−.003 (.074)	.137 (.083)	−.123 (.169)	.121 (.143)
Managerial position (yes = 1)	.120** (.037)	−.110* (.042)	.027 (.086)	.066 (.072)
Tenure in job	−.004 (.009)	.011 (.011)	.008 (.022)	.006 (.018)
Gender (women = 1)	−.003 (.039)	.033 (.044)	.015 (.088)	−.216** (.076)
Race (White = 1)	−.067 (.039)	.088* (.041)	.331*** (.089)	.244** (.076)
Age	.002 (.002)	.006** (.002)	−.011** (.004)	−.007* (.003)
Married (yes = 1)	−.009 (.035)	.011 (.039)	−.322*** (.079)	−.248*** (.068)
Minor children in the home (yes = 1)	.035 (.037)	.025 (.041)	−.002 (.084)	−.001 (.072)
Education	.004 (.013)	−.030* (.015)	−.011 (.029)	.000 (.025)
Adjusted R²	.089	.393	.101	.233

Note: Values are unstandardized regression coefficients. Standard errors are shown in parentheses.

Reference category is non-White; ^breference category is social, human or employment services.

***

p < .001. **p < .01. *p < .05.

There are no significant differences in surface acting by occupational group, which suggests that the performance of surface acting among public service workers is not limited to a particular occupational group, as distinguished here. Furthermore, net of the job characteristic variables, none of the demographic control variables significantly predict surface acting.

Consequences of Surface Acting

The regressions of the expected consequences of emotional labor are also presented in Table 3. Net of the demographic and job characteristic variables, extensive surface acting significantly reduces job satisfaction (b = –.181) and significantly increases psychological distress (b = .447). In addition, worker autonomy and job complexity are associated with job satisfaction and psychological distress in the expected directions. Both job characteristics positively affect worker well-being. It is notable that the full job satisfaction model explains almost 40% of the variance in job satisfaction. Approximately 10% of the variance in psychological distress is explained by the included variables.⁶ These findings lend support to Hypothesis 1.

Column 4 of Table 3 reports the regression of self-estrangement on surface acting and the control variables. If a mediation effect of self-estrangement on the relationship between surface acting and its consequences is present, we would first expect surface acting to be a significant predictor of self-estrangement. Indeed, net of the control variables, surface acting significantly heightens feelings of self-estrangement (b = .671). A comparison of the standardized regression coefficients (not shown) reveals that surface acting is the most robust predicator of self-estrangement in the model (β = .320).

The test of the mediation effect of self-estrangement is presented in Table 4. Consistent with Hypothesis 2, self-estrangement fully mediates the effects of surface acting on job satisfaction and psychological distress, net of the control variables. When self-estrangement is added to the regressions of job satisfaction and psychological distress on surface acting, the coefficients for surface acting lose statistical significance. This suggests that surface acting is associated with negative consequences for workers because it increases feelings of self-estrangement. Thus, efforts to reduce the negative impact of surface acting on workers should examine the conditions under which surface acting produces feelings of self-estrangement. The next set of analyses represents an initial attempt to determine how the feelings of self-estrangement associated with surface acting among public service workers may be lessened.

Table 4.

The Mediating Effect of Self-Estrangement (n = 1,395).

	Job satisfaction		Psychological distress
	Model 1	Model 2	Model 3	Model 4
Surface acting	−.181*** (.030)	−.054 (.032)	.447*** (.061)	.086 (.058)
Self-estrangement		−.186*** (.015)		.537*** (.029)
Adjusted R²	.393	.454	.101	.287

Note: Values are unstandardized regression coefficients. Standard errors are shown in parentheses.

All models control for the following variables: Gender, race, age, marital status, having children, education, working in a managerial position, tenure in the job, pay grade, occupational group, amount of time spent with customers, coworkers, and supervisors, autonomy, and complexity.

***

p < .001. **p < .01. *p < .05.

The Role of Self-Efficacy

Table 5 displays the test of the moderating role of self-efficacy in the relationship between surface acting and self-estrangement. Model 1 includes the self-efficacy measures, surface acting, and the control variables. As shown, while surface acting is associated with increased self-estrangement, service self-efficacy and perceived influence over others lessen feelings of estrangement. Workers who feel confident in their ability to provide good service to customers and workers who believe they can influence others in the workplace report lower levels of self-estrangement than workers who feel less effective in these areas. Relational self-efficacy, however, does not affect feelings of self-estrangement.

Table 5.

Regression of Self-Estrangement on Surface Acting, Self-Efficacy, Interaction Terms and Controls (n = 1,395).

	Model 1	Model 2
Surface acting	1.100*** (.251)	1.740*** (.381)
Service self-efficacy	−.338*** (.051)	.181 (.238)
Relational self-efficacy	−.062 (.231)	−.077 (.231)
Perceived influence	.413* (.184)	.385* (.184)
Interaction terms
SA × Service efficacy		−.160* (.071)
SA × Relational efficacy		.025 (.069)
SA × Perceived influence		−.139** (.053)
Adjusted R²	.274	.281

Note: Values are unstandardized regression coefficients. Standard errors are shown in parentheses.

SA = surface acting. All models control for the following variables: Gender, race, age, marital status, having children, education, working in a managerial position, tenure in the job, pay grade, occupational group, amount of time spent with customers, coworkers, and supervisors, autonomy, and complexity

***

p < .001. ** p< .01. *p < .05.

The multiplicative interactions between the self-efficacy measures and surface acting, along with the controls, are added in Model 2. The significant negative coefficient for the surface acting by service self-efficacy interaction indicates that the relationship between surface acting and self-estrangement depends on a worker’s service self-efficacy (b = −.160). That is, feeling effective in the service role lessens the association between surface acting and self-estrangement. A graphical depiction of this interaction effect is shown in Figure 1.

Figure 1.

The interaction effect between surface acting and service self-efficacy on self-estrangement.

Likewise, the significant negative coefficient for the surface acting by perceived influence interaction shows that perceived influence mitigates the effect of surface acting on self-estrangement (b = −.139). Thus, workers who believe that they have influence over others at work experience less self-estrangement from surface acting than those who perceive less influence over others. This effect is depicted in Figure 2. Thus, Hypothesis 3 receives partial support—two of the self-efficacy measures, service self-efficacy and perceived influence over others, moderate the relationship between surface acting and self-estrangement.

Figure 2.

The interaction effect between surface acting and perceived influence over others on self-estrangement.

Discussion and Conclusion

It has been almost thirty years since Hochschild (1983) identified emotional labor as a key feature of service work and called attention to its potential costs to worker well-being. Since that time, an extensive amount of research has examined the impact of emotional labor on workers and has revealed both positive and negative effects on worker well-being. More recently, scholars have found emotional labor to be a necessary, yet often unrecognized, component of work in the public sector (e.g., Guy et al., 2008). Identifying ways that the negative consequences of emotional labor can be prevented, or at least lessened, has the potential to improve the experience of work for public employees. The current study extends previous research by illustrating that emotional labor has negative consequences for worker well-being and job satisfaction when it is associated with feelings of self-estrangement and by demonstrating the moderating role of self-efficacy in this relationship in a large sample of public service workers from a variety of occupations.

First, in order to further understand the detrimental outcomes of surface acting, the potential mediating role of self-estrangement in the relationship between surface acting and job satisfaction and psychological distress was examined. The analyses suggest that surface acting—suppressing felt emotions while acting artificially friendly to others—has negative effects on workers because it increases feelings of self-estrangement. Workers who perform surface acting at work are unable to express how they really feel while working and the expressions they display in the workplace do not represent their true selves. Surface acting, then, distances a worker from her true self and real feelings, and this experience is distressing for workers. This type of emotional labor, however, is often essential to work that involves interpersonal communication (Hsieh & Guy, 2009). Thus, workers may be unable to stop engaging in surface acting to prevent its negative effects. The mediating effect of self-estrangement on the relationship between surface acting and its negative consequences suggests that efforts to improve the working experiences of public service workers should focus on ways to prevent workers from experiencing the feelings of self-estrangement that are associated with surface acting.

Accordingly, I investigated the moderating role of self-efficacy in the relationship between surface acting and self-estrangement. A significant moderating effect was found for two of the three self-efficacy indicators, service self-efficacy and perceived influence over others. In both cases, self-efficacy buffered the relationship between surface acting and self-estrangement. Feelings of self-estrangement for workers who perform extensive surface acting were less intense when the workers felt effective in their service role or felt that they have influence over others. Although public service workers may need to surface act in their jobs, feeling effective in providing a service to others can decrease the sense of self-estrangement that accompanies surface acting. Likewise, the perception that the worker has influence over others at work can reduce feelings of self-estrangement. Workers who surface act at work but feel effective in their work role do not experience this emotional labor as inauthentic to the same extent as worker who feel less effective. These findings suggest that one way for managers to limit decreased well-being among workers who must engage in extensive surface acting is to develop efforts aimed at enhancing the self-efficacy of those workers. Furthermore, where possible, managers may be able to lessen the need for detrimental surface acting in the first place. Effectively run organizations can reduce the potential for negative customer experiences and, in turn, negative interactions between workers and customers, thus minimizing the need for surface acting (e.g., Lopez, 2006).

The other type of self-efficacy, relational self-efficacy, did not influence the relationship between surface acting and self-estrangement. Although a worker’s confidence in her ability to manage the emotions of coworkers has been linked with increased job satisfaction and reduced job burnout in previous research, relational self-efficacy does not appear to affect feelings of self-estrangement (Guy et al., 2008). Relational self-efficacy may relate to more emotionally intense forms of emotional labor and be a more important factor to consider in examinations of deep acting emotional labor (Hsieh et al., 2012).

While the analyses did not report occupational differences in the extent of surface acting reported by workers, workers in a managerial position reported greater amounts of surface acting than workers without supervisory responsibilities. This effect held net of job complexity and control over work. Despite working in more complex jobs and having control of their own work, managers still must engage in extensive emotion management. Managers are in a position where they must deal with the needs of the public and the employees, while maintaining an affectively “professional” demeanor. Future investigations may consider focusing on the emotional labor of managers as a group and examining how it relates to the performance of their subordinates (see Bolton & Houlihan, 2010).

In sum, this research represents an initial effort to test the moderating role of self-efficacy on the relationship between surface acting and self-estrangement in a large sample of public service workers. The findings suggest that research efforts should be aimed at identifying additional ways to lessen the likelihood that surface acting will lead to feelings of self-estrangement. For instance, worker autonomy is associated with decreased feelings of self-estrangement (Bulan et al., 1997; Erickson & Ritter, 2001). It is possible that the feeling that a worker is in control of her emotional labor may reduce the likelihood that self-estrangement will result from surface acting (Guy et al., 2008).⁷ In addition, Hsieh and colleagues (2012) found that a greater engagement in emotion work can increase worker’s ability to successfully engage in surface acting. Thus, workers who engage in more emotionally intense forms of emotional labor (e.g., deep acting) in addition to surface acting may be at a lowered risk for negative consequences. While engaging in surface acting may be unavoidable for many public service workers, research directed towards furthering our understanding of the link between surface acting and self-estrangement can provide insight into ways to improve worker well-being.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was funded in part by a National Science Foundation Dissertation Improvement Grant (#0424969).

Notes

Author Biography

Melissa M. Sloan is an assistant professor of interdisciplinary social sciences and sociology at the University of South Florida Sarasota-Manatee. Her research interests include the sociology of emotion, work and occupations, mental health, and survey research.

References

Bandura

(1996). Social learning theory. Englewood Cliffs, NJ: Prentice-Hall.

Bolger

DeLongis

Kessler

R. C.

Schilling

E. A.

(1989). Effects of daily stress on negative mood. Journal of Personality and Social Psychology, 57, 808-818.

Bolton

S. C.

Boyd

(2003). Trolley dolly or skilled emotion manager? Moving on from Hochschild’s managed heart. Work, Employment & Society, 17(2), 289-308.

Bolton

S. C.

Houlihan

(2010). Bermuda revisited? Management power and powerlessness in the worker-manager-customer triangle. Work and Occupations, 37, 404-432.

Brotheridge

C. M.

Grandey

A. A.

(2002). Emotional labor and burnout: Comparing two perspectives of “people work.” Journal of Vocational Behavior, 60, 17-39.

Bureau of Economic Analysis (BEA). (2007). State economic growth slowed in 2007: Advance 2007 and revised 2004-2006 GDP-by-state estimates. U.S. Department of Commerce. Retrieved from http://www.bea.gov/newsreleases/regional/gdp_state/2008/gsp0608.htm

Bureau of Labor Statistics (BLS). (2004). Regional and state employment and unemployment. November 2004. Retrieved from http://www.infran.ru/history_eng.htm

Bureau of Labor Statistics (BLS). (2005). Union members in 2004. U.S. Department of Labor. Retrieved from http://www.bls.gov/news.release/archives/union2_01272005.pdf

Dillman

D. A.

(2000). Mail and internet surveys: The tailored design method (2nd ed). New York: Wiley.

10.

Erickson

R. J.

(1995). The importance of authenticity for self and society. Symbolic Interaction, 18(2), 121-144.

11.

Erickson

R. J.

Ritter

(2001). Emotional labor, burnout, and inauthenticity: does gender matter? Social Psychology Quarterly, 64(2), 146-163.

12.

Erickson

R. J.

Wharton

A. S.

(1997). Inauthenticity and depression: Assessing the consequences of interactive service work. Work and Occupations, 24(2), 188-213.

13.

Glasø

Einarsen

(2008). Emotion regulation in leader–follower relationships. European Journal of Work and Organizational Psychology, 17, 482-500.

14.

Glomb

T. M.

Tews

M. J.

(2004). Emotional labor: A conceptualization and scale development. Journal of Vocational Behavior, 64, 1-23.

15.

Grandey

A. A.

(2000). Emotion regulation in the workplace: A new way to conceptualize emotional labor. Journal of Occupational Health Psychology, 5(1), 95-110.

16.

Grandey

A. A.

(2003). When “the show must go on”: Surface acting and deep acting as determinants of emotional exhaustion and peer-rated service delivery. Academy of Management Journal, 46(1), 86-96.

17.

Guy

M. E.

Newman

M. A.

(2004). Women’s jobs, men’s jobs: Sex segregation and emotional labor. Public Administration Review, 64, 289-298.

18.

Guy

M. E.

Newman

M. A

Mastracci

S. H.

(2008). Emotional labor: Putting the service in public service. Armonk, NY: M. E. Sharpe.

19.

Hall

E. J.

(1993). Smiling, deferring, and flirting: Doing gender by giving “good service.” Work and Occupations, 20, 452-471.

20.

Heuven

Bakker

A. B.

Schaufeli

W. B.

Huisman

(2006). The role of self-efficacy in performing emotion work. Journal of Vocational Behavior, 69, 222-235.

21.

Hochschild

A. R.

(1979). Emotion work, feeling rules, and social structure. American journal of Sociology, 85, 551-575.

22.

Hochschild

A. R.

(1983). The managed heart. Berkeley, CA: University of California Press.

23.

Hsieh

C-W.

Guy

M. E.

(2009). Performance outcomes: The relationship between managing the “heart” and managing client satisfaction. Review of Public Personnel Administration, 29(1), 41-57.

24.

Hsieh

C-W

Jin

M. H.

Guy

M. E.

(2012). Consequences of work-related emotions: Analysis of a cross-section of public service workers. American Review of Public Administration, 42(1), 39-53.

25.

Hülsheger

U. R.

Lang

J. W. B.

Maier

G. W.

(2010). Emotional labor, strain, and performance: Testing reciprocal relationships in a longitudinal panel study. Journal of Occupational Health Psychology, 15, 505-521.

26.

Hülsheger

U. R.

Schewe

A. F.

(2011). On the costs and benefits of emotional labor: A meta-analysis of three decades of research. Journal of Occupational Health Psychology, 16, 361-389.

27.

Jin

M. H.

Guy

M. E.

(2009). How emotional labor influences worker pride, job satisfaction, and burnout: An examination of consumer complaint workers. Public Performance & Management Review, 33(1), 88-105.

28.

Kohn

M. L.

Schooler

(1973). Occupational experience and psychological functioning: An assessment of reciprocal effects. American Sociological Review, 38, 97-118.

29.

Leidner

R. L.

(1993). Fast food, fast talk: Service work and the routinization of everyday life. Berkeley, CA: University of California Press.

30.

Lively

K. J.

(2000). Reciprocal emotion management. Work and Occupations, 27(1), 32-63.

31.

Lively

K. J.

(2006). Emotions in the workplace. In Stets

J. E.

Turner

J. H.

(Eds.), Handbook of the sociology of emotions (pp. 569-590). New York, NY: Springer.

32.

Lopez

S. H.

(2006). Emotional labor and organized emotional care: Conceptualizing nursing home care work. Work and Occupations, 33(2), 133-160.

33.

Lorens

J. L.

Wenger

J. B.

Kellough

E. J.

(2008). Choosing public sector employment: The impact of wages on the representation of women and minorities in state bureaucracies. Journal of Public Administration Research and Theory, 18, 397-413.

34.

MacRae

(1998). Managing feelings: Caregiving as emotion work. Research on Aging, 20, 137-160.

35.

Maslach

Schaufeli

W. B.

Leiter

M. P.

(2001). Job burnout. Annual Review of Psychology, 52, 397-422.

36.

Mastracci

S. H.

Newman

M. A.

Guy

M. E.

(2006). Appraising emotion work: Determining whether emotional labor is valued in government jobs. American Review of Public Administration, 36(2), 123-138.

37.

Meier

K. J

Mastracci

S. H.

Wilson

(2006). Gender and emotional labor in public organizations: An empirical examination of the link to performance. Public Administration Review, 66, 899-909.

38.

Mirowsky

Ross

C. E.

(2003). Education, social status, and health. New York: Aldine de Gruyter.

39.

Mirowsky

Ross

C. E.

(2007). Creative work and health. Journal of Health and Social Behavior, 48(4), 385-403.

40.

Morris

J. A

Feldman

D. C.

(1997). Managing emotions in the workplace. Journal Managerial Issues, 9, 257-274.

41.

Pugh

S. D.

Groth

Hennig-Thurau

(2011). Willing and able to fake emotions: A closer examination of the link between emotional dissonance and employee well-being. Journal of Applied Psychology, 96, 377-390.

42.

Pugliesi

(1999). The consequences of emotional labor: Effects on work stress, job satisfaction, and well-being. Motivation and Emotion, 23(2), 125-154.

43.

Pugliesi

Shook

(1997). Gender, jobs, and emotional labor in a complex organization. Social Perspectives on Emotion, 4, 283-316.

44.

Sloan

M. M.

(2008). Emotion management and workplace status: Consequences for well-being. International Journal of Work Organisation and Emotion, 2(2), 236-255.

45.

Steinberg

R. J.

Figart

D. H.

(1999). Emotional demands at work: A job content analysis. Annals of The American Academy of Political and Social Sciences, 561, 177-191.

46.

Taylor

(2000). Emotions and identity in women’s self-help movements. In Stryker

Owens

T. J.

(Eds.), Self, identity and social movements (pp. 271-299). Minneapolis: University of Minnesota Press.

47.

Thoits

P. A.

(2002). Emotional deviance: Some antecedents and consequences. Paper presented at 2002 Annual Meeting of the American Sociological Association, Chicago, IL.

48.

Thoits

P. A.

(2004). Emotion norms, emotion work, and social order. In Manstead

A. S. R.

Frijda

N. H.

Fischer

A. H.

(Eds.), Feelings and emotions: The Amsterdam symposium (pp. 357-376). Cambridge: Cambridge University Press.

49.

Tolich

M. B.

(1993). Alienation and liberating emotions at work. Journal of Contemporary Ethnography, 22, 361-381.

50.

Wharton

(1993). The affective consequences of service work: Managing emotions on the job. Work and Occupations, 20, 205-232.

51.

Wharton

A. S.

(2009). The sociology of emotional labor. Annual Review of Sociology, 35, 147-165.

52.

Wolpin

Burke

R. J.

Greenglass

E. R.

(1991). Is job satisfaction an antecedent or a consequence of psychological burnout? Human Relations, 44, 193-209.

53.

Zapf

Holz

(2006). On the positive and negative effects of emotion work in organizations. European Journal of Work and Organizational Psychology, 15, 1-28.