Emotional Labor in the Airlines: Relation between Psychological Contract Violation, Job Satisfaction, and Empathy

Abstract

With the increasing domination of the service sector in the globalized world economy, this research examines the hypotheses that empathy as emotional labor practiced by human service employees moderate the decreasing propensity of job satisfaction. The background is the changed organizational climate, where the human service employees may perceive a psychological contract violation (PCV). The questionnaire surveys were administered to a total of 827 flight attendants working for two airlines, one is a European and another is an Asian airline. The result shows that, first, for both airlines, PCV perceived by the employees decrease job satisfaction. Second, empathy as emotional labor practiced by the employees moderates the decreasing propensity of job satisfaction. Third, the cross-cultural difference is observed how to moderate or repair the decreasing propensity of job satisfaction between European and Asian airline employees. As a practical implication to discuss, emphatically or emotionally competent employees might harmoniously work with people as well as the automated machines, IT and AI in the stressful workplace, thus, contribute a long-term growth of the human service organization.

1. Introduction

The world economy is becoming increasingly dominated by the service sector. For example, the population in professional and business services increased by 11.75% between 2000 and 2010 in the United States [1]. Organizational researchers are increasingly paying attention to the unique problem and issues involved in managing service providers, organizations, and employees.

In the airline industry, many airlines use information technology (IT) and automated machinery for reservation, airport check-ins, and other operations. While it is important for airlines to adapt the competitive environment, IT and the automated machine perform the work previously achieved by human contact employees. As a result, downsizing and the introduction of early retirement program are a trend [2]. Moreover, the advanced technology made many recent airports possible to operate 24 hours a day and 7 days a week and generate challenges for flight attendants to work in the extended duty periods, highly variable schedule, frequent time changes, and increased passenger load. Thus, the external competitive environment and the internal managerial changes raise a new challenge to employees.

As a result, the professional roles expected of modern flight attendants gradually changed from what they were in past decades. Thus, the service-oriented employees might perceive a psychological contract violation (PCV) in the workplace. PCV is an employee’s feeling of disappointment and betrayal arising, when they believe that their organization has broken its work-related promises [3]. Conversely, interest in emotions in the workplace has accelerated rapidly over the past decade (e.g., [4, 5]).

The purpose of the present research is, first, to test the hypothesis that psychological contract violation perceived by the human service employees (flight attendants) decreases their job satisfaction. Second, the present research tests the hypotheses whether empathy as emotional labor practiced by the flight attendants moderates the decreasing propensity of job satisfaction.

2. Literature review

2.1 Emotional labor

Emotional labor refers to the process by which workers are expected to manage their feelings in accordance with organizationally defined rules and guidelines [6]. Emotional labor is considered an essential part of service-oriented employees. Hochschild [7] classified occupations requiring a higher degree of emotional labor, which include personal service workers and flight attendant as well as health service workers and some other customer-oriented workers. Hochschild [7] has practiced many fieldworks in an airline in the United States and observed the recruiting, training, and other processes of flight attendant. For example, Hochschild [7] introduced an episode in the training that a pilot spoke of the smile as the flight attendant’s assets. According to Hochschild [7], organizations are increasingly willing to direct and control how employees present themselves to others.

Numerous scholars have investigated the role of emotional labor. Emotional labor, which was the idea of managing with others as part of the work role, was proposed in sociology in the 1980s. Slowly, the organizational behavior (OB) and organizational psychology (OP) literature began recognizing the value of understanding emotions at work, and emotional labor became a focal area of study [8].

Emotional regulation: Emotional regulation is defined as “the processes by which individuals influence which emotions they have, when they have them, and how they experience and express these emotions” ([9], p. 275). Figure 1 shows that the consensual process model of emotional regulation described by Gross [10]. This model suggests that emotions may be regulated either by manipulating the input to the system (antecedent-focused emotional regulation) or by manipulating its output (response-focused emotional regulation).

Figure 1.

Emotional regulation. A consensual process model of based on Gross [10] adapted for human service employees by the author.

Antecedent-focused emotion regulation: Antecedent-focused emotion regulation concerns the manipulation of the input to the system [10]. An example of antecedent-focused emotion regulation is situational selection, in which one approaches or avoids certain people or situations on the basis of their likely emotional impact [10]. Taking a different route to the store to avoid a neighbor who tells offensive jokes [11] and seeking out a friend with whom one can have a good cry [9] are examples of situational selection. Another example of antecedent-focused emotion regulation is situation modification which is a potentially emotion-eliciting situation, for example, whether a flat tire on the way to an important appointment or loud music next door at 3:00 a.m. does not ineluctably call forth emotion. One may convert a meeting into a phone conference or convince a neighbor to tone down a raucous party [9].

Response-focused emotion regulation: Response-focused emotion regulation, in contrast, concerns the manipulation of output from the system [10], that is, response modulation or manipulations of physiology, which include facial expression, behavior, and cognition once an emotion is experienced; for example, subordinates tend to hide their experienced anger from their bosses [12]. Another example of response-focused emotion regulation when individuals exhibit more elevated signs of strain is that individuals try not to show any feelings while watching a terrifying or sad movie which make people cry (e.g., [13]).

Emotional regulation in organization: Emotional regulation involves the employees in displaying the organizationally desired emotion [14] and inducing or suppressing feelings to sustain the outward countenance that produces the proper state of mind in others [7]. For example, flight attendants act cheerfully and friendly (e.g., [15]) and put on a smile while dealing with customers, because it is a part of their job [16]. Emotion regulatory process may be automatic or controlled and conscious or unconscious and may have their effects at one or more points in the emotion generative process [9]. The capacity of emotion to promote or undermine constructive functioning depends on the extent to which emotional arousal is monitored, evaluated, and controlled by the individual [17] or the latency, rise time, magnitude, duration, and offset of responses in behavioral, experiential, or physiological domains [9]. Psychologically, emotional regulation is a painstaking developmental process, because it requires intervening in phylogenetically deeply rooted affect systems with the psychologically complex control mechanism. For this reason, the management of emotion is an important component of “emotional maturity” [18] and “emotional competence” [19].

Emotional display rules: Although emotions have long been a topic of interest to sociologists and psychologists [7], the display of emotions in organizations has been become a topic of greater interest to organizational scholars [4, 20, 21, 22]. Hochschild [7] argued that common expectations exist concerning the appropriate emotional reactions of individuals involved in service transaction. In the emotional labor literature, the focus is customer service, where interactions are less spontaneously “emotional,” yet high emotional control is needed to maintain positive emotions with customers across time and situations [7]. Human service employees act as emotional labor in order to conform to the emotional regulation and the display rules required by the organizations (Figure 2).

Figure 2.

Conceptual model of the direct and the moderating effects between psychological contract violation (PCV) and job satisfaction.

Emotional display for organizational purposes has been referred to as “display rules” [23]. Display rules are standards of behavior that indicate not only which emotions are appropriate in a given situation but also how those emotions should be conveyed or publicly expressed [24]. Thus, display rules are norms and standards of behavior indicating what emotions are appropriate in a given situation (Figure 3). For example, flight attendants are encouraged to smile, while lawyers use an aggressive and angry tone to encourage compliance in adversaries [25].

Figure 3.

Study 1: interaction of PCV and affective delivery on job satisfaction.

Affective delivery: Affective delivery or expressing positive emotions in service interactions promotes customer satisfaction [26]. “Employee affective delivery” refers to an employee’s “act of expressing socially desired emotions during service transaction” [15]. An affective service delivery is perceived as friendly and warm, which is related to desirable outcomes [27].

Emotional strategies: Hochschild [7] identified two emotional strategies that may be used by employees to manage their emotions: surface acting and deep acting.

1. Surface acting is an emotional strategy in which employees modify their behavioral displays without changing their inner feelings and employees conform to the rules to retain their job, not to help the customer or organization [26]. Surface acting corresponds to the managing observable expression (e.g., facial or behavioral expressions) to obey display rules.

2. Deep acting is another emotional strategy in which the process of controlling internal thoughts and feelings to meet the mandated display. Emotions involve physiological arousal and conditions, and deep acting works on modifying arousal or cognitions through a variety of techniques [28].

2.2 Psychological contract violation

Contracts: Contracts are important features of exchange agreements. The contract binds the transacting parties as well as the employer and employees and regulates their activities (Figure 4).

Figure 4.

Study 2: interaction of PCV and surface acting on job satisfaction.

Psychological contracts: The term psychological contract was introduced in the early 1960s by Argyris [29], Levinson et al. [30] and by Schein [31]. While the economic and legal contract can be explicitly described in formal legal terms, Rousseau [32] defines the psychological contracts as an individual’s belief regarding the term and conditions of a reciprocal exchange agreement between that focal person and another party.

Psychological contract violation (PCV): A psychological contract violation occurs when one party in a relationship perceives another to have failed to fulfill promised obligations [33]. Since contracts emerge under assumptions of good faith and fair dealing and involve reliance by parties on the promises of the others, violations can lead to serious consequences for the parties involved. Psychological contract violation is distinct from unmet expectations and perceptions of inequity [33]. Employees initially hold unrealistic expectations, and when these expectations go unmet, employees may become less satisfied, perform less well, and become more likely to leave their employer [34].

3. Research hypothesis construction

3.1 Direct effects between an antecedent (PCV) and an outcome (job satisfaction)

Job satisfaction is the pleasurable emotional state resulting from appraisal of one’s job as achieving of facilitating one’s job value [35] and a measure of the employee’s evaluation of the job and has often been used as a proxy for employee well-being at work [36]. I suppose the human service employee (flight attendants in this study) may perceive a psychological contract violation in the recent changing industrial climate, and then, the situation may deteriorate the job satisfaction of the employees. Therefore, I propose the following direct effect hypothesis:

Hypothesis 1: Psychological contract violation (PCV) perceived by human service employee negatively relates to job satisfaction.

3.2 Moderating effects of emotional labor aspects between an antecedent (PCV) and an outcome (job satisfaction)

A moderator variable specifies when and under what conditions a predictor variable influences a dependent variable [37]. A moderator variable may reduce or enhance the direction of the relationship between a predictor variable and a dependent variable, or it may even change the direction of the relationship between the two variables from positive to negative or vice versa. I propose the following moderating effect hypotheses:

Hypothesis 2: Affective delivery of customer service employees moderates the negative relationship between psychological contract violation (PCV) and job satisfaction.

Hypothesis 3: Surface acting of customer service employees moderates the negative relationship between psychological contract violation (PCV) and job satisfaction.

Hypothesis 4: Deep acting of customer service employees moderates the negative relationship between psychological contract violation (PCV) and job satisfaction.

4. General methods

4.1 Participants and procedures

The questionnaire surveys were administered to a total of 827 flight attendants, 414 flight attendants working for a European airline (Study 1), and 413 flight attendants working for an Asian airline (Study 2). A five-point Likert-type scale of questionnaire was developed based on a review of the literature to assess the emotional labor aspects and other variables. The questionnaire was randomly distributed. The researcher explained the purpose of the survey to the participants, emphasizing the anonymity and confidentiality of the data. Participants were asked to complete and return the questionnaire to the researcher directly on the place or by using a provided stamped envelope.

4.2 Measures

Affective delivery (an emotional labor aspect) was measured by using three items (α = 0.82) derived from the bases of Grandey [26]. Surface acting (an emotional labor aspect) was measured by using four items (α = 0.89) derived from the bases of Brotheridge and Lee [38]. Deep acting (an emotional labor aspect) was measured by using three items (α = 0.91) of Brotheridge and Lee [38]. Psychological contract violation (PCV) was measured by using two items (α = 0.84) derived from Robinson and Rousseau [33]. Job satisfaction was measured by two items (α = 0.86) derived from Robinson and Rousseau [33]. Careerism orientation was measured by using five items (α = 0.91) derived from Robinson and Rousseau [33]. Task performance was measured by using three items (α = 0.82) derived from Williams and Anderson [39] and Brown et al. [40]. Emotional exhaustion was measured by using four items (α = 0.93) derived from Pines & Aronson [41].

4.3 Data analysis

Cronbach’s α is the most widely used index of the reliability of a scale [42] and is an important concept in the evaluation of assessments and questionnaires. In the descriptive statistics, Cronbach’s α and intercorrelations were calculated (Table 1 for Study 1 and Table 3 for Study 2). Then, hierarchical regression analyses were conducted to test the study hypotheses, the direct effects of the antecedents (PCV) on the consequence (job satisfaction), and the moderating effects of PCV and emotional labor aspects (affective delivery, surface acting, and deep acting) on the relations between the antecedents and the consequence (Table 2 for Study 2 and Table 4 for Study 2).

		Mean	SD	α4	1	2	3	4	5	6	7	8	9	10	11
1	Gender 1	.75	.44
2	Tenure 2	4.11	1.51		.13**	—
3	Age 3	3.98	.79		.01	.76***	—
Variables in psychological contract context
4	PCV	2.50	.76	.84	−.08*	−.07	−.43	—
5	Job satisfaction	4.08	.71	.86	.04	.14***	.11**	−.22***	—
6	Careerism orientation	2.18	.67	.91	.06	−.10**	−.07	.11*	−.25***	—
7	Task performance	4.25	.52	.82	.10*	.12**	.09	.01	.10**	−.08	—
8	Emotional exhaustion	3.43	.88	.93	−.16***	−.03	.01	.11*	−.21***	−.04	−.17***	—
Variables in empathy as emotional labor
9	Affective delivery	4.50	.52	.82	−.14**	.14***	.22***	.06	.04	−.02	.36***	−.06	—
10	Surface acting	3.90	.68	.89	.13**	.03	.08	.03	.01	.08*	.26***	.07	.19***	—
11	Deep acting	3.65	.90	.91	.01	.02	−.02	−.03	.02	.00	−.04***	.18***	.13***	.34***	—

Table 1.

Study 1: descriptive statistics, reliability, and intercorrelations.

¹

Gender: coded as male = 0, female = 1.

²

Job tenure: coded as 1 = 0–5 years, 2 = 6–10 years, 3 = 11–15 years, 4 = 16–20 years, 5 = 21–25 years, 6 = 26–30 years, 7 = more than 30 years.

³

Age: coded as 1 = less than 20, 2 = 21–30, 3 = 31–40, 4 = 41–50, 5 = 51–60, 6 = more than 60.

⁴

α: Reliability is denoted by Cronbach alpha coefficients.

Note: *** p < .001, **p < .01, *p < .05. N = 414.

Table 2.

Study 1: hierarchical regressions analyses.

5. Results

5.1 Study 1

The participants of the Study 1 are flight attendants working for a European airline. The participants usually work on the long-haul flights, of which the durations are between 6 and 12 hours. After explaining the confidentiality of data, approximately 500 questionnaires were randomly distributed during the field works (10 times) between 2015 and 2017, and 414 responses were received, resulting in a valid response rate of approximately 82.8%.

Profile of the respondents: Regarding gender, 307 respondents were female (74.2%) and 107 respondents were male (25.8%). Regarding tenure, the most frequent respondents answered that their work experience was from 16 to 20 years (26.6%), followed by the respondents who have work experience from 11 to 15 years (19.6%), and from 26 to 30 years (18.6%). Regarding age, the most frequent responses answered that their ages were from 41 to 50 years (53.6%), followed by the respondents aged from 51 to 60 years (22.0%), and from 31 to 40 years (20.0%).

Table 1 presents the descriptive statistics, reliability, and intercorrelations. Hypothesis 1 (study 1) proposed that psychological contract violation (PCV) perceived by the human service employee is negatively related to job satisfaction. As predicted by Hypothesis 1 (study 1), Table 1 presents that PCV is significantly and negatively relates to job satisfaction (r = − .22, p < .001), supporting H1 (study 1).

Table 2 presents the summary of the hierarchical regression analyses. In step 1, the control variables, including gender, tenure, age, and the additional independent variables, including PCV, careerism orientation, task performance, and emotional exhaustion, are inserted into the regression equation to eliminate alternative explanations. In the step 2, the independent variables of emotional labor aspects (affective delivery, surface acting, and deep acting) are inserted into the regression equation. Table 2 presents that PCV is negatively related to job satisfaction (β = −.17, p < 0.01, both in the step 1 and step 2), also supporting Hypothesis 1 (study 1).

Hypotheses 2 (study 1) proposed that affective delivery of human service employees moderates the negative relationship between PCV and job satisfaction. Table 2 presents that, when the interaction term (PCV × Affective Delivery) is inserted into the regression equation in the step 3, the interaction was significant [F (12, 401) = 6.640, p < .001, ∆R²  = −.001], supporting Hypothesis 2. Moreover, while PCV was negative predictor on job satisfaction (β = −.17, p < 0.001, both in the step 1 and step 2; β = −.85, p < 0.01 in the basic equation of step 3), when the interaction term (1) (PCV × Affective Delivery) was inserted into the equation in the step 3, significantly positive beta appeared (β = .74, p < .05), supporting Hypotheses 2 (study 1). This interaction term (or moderator variable) changes the relationship between the antecedent (or predictor, PCV) and the outcome (job satisfaction) from negative to positive. Conversely, Hypotheses 3 (interaction term (2) PCV and surface acting) and Hypotheses 4 (interaction term (3) PCV and deep acting) were not supported.

5.2 Study 2

The participants of the study 2 are flight attendants working for an Asian airline. The participants usually work on the medium-haul flights, of which the durations are between 3 and 6 hours. After explaining the confidentiality of data, approximately 500 questionnaires were randomly distributed during the field works (8 times) between 2015 and 2017, and 413 responses were received, resulting in a valid response rate of approximately 82.6%.

Profile of the respondents: Regarding gender, 322 respondents were female (78.0%), and 91 respondents were male (22.0%). Regarding tenure, the most frequent respondents answered that they work less than 5 years (33.7%), followed by the respondents who have work experience from 6 to 10 years (25.2%), and from 11 to 15 years (13.1%). Regarding age, the most frequent responses answered that their ages were from 21 to 30 years (33.9%), followed by the respondents aged from 31 to 40 years (29.1%), and from 41 to 50 years (20.1%).

Table 3 presents the descriptive statistics, reliability, and intercorrelations. Hypothesis 1 (study 2) proposed that psychological contract violation (PCV) perceived by human service employee is negatively related to job satisfaction. As predicted by Hypothesis 1 (study 2), Table 3 presents that PCV is significantly and negatively relates to job satisfaction (r = − .45, p < .001), supporting Hypothesis 1 (study 2). Table 4 also presents that PCV is negatively related to job satisfaction (β = −.46, p < 0.001, both in the step 1 and step 2), supporting Hypothesis 1 (study 2).

Table 3.

Study 2: descriptive statistics, reliability, and intercorrelations.

Table 4.

Study 2: hierarchical regressions analyses.

Table 4 presents the summary of hierarchical regression analyses. In the step 1, the control variables, including gender, tenure, age, and the additional independent variables, including PCV, careerism orientation, task performance, and emotional exhaustion, are inserted into the regression equation to eliminate alternative explanations. In the step 2, the independent variables of emotional labor aspects (affective delivery, surface acting, and deep acting) are inserted into the regression equation.

Hypotheses 3 (study 2) proposed that surface acting of human service employees moderates the negative relationship between PCV and job satisfaction. Table 4 presents that, when the interaction term (PCV × surface acting) is inserted into the regression equation in the step 4, the interaction was significant [F(13, 399) = 11.781, p < .001, ∆R ² = −0.010], supporting Hypothesis 3 (study 2). Moreover, while PCV was negative predictor on job satisfaction (β = −.46, p < 0.001, both in the step 1 and step 2), when the interaction term (2) (PCV × surface acting) was inserted into the equation in the step 4, significantly positive beta appeared (β = .88, p < 0.05). This interaction term (or moderator variable) changes the relationship between the antecedent (or predictor: PCV) and the outcome (job satisfaction) from negative to positive, also supporting Hypothesis 3 (study 2). Conversely, Hypotheses 2 [interaction term (1) PCV and affective delivery] and Hypotheses 4 [interaction term (3) PCV and deep acting] were not significant, and not supporting Hypotheses 2 (study 2) and 4 (study 2).

6. Discussion

In the recent airline industry, many airlines must reduce their costs in order to adapt to the competitive business environment. As a result, the airlines also change the employment relationships. Security, safety, and customer service are the main tasks for flight attendant; irrespective of the airline company is a low-cost carrier (LCC) or traditional airlines. Although much of the work carried out in an aircraft today is nearly the same concerning security and safety as it was in the past days, the working conditions and the expected customer service roles have changed dramatically. Such might generate psychological contract violation in the workplace.

In this research, the self-evaluated questionnaire survey with a five-point Likert-type scale was administered in order to estimate the levels of perceptions of human service employees (flight attendants) concerning PCV-related variables and empathy as emotional labor variables. Table 5 shows the summary of levels concerning job-related perception and empathy as emotional labor. Concerning job-related perception, the level of PCV perceived by the employees is from medium to low (average, 2.79), the level of job satisfaction is from medium to high (average, 3.87), and the level of emotional exhaustion is from medium to high (average, 3.63). Concerning empathy as emotional labor, the level of affective delivery of the employees is high (average, 4.50), the level of surface acting is high (average, 4.05), and the level of deep acting is from medium to high (average, 3.74). In summary, the descriptive statistics show that the level of PCV perceived by the employees is from medium to low and the level of empathy as emotional labor is high.

	Job-related perception			Empathy as emotional labor
	PCV	Job satisfaction	Emotional exhaustion	Affective delivery	Surface acting	Deep acting
Study 1	2.49	4.08	3.45	4.50	3.92	3.65
Study 2	3.09	3.66	3.81	4.49	4.17	3.83
Average	2.79	3.87	3.63	4.50	4.05	3.74

Table 5.

Summary of levels concerning empathy as emotional labor and other job-related variables.

The questionnaire was developed with a five-point Likert-type scale.

The result also shows that, first, for both airlines, PCV perceived by the service-oriented and/or human service employees decreases job satisfaction. Second, empathy as emotional labor practiced by the flight attendant moderates or repairs the decreasing propensity of job satisfaction. Third, while the employees of a European airline who work on the long-haul flights (6–12 hours flights) tend to moderate the decreasing propensity of job satisfaction by practicing “affective delivery” (an aspect of emotional labor), the employees of an Asian airline who work on the medium-haul flights (3–6 hours flights) tend to moderate the decreasing propensity of job satisfaction by practicing surface acting (another aspect of emotional labor). Thus, the cross-cultural difference is observed on how to moderate or repair the decreasing propensity of job satisfaction between the European airline employees and the Asian airline employees.

7. Practical implication

The practical implication is, first, about the duration of duty and the amount of tasks. When the duration of the duty of human contact is long enough to treat people, the employees might use affective delivery (an aspect of emotional labor) to moderate or repair the decreasing propensity of job satisfaction. Conversely, when the employees have to do many tasks and treat people in short/medium time, the employees might use surface acting to moderate or repair the decreasing propensity of job satisfaction.

Second, the result of this study may apply to the occupations classified by Hochschild [7] requiring a higher degree of emotional labor, which include flight attendant as well as health service workers, including physicians, nurses, therapists, dentists, dental hygienists, and any other customer-oriented workers, including teachers and lawyers.

Third, empathy affects a wide range of work behavior such as teamwork and helps the employees keep hold their job satisfaction in the organizational changes.

Finally, emotionally competent human service employees might harmoniously work with people in a specific environmental workplace, thus, contributing a long-term growth of the organization.

8. Limitations

In the cross-sectional design, the use of only the self-evaluated responses of emotional labors may be considered the first limitations of this research. However, self-reports of these psychological variable provide accurate measurement, since it would be difficult for a co-worker or supervisor to accurately estimate whether the employees perceive a certain emotion and psychological aspects.

Another limitation may be that individual characteristics may also account for alternative explanations. For example, employees with some characteristics (e.g., a frank and open character versus a timid and close character), employees with any ability (high/low cognitive ability or high/low self-evaluation), employees with some physical ability (physical fitness), as well as employees’ gender (female versus male), tenure, age (experienced versus novice), culture, and nationality may also account for alternative explanations.

In addition, the present research exclusively focused on flight attendants and limits the generalizability of the findings. For example, the duration and/or frequency of contact with customers would be comparatively shorter for flight attendants than the duration and/or frequency of contact with hospitalized patients for physicians and nurses in the hospital. It limits the generalizability of the findings.

9. Suggestion for future research

A suggestion for future research direction would be an investigation of empathy as emotional labor in the different organizations. Another future research direction would be exploring the optimal level of moderating or repairing effect by empathy as emotional labor. Though this research showed that empathy as emotional labor could moderate or repair the decreasing propensity of job satisfaction, it is uncertain the optimal level of moderation or repair.

In addition, the future research direction could be exploring how to improve the empathic competence in an organization. Finally, empathy as emotional labor may evoke a new stream of explanation for the effectiveness of human service employees in the organization.