In this chapter, it is divided into the research framework, research hypotheses, sample, data collection, pilot test, measurement, control variables and data analysis.
Research Framework
The present study proposed emotional labor with its dimensions surface and deep acting as independent variable, while burnout dimensions as dependent variable: emotional exhaustion, depersonalization, and personal accomplishment. Customer mistreatment and nationality played as a moderator on the relationship between emotional labor and burnout.
The research framework is shown in Figure 3.1.
Emotional labor
28
Research Hypotheses
Based on the literature review and the objective of this study, the following hypotheses have been proposed:
H1: Surface Acting is positively related to burnout.
H1a: Surface acting is positively related to emotional exhaustion.
H1b: Surface acting is positively related to depersonalization.
H1c: Surface acting is positively related to personal accomplishment.
H2: Deep acting is negatively related to burnout.
H2a: Deep acting is negatively related to emotional exhaustion.
H2b: Deep acting is negatively related to depersonalization.
H2c: Deep acting is negatively related to personal accomplishment.
H3: The positive relationship between surface acting and burnout is stronger for Salvadorean service workers than Taiwanese.
H3a: The positive relationship between surface acting and emotional exhaustion is stronger for Salvadorean service workers than Taiwanese.
H3b: The positive relationship between surface acting and depersonalization is stronger for Salvadorean service workers than Taiwanese.
H3c: The positive relationship between surface acting and personal accomplishment is stronger for Salvadorean service workers than Taiwanese.
H4: The negative relationship between deep acting and burnout is weaker for Salvadorean service workers than Taiwanese.
H4a: The negative relationship between deep acting and emotional exhaustion is weaker for Salvadorean service workers than Taiwanese.
29
H4b: The negative relationship between deep acting and depersonalization is weaker for Salvadorean service workers than Taiwanese.
H4c: The negative relationship between deep acting and lack of personal accomplishment is weaker for Salvadorean service workers than Taiwanese.
H5: Customer mistreatment has a positive moderating effect of surface acting on burnout
H5a: Customer mistreatment has a positive moderating effect of surface acting on burnout.
H5b: Customer mistreatment has a positive moderating effect of surface acting on depersonalization.
H5c: Customer mistreatment has a positive moderating effect of surface acting on personal accomplishment.
H6: Customer mistreatment has a negative moderating effect of deep acting on burnout H6a: Customer mistreatment has a negative moderating effect of deep acting
on emotional exhaustion.
H6b: Customer mistreatment has a negative moderating effect of deep acting on depersonalization.
H6c: Customer mistreatment has a negative moderating effect of deep acting on personal accomplishment.
H7a: Customer mistreatment towards Salvadorean service workers has a more positive moderating effect of surface acting on burnout than Taiwanese service workers.
30
H7b: Customer mistreatment towards Taiwanese service workers has a more negative moderating effect of deep acting on burnout than Salvadorean service workers.
Sample
The target sample of this study are employees working in the service industry in El Salvador and Taiwan. One of the strongest economy sectors in both countries is the service industry; therefore, this population was chosen because of the high interaction that service employees have with customers. Furthermore, these two samples represent a great interest for the researcher to compare the differences between developed and developing countries, being both collectivist countries. The sample included all employees who are currently working on the service industry and have frequent interaction with clients.
Pilot Study
A pilot study was conducted before the data collection. The purpose was to identify the suitability and accuracy of the terms and the set of questions for employees in the service industry. A total of 100 responses were used for the pilot test. Responses were captured from Salvadorean employees in the service industry. With the pilot test was determined the reliability and validity of the instrument. According to Nunnally (1978), the accepted level for reliability is ≥ 0.70. For the pilot test, Cronbach’s alpha for emotional labor was 0.83, for customer mistreatment was 0.96, and burnout was 0.84.
Data Collection Process
A total of 289 valid responses were collected during the period of April-May 2018. Due to the difficulty that the researcher presented in collecting data due to the location, an online
31
questionnaire was implemented to get responses for Salvadorean employees. Snowball and convenient sampling approach were used for data collection. The sample came from different subsectors included in the service industry. Participants received an e-mail with the link directing to the website containing the online survey. Also, through this e-mail was encouraged them forward the e-mail to friends who work in the service industry. For the Taiwanese sample both -online questionnaire and paper-based questionnaire- were implemented. A total of 50 responses were paper based. The questionnaire included a cover letter, in which was explained the purpose of the research, and expressing the confidentiality of the responses. Furthermore, the questionnaire included screening question to ensure that the obtained responses were just from employees in the service industry and who have frequent interaction with customers.
Questionnaire Design
This questionnaire used scales developed originally in English with established reliability and validity. The questionnaire consisted of 48 items to measure the three variables, as well as 13 measuring social desirability as a control variable, and 7 demographic questions.
Since participants are not native English speakers, a back-translation (Brislin, 1980) procedure was performed. The researcher asked native Spanish and Chinese speakers to translate both versions. Screening questions are also included at the beginning of the questionnaire, to identify valid participants that are working in the service industry and that have frequent interaction with customers. Also, since all the questions in the different variables will be answered for one person, the scores were rated with 5-point or 7-point Likert Scales to avoid the Common Method Variance (CMV) (Podsakoff, Mackenzie, Lee, &
Podsakoff, 2003).
32
Measurement
The measurement in this study comprised of five parts: emotional labor, customer mistreatment, burnout, social desirability and demographic information. These measures are described below:
Emotional Labor
The instrument for this variable was adopted from Grandey (2003). This instrument includes 5 items measuring Surface acting (α = .83), and 3 items measuring deep acting (α
= .86). Emotional labor was measured using five-point Likert scale (1 = never, 5 = always).
Sample items for surface acting were: “I pretend to have the emotions I need to display for my job”; and for deep acting “I work hard to feel the emotions that I need to show to others”.
The total items of emotional labor were drawn on the following table (Table 3.1).
Customer Mistreatment
Customer mistreatment instrument was adopted from Wang, et al. (2011) adaptation.
The adaptation made by Wang et al. (2011) used items from Dorman and Zapf (2004) and Sharlicki et. al (2008). This instrument includes 18 items assessing customer-related social stressors, and customer interpersonal injustice behaviors. This instrument used five-point Likert scale (1 = never, 5 = all the time). Some sample items were: “Customers demanded special treatment”, “Customers refused to listen to me” and “Customers used condescending language to me”. The total items of customer mistreatment were drawn on the following table (Table 3.2).
33 Table 3.1.
Emotional Labor Scale Emotional Labor
1 I put on a show or performance when interacting with customers.
2 I put on a “mask” in order to express the right emotions for my job.
3 I pretend to have the emotions I need to display for my job.
4 I put on an act in order to deal with customers in an appropriate way.
5 I fake a good mood when interacting with customers
6 I work hard to feel the emotions that I need to show to others.
7 I make an effort to actually feel the emotions I need to display toward others.
8 I try to actually experience the emotions that I must show.
Burnout
Maslach Burnout Inventory (Maslach, Jackson, & Leiter, 1996) was used to measure burnout. The scale is seven-point Likert scale (0 = never, 6 = every day), including 22 items measuring three dimensions: emotional exhaustion (α = 0.90), depersonalization (α = 0.79), and personal accomplishment (α = 0.71). Some of the items included for emotional exhaustion is: “I feel I’m working too hard on my job”; for depersonalization: “I do not care what happens to the customers”, and for personal accomplishment: “I deal effectively with customers problems”. The total items of burnout were drawn on the following table (Table 3.3).
34 Table 3.2.
Customer Mistreatment Scale Customer Mistreatment
1 Customers demanded special treatment.
2 Customers thought they were more important than others.
3 Customers asked me to do things they could do by themselves.
4 Customers vented their bad mood out on me.
5 Customers did not understand that I had to comply with certain rules.
6 Customers complained without reasons.
7 Customers made exorbitant demands.
8 Customers were impatient.
9 Customers yelled at me.
10 Customers spoke aggressively to me.
11 Customers got angry at me even over minor matters.
12 Customers argued with me the whole time.
13 Customers refused to listen to me.
14 Customers cut me off mid-sentence.
15 Customers made demands that I could not deliver.
16 Customers insisted on demands that are irrelevant to my service.
17 Customers doubted my ability.
18 Customers used condescending language to me.
35 Table 3.3.
Burnout Scale Burnout
1 I feel used up at the end of the workday.
2 I feel I’m working too hard on my job.
3 I feel emotionally drained from work.
4 I feel frustrated by my job.
5 I feel fatigued when I get up in the morning and have to face another day on the job.
6 I feel burn out from my work.
7 I feel working with people is a strain for me.
8 I feel like I am at the end of my rope.
9 I feel working with people directly is too stressful.
10 I feel recipients blame me for some of their problems.
11 I feel I treat some customers like impersonal objects.
12 I act more callously toward people since this job.
13 I worry that this job is hardening me emotionally.
14 I worry that this job is hardening me emotionally.
15 I deal effectively with customers problems.
16 I easily understand how customers feel.
17 I can easily create a relax atmosphere with customers.
18 I positively influence others through my work.
19 I have accomplished many worthwhile things on my job.
(continued)
36 Table 3.3. (continued)
Burnout
20 I deal calmly with emotional problems at work.
21 I feel very energetic
22 I feel exhilarated after closing work with customers.
Control Variables Age
Previous studies exhibited that age was pertinent on the selection of emotional labor strategies during the interaction between customer and service employee (Cheung & Tang, 2010; Dahling & Perez, 2010; Sliter et al., 2010; Yeung, Wong, & Lok, 2011). As individuals age, they turn out to be progressively spurred to boost experiences which deliver positive feelings and limit those related to negative emotions, which may impact the selection of emotional labor. (Charles & Carstensen, 2007; Dahling & Perez, 2010). Besides, some studies have demonstrated that older people comprehend and control their feelings better than youthful individuals (Ng & Feldman, 2009).
Social Desirability
Social Desirability is one of control variables. Due to the self-report nature of the responds, social desirability was included as a control variable in order to minimize the influence of “faking” response, since the independent variable “emotional labor” has some sensitive questions and is necessary to obtain honest answers from the participants. This measure was adopted from Marlowe-Crowne Social Desirability questionnaire (Reynolds, 1982). This instrument includes 13 true-false items. Cronbach Alpha for this measurement was α = .76. Examples of items included: “It is sometimes hard for me to go on with my
37
work if I am not encouraged”; and “There have been occasions when I took advantage of someone.”
Organizational Level
Four organizational levels are considered in this research: staff, first line manager, middle level manager, and high-level manager. Some studies (Beehr and Drexler, 1986;
Westman, 1992), proposed that the management of emotions toward customers may depend on the organizational level that the employee belongs, and this affect more to low-level employees than managerial positions.
Data Analysis
IBM SPSS 23.0 and AMOS 22.0 were used to perform data analysis. Other statistical techniques as descriptive statistics analysis, Pearson’s correlation analysis, confirmatory factor analysis (CFA), Hierarchical Regression, and structure equation modeling (SEM) were carried out to test the hypotheses. Also, a comparison of the results between El Salvador and Taiwan were performed using multi-group SEM.
Descriptive Analysis
This type of analysis helps to have an overview of the characteristics of the participants regarding of frequency and percentages. Descriptive statistics in this study included background information such as gender, age, education level, type of company they are working, job position level, and time working at the current company. The mean and standard deviation is useful to investigate all the variables in this study, including emotional labor, burnout, customer mistreatment and control variables. The Table 3.4 showed the results for El Salvador and Taiwan. From the data collected, 60.9% (176) were female participants and
38
39.1% (113) were male. Making a division just for female, 28.4% (82) were Salvadorean and 32.5% (94) Taiwanese. Males in El Salvador were 24.9% (72) and for Taiwan, 14.2% (41).
Most of the participants were still very young with 34.9% at the age of 18-25 years old, 31.5% at the age of 26-31 years old. For El Salvador, the ages of 18-25 represent the majority of responses with 21.1%, the same as Taiwan with 13.8%. There is also a great participation with ages of more than 41 years old in Taiwan, with 13.5%. For the education level, 28.4%
had high school degree, 14.5% technical degree, 46.7% Bachelor degree and 10.4% Master’s degree. For both countries, the majority of responses had a Bachelor degree, with 23.2% for El Salvador, and 23.5% for Taiwan. Most of the respondents are working in the commerce sub-sector, with a total of 76.1%. Results also show that 88% of respondents are working as a staff. Most of the employees expressed working at their current company for a short period, 22.1% have been working for less than one year in the company; 45% of employees have worked between 1-3 years, and 11.1% had been working at the current company for 4-6 years.
Correlation and Reliability Analysis
To have a better understanding of the relationship between variables, the correlation analysis was conducted. Table 3.5 and Table 3.6 presented the means, standard deviations, correlations and reliabilities of gender, age, social desirability, emotional labor, customer mistreatment, and burnout for both countries. The acceptable Cronbach’s alpha coefficient is
≤ 0.7 (Nunnally, 1978). The Cronbach alpha in Taiwan for surface acting is 0.90, deep acting, 0.78, emotional exhaustion 0.94, depersonalization 0.93, personal accomplishment 0.90, and customer mistreatment 0.96. For El Salvador’s sample, surface acting has a Cronbach alpha of 0.80, deep acting, 0.88, emotional exhaustion 0.94, depersonalization 0.85, personal accomplishment 0.84, and customer mistreatment 0.97. According to Table 3.5 and Table 3.6, most of the variables have significant correlations in this study. The correlation (r) help us to
39
identify the relationship and strength between emotional labor and burnout, and the moderating effect of customer mistreatment. The correlations revealed that surface acting in Taiwan has a positive significant relationship with deep acting (r = .52, p < 0.01), emotional exhaustion (r = .30, p < .01), depersonalization (r = .24, p < .01), personal accomplishment (r
= .20, p < 0.05), and customer mistreatment (r = .19, p < .01). Emotional exhaustion has a positive correlation with depersonalization (r = .74, p < .01), personal accomplishment (r
= .31, p < .01), and customer mistreatment (r = .55, p < .01). And depersonalization is positive correlated with personal accomplishment (r = .27, p < .01), and customer mistreatment (r = .67, p < .01). In El Salvador, all the variable for emotional labor, burnout and customer mistreatment are positively correlated as shown in Table 3.5.
Table 3.4.
Descriptive Statistics (N=289)
(continued)
Item El Salvador Taiwan Total
1. Gender Frequency Percentage Frequency Percentage Frequency Percentage
Female 82 28.4% 94 32.5% 176 60.9%
Male 72 24.9% 41 14.2% 113 39.1%
Total 154 53.3% 135 46.7% 289 100%
2. Age (years)
18-25 61 21.1% 40 13.8% 101 34.9%
26-31 63 21.8% 28 9.7% 90 31.5%
32-41 16 5.5% 28 9.7% 44 15.2%
>41 14 4.8% 39 13.5% 53 18.3%
Total 154 53.3% 135 46.7% 289 100.0%
40 Table 3.4. (continued)
(continued)
Item El Salvador Taiwan Total
3. Education Freq. Percentage Freq. Percentage Freq. Percentage
High school 58 20.1% 24 8.3% 82 28.4%
Technical 25 8.7% 17 5.9% 42 14.5%
Bachelor 67 23.2% 68 23.5% 135 46.7%
Master 4 1.4% 26 9.0% 30 10.4%
Total 154 53.3% 135 46.7% 289 100.0%
4. Type of company
Commerce 133 46.0% 87 30.1% 220 76.1%
Communica. 6 2.1% 17 5.9% 23 8.0%
Community 4 1.4% 19 6.6% 23 8.0%
Finances 4 1.4% 7 2.4% 11 3.8%
Hospitality 6 2.1% 3 1.0% 9 3.1%
Outsource 1 0.3% 0 0.0% 1 0.3%
Transport 0 0.0% 2 0.7% 2 0.7%
Total 154 53.3% 135 46.7% 289 100.0%
5. Organizational level
Staff 136 47.1% 85 29.4% 221 76.5%
First-line 3 1.0% 22 7.6% 25 8.7%
Middle-level 11 3.8% 18 6.2% 29 10.0%
High-level 4 1.4% 10 3.5% 14 4.8%
Total 154 53.3% 135 46.7% 289 100.0%
41 Table 3.4. (continued)
Item El Salvador Taiwan Total
7. Tenure Freq. Percentage Freq. Percentage Freq. Percentage Less than 1
year
50
17.3%
14
4.8% 64 22.1%
1-3 years 68 23.5% 62 21.5% 130 45.0%
4-6 years 8 2.8% 24 8.3% 32 11.1%
7-9 years 3 1.0% 13 4.5% 16 5.5%
10-12 years 3 1.0% 9 3.1% 12 4.2%
13-15 years 2 0.7% 0 0.0% 2 0.7%
More than 15 years
20
6.9%
13
4.5% 33 11.4%
Total 154 53.3% 135 46.7% 289 100.0%
42 Table 3.5.
Mean, Standard Deviations, Correlations, and Reliability for El Salvador (N=154)
Variables Mean S.D. 1 2 3 4 5 6 7 8 9
1 Age 29 8.37
2 Organizational Level 1.24 0.70 .12
3 Social Desirability 0.54 0.19 -.12 -.21*
4 Surface Acting 2.81 1.11 -.14 -.16* .93** (.80)
5 Deep Acting 2.71 1.45 -.06 -.21** .88** .64** (.88)
6 Emotional Exhaustion 3.40 3.4 -.03 -.154 .65** .63** .53** (.94)
7 Depersonalization 1.88 1.88 .02 -.156 .57** .55** .48** .71** (.85)
8 Personal Accomplishment 1.62 1.62 -.25** -.158 .31** .34** .20* .43** .45** (.84)
9 Customer Mistreatment 2.38 2.38 -.15 -.24** .65** .58** .59** .67** .61** .40** (.97) Note. Cronbach’s Alpha are in boldface.
*p < .05, **p < .01, ***p < .001
43 Table 3.6.
Mean, Standard Deviations, Correlations, and Reliability for Taiwan (N=135)
Mean S.D. 1 2 3 4 5 6 7 8 9
1 Age 35 11.76
2 Organizational Level 1.7 1.07 .47**
3 Social Desirability 0.58 0.15 -.12 -.04
4 Surface Acting 2.46 0.99 -.30** -.21* -.02 (.90)
5 Deep Acting 2.93 0.82 -.19* -.11 -.04 .52** (.78)
6 Emotional Exhaustion 2.99 1.47 -.31** -.19* .07 .30** .10 (.94)
7 Depersonalization 2.11 1.65 -.28** -.13 .12 .24** .09 .74** (.93)
8 Personal Accomplishment 2.18 1.19 -.22* -.31** -.05 .20* .02 .31** .27** (.90)
9 Customer Mistreatment 1.79 0.928 -.20* -.06 .18* .19* .10 .55** .67** .05 (.96) Note. Cronbach’s Alpha are in boldface.
*p < .05, **p < .01, ***p < .001
44