Inglês (pdf)
Artigo em XML
Referências do artigo
Tradução automática
Enviar este artigo por email
Citado por SciELO 
Similares em
SciELO 
Permalink
1. Introduction
Quality of service has long been suggested to be a critical source of competitive advantage in hospitality industry as it has significant influence on business outcomes (Cronin & Taylor, 1992; Gruen, Summers & Acito, 2000; Priporas, Stylos, Vedanthachari & Santiwatana, 2017). However, through the daily face-to-face service delivery, service failures which fail to satisfy customers’ expectations are inevitable in this industry (Babakus, Yavas, Karatepe & Avci, 2003; Luo, Guchait, Lee & Madera, 2019). Therefore being able to provide successful service recovery is the key to survive from competitive market. In fact, service recovery performance (SRP) has long been emphasized by large amounts of scholars, and demonstrated to relate to multiple positive outcomes in workplace (Chou, 2015; Daskin & Kasim, 2016; Ha & Jang, 2009; Jung & Seock, 2017; Liao, 2007; Liat, Mansori, Chuan & Imrie, 2017; Lin, McCain & Lolli, 2016; Lin, 2010). Given the critical contributions to business success, prior research has identified that employees’ engagement of SRP could be motivated by multiple factors, including intrinsic (individual characteristics, such as self-efficacy and emotional exhaustion; Choi, Kim, Lee & Lee, 2014; Guchait, 2017; Guchait, Paşamehmetoğlu & Dawson, 2014; Wang, Guchait, Khoa, Paşamehmetoğlu & Wen, 2022), and extrinsic factors (such as leadership, organizational resources; Boshoff & Allen, 2000; Guchait, 2017; Guchait, Abbott, Lee, Back & Manoharan, 2019).
However, majority of prior studies focus either on intrinsic factors while ignoring the influence of extrinsic factors (Karatepe & Karadas, 2012; Karatepe & Tekinkus, 2006; Zapata-Phelan, Colquitt, Scott & Livingston, 2009), or on extrinsic factors while ignoring the influence of intrinsic factors (Dysvik & Kuvaas, 2011; Yavas, Karatepe, Avci & Tekinkus, 2003). In fact, theory and research suggest that both intrinsic and extrinsic factors coexist in workplace (Amabile, 1993; Gong, Wu, Song & Zhang, 2017; Moneta, & Spada, 2009 ) and hence, there is a growing interest these years which calls for research focus on more specific and comprehensive predictors to examine the triggering mechanism (Guchait, 2017), such as involving both intrinsic and extrinsic factors simultaneously.
According to resource ecology perspective of COR theory (Hobfoll, 2011, 2012), organizations can be considered as a resource bank and can also strategically design the salience of resources to shape or promote expected employee outcomes. Using the lens of resource ecology, we address a comprehensive triggering mechanism for engagement of SRP, in which we shape employees’ engagement in SRP by offering designed extrinsic and intrinsic factors. Firstly, given the significant importance of leadership style and job characteristics in workplace context (Babakus et al., 2003; Crant, 2000; Grant & Ashford, 2008; Guchait et al., 2014; Khorakian et al., 2021; Rahimizhian & Irani, 2021; Yilmaz & Konaklioglu, 2022), we proposed both selected inclusive leadership and job autonomy to be the extrinsic factors to develop potential resource ecology.
Secondly, we further explored the moderating role of surface acting as the intrinsic factor within the designed resource ecology (Hobfoll, 2011, 2012). In fact, emotional expression matters where with a strong focus on daily face-to-face service delivery in hospitality industry (del Río-Lanza, Vázquez-Casielles & Díaz-Martín, 2009; Luo et al., 2019; Wang & Xie, 2020). According to emotional labor theory (Lee & Madera, 2019), in order to fit for expectations of their work environment, individuals may apply emotional labor strategies - surface acting (individuals must hide or fake felt emotions) and deep acting (experience the desired emotions) - to manage emotion expressions and regulate how they feel (Brotheridge & Lee, 2002; Hochschild, 1983). Prior theoretical and empirical evidence shows that emotional labor strategies are closely linked to divergent outcomes in workplace, particularly regarding to employees’ psychological well-being (Grandey, 2003; Grandey & Sayre, 2019; Judge, Woolf & Hurst, 2009; Lee & Madera, 2019). It’s reasonable to propose that whether both inclusive leadership and job autonomy could successfully trigger great self-efficacy and afterward enhance engagement of SRP depends on how employees regulate their emotion expression. Additionally, some empirical evidences show that compared with deep acting, the impacts of surface acting are more consistent (Brotheridge & Lee, 2003; Goodwin, Groth & Frenkel, 2011; Judge et al., 2009; Luo et al., 2019; Totterdell & Holman, 2003). Taken these together, we propose surface acting plays the role to moderate the relationships mentioned previously in this study.
Overall, the current study aims to fill those aforementioned gaps and contribute to the hospitality literature in several ways. Firstly, as mentioned previously, studies on engagement of SRP have long focused on singular antecedents (extrinsic or intrinsic). The current study fills this gap and provide a comprehensive perspective by empirically integrating both extrinsic (inclusive leadership and job autonomy) and intrinsic (surface acting) factors together to explore SRP engagement in hospitality environment. Secondly, different from recent research based on traditional COR theory, we highlight a new perspective of resource ecology in literature of the COR theory (Hobfoll, 2011), through proposing a potentially practical resource ecology that may promote employees’ engagement in SRP, which may shed light to the future research to take an active role to offer necessary resources in workplace to achieve the goal. Thirdly, as our investigation proposing self-efficacy as the mediating mechanism to explore SRP engagement, we involve emotional labors (surface acting) as the boundary condition, which we consider to add to the literature on self-efficacy and its boundary conditions.
2. Theoretical background and hypothesis
Resource ecology perspective (Hobfoll, 2011, 2012) of COR theory shed light to develop the conceptual framework in this study. Hobfoll (2011) addressed that organizations can strategically arrange the salient factors for expected outcomes by building resource ecology in workplace, where employees gain psychological resources (Colbert, 2004; Hobfoll, 2011, 2012; Hobfoll, Halbesleben, Neveu & Westman, 2018). In line with this logic, we answered the call by proposing a potential resource ecology for engagement of SRP, which involve the salience of designed extrinsic (leadership and job characteristic) and intrinsic (individual differences) factors. To better understand the proposed hypothesis framework, we developed a conceptual model.
2.1. The relationship between self-efficacy and service recovery performance
Service recovery performance, as mentioned previously, refers to employees’ ability and actions taken after service failures to meet what customers expect for (Choi & La, 2013; Miller, Craighead & Karwan, 2000; Van Vaerenbergh & Orsingher, 2016). Given the highlights of service on inseparability of consumption and production and coproduction, service failures to satisfy customers and meet their expectations are inevitable, which makes SRP very important (Babakus et al., 2003; Luo et al., 2019; Yoo, Shin & Yang, 2006).
According to social cognitive theory (Bandura, 2010), individuals who are with great self-efficacy may strongly believe that they are capable to complete tasks, conduct better performance, as well as be influential on what they are experiencing (Bandura, 2013). Large number of research has been conducted to identified outcomes of self-efficacy, including better work-related performance, higher levels of job satisfaction, and lower motivation for leaving the position (Federici & Skaalvik, 2012; Judge, Jackson, Shaw, Scott & Rich, 2007). Therefore, we expected employees with great self-efficacy are likely to believe they are capable to handle service failure, as well as recover customers’ negative feeling. Thus, we propose the following hypothesis:
H1: Self-efficacy is positively related to engagement of SRP.
2.2. The mediating role of self-efficacy between antecedents and SRP
As mentioned above, both extrinsic and intrinsic factors are recognized as antecedents of SRP. In this section, we propose both inclusive leadership and job autonomy working as extrinsic factors to promote engagement of SRP via self-efficacy.
2.2.1 Inclusive leadership, Job autonomy and self-efficacy
Inclusive leadership captures leader behaviors that ensure members perceive that they both belong and are valued for their uniqueness, which may enhance the employees’ perceptions of inclusion (Randel et al., 2018). A handful of studies have demonstrated that inclusive leadership positively related to multiple positive outcomes in workplace (Fang, Chen, Wang & Chen, 2019; Jia, Jiao & Han, 2021; Randel et al., 2018). We propose in this study that inclusive leadership may trigger employees’ development of self-efficacy in workplace.
According to Wang and colleagues (2022), self-efficacy is suggested as a salient self-belief that may likely be motivated by contextual cues in workplace. Social persuasion is regarded as one of the four potential sources of self-efficacy (the other three sources are mastery experience, vicarious modelling, and emotional state; Wang et al., 2022), which is regarded as environmental force that enhance ones’ motivation to spent more effort. Specifically, inclusive leader behaviors provide cues to employees and likely make them perceive being valued, and available with supports and resources. Thus, employees more likely to develop strong self-belief about their capability and influence, and be motivated to work harder, spend more efforts, as well as not afraid of difficulties (Bandura, 1977, 1997). In fact, this line of reasoning is consistent with research indicating the perception of inclusive leadership may enhance employees’ psychological empowerment (Randel et al., 2018), as well as perceptions of competence and control (Boudrias, Morin & LaJoie, 2014). Therefore, we propose inclusive leadership is positively related to employee self-efficacy.
H2a: Inclusive leadership is positively related to employee self-efficacy.
Meanwhile, we also expect job autonomy contributes to the development of self-efficacy. Job autonomy refers to the extent to which a job offers employee latitude, independence, and freedom to arrange work schedule, select related skills and techniques, as well as make job-related decisions in workplace (Dysvik & Kuvaas, 2011; Humphrey, Nahrgang & Morgeson, 2007). Job autonomy, considered as one of the most important job characteristics, has been linked to multiple positive work-related outcomes because its effectively offer a safe and supportive context (Parker, Axtell & Turner, 2001). In line with logic between social persuasion and self-efficacy mentioned above, and given the crucial importance of job autonomy in workplace context, we argue that job autonomy may trigger the development of employee self-efficacy. Therefore, we propose that both inclusive leadership and job autonomy may enhance the development of self-efficacy, and we propose the following:
H2b: Job autonomy is positively related to employee self-efficacy.
2.2.2. The mediating role of self-efficacy
In current study we further propose that self-efficacy mediate the relationships between 1) inclusive leadership and engagement of SRP; 2) job autonomy and engagement of SRP. Indeed, the relationship between inclusive leadership and engagement of SRP via self-efficacy has been indicated by scholars. As mentioned previously, when surrounded by inclusive and supportive workplace via inclusive leaders’ availability and resources to make job-related decisions, employees are more likely to perceive cues of being valued, available with supports and resources (Randels et al., 2018). Further, employees are more likely to feel safe and confidence to complete their performance and have more influence on ones’ own job performance, and then are more motivated to engage in SRP. Given the relationship discussed above, we proposed that inclusive leadership may trigger the development of self-efficacy, and then enhance the engagement of SRP.
H3a: The positive relationship between inclusive leadership and SRP is mediated by employee self-efficacy.
In addition, research on both job autonomy and SRP has shown the potential positive relationship between job autonomy and engagement in SRP via self-efficacy (Fiernaningsih, Herijanto & Maskur, 2021; Wang & Netemeyer, 2002; Zakeri & Shahtalebi, 2014). Specifically, job autonomy may offer more opportunities for employees to make decisions and take charge, moreover provide employees a sense of confidence to complete their performance and influence ones’ own job-related outcomes (Fiernaningsih et al., 2021). Further, we expected that when strong belief about ones’ capability and potential influence has been made, employees may feel psychological motivated to engage in more efforts to meet their role expectations, perform better, and persevere longer when face difficulties (Bandura, 1977, 1997). Given the discussed relationship between self-efficacy and engagement of SRP, we also proposed that job autonomy may trigger the development of self-efficacy, and then enhance the engagement of SRP. Thus, we proposed the following:
H3b: The positive relationship between job autonomy and SRP is mediated by employee self-efficacy.
2.3. Moderating role of surface acting in the relationship between antecedents and self-efficacy
2.3.1. Inclusive leadership, self-efficacy and surface acting
Both extrinsic and intrinsic factors coexist and play very important roles in workplace (Gong et al., 2017). Research suggests that more studies should place a focus on the interaction of both extrinsic and intrinsic factors rather than singular focus on effect of just one (George, 2007; Gong et al., 2017). Surface acting is regarded as a type of strategical performance which contribute to develop an expected image in workplace by regulating employees’ behavioural display (Hochschild, 1983). Whereas the importance of service interaction between customers and frontline employees, growing research interest in the impact and influence of surface acting concerns about hospitality context (Lee & Madera, 2019; Lennard, Scott & Johnson, 2019; Mo & Shi, 2017). Thus, we expected surface acting may work the role of moderator between inclusive leadership and self-efficacy. According to the COR theory, employee engaging in surface acting may experience the lack of psychological resources, and great uncertainty about their thoughts and feelings (Brotheridge & Lee, 2003; Halbesleben, Neveu, Paustian-Underdahl & Westman, 2014; Ozcelik, 2013). Research has suggested that employees experiencing surface acting are likely to become emotionally burn out and psychological loss of well-being, which in turn influence behavioural outcomes and individual performance (Lennard et al., 2019; Thoresen, Kaplan, Barsky, Warren & de Chermont, 2003).
As previously argues, employees are likely to trigger the development of self-efficacy with an inclusive leader who provides them a sense of being supported, available of resources, and are likely to experience greater confident and control in workplace. However, we argued that this relationship might be regulated by the employees’ engagement of surface acting. Indeed, when experiencing greater engagement in surface acting, compared with when lower engagement, employees are more likely to feel anxious, stress and uncertainty since they are facing the loss of psychological resources (Geng, Liu, Liu & Feng, 2014; Mo & Shi, 2017). Consequently, they are less sensitive to the cues and signals from inclusive leader behaviors. Thus, these employees may be less likely to trigger further development of self-efficacy since that they are motivated to reduce cognitive effort and volition (Ozcelik, 2013). This is in line with previous empirical research, which shown the negative effects of surface acting on outcomes in workplace, including psychological and behavior outcomes (Mo & Shi, 2017). Thus, we suggest that when employee apply surface acting strategy in workplace, inclusive leader behaviors may be less likely to promote the development of self-efficacy, and we propose the following:
H4a: Surface acting moderates the relationship between inclusive leadership and self-efficacy such that the positive relationship is weaker when the surface acting is higher.
2.3.2. Job autonomy, self-efficacy and surface acting
Likewise, we also expected that surface acting moderates the relationship between job autonomy and self-efficacy. To repeat, by providing greater freedom and latitude to employees in workplace, job autonomy provides employees opportunities to make job-related decisions and trigger the sense of confidence and control (Hackman & Oldham, 1975). According to trait activation theory, how employees perceive and act on the opportunities of decision-making provided by autonomous job, significantly depends on their personal traits, such as personality, and self-regulatory mechanism (Ng, Ang & Chan, 2008). In consistent with this, we proposed that when employees with greater surface acting, employees may experience loss of psychological resources, as well as the increase of stress and uncertainty, which simultaneously offer less possibilities for employees sense the opportunities available and develop strong confidence in their jobs and control on job-related performance. Compared to this, when lower engagement in surface acting, employees are likely to sense these opportunities and trigger the development mechanism of self-efficacy. Accordingly, we proposed the following:
H4b: Surface acting moderates the relationship between job autonomy and self-efficacy such that the positive relationship is weaker when the surface acting is higher. Figure 1
3. Method
3.1. Participants and procedure
The data for this study was collected from eight privately owned hotels in different regions of China, including Shenzhen, Guangzhou, Hangzhou, Wuhan, and Xi’an. Authors use personal connections to get in touch with, and receive great supports from, selected hotel human resources (HR) managers. 83 supervisors and 350 frontline employees were invited to complete the questionnaires for this study. Due to the impact of COVID-19 pandemic, the questionnaire was sent online. In order to reduce participants’ concerns about privacy leakage and ensure the accuracy of questionnaire data, researchers promised to keep confidential for all the information, and the data obtained would only be used for academic research. Meanwhile, in order to enhance the participants’ cooperation and improve the response rate of the obtained data, researchers provided a WeChat red packet of RMB 20 (about USD 2.76) for each respondent who completes the questionnaires. Additionally, to facilitate the matching of all employee responses, we asked the interviewed subordinates to fill in the last four digits of their mobile phone number and the initials of given names. Likewise, for the purpose of matching the supervisor questionnaire with the employee questionnaire, researchers also asked the respondent supervisors to fill in the last four digits of their direct subordinates' mobile phones and the initials of given names.
The data were collected in three phases over three months from November 2021 to January 2022 to reduce the generation of common methods variance (CMV). In the first phase (November 2021), researchers completed the collection of demographic information from 350 frontline employees and their ratings for inclusive leadership and job autonomy. The second phase got started one month later (December 2021) for them to complete the evaluation of self-efficacy and surface acting. The ratings from their direct supervisors on service recovery performance was finished during the third phase which got started one month later (January 2022). At last, researchers got a sample of 305 supervisor-subordinate dyads which representing an 87.1% response rate. Since there are 34 measurable items for this model, we collected 350 questionnaires and 304 valid questionnaires were selected, the ratio approach 9:1 that sample size is appropriate for data analysis (Hair et al., 2010).
As shown in Table 1, among the participants who participated in the final sample survey, 61.6% were female frontline employees, 38.4% were male frontline employees, 54.8% were female supervisors and 45.2% were male supervisors. In terms of education, 90.4% of the frontline employees and 90.1% of supervisors had a college education or above. In terms of age, frontline employees are mainly concentrated in two age groups: under 25 years old (45.2%) and 26-35 years old (32.1%), and supervisors are mainly concentrated in the 26-35 (41.4%) and 36-45 (31.5%) age groups. In terms of years of cooperation between supervisors and subordinates, two groups dominate: more than three months and less than one year (39.3%), and between one and three years (23.6%).
Table 1 Demographic Characteristics of the Sample
| Variable | Subordinate（%） | Supervisor(%) | |
| Gender | Male | 38.4 | 45.2 |
| Female | 61.6 | 54.8 | |
| Age | ＜25 | 45.2 | 15 |
| 26-35 | 32.1 | 41.4 | |
| 36-45 | 11.1 | 31.5 | |
| ＞45 | 11.6 | 12.1 | |
| Education | Junior middle school or below | 1.6 | 0 |
| High school | 8.0 | 9.9 | |
| College/Undergraduate | 79.3 | 86.6 | |
| Postgraduate or above | 11.1 | 3.5 | |
| Years of cooperation between superior and subordinate | ＜3 months | 10.7 | 10.7 |
| 3 months-1 year | 39.3 | 39.3 | |
| 1 year-3 years | 23.6 | 23.6 | |
| 3 years-5 years | 9.6 | 9.6 | |
| ＞5 years | 16.9 | 16.9 | |
3.2 Measures
This questionnaire is composed of seven parts including matching information, demographic information, inclusive leadership, job autonomy, self-efficacy, service recovery performance and surface acting. Given that all the scales for variables in current study were developed in English, we use a translation/back-translation procedure (Brislin, 1980) to translate them into Chinese. The rest of the questionnaire, except demographic statistics information, was measured using a five-point Likert scale (anchored by 1 = strongly disagree to 5 = strongly agree).
As shown in Table 2, This research adopted the nine-item scale of Carmeil, ReiterPalmon and Ziv (2010) to measure inclusive leadership, three-item scale of Hackman and Oldham (1980) to measure job autonomy, ten-item scale of Schwarzer, Bäßler, Kwiatek, Schröder and Zhang (1997) to measure self-efficacy, five-item scale of Boshoff and Allen (2000) to measure service recovery performance and seven-item scale of Diefendorff, Croyle and Gosserand (2005) to measure surface acting.
Table 2 Measurement scales
| Construct | Item | Reference |
| Inclusive leadership (subordinate rating) | IL1.The manager is open to hearing new ideas. | Carmeil et al., (2010) |
| IL2.The manager is attentive to new opportunities to improve work processes. | ||
| IL3.The manager is open to discuss the desired goals and new ways to achieve them. | ||
| IL4.The manager is available for consultation on problems. | ||
| IL5.The manager is an ongoing 'presence' in this team-someone who is readily available. | ||
| IL6.The manager is available for professional questions I would like to consult with him/her. | ||
| IL7.The manager is ready to listen to my requests. | ||
| IL8.The manager encourages me to access him/her on emerging issues. | ||
| IL9.The manager is accessible for discussing emerging problems. | ||
| Job autonomy (subordinate rating) | JA1.I have significant autonomy in deter mining how I do my job. | Hackman & Oldham (1980) |
| JA2.I can decide on my own how to go about doing my work. | ||
| JA3.I have considerable opportunity for independence and freedom in how I do my job. | ||
| Self-efficacy (subordinate rating) | SE1.I can always manage to solve difficult problems if I try hard enough. | Schwarzer et al. (1997) |
| SE2.If someone opposes me. I can find means and ways to get what I want. | ||
| SE3.It is easy for me to stick to my aims and accomplish my goals. | ||
| SE4.I am confident that I could deal efficiently with unexpected events. | ||
| SE5.Thanks to my resourcefulness, I know how to handle unforeseen situations. | ||
| SE6.I can solve most problems if I invest the necessary effort. | ||
| SE7.I can remain calm when facing difficulties because I can rely on my coping abilities. | ||
| SE8.When I am confronted with a problem, I can usually find several solutions. | ||
| SE9.If I am in a bind, 1 can usually think of something to do. | ||
| SE10.No matter what comes my way, I’m usually able to handle it. | ||
| Service recovery performance (supervisor rating) | SRP1.Considering all the things I do, I handle dissatisfied customers quite well. | Boshoff & Allen (2000) |
| SRP2.I don’t mind dealing with complaining customers. | ||
| SRP3.No customer I deal with leaves with problems unresolved. | ||
| SRP4.Satisfying complaining customers is a great thrill to me. | ||
| SRP5.Complaining customers I have dealt with in the past are among today’s most loyal customer. | ||
| Surface acting (subordiate ratingn) | SA1.I put on an act in order to deal with customers in an appropriate way. | Diefendorff et al. (2005) |
| SA2.I fake a good mood when interacting with customers. | ||
| SA3.I put on a “show” or “performance” when interacting with customers. | ||
| SA4.I just pretend to have the emotions I need to display for my job. | ||
| SA5.I put on a “mask” in order to display the emotions I need for the job. | ||
| SA6.I show feelings to customers that are different from what I feel inside. | ||
| SA7.I fake the emotions I show when dealing with customers. |
4. Result
In this study, Smart PLS 4 and SPSS 25 were used for data analysis. The demographic information of the participants was descriptively analysed by SPSS. Smart PLS was used to evaluate the reliability and validity of the measurement model, as well as the hypothesis testing. Partial-least-squares (PLS) is a soft modelling method that is well-suited for optimizing predictions (Latan & Noonan, 2018). The study's exploratory nature, complex model, and non-normally distributed data required the use of the Partial-least-squares structural equation modelling (PLS-SEM) technique (Akter, D'Ambra & Ray, 2011; Hair, Hult, Ringle & Sarstedt, 2017). Another reason is that PLS can handle formative and reflective structures compatibly (Chin, 1998). In this research, the five constructs (inclusive leadership, job autonomy, self-efficacy, services recovery performance, surface acting)both fit the criteria of reflective ones, which are to“being caused by underlying construct”, instead of “define or cause the construct” (Hulland, 1999, p. 201).
4.1 Reliability and validity analysis
As shown in Table 3, the Cronbach's alpha (α) and composite reliability (CR) values of the model structure are both greater than 0.80, indicating that the measurement model has good reliability. Convergent validity can be measured with the average variance extraction (AVE) of the construct. As shown in Table 4, all items exceed the minimum threshold of 0.5 (Bagozzi & Yi, 2012; Henseler, Ringle & Sinkovics, 2009). The factor loadings (Appendix )for all indicators were above 0.7 (p < 0.001). Additionally, as shown in Table 3, all constructs exhibited acceptable discriminant validity (Fornell & Larcker, 1981). Additionally, in Table 3, all Heterotrait-Monotrait (HTMT) correlations should have values less than 0.90 (Henseler, Ringle & Sarstedt, 2015). If the corresponding 5,000 bootstrap confidence intervals (95%, two-tailed) of the HTMT criterion do not include the value 1, then all HTMT values are significantly lower than 0.9, indicating sufficient discriminant validity (Hair et al., 2017). Based on these criteria, it can be concluded that the discriminant validity of the constructs is satisfactory.
Table 3 Measurement properties
| Measurement | Mean | S.D. | Cronbach's alpha | CR | AVE | Heterotrait-Monotrait ratio IL JA SA SE | |||
| Inclusive leadership(IL) | 4.293 | 0.909 | 0.967 | 0.965 | 0.759 | ||||
| Job autonomy (JA) | 3.497 | 1.051 | 0.890 | 0.891 | 0.732 | 0.135 | |||
| Surface acting(SA) | 3.172 | 1.217 | 0.952 | 0.952 | 0.739 | 0.081 | 0.300 | ||
| Self-efficacy(SE) | 3.813 | 0.809 | 0.939 | 0.939 | 0.609 | 0.230 | 0.573 | 0.344 | |
| Service recovery performance(SRP) | 4.026 | 0.763 | 0.892 | 0.870 | 0.604 | 0.147 | 0.117 | 0.046 | 0.163 |
Note(s): S.D.= Standard Deviation; CR = Composite Reliability; AVE = Average Variance Extracted; The confidence intervals (bias corrected) of HTMT criterion: 5,000 bootstrapping subsamples, 95% significance level, two-tailed.
Table 4 Descriptive data, inter-construct correlations and the square root of AVE
| Inclusive leadership | Job autonomy | Surface acting | Self-efficacy | Service recovery performance | |
| Inclusive leadership | 1.000 | ||||
| Job autonomy | 0.140 | 1.000 | |||
| Surface acting | -0.055 | 0.299 | 1.000 | ||
| Self-efficacy | 0.235 | 0.573 | 0.345 | 1.000 | |
| Service recovery performance | 0.139 | 0.122 | 0.030 | 0.176 | 1.000 |
| √AVE | 0.871 | 0.855 | 0.860 | 0.781 | 0.777 |
4.2. Structural model analysis and hypothesis testing
As shown in Table 5 and Figure 2, the researchers obtained the test results of the structural model after sampling 5000 times from the original sample for calculation using Bootstrapping. The results showed that all the other hypotheses were valid except H4b.
Hypothesis test results showed the positive relationship between self-efficacy and services recovery performance (β=0.168, p＜0.005, t=2.926). Thus, H1 supported. Hypothesis test results confirm significant positive effects of inclusive leadership and job autonomy on self-efficacy (β=0.186, p＜0.001, t=3.939 ; β=0.414, p＜0.001, t=8.023). Thus, H2a and H2b were validated. At the same time, the mediating role of self-efficacy between inclusive leadership and service recovery performance was also confirmed (β=0.031, p＜0.05, t=5.151), and similarly, self-efficacy was confirmed to have a mediating role between job autonomy and service recovery performance (β=0.070, p＜0.05, t=2.926). In this research, we presented H3a and tested a model that sought to better understand the effects of self-efficacy, which mediates between inclusive leadership and services recovery performance. We found (β =0.031, p < 0.05, t=2.206), thus the H3a supported because there the relationship was significant with a p-value <0.05. Similarly, H3b predicted the mediating role of self-efficacy between job autonomy and services recovery performance. The significant results (β = 0.070, p < 0.05, t=2.644) was found, thus H3b is accepted. In addition, according to Table 5 and Figure 3, it is not difficult to find that surface acting has a significant negative moderating effect on the relationship between inclusive leadership and self-efficacy (β=-0.208, p＜0.001, t=4.941). The relationship between inclusive leadership and self-efficacy is weakened when employees have higher levels of surface acting, and vice versa. Therefore, H4a is confirmed. Similarly, by observing Table 5 and Figure 4, it can be concluded that surface acting has no significant moderating effect on the relationship between job autonomy and self-efficacy (β=-0.010, p＞0.05, t=0.188), so H4b is not supported.
The model fit of PLS SEM can be evaluated by goodness of fit index (GFI) (Tenenhaus, Vinzi, Chatelin & Lauro, 2005), Variance Infation Factor (VIF), R2 and Q2 (Hair, Risher, Sarstedt & Ringle, 2019). The goodness of fit index is the geometric mean of the product of the average AVE value of the facet and the average R² (GFI= √AVE×R²). As shown in Table 6, the goodness of fit value of the model in this study was 0.378. According to the Cudeck and Browne (1992) , GFIsmall= 0.1, GFImedium= 0.25, GFIlarge= 0.36, thus, the model has a good fitting effect. In addition, according to the standard of Hair et al. (2019), VIF index should be less than 3.0 to rule out the possibility of multi-collinearity among constructs. In this study, the VIF values of inclusive leadership, job autonomy and surface acting were 1.029, 1.176 and 1.123, respectively, both of which were less than 3.0.
The explanatory power of endogenous latent variables is usually estimated by R2 (Henseler et al., 2009), which is suggested to be acceptable by Hair and colleagues (Hair et al., 2019). As shown in Table 7, the value of R2 of self-efficacy and service recovery performance are 0.386 and 0.028, respectively. Although according to the standard of Hair et al. ’s (2019), R2 estimates were not moderate (＞0.5) or substantial (＞0.75), it still meets the Cohen’s (2013) criteria (＞0.02 for weak and ＞0.13 for moderate).
The predictive validity of exogenous latent variables is accustomed to be estimated by Q2 (Geisser, 1974; Stone, 1974), which is aimed to test the extent to which each endogenous latent structure can be predicted by endogenous latent structures in the proposed model (Hair, Ringle & Sarstedt, 2011). The value of Q2 greater than zero indicates well-reconstructed observed value and predictive relevance the exogenous constructs have with the considered endogenous construct. As shown in Table 7, all values of Q2 are significantly above zero, which indicates that the model has predictive relevance.
Table 5 Results of hypothesis testing
| Path coefficients | P values | Hypothesis | |
| H1(+) Self-efficacy --> Service recovery performance | 0.168 | 0.003** | Supported |
| H2a(+) Inclusive leadership --> Self-efficacy | 0.186 | 0.000*** | Supported |
| H2b(+) Job autonomy --> Self-efficacy | 0.414 | 0.000*** | Supported |
| H3a(+) Inclusive leadership --> Self-efficacy--> Service recovery performance | 0.031 | 0.027* | Supported |
| H3b(+) Job autonomy --> Self-efficacy--> Service recovery performance | 0.070 | 0.008* | Supported |
| H4a(-) Surface acting x Inclusive leadership --> Self-efficacy | -0.208 | 0.000*** | Supported |
| H4b(-) Surface acting x Job autonomy --> Self-efficacy | -0.010 | 0.851 | Not supported |
***p <0.001; **p <0.005; *p <0.05
Figure 2 Results of PLS-SEM analysis. Notes: SA=surfacing acting; IL= inclusive leadership; JA= job autonomy; SE= self-efficacy; SRP= service recovery performance.
Table 6 Goodness of fit index
| Constructs | AVE | R2 |
| Inclusive leadership | 0.759 | |
| Job autonomy | 0.732 | |
| Surface acting | 0.739 | |
| Self-efficacy | 0.609 | 0.245 |
| Service recovery performance | 0.604 | 0.015 |
| Average scores | 0.689 | 0.207 |
| AVE×R² | 0.143 | |
| (GFI= √ AVE×R²) | 0.378 | |
Table 7 The results of R2 and Q2.
| Constructs | R2 | Q2 |
| Self-efficacy | 0.386 | 0.245 |
| Service recovery performance | 0.028 | 0.015 |
In addition, it is worth noting the moderating role of surface acting between inclusive leadership and self-efficacy as well as between job autonomy and self-efficacy. H4a states that surface acting buffer the positive relationship between inclusive leadership and self-efficacy, such that the relationship is weaker among dyads with higher surface acting, compared to those with weaker surface acting. In Figure 3, results from simple slope analysis showed that, when surface acting was low, the relationship between inclusive leadership and self-efficacy was positive; when surface acting was high the simple slope much higher, surface acting has the reinforcement interaction. Therefore, H4a was supported.
Figure 3 Interaction of inclusive leadership and surface acting on self-efficacy. Note: SE= self-efficacy
H4b states that surface acting buffer the positive relationship between job autonomy and self-efficacy, such that the relationship is weaker among dyads with higher surface acting, compared to those with weaker surface acting. However, in Figure 4, results from simple slope analysis showed that, whatever surface acting was low or high, the simple slope have no change. Therefore, H4b was not supported.
5. Discussion and conclusion
This study focuses on service recovery performance, which is considered as a crucial factor for business success (Guchait et al., 2019; Karatepe & Karadas, 2012). Both extrinsic (leadership style and job characteristics) and intrinsic (surface acting) factors are interactively involved in the tested model, as well as self-efficacy working as the mediator. Firstly, this study confirmed both inclusive leadership and job autonomy interactively contribute to engagement of SRP via self-efficacy in hospitality industry. Secondly, this study is the one of the first study to introduce the resource ecology perspective of COR theory (Hobfoll, 2011) to the literature of SRP, also the first one to involve surface acting as the boundary condition to understand the development of self-efficacy, which add to both self-efficacy and emotional labor literatures. Lastly, the study also adds to the self-efficacy literature by acting as potential pathway to bridge the interaction between extrinsic and intrinsic motivators and SRP engagement, which extends prior research (Wang et al., 2022)
5.1. Theoretical implications
As anticipated, we found that self-efficacy mediated the relationships between antecedents (inclusive leadership and job autonomy) and SRP. Meanwhile, we found surface acting negatively moderates the relationship between inclusive leadership and self-efficacy. Different from our hypotheses, we didn’t find support for the moderating role of surface acting between job autonomy and self-efficacy.
First, while most past studies focus on singular factors, we add to the SRP literatures by taking an interactive perspective to examine a theoretically plausible pathway for the triggering mechanism for engagement of SRP in workplace. Specifically, results in this study shows the underlying mechanisms of engagement of SRP by demonstrating the interactive effects of both extrinsic (inclusive leadership and job autonomy, working as antecedents) and intrinsic (surface acting, working as moderator) factors contributing to engagement of SRP via self-efficacy. To the best of our knowledge, the current research is the first to comprehensively explain this psychological process. This finding sheds light to explain the inconsistent findings in this is this area (Van Vaerenbergh & Orsingher, 2016). Thus, we consider this contribution answer Guchait’s call (2017), which argues that more research should focus on specific and comprehensive situation rather than focusing on generic antecedents.
Secondly, our findings are also consistent with research on the resource ecology in literature of the COR theory (Hobfoll, 2011). Given organizations relying more on employees to initiatively gain resources from context, our results indicate that firms actively and strategically design resource ecology, as well as offer necessary resources available are equally important for motivating engagement of SRP. This result further reinforces the notion of the initiative role of organizations to strategically select, develop, combine, and deploy current resources by building resource ecology (Colbert, 2004; Hobfoll, 2011, 2012; Hobfoll et al., 2018). In current study we propose that organizations can strategically design and arrange the context where employees gain resources. That is, by design, organization may make some crucial contextual factors salient to enrich employees resource gains, which in turn contributes to positive psychological state or behavioural outcomes (Hobfoll, 2011; Hobfoll et al., 2018). Specifically, we introduce a plausible strategy in which organizations offer both designed leadership style (inclusive leadership) and job characteristics (high job autonomy) to shape greater employees’ self-efficacy, and then enhance the engagement of SRP in hospitality sector.
Thirdly, by considering the moderator effect of surface acting on the prediction of self-efficacy, the current study also adds to the literature on self-efficacy and its boundary conditions. Although previous research has indicated the importance of individual characteristics (He, Zhou, Zhao, Jiang & Wu, 2020; Judge et al., 2007; Şahin, Karadağ & Tuncer, 2019), to the best of our knowledge prior research has not yet involve emotional labor strategy - specifically surface acting - as the potential moderator on the relationship between antecedents and self-efficacy in hospitality industry. According to social information processing theory (Salancik & Pfeffer, 1978), self-identity is closely related to individual differences as these differences, such as personalities and trait foci, which may guide and regulate how one thinks and behave when they get cues from social environment（Dinh & Lord, 2012). In this study, we propose that when employees receive cues from inclusive leaders and jobs full of autonomy, their perceptions about self-efficacy may be regulated by the selecting of emotional labor strategy. Thus, these results demonstrated the importance of emotional concepts to trigger the mechanism of psychological social-self and self-identity under the context of inclusive leadership and job autonomy in workplace（Shah, Ou, Attiq, Umer & Wong, 2022; Shore & Chung, 2022).
Fourthly, we consider our research also add to inclusive leadership research by showing that, as a relation-oriented leadership, developing high-quality relationship with subordinates is not the only way for inclusive leaders to exert impact, they may also inspire how they perceive themselves. In fact, scholars have long-held notion about inclusive leaders consistently care about relationship quality with subordinates (Randel et al., 2018). Our findings suggest that inclusive leadership may also exert profound impact to influence subordinates’ self-concepts.
Lastly, we found support for the moderating role of surface acting between inclusive leadership and self-efficacy, but null finding for the moderating role between job autonomy and self-efficacy. Nevertheless, we consider our null finding echo Ozcelik’s and Mo’s researches (Mo & Shi, 2017; Ozcelik, 2013) that have found surface acting is typically interactive relationship-oriented, and individuals with surface acting are likely to maintain a particular interpersonal acceptance and belonging. Another potential explanation for our null findings is that there might be unidentified boundary condition within this relationship (Goodwin et al., 2011).
5.2. Practical implications
The current study attempted to take an interactive perspective which involve contextual and individual factors together to examine their interaction effects on employees’ engagement in SRP via self-efficacy. Results demonstrated that both inclusive leadership and job autonomy enhance SRP via self-efficacy, and employees’ surface acting regulates the influences on inclusive leadership and job autonomy on self-efficacy. These findings provide several implications on practical human resource management in hospitality industry.
First, our findings suggested that, compared to focus only on recruiting employees who is willing to proactively perform high quality service recovery behaviors, strategical arrangement and deployment of contextual factors is now becoming more important to motivate the engagement of SRP. For instance, leadership style, job characteristics design, as well as other human resources management strategies should all be taken into consideration to make a systemic resource ecology. Such resource ecology may help employees to enrich their psychological resources to enhance greater SRP.
Secondly, managers should understand that, in order to enhance employees’ engagement in SRP, both strategically organized workplace context and initiative role of individual differences matter. That is, only considering about selecting, developing, combining, and deploying different contextual factors, while ignoring the effect of initiative role of individual differences, or only considering about the latter, while ignoring effects of the former, organizations may not achieve final goals. Specifically, in hospitality sector, since inclusive leader behaviors and jobs designed with great autonomy are found to be able to motivate employees to engage more service recovery behaviors, most managers only pay all attention to strategically arrange these contextual factors. Rather, the current study implies individual differences may regulate the influence of contextual factors which make these arrangements to a risky decision. Our research highlights the interaction of both two different dimensions that, in order to enhance greater engagement of SRP, both inclusive leadership style and autonomous job design could be effective since they all motivate the development of employees’ self-efficacy. However, such arrangement may be accompanied with some trainings which may help employees reduce the possibilities to select and conduct surface acting since these acting labours may conserve emotional resources.
5.3. Limitations and directions for future study
First, the current study did not take deep acting, which is another typical emotion labor, into consideration, since the measure originally developed by Brotheridge and Lee (1998) is suggested to have limitations (Goodwin et al., 2011; Luo et al., 2019). Whereas deep acting has long been suggested to be a more beneficial emotional labor strategy to regulate emotions in workplace. Thus, future research may work on more effective scale development and this research may pave the way for the future studies of deep acting in literature of both SRP and emotional labours.
Secondly, this study only focuses on frontline employees in hospitality sectors since their daily jobs need them face customers and deal with service failure every day. However, service failures won’t just happen between frontline employees and customers, but also within organizations. Thus, future research could examine more on non-frontline employees’ relationship.
Thirdly, results and findings in current study are based on Chinese context. Given previous studies has suggested that self-efficacy is strongly associated to ‘social self’ (Anderson & Betz, 2001), different cultural background may affect how this ‘social-self’ related concept developed. Therefore, future studies should examine the model within different culture background to increase generalizability, or consider taking some cultural moderators into study.
In conclusion, the current study demonstrates the importance of considering both contextual and individual factors in promoting engagement in SRP in the hospitality industry. By introducing an interactive perspective, the study highlights the roles of both inclusive leadership and job autonomy in predicting employee self-efficacy, which in turn enhances engagement in SRP. The study also identifies surface acting as a negative regulator in these relationships. The findings suggest that organizations and managers can take specific actions to enhance employees' self-efficacy and engagement in SRP. Overall, the study makes important contributions to the SRP literature by providing insights into the complex mechanisms underlying engagement in SRP and identifying practical implications for organizations and managers.
Acknowledgement
The authors would like to express our great appreciation to Dr. Kelvin Zhang, for his valuable time in proofreading this manuscript.
