Employee engagement in CSR initiatives and customer-directed counterproductive work behavior (CWB): The mediating roles of organizational civility norms and job calling

This study aims to examine how service employees' perceptions of corporate social responsibility (CSR) affect their customer‐directed counterproductive work behavior (CWB) and the mediation of this link through their organizational civility norms and job calling. Working with a sample of 252 frontline employees in South Korea hotels, structural equation modeling is employed to test research hypotheses. The results of this study suggest that service employees' perceptions of CSR are negatively related to their customer‐directed CWB. Second, service employees' organizational civility norms mediated the negative relationship between service employees' perceptions of CSR and customer‐directed CWB. Third, service employees' job calling also mediated the negative relationship between their perceptions of CSR and customer‐directed CWB. Finally, the relationship between service employees' perceptions of CSR and customer‐directed CWB is sequentially and fully mediated by organizational civility norms and job calling. The theoretical and managerial implications of the results and limitations of the study are discussed, and future research directions are suggested.

Keywords: customer‐directed counterproductive work behavior (CWB); employees' CSR perceptions; job calling; organizational civility norms

Although most corporate social responsibility (CSR) research is rooted in the macro (i.e. organizational) level (Lee, [62]), there has been a growing interest in recent years in the micro level of CSR, which seeks to explain how CSR influences the attitudes and behaviors of employees (Aguinis & Glavas, [2]; Glavas, [42]; Kim & Kim, [58]; Rupp & Mallory, [89]; Temminck, Mearns, & Fruhen, [105]; Zhou, Luo, & Tang, [119]; Zhu, Yin, Liu, & Lai, [120]). Micro‐CSR is defined as 'the study of the effects and experiences of CSR on individuals as examined at the individual level of analysis' (Rupp & Mallory, [89], p. 216). Correspondingly, recent research studies within the organizational behavior (OB) and business ethics literature have examined the effects of employees' sense‐making of a firm's CSR activities (hereafter referred to as their CSR perceptions) on their attitudes, behaviors and performance (Aguinis & Glavas, [2]; De Roeck & Maon, [25]; Farooq, Payaud, Merunka, & Valette‐Florence, [37]; Glavas, [42]; Hillenbrand, Money, & Pavelin, [50]; Hur, Moon, & Ko, [55]; Kim & Kim, [58]; Temminck et al., [105]; Zhou et al., [119]; Zhu et al., [120]).

Micro‐CSR assumes that an employee's perceptions concerning the organization's activities have implications for employee outcomes such as work roles, feelings and attitudes, and job performance (Cropanzano, Byrne, Bobocel, & Rupp, [24]). Positive perceptions shaped by their firm's CSR activities tend to promote positive attitudinal and behavioral changes among employees (Rupp, Ganapathi, Aguilera, & Williams, [88]). For example, research in the area of CSR perceptions among employees has found a positive relationship with several employee outcomes, including affective organizational commitment (AOC), job satisfaction, organizational citizenship behavior (OCB), compassion, creativity and job performance (Aguilera, Rupp, Williams, & Ganapathi, [1]; De Roeck, Marique, Stinglhamber, & Swaen, [26]; Dhanesh, [27]; Hur et al., [55]; Lin, Lyau, Tsai, Chen, & Chiu, [67]; Moon, Hur, Ko, Kim, & Yoon, [71]; Rego, Leal, Cunha, Faria, & Pinho, [83]; Temminck et al., [105]; Zhu et al., [120]).

Despite the increasing body of micro‐CSR studies, researchers have predominantly focused the impact of CSR perceptions on positive employee outcomes (i.e. job satisfaction, affective commitment, creativity and OCB), and have rarely been concerned with negative employee outcomes (Sheel & Vohra, [93]). Past research reveals that negative work experiences have been found to be a critical factor that prevents organizations from building and maintaining a sound workplace and promoting employees' well‐being (Lim & Cortina, [65]; Lim, Cortina, & Magley, [66]), which together outweigh the benefits gained from positive work experiences (Baumeister, Bratslavsky, Finkenauer, & Vohs, [12]; Felps, Mitchell, & Byington, [38]). Furthermore, recent studies point out that micro‐CSR research must extend beyond the outcome variables that the literature has addressed to date in order to achieve a more comprehensive understanding of the effects of CSR perceptions on employee outcomes (Aguilera et al., [1]; Glavas, [42]). For instance, Gond, El Akremi, Swaen, and Babu ([46]) suggest that the micro‐CSR literature should include not only positive OB outcomes, but also negative and destructive outcomes, such as violence, deviance, sabotage, revenge or burnout. In so doing, the micro‐CSR research could tell whether CSR perceptions boost positive employee outcomes, and/or potentially prevent the development of negative employee outcomes (Gond et al., [46]).

Among the various negative employee outcomes that may be affected by CSR perceptions, employees' counterproductive work behavior (CWB) – referred to as employees' intentional behaviors that harm the organization or stakeholders in organizations, including employees, customers or clients (Spector & Fox, [97]) – has recently attracted considerable attention from scholars and practitioners due to the substantial costs incurred by employee misbehaviors (Tepper, [106]), and has not yet been considered an employee outcome in the micro‐CSR literature. In particular, little is known about how CSR perceptions influence customer‐directed CWB (e.g. employee rudeness, unresponsiveness and mistreatment of customers). In this paper, we contribute to the extant micro‐CSR literature by developing understanding of the effects of CSR perceptions among employees on customer‐directed CWB.

Much research in the ethics literature has demonstrated that how employees perceive the ethical climate, as shaped by an organization's dedicated efforts and resources, increases the positive job attitudes and beneficial conduct of these employees (e.g. Baker, Hunt, & Andrews, [10]; Valentine, Godkin, & Lucero, [109]) whilst also decreasing the negative emotions triggered by their job (Mulki, Jaramillo, & Locander, [74]) and the negative affective reaction that makes them become deviant (Appelbaum, Iaconi, & Matousek, [6]; Yang & Diefendorff, [118]). Since CSR constitutes any discretionary corporate activity aimed at positive social change beyond the narrow economic, technical and legal interests of the firm by focusing on the relationships with and responsibilities toward various stakeholder groups (Aguilera et al., [1]), CSR perceptions may develop an ethical context in an organization that reduces customer‐directed CWB.

Beyond examining the direct effect, our study will contribute to the existing literature by exploring the underlying mechanism through which CSR perceptions reduce employees' customer‐directed CWB. Glavas ([42]) suggests that there is a lack of meaningful exploration of the mediators that bridge CSR perceptions and employee outcomes. Thus, the primary purpose of our research is to show how the organization's CSR acts affect employees' sense‐making, which ultimately attenuates employees' customer‐directed CWB. In line with the self‐enhancement motivation underlying social identity theory (Ashforth, Harrison, & Corley, [8]; Ashforth & Mael, [9]), we suggest that employees' CSR perceptions are more likely to foster organizational civility norms among employees, referred to as 'the norms for mutual respect at work [comprising] behaviors that are fundamental to positively connecting with another, building relationships, and empathizing' (Pearson, Andersson, & Porath, [77], p. 125). Positive perceptions about a firm shaped by its CSR activities promote employees' pro‐social norms and behaviors due to their pride in their firm and intention to sustain its positive social reputation (Dutton, Roberts, & Bednar, [33]; Ellemers, De Gilder, & Haslam, [35]). Walsh et al. ([114]) found that the CSR perceptions among employees develop social norms for similar attitudes and socially acceptable behaviors through a positive sense‐making process, which does not tolerate deviant and uncivil behaviors. Thus, it is expected that organizational civility norms will perform a role as an important mediator in the effect of CSR perceptions on employees' customer‐directed CWB.

Another possible mediator on the link between CSR perceptions and employees' customer‐directed CWB may be job calling, defined as a work orientation to fulfill a meaningful passion for a career domain (Dobrow & Tosti‐Kharas, [28]). In this article, we rely on signaling theory to explain how CSR perceptions affect employees' customer‐directed CWB via job calling. Signaling theory suggests that CSR acts depict the organization in a positive light by reflecting its socially desired values (Carroll, [22]), which exert a considerable influence over employees' attitude and behaviors toward their organization (Moore, [72]). According to signaling theory, a particular signal can affect an individual's attitude and behavior toward the sender (Rynes, [91]). CSR practices signal the characteristics of a firm, leading employees to perceive their organization as being responsible, caring and compassionate toward society (Moore, [72]), and thus enabling employees to find greater meaningfulness and purpose at work (Glavas & Kelley, [44]). Positive signals triggered by the firm's CSR foster employees' fulfillment of their own value at work (Glavas & Piderit, [45]), which may develop job calling. Since employees who perceive their work as a calling are more likely to have a more positive attitude toward the organization (Cardador, Dane, & Pratt, [20]), employees' customer‐directed CWB will be diminished under circumstances in which a firm is engaged in CSR. Therefore, we suggest that job calling will perform a role as another crucial mediator in the association between CSR perceptions and employees' customer‐directed CWB.

Finally, based on the combined rational of social identity theory and signaling theory, our study will make an empirical contribution to the literature by investigating whether the serial multiple mediation effects on the relationship between CSR perceptions and employees' customer‐directed CWB are sequentially mediated by organizational civility norms and job calling. In sum, the primary objective of our research is to offer an empirical framework for how CSR perceptions influence employees' customer‐directed CWB by incorporating the mediating variables of organizational civility norms and job calling.

The dominant focus in the micro‐CSR literature has been on positive employee outcomes (Gond et al., [46]). For example, a sizeable volume of micro‐CSR research has shown a positive association between CSR perceptions and several positive outcomes, such as organizational and work commitment (Dhanesh, [27]; Farooq et al., [37]; Hofman & Newman, [52]), job satisfaction (De Roeck et al., [26]; Dhanesh, [27]), organizational identification (Carmeli, Gilat, & Waldman, [21]; Kim, Lee, Lee, & Kim, [57]), OCB (Hansen, Dunford, Boss, Boss, & Angermeier, [48]; Rupp, Shao, Thornton, & Skarlicki, [90]), improved employee relations (Glavas & Piderit, [45]) and work engagement (Caligiuri, Mencin, & Jiang, [19]; Glavas & Piderit, [45]). Despite the growing body of research on the role of CSR perceptions in relation to positive employee outcomes, little is known about how CSR perceptions affect negative employee outcomes, such as employees' deviant attitudes and behaviors (Gond et al., [46]). For example, a few studies have found that CSR perceptions reduce employees' emotional exhaustion (Watkins et al., [115]) and prevent cynicism (Evans, Goodman, & Davis, [36]; Sheel & Vohra, [93]). Although the research of Viswesvaran, Deshpande, and Milman ([111]) has found a negative relationship between CSR perceptions and CWB, their sample is limited to college students, and there is still little understanding about how and why employees' CSR perceptions affect CWB. Thus, we attempt to fill this gap by examining the serial multiple mediation effects of organizational civility norms and job calling on the relationship between CSR perceptions and customer‐directed CWB.

Extant studies in the ethics literature have demonstrated that ethical climate in a firm as perceived by its employees can prevent their deviant workplace behaviors (Andreoli & Lefkowitz, [5]; Bulutlar & Öz, [18]; Treviño, Butterfield, & McCabe, [107]). A caring and supportive climate, which benefits others in an organization (i.e. a firm's engagement in CSR acts), encourages employees to make ethical decisions (Treviño et al., [107]). Research has shown that employees' sense‐making of the firm's actions heavily influences their attitudes and behaviors (Cropanzano et al., [24]). Since CSR activities reflect socially desired values in a firm (Carroll, [22]), employees whose firms address social issues through CSR are more likely to see their company as being responsible, caring and compassionate toward society, which in turn improves their attitudes and behaviors in the workplace (Moore, [72]).

In a similar vein, previous research has shown that perceptions of an ethical climate amongst employees decrease the negative emotions they experience on the job and mitigate negative affective responses (Mulki et al., [74]), in turn reducing deviant behaviors (Appelbaum et al., [6]; Yang & Diefendorff, [118]). For instance, organizational virtuousness develops positive emotions and a sense of happiness amongst employees (Rego, Ribeiro, & Cunha, [84]), leading to a decrease in workplace deviance (Lyons & Scott, [69]). Likewise, employees who perceive their firm positively for its CSR activities are likely to be more positive emotionally and possess a stronger identity with regard to their membership of that organization (Ellemers et al., [35]), which induces employees to refrain from CWB (Viswesvaran et al., [111]), those intentional behaviors of employees that harm the organization or stakeholders in organizations, such as taking excessively long breaks, loafing, sabotage, verbal abuse, theft, lying or engaging in interpersonal hostility (Spector & Fox, [97]). If employees perceive their organization as caring about social justice and being socially responsible, this shapes a positive organizational identity that guides their behaviors at work and results in them desisting from CWB (Viswesvaran et al., [111]). In contrast, employees who perceive their organization as a careless system concerned only with monetary outcomes are likely to be more vulnerable to CWB, since they are reluctant to see themselves as a part of this organization (Ellemers et al., [35]; Viswesvaran et al., [111]). When negative sense‐making about the organization undermines organizational identity, employees are less likely to engage in work‐related effort and more likely to be permissive of CWB. In this article, we focus on how CSR perceptions influence employees' customer‐directed CWB (e.g. employee rudeness, unresponsiveness and mistreatment of customers). Based on the preceding discussion, we advance the following hypothesis.

CSR perceptions are negatively related to employees' customer‐directed CWB.

Although much research has focused on the effect of CSR perceptions on employee outcomes, there is still little understanding of how CSR perceptions actually influence employee outcomes (Glavas, [42]). Thus, beyond investigating the direct impact of CSR perceptions on employees' customer‐directed CWB, the purpose of this study is to explore an underlying mechanism through which employees' sense‐making of CSR affects their customer‐directed CWB. The most frequently used underlying mechanism to explain the relationship between employees' perceptions of CSR and employee outcomes is social identity theory (Farooq et al., [37]). According to social identity theory (Tajfel, [102], [103]), employees who perceive their organizations as being highly prestigious with a positive and attractive image tend to remain with such high‐status organizations, since they feel proud about their organizations and are motivated to sustain their positive social reputation (Ellemers et al., [35]). Since the CSR activities of a firm enhance its internally perceived image and reputation (Turban & Greening, [108]), employees are more likely to develop and maintain a favorable sense of self‐worth and organizational identity (Ellemers et al., [35]). Social identity theory suggests that CSR perceptions induce employees' intrinsic motivations for the further improvement of organizational identity in a positive manner, since they can promote employees' sense of self‐worth, meeting their needs for self‐enhancement and developing their self‐esteem and identification (Collier & Esteban, [23]). Employees who perceive their organization as a caring system with social support will be proud to identify with the organization, which in turn enhances their attitudes and behaviors in the workplace (Dutton et al., [32]). In order to explain the underlying effects of ethical climate on employees' deviant behaviors, Treviño et al. ([107]) and Valentine, Greller, and Richtermeyer ([110]) suggest that the perceived ethical context (i.e. CSR perceptions) shapes a social norm to guide employees' decision‐making and to improve their ethics, which functions as a normative constraining force to refrain employees from behaving in deviant manners. In a similar vein, we argue that CSR perceptions tend to develop a civil norm in an organization that further reinforces employees' perceptions of respect and status within the organization, ultimately discouraging deviant behaviors toward one another within the organization. Based on social identity theory, a firm's CSR acts encourage employees to develop organizational civil norms in order to further develop or sustain a positive organizational identity, which may prevent employees from engagement in CWB.

Civility norms serve as an informal guide to how employees should act that designates work behaviors for respectful treatment among workgroup members (Walsh et al., [114]) and that constrains deviant behaviors such as CWB (Pearson et al., [77]). The existence of civility norms has the potential to influence workplace behavior for mutual respect, leading employees to minimize the level of disrespectful behavior or workplace incivility (Andersson & Pearson, [4]). Gill and Sypher ([41], p.55) suggest that 'Civility demands that one speaks in ways that are respectful, responsible, restrained, and principled, and avoid that which is offensive, rude, demeaning, and threatening'. Taken together, we suggest that CSR perceptions influence employees' customer‐directed CWB through the mediation of civility norms. Based on the preceding discussion, we advance the following hypothesis.

The relationship between CSR perceptions and employees' customer‐directed CWB is mediated by their organizational civility norms.

Besides social identity theory, we argue that signaling theory helps explain the effect of CSR perceptions on employees' customer‐directed CWB. Research based on social identity theory suggests that CSR perceptions decrease employees' customer‐directed CWB since it changes their attitudes and behaviors in a positive manner through their positive sense‐making of their firm. On the other hand, the signaling theory mechanism suggests that signals reflected from the firms' CSR activities can enhance its recognition among major stakeholders (Riordan, Gatewood, & Bill, [86]), leading to changes in employees' attitudes and behaviors toward their organization (Moore, [72]). According to signaling theory (Spence, [98], [99]), the firm sends a direct or indirect signal indicating its intentions, values and missions in order to influence employees and other crucial stakeholders to make decisions in favor of the company (Kooij, Jansen, Dikkers, & De Lange, [60]; Turban & Greening, [108]).

In the HRM literature, signaling theory is often used to explain the effects of a firm's policies on its reputation or organizational attractiveness (e.g. Aguilera et al., [1]; Suazo, Martinez, & Sandoval, [100], [101]). For example, a firm's CSR policies convey a positive signal to the various stakeholders regarding the actions of the organization, leading to the improvement of its reputation. For instance, job applicants are more likely to apply for a job at a firm with a better reputation since, whilst they may not have complete information about their potential employer, they are able to pick up on the signals that are reflected from the activities and policies offered by the organization (Baum & Kabst, [11]). Similarly, employees also depend on signals generated by their firm's policies, so that those who work in a firm with a positive reputation tend to be more committed to their work and generally seek greater meaningfulness and purpose at work (Suazo et al., [100]). In this respect, firms with a stronger reputation shaped by positive signals through CSR acts are more likely to promote employees' fulfillment of their own value at work, such as job calling. Scholars have suggested that the effects of CSR perceptions on employee outcomes are affected by how employees perceive their job (e.g. Bridoux, Stofberg, & Den Hartog, [16]; Rupp et al., [90]). Accordingly, Glavas and Godwin ([43]) propose that both CSR perceptions and the attendant consequences are influenced by employees' calling orientation, the extent to which employees perceive their job as a 'calling' rather than simply a 'job'. Thus, we propose that CSR perceptions convey a positive signal to employees, which enables them to perceive their work as a calling rather than merely a job.

Previous research has shown that employees with calling are likely to be more committed to a job with meaningful purpose and to overcome the various difficulties and obstacles they face, since they are able to see the new aspects of their work (Elangovan, Pinder, & McLean, [34]). Job calling has been closely related to positive employee outcomes, such as job satisfaction (Peterson, Park, Hall, & Seligman, [79]), career commitment (Duffy, Dik, & Steger, [31]) and work meaning (Duffy, Allan, & Bott, [29]). Employees who perceive their job as a calling generally have a strong motivational and psychological capability to pursue a meaningful purpose at work, high productivity and job satisfaction (Fock, Yim, & Rodriguez, [39]), which disinclines employees from engaging in deviant behaviors, such as CWB. Taken together, we suggest that CSR perceptions reduce employees' customer‐directed CWB through the mediation of employees' job calling. Based on the preceding discussion, we advance the following hypothesis.

The relationship between CSR perceptions and employees' customer‐directed CWB is mediated by the employees' job calling.

Although CSR perceptions have been shown to reduce CWB in the literature (Viswesvaran et al., [111]), little is known about the sequence by which employees' CSR perceptions diminish CWB. The extant research has found various antecedents of CWB, including employees' personal traits, organizational constraints, organizational injustice and work experiences (Bennett & Robinson, [13]; Mount, Ilies, & Johnson, [73]; Penney & Spector, [78]). According to Bennett and Robinson ([13]), CWB is generally triggered by both a reaction to employees' experiences at work and a reflection of the employees' personalities. In this respect, the current study also proposes an integrative model that considers both work experiences (i.e. civil norms shaped by CSR) and personal traits (i.e. work orientations – 'job calling') as predictors of CWB.

Civil norms at work and job calling are thus considered possible serial multiple mediators between CSR perceptions and employees' customer‐directed CWB. Combined with social identity theory and signaling theory, our study attempts to examine the serial multiple mediation effects of organizational civil norms and job calling on the relationship between CSR perceptions and employees' customer‐directed CWB. Drawing upon social identity theory, a firm's involvement in CSR acts enhances its reputation (Turban & Greening, [108]), which positively influences employees' organizational identity and work experiences, which in turn develops civil norms at work. Signaling theory suggests that a positive signal reflected from the firm's CSR acts enable employees to see their work as giving their lives meaning or a greater purpose rather than merely being a means of getting paid or seeking promotion. CSR perceptions may actually develop civil norms in an organization, which allows employees to see their work as a 'calling' and thereby mitigates their involvement in customer‐directed CWB. That is, our study attempts to explore how CSR perceptions shape the civil norm at an organizational level, which then enables employees to perceive their job as a calling and in turn decreases customer‐directed CWB. Accordingly, the following is hypothesized.

The relationship between CSR perceptions and employees' customer‐directed CWB is sequentially mediated by their organizational civility norms and job calling.

A total of 300 frontline employees in six five‐star hotels were invited to participate in the study. Checks were made to see whether the hotels selected for our research sample explicitly engaged in CSR activities, and we found that each variously communicated its CSR activities (e.g. greening, greater efforts for energy efficiency, water conservation, climate change awareness campaigns, donations to the community and so on) via its homepage and through different media, including newspapers, television and social networking services. We contacted the human resource (HR) managers of the companies to obtain permission for the data collection. The HR managers of these hotels then administered the survey to their frontline employees. The respondents received a packet that contained a cover letter, a self‐administered questionnaire, and a stamped pre‐addressed envelope. The cover letter clearly indicated that participation was voluntary and emphasized that all responses would be kept confidential and anonymous. To ensure confidentiality, the respondents were solicited to seal the completed survey in the pre‐addressed envelope and return it directly to the researchers. They received a gift voucher in return for their participation in the research.

A total of 252 questionnaires were returned, yielding a response rate of 80.4%. We used the full‐information maximum likelihood (FIML) technique to deal with missing values. FIML estimation is desirable in that other techniques (i.e. listwise‐deletion) can lead to biased results (Asendorpf, Van De Schoot, Denissen, & Hutteman, [7]). The average age and organizational tenure of the respondents were 30.43 (SD = 6.80) years and 6.04 (SD = 5.58) years respectively, with 55.2% of the respondents identifying themselves as female. The education levels of the respondents varied: high school (0.8%), two‐year vocational college (34.3%), four‐year university (58.1%) and graduate school (6.9%).

According to Brislin's ([17]) back‐translation procedure, the original survey items were translated into Korean and then back‐translated into English. The back‐translated version of the survey items was reviewed by four management scholars and found to be equivalent to the original items, indicating that the Korean version was acceptable for use.

The employees' CSR perceptions were measured using three items adjusted from Hur, Kim, and Jang ([54]), Ko, Moon, and Hur ([59]) and Wagner, Lutz, and Weitz ([113]). This scale, when applied to employees, measures employees' perceptions of their organization's overall external CSR management. It has been widely used in prior studies (Brammer, He, & Mellahi, [15]; Kim et al., [57]; Vlachos, Panagopoulos, & Rapp, [112]).

Organizational civility norms were assessed with four items adapted from the scales of Walsh et al. ([114]). Sample items are 'Rude behavior is not accepted by my coworkers' and 'Respectful treatment is the norm in your unit/workgroup'. The items were rated on a five‐point Likert‐type scale (1, 'strongly disagree', to 5, 'strongly agree').

We used the 12‐item scale of Dobrow and Tosti‐Kharas ([28]) to assess calling toward a domain (e.g. service work). Sample items are 'I am passionate about being a hotel employee' and 'The first thing I often think about when I describe myself to others is that I'm a hotel employee'. The items were rated on a five‐point Likert‐type scale (1, 'strongly disagree', to 5, 'strongly agree').

Customer‐directed CWB was measured with 10 items adapted from Hunter and Penney ([53]). Sample items are 'I ignored a customer' and 'I acted rudely toward a customer'. Each of the seven items was rated on a five‐point Likert‐type scale (1, 'never', 2, 'once or twice', 3, 'once or twice a month', 4, 'once or twice a week', and 5, 'every day').

In testing the hypotheses, we controlled for age, gender and work experience (years) in all analyses. These variables were controlled because demographic variables are related to levels of civility norms (see, e.g., Walsh et al., [114]), job calling (see, e.g., Xie, Xia, Xin, & Zhou, [116]; Xie, Zhou, Huang, & Xia, [117]) and customer‐directed CWB (see, e.g., Penney & Spector, [78]; Sakurai & Jex, [92]; Skarlicki, van Jaarsveld, Shao, Song, & Wang, [96]).

The means, standard deviations, intercorrelations and reliabilities of the variables for the study are reported in Table 1. First, we evaluated the reliability of the constructs using Cronbach's alpha coefficients. The reliability coefficients for the variables ranged from 0.86 to 0.96, which exceed the criterion of 0.70 (Nunnally, [75]). Next, we conducted a CFA to verify the convergent and discriminant validity of the measures using M‐plus 7.4 software. Noting that the validity of the model evaluation can be adversely affected in situations where there is a low ratio of sample size to number of free parameters, particularly in cases where there are a significant number of latent variable indicators (Bentler & Chou, [14]), item parcels were devised to operate as pointers of customer‐directed CWB and job calling, both of which were treated as being uni‐dimensional in cases where there were more than six items (Little, Cunningham, Shahar, & Widaman, [68]). The item‐to‐construct balance approach (Little et al., [68]) was utilized for the parceling process, which produced three‐item parcels for the two constructs. In our measurement models, factor and item loadings all exceeded 0.55, with all t‐values greater than 2.58, confirming the convergent validity of our measures. All measures exhibited strong reliability, with composite reliabilities ranging from 0.86 to 0.95 (Table 1). Finally, we evaluated the discriminant validity among the variables (Fornell & Larcker, [40]). All average variance extracted (AVE) values were larger than the squared correlation between the variable and any others (Table 1). Taken together, these results indicate that our constructs possessed sufficient reliability and validity.

Construct means, standard deviations, and correlations

• 15220001 † p < 0.10.
• 15220002 * p < 0.05.
• 15220003 ** p < 0.01; CR, composite reliability; the number on the diagonal is the AVE (average variance extracted).

Because our data were based on self‐report, it was necessary to assess potential biases resulting from common method variance (CMV). Following the recommendations of Podsakoff, MacKenzie, and Podsakoff ([80]), we took several precautions to reduce CMV, such as protecting respondent anonymity, reducing evaluation apprehension, improving item wording and separating the measurement of predictor and outcome variables. Furthermore, as a statistical procedure, we performed Harman's one‐factor analysis. All fit indices demonstrated a worse fit for the one‐factor model than for our original measurement model (χ2(65) = 1468.69; p < 0.05, CFI = 0.44, TLI = 0.33, RMSEA = 0.29, SRMR = 0.17). We used another procedure that introduces an additional latent common method factor into the measurement model (Podsakoff et al., [80]). This factor did not account for any substantial variance in the indicator variables (2.1%), given that an average of 18–32% of the variance in a typical measure is attributable to method variance (Podsakoff et al., [80]). Furthermore, the standardized factor loadings of all items were below 0.30 for the latent common method factor and therefore statistically not significant. Taken together, these findings provide evidence that our data were not affected by CMV.

We estimated the path coefficients in our structural model. The results of structural equation modeling (SEM) are reported in the upper section of Table 2. We specified and tested a multiple sequential mediation model (Lau & Cheung, [61]; Taylor, MacKinnon, & Tein, [104]). The advantage of this approach is that we could isolate and estimate the indirect effect of the two mediators (Lau & Cheung, [61]; Taylor et al., [104]). Figure 1 illustrates the results of the sequential mediation model. The hypothesized model demonstrated an acceptable fit to the data (χ2(86) = 160.46; p < 0.05, CFI = 0.97, TLI = 0.96, RMSEA = 0.06, SRMR = 0.05) and accounted for 27.4% of the variance in civility norms, 41.6% of the variance in job calling and 9.8% of the variance in customer‐directed CWB (Table 2).

Path coefficients and indirect effects for mediation models

• 15220004 † p < 0.10.
• 15220005 * p < 0.05.
• 15220006 ** p < 0.01; tests of path coefficients are two‐tailed tests; CI, 95% confidence level; CSR = employees' perceptions of CSR; CN = organizational civility norms; JC = job calling; CD‐CWB = customer‐directed CWB; b = unstandardized coefficient.

We conducted a three‐step process to test the hypotheses that employees' civility norms in their organization and job calling would fully or partially mediate the relationship between employees' CSR perceptions and their customer‐directed CWB. As a first step, we predicted that employees' perceptions of CSR were negatively related to customer‐directed CWB, and Hypothesis 1 was supported (b = −0.20, p < 0.01), as indicated in the upper part of Figure 1. Second, to investigate the three mediation effects, six paths were examined in the model in the lower part of the figure. Third, we estimated these indirect effects, along with the symmetric and 95% bias‐corrected bootstrapped confidence intervals for our path estimates (N = 5000; Shrout & Bolger, [95]; Hayes, [49]) in Table 2. Hypothesis 2 stated that the relationship between employees' CSR perceptions and their customer‐directed CWB would be mediated by organizational civility norms. The indirect effect of employees' CSR perceptions on customer‐directed CWB via civility norms was moderate (b = −0.08), and a bootstrapped estimate of the indirect effect confirmed that it was statistically significant at 95% CI [−0.18, −0.01], supporting Hypothesis 2. Hypothesis 3 stated that employees' job calling would mediate the path between employees' CSR perceptions and customer‐directed CWB, and this hypothesis was supported (b = −0.07, 95% CI [−0.15, −0.01]). Finally, to test Hypothesis 4, we estimated the serial multiple mediation effect throughout organizational civility norms and employees' job calling, specifically the indirect effect from employees' CSR perceptions on customer‐directed CWB through organizational civility norms and employees' job calling, which was also found to be statistically significant (b = −0.02, 95% CI [−0.06, −0.01]). These results indicate that employees' CSR perceptions are associated with lower organizational civility norms and job calling, which is related to lower CWB levels toward customers. In addition, controlling for the sequential mediators, employees' CSR perceptions were found to no longer be a statistically significant predictor of customer‐directed CWB (b = −0.03, 95% CI [−0.21, 0.16]). In sum, we confirmed that the positive relationship between employees' CSR perceptions and customer‐directed CWB was fully and sequentially mediated by organizational civility norms and employees' job calling.

Beyond investigating the relationship between employees' CSR perceptions and customer‐directed CWB, our study suggested an underlying sequential mediation mechanism by which organizational civility norms and job calling relate to customer‐directed CWB. As predicted, employees' CSR perceptions were negatively associated with customer‐directed CWB. Furthermore, it intervened in the relationship between employees' CSR perceptions and customer‐directed CWB through the intermediary process of organizational civility norms and job calling. The results of this research extend previous studies into employees' CSR perceptions and employee outcomes by not only highlighting the attenuating role of CSR perceptions on negative employee outcomes, such as customer‐directed CWB, but also incorporating social identity theory and signaling theory to explore the underlying mechanisms of CSR perceptions and customer‐directed CWB linkage. Our findings have several theoretical and practical implications.

Our study brings new insight to the micro‐CSR literature by exploring the undiscovered path between employees' CSR perceptions and customer‐directed CWB. Although a growing body of research has examined the role of employees' perceptions of CSR on organizational and employee outcomes, prior research has mainly focused on positive employee outcomes (e.g. compassion, job performance, creativity and organizational citizenship behavior) as a consequence of employees' CSR perceptions (e.g. Hur, Moon, et al., 2016; Ko et al., [59]; Moon et al., [71]). Our study is the first to demonstrate the negative ramifications of employees' CSR perceptions on service employees' in‐role task behaviors (i.e. customer‐directed CWB) in terms of the social identity and signaling perspectives. Complementing the extant body of research on employees' CSR perceptions, the findings of our study suggest that ethical norms pursued by organizations and the desirable duties sought by individuals play a critical role in decreasing engagement in unethical behavior toward customers. Thus, by demonstrating the significance of employees' CSR perceptions as an important antecedent that reduces service employees' customer‐directed CWB, our study enriches the knowledge of the relationship between employees' CSR perceptions and employee outcomes.

Another theoretical contribution lies in the exploration of the intermediary mechanisms involving employees' CSR perceptions and customer‐directed CWB. The sequential mediation of organizational civility norms and employees' job calling validates social identity theory and the signaling‐theory‐based models of motivation as pertinent frameworks for explaining how employees' CSR perceptions relate to customer‐directed CWB. As postulated by social identity theory, CSR perceptions influence employees to cultivate the organizational civil norms in order to further enhance their positive identification with their organization, which may prevent service employees from engaging in customer‐directed CWB. Furthermore, as predicted by the signaling‐theory‐based model of motivation, employees' CSR perceptions and the associated fulfillment of their own value at work engender job calling, which works to preclude employees actively engaging in customer‐directed CWB. Thus, by presenting and validating the relevant frameworks for the mediating mechanisms of the employees' CSR perceptions–customer‐directed CWB relationship, our study makes theoretical contributions to research on employees' CSR perceptions and employees' unethical behavior toward external stakeholders (i.e. customers).

Finally, our study contributes to the ethics literature by investigating the association between the ethical context shaped by a firm's CSR and employee's customer‐directed CWB. The findings suggest that an ethical work climate, for example organizational civil norms developed by the firm's CSR, combined with employees' job calling, is crucial in reducing employees' deviant behaviors, which has important implications for the ethics literature. An ethical work context generated from the firms' CSR helps lessen the possibility that employees engage in customer‐directed CWB. Our study provides evidence that an ethical context, coupled with job calling, attenuates employees' customer‐directed CWB.

Our findings suggest that service employees' CSR perceptions decrease their customer‐directed CWB at work. Managers striving to reduce service employees' customer‐directed CWB should take notice of the role of employees' CSR perceptions as a pivotal antecedent. In particular, given the mounting evidence that CSR perceptions of employees reduced their negative attitudes and behaviors (see, e.g., Sheel & Vohra, [93]), managers need to care about employees' CSR perceptions of their organization. These managers should pay attention to how employees understand and respond to their CSR initiatives and policies. Furthermore, the narratives of a firm's CSR activities develop a shared recognition of corporate value, such as an overall sense of social justice, compassion, virtue and equity toward society, which directly strengthens employees' positive sense‐making about their organization (Ko et al., [59]). Seeing their organization as a care‐providing system and a source of social support and healing (Kahn, [56]) shapes civil norms among employees and leads to stronger calling toward their job, which subsequently lessens employees' customer‐directed CWB. In addition, the use of various communication channels, such as an intranet, staff letters, employee meetings and team briefings, helps employees take notice of the firm's engagement in CSR, and encourages them to behave in accordance with expectations of high quality service delivery to customers (Ko et al., [59]).

Based on the finding that the organizational civil norm is a pivotal mediator of employees' CSR perceptions and the customer‐directed CWB relationship, managers would be advised to enhance organizational civil norms as a means of reducing customer‐directed CWB. Therefore, an organization's managers need to not only instill the organization's CSR initiatives and policies into the employees' perceptions, but also promote a civil work context at work. To develop a civil climate within an organization, scholars have suggested that organizations need to foster respectful, trustful, courteous and compassionate behavior among their employees (e.g. Porath & Pearson, [81], [82]). Much research has shown that employee civility can be developed by way of organizational interventions. For example, the Civility, Respect, and Engagement at Work (CREW) intervention program has enhanced the quality of employees' social interactions as well as having several other important outcomes (Leiter, Day, Gilin Oore, & Spence Laschinger, [63]; Leiter, Spence Laschinger, Day, & Gilin Oore, [64]; Osatuke, Moore, Ward, Dyrenforth, & Belton, [76]).

Third, considering the significance of employees' job calling as a proximal antecedent of customer‐directed CWB, organizations should cultivate employees' job calling in order to prevent customer‐directed CWB. Since a calling can provide human beings with a deep sense of meaning and purpose in their work (Hirschi, [51]; Rosso, Dekas, & Wrzesniewski, [87]), organizations should help employees to connect their work with a prosocially oriented purpose to foster this sense of calling (Duffy & Dik, [30]). Therefore, a firm's efforts to better communicate its CSR activities and promote such activities as core organizational values to its employees are necessary if the organization is to successfully reduce customer‐directed CWB amongst its employees due to a civil climate in the workplace.

Finally, a more sustained way to prevent employees' CWB may include establishing systematic and institutionalized practices and policies for developing an ethical climate. For instance, training and development programs for facilitating a firm's CSR activities can shape a civil climate within the organization, coupled with the development of job calling, which leads to organization‐wide sensitivity to the issue of employees' deviant behaviors (Reio & Ghosh, [85]). Thus, we suggest that practitioners (i.e. CEOs and senior managers) should build up a work climate that derives a civil norm among employees and job calling within an organization in order to prevent customer‐directed CWB. Our study prompts practitioners to reconsider the role of organizational civil norms and job calling as a way of generating synergy effects that discourage employees' engagement in CWB. Accordingly, managers should closely monitor how employees' CSR perceptions can be changed over time to promote organizational civil norms and the sense of job calling, and thus prevent unfavorable motivations and behaviors among employees.

The contributions of our study should be qualified by the following limitations. First, the use of self‐reported, cross‐sectional data poses a limitation on the causal conclusion made between the variables. We recommend that future researchers need to use a longitudinal design to better determine the causality among the research constructs. In addition, obtaining a measure of customer‐directed CWB from the target employee's supervisors or customers can help reduce biases resulting from CMV and social desirability. Second, although organizational civility norms are an organizational‐level property, we measured them through employee perceptions. For this reason, we could not capture how civility norms at the organizational level affected employees' job calling and customer‐directed CWB. Although we observed an acceptable level of variance between employees' responses regarding organizational civility norms, shared perceptions of organizational civility norms can better represent phenomena taking place at the organizational level. Therefore, a future study might explore the multi‐level or cross‐relationships between organizational‐level civility norms and job calling and customer‐directed CWB by collecting data from multiple organizations. Third, for the sake of model parsimony, we did not specify boundary conditions at the organizational or individual level that may affect the employees' CSR perceptions–customer‐directed CWB relationship. In particular, organizational characteristics (e.g. ethical leadership, organizational support, ethical climate and organizational culture) or individual characteristics (e.g. demographics, moral identity and environmental values) can be potential moderators that strengthen or weaken the linkage between CSR perceptions and employees' customer‐directed CWB. For instance, personal attitudes or values concerning environmental issues would be an important moderator of the relationship between CSR perceptions and employees' customer‐directed CWB. Employees' environmental values are referred to as the 'desired end state of natural systems integrity and the means of human adaptation to, rather than domination over, the natural environment' (Marcus, MacDonald, & Sulsky, [70], p. 464). Since employees with strong environmental values are more likely to have values pertaining to the conservation and protection of the natural environment (Aguinis & Glavas, [3]), environmental values will mitigate the relations between CSR perceptions and employees' customer‐directed CWB. Thus, future work could explore such moderators and test the moderating relationships at different levels of the organizations or at the individual level. Fourth, our research did not separate the types of CSR perception (e.g. internal and external CSR) (Hameed, Riaz, Arain, & Farooq, [47]); future studies need to identify each different effect of specific types of CSR in order to properly investigate whether the different types of CSR have similar or different effects on mediators or outcomes. Fifth, it should be noted that our sample consisted only of frontline employees in South Korean hotels. Although we assume that the proposed relationships between employees' CSR perceptions, organizational civility norms, job calling and customer‐directed CWB hold across different cultures, industries and organizations, the findings of our research need to be replicated in other contexts and samples to confirm their external validity. In addition, the majority of our sample was female employees. Although this represents the typical gender balance of hotel employees in Asian countries (see, e.g., Hur, Moon, & Ko, [55]; Ko et al., [59]; Shin, Hur, & Choi, [94]), future research may need to collect data and replicate the current findings for a more gender‐balanced sample. Furthermore, the employees were all recruited from luxury hotels due to the fact that these hotels explicitly communicate their CSR activities via their homepages and through different media, including newspapers, TV and SNS in South Korea (Hur, Moon, & Ko, [55]). However, future research may need to use employees from a full range of hotel grades in order to confirm our results. Sixth, the effect of CSR perceptions on employee outcomes varies greatly depending on the level of CSR proactivity in different hotels and the level of communication to employees relating to CSR. It is thus advisable that future studies should measure the extent to which employees perceive their firm's CSR proactivity in relation to its CSR communications (e.g. variety of communication channels, such as an intranet, staff letters, employee meetings and team briefings). This is an area in which additional research is required. Finally, our study did not control for organizational effects such as 'well‐managed companies' or 'ethical organizational climate'. For instance, without the firms' engagement in CSR, the hotels are just well managed and the employees are happy, making it less likely that the workforce will experience negative actions such as CWB. Thus, future study should include organizational effects as control variables, such as organizational climate, the service environment, the frequency of interaction and the duration of employee–customer rapport.