Hidden Roles of CSR : Perceived Corporate Social Responsibility as a Preventive against Counterproductive Work Behaviors

The purpose of this study is to identify how the degree to which employees perceive that their organization is socially responsible influences their counterproductive behaviors at work. We developed a research model capturing the relationships between perceived corporate social responsibility (CSR), affective organizational commitment (AOC), and organization-, colleague-, and customer-directed counterproductive work behaviors (CWBs). Drawing on social identity and social bonding theories, we hypothesized that employees’ perceptions of CSR would enhance their levels of AOC and thereby hinder the three types of CWBs. Survey data from 252 frontline employees of six luxury hotels in South Korea were examined. We found that perceived CSR is indirectly and negatively related to the three types of CWBs through AOC. We discussed the theoretical contributions and practical implications of this study.


Introduction
It has been frequently reported that many firms experience disadvantages caused by volitional behavior by employees that has a detrimental effect on their organizations and stakeholders [1,2].Such behavior may be the source of internal operational risk, which implies the potential for people failures in the course of regular business operations that causes financial damage [3,4].Indeed, research has shown that negative behavior conducted by members of an organization-so-called deviant behavior or counterproductive work behavior (CWB)-poses expensive problems to that organization, including increased absenteeism and turnover, as well as decreased satisfaction and productivity (e.g., [5,6]).To address these issues, organizations are increasingly seeking to deter or prevent such behavior.Scholarly interest in this phenomenon has suggested various situational reasons for the development of CWB amongst employees in the workplace, such as organizational injustice, interpersonal conflict, and emotional stressors (e.g., [7][8][9]).Although existing studies have contributed to increasing our understanding in this area, they have tended to pay attention to examining what causes CWB.However, given the explicit threat and cost of CWB, it seems desirable to shift the focus from investigating the triggers of CWB to exploring potential constraints against it.In short, we believe that identifying the negative rather than the positive antecedents of CWB would be useful to both academic researchers and organizational practitioners.
The current study seeks to explore the organizational conditions which play a pivotal role in inhibiting destructive employee behavior at work.In doing so, we focus on corporate social responsibility (CSR), those organizational actions and policies which go beyond the pursuit of financial gains and legal responsibilities, and take into account stakeholders' expectations [10][11][12].It is well known that the degree to which employees perceive whether their organization participates in CSR activities and policies contributes to organizational effectiveness by promoting their outcomes [13][14][15].Indeed, numerous studies have demonstrated that employees' beliefs about their organization being socially responsible exert influence over a variety of their outcomes, including job satisfaction, employee engagement, organizational identification, and job performance (e.g., [16][17][18][19].Even if the significant effects of perceived CSR on employees' attitudes and behaviors lead us to expect that it has sufficient potential to determine their levels of CWB, there is little information regarding how perceived CSR is actually associated with CWB. Accordingly, the current study aims to examine links between perceived CSR and CWB.While the latter is known for the harmful behavior which can be directed at the employing organization and individuals within that organization, it has only recently been suggested to include detrimental behavior directed at customers outside the organization [8].Consequently, CWB may be categorized as one of three types according to its target: organization-directed CWB (e.g., purposely working slowly), colleague-directed CWB (e.g., ignoring a coworker at work), or customer-directed CWB (e.g., acting rudely toward a customer).The research model employed in the current study incorporates all three types.We thus seek to clarify how employees' perceptions of CSR relate to these types of CWBs by identifying a mechanism responsible for these links.
We expect this study to make a contribution to the relevant literature in two key regards.This study first uncovers the hidden roles of CSR that have been neglected to date by linking employees' perceptions of CSR to their CWBs.This study also suggests a research model which focuses on the bridging role of employee attitudes toward the organization found in the theoretical CSR-CWB relationship.In the following sections, we first suggest the theoretical framework for this study and develop the specific hypotheses.Next, we propose a suitable methodology to test our research model.Lastly, we suggest the empirical results of this study and conclude with discussions on the contributions and implications of those findings.

Theoretical Framework
Since the main reason for an organization's CSR activities is not to facilitate the productive outcomes of its members or to restrain their counterproductive outcomes [18], it is difficult for us to believe that perceived CSR is likely to have a direct influence on their CWB.To investigate how and why employees' perceptions of CSR are related to their CWB, the current study relies on the stimulus-organism-response model that has been widely used to explain the mechanisms and reasons that stimuli and responses are related, and to identify the fundamental processes underlying the relationship between organizational contexts and individual behavior [20].This model is applied to the present study to examine the organism through which perceived CSR as the stimulus have an effect on CWB as the response.Employees' attitudes at work have received much attention as the organism responsible for their behavior (e.g., [21]), leading us to expect one's attitude toward his or her organization to be a potential organism in the linkage between perceived CSR and CWB.
A meta-analytic review of the organizational commitment literature [22] has shown that affective commitment to the organization has greater explanatory power, compared to the other sub-dimensions such as normative and continuance organizational commitment in predicting a variety of employee work behaviors.We thus expect that affective organizational commitment (AOC), which is defined as employees' emotional attachment to, involvement in, and identification with their organization [23], might serve as a significant mediating variable between employees' perceptions of CSR and their CWBs.With these considerations, we present the research model for this study in Figure 1.In the following section, drawing on social identity and social bonding theories, we would hypothesize that employees' perceptions of CSR would enhance their levels of AOC and thereby hinder the three types of CWBs.

Hypothesis Development
Social identity theory suggests that individuals are inclined to distinguish themselves and others based on social categories, and to define their view of themselves-the so-called self-concept-with their membership of a specific social group [24,25].Since a corporate organization is one of the representative social groups that have significant impacts on its members, organizational membership functions as an important dimension that describes their identity.In other words, the perceived identity of a company could affect the formation and maintenance of employees' self-concept.
Social identity theory also implies that persons are able to enhance their self-concept by recognizing that their own group is better than other groups [26].When internal stakeholders such as organizational members evaluate the organization to which they belong, they have a tendency to place great importance on the social activities carried out by their organization [27].For example, it is known that an organization's interest in social issues and evidence of its ethical responsibility generally improve its reputation, and that employees tend to be proud to identify themselves with an organization possessing a favorable reputation [28,29].This suggests that CSR activities could meet the requirements of society and also fulfill the needs of employees who want to belong to a desirable group.Accordingly, if an organization is perceived to be different from and superior to others in terms of its socially responsible actions and policies, the employees of that organization are likely to identify strongly with their organization and to consequently possess high levels of attachment and belonging to, and loyalty for that organization.Based on this rationale, scholars have demonstrated that when employees recognize that their organization participates in CSR activities, they tend to be affectively immersed in the organization (e.g., [13,30,31]).
Meanwhile, social bonding theory is helpful to understand why some people do not exhibit social conformity [30].According to social bonding theory, the higher the social bond, the higher the likelihood of conformity to social expectations.This implies that employees with a stronger social bond are less likely to deviate from social group expectations.Hollinger [32] proposed that this theory could be directly applied to the explanation of employee deviance, and that attachment-one of the elements of the social bond that Hirschi [33] had proposed-should be regarded as its affective component.Based on this proposition, Sims [34] found that affective commitment to the organization as another measure of attachment has a significant influence on ethical rule breaking, namely employee misconduct involved with unethical practices that disobey the organization's norms.In his meta-analytic review, Dalal [6] showed that the level of organizational commitment is negatively related to the degree of deviant or counterproductive behavior.Although he did not report the specific dimensions of commitment, it could be inferred that almost all of the measures

Hypothesis Development
Social identity theory suggests that individuals are inclined to distinguish themselves and others based on social categories, and to define their view of themselves-the so-called self-concept-with their membership of a specific social group [24,25].Since a corporate organization is one of the representative social groups that have significant impacts on its members, organizational membership functions as an important dimension that describes their identity.In other words, the perceived identity of a company could affect the formation and maintenance of employees' self-concept.
Social identity theory also implies that persons are able to enhance their self-concept by recognizing that their own group is better than other groups [26].When internal stakeholders such as organizational members evaluate the organization to which they belong, they have a tendency to place great importance on the social activities carried out by their organization [27].For example, it is known that an organization's interest in social issues and evidence of its ethical responsibility generally improve its reputation, and that employees tend to be proud to identify themselves with an organization possessing a favorable reputation [28,29].This suggests that CSR activities could meet the requirements of society and also fulfill the needs of employees who want to belong to a desirable group.Accordingly, if an organization is perceived to be different from and superior to others in terms of its socially responsible actions and policies, the employees of that organization are likely to identify strongly with their organization and to consequently possess high levels of attachment and belonging to, and loyalty for that organization.Based on this rationale, scholars have demonstrated that when employees recognize that their organization participates in CSR activities, they tend to be affectively immersed in the organization (e.g., [13,30,31]).
Meanwhile, social bonding theory is helpful to understand why some people do not exhibit social conformity [30].According to social bonding theory, the higher the social bond, the higher the likelihood of conformity to social expectations.This implies that employees with a stronger social bond are less likely to deviate from social group expectations.Hollinger [32] proposed that this theory could be directly applied to the explanation of employee deviance, and that attachment-one of the elements of the social bond that Hirschi [33] had proposed-should be regarded as its affective component.Based on this proposition, Sims [34] found that affective commitment to the organization as another measure of attachment has a significant influence on ethical rule breaking, namely employee misconduct involved with unethical practices that disobey the organization's norms.In his meta-analytic review, Dalal [6] showed that the level of organizational commitment is negatively related to the degree of deviant or counterproductive behavior.Although he did not report the specific dimensions of commitment, it could be inferred that almost all of the measures included in his research would reflect affective commitment to the organization founded in its prevalence [35].It is thus anticipated that affective commitment to the organization will likely reduce the propensity to exhibit counterproductive behavior at work.
In summing up the above discussion, employees' perceptions of CSR would be positively related to AOC and, in turn, negatively influence CWB.The relationships between perceived CSR, AOC, and three types of CWBs will be described in more detail below.A strong emotional bond with an organization serves a key role in maintaining an employee's close relations with his or her organization [36].Employees with high levels of affective commitment to their organization tend to identify themselves with the organization, which, as a result, leads them to behave in ways that contribute to the smooth operation of the organization and the achievement of its goals [22,36].As a result, it is expected that employees who are affectively involved in their organization are willing to conform to the norms and regulations of the organization rather than to act against them.As discussed above, when employees consider that their organization invests considerable effort in fulfilling its social responsibilities, they are likely to strengthen their affective commitment to their organization.Furthermore, employees who are affectively committed to such socially responsible organizations are unlikely to perpetrate deviant behavior to harm or otherwise violate the norms of those organizations.Accordingly, we present the following hypothesis: Hypothesis 1 (H1).Perceived CSR will be indirectly and negatively related to organization-directed CWB mediated through AOC.
One might view coworkers as agents of their organization and consider the organization to be a collection of its members [37].Accordingly, employees with high levels of AOC are likely to show not only their identification to the organization but also their attachment to their peers.Previous research has argued that affective commitment to the organization is conducive to prosocial behavior among organizational members, such as interpersonal assistance in the organization instead of antisocial behavior targeted at coworkers in the workplace (e.g., [14]).In this regard, colleague-directed CWB would be less frequently manifested by employees high in AOC compared to those low in AOC.Indeed, scholars have found that organizational commitment negatively predicts interpersonal deviance in the organization (e.g., [37]).Consequently, it is expected that perceived CSR will be positively associated with an emotional dedication to the organization, which would in turn be negatively associated with counterproductive behavior toward colleagues in the workplace.We thus suggest the following hypothesis: Hypothesis 2 (H2).Perceived CSR will be indirectly and negatively related to colleague-directed CWB mediated through AOC.
Organizational commitment consists of an individual's acceptance of organizational goals and values as well as his or her willingness to make endeavors on behalf of the organization [23,38].Employees at customer contact points could be aware that the performance of their work, including interactions with customers, is directly linked to customer satisfaction and loyalty, and is further critical to organizational effectiveness and reputation.In this situation, employees who are emotionally immersed in the organization would strive to serve customers' needs and to refrain from behaving in ways that would inadequately deal with customers because they might consider their own efforts to help organizational success as well as individual performance [39].In summing up this discussion and the assumption about the positive relationship between perceived CSR and AOC presented earlier, we expect that employees working for a socially responsible organization will not engage in harmful behavior toward customers due to their high affective commitment to their organization.Accordingly, we present the following hypothesis: Hypothesis 3 (H3).Perceived CSR will be indirectly and negatively related to customer-directed CWB mediated through AOC.

Sampling Procedures and Participant Characteristics
Firms in the service industry are susceptible to operational failures that can cause major damage, and hotels are particularly firms with significant operational risk [3,40].This operational risk is likely to be caused by the deviant behavior of the internal members.Hotels are also recognized to be sensitive to changes originating from CSR issues [41].Hotel employees are thus considered to be suitable for the objective of our research to examine the CSR-CWB relationship.For sampling purposes, we undertook to invite 300 frontline employees engaged with six full-service luxury hotels recognized for their involvement with CSR activities.Prior to approaching these staff, careful consideration was first made to establish that each of these hotels was actively and visibly participating in a range of regular CSR initiatives, such as water conservation efforts, energy efficiency programs, environmental awareness campaigns, and charitable work within their respective communities.To this end, we sought to verify that these hotels were proactive in communicating their various CSR efforts, which included a variety of recycling and environmental protection schemes, through their homepages and via social and other media.To confirm confidentially, we asked the respondents to seal the completed questionnaires in envelopes with the address and to send them to us by mail.A total of 256 completed questionnaire packets were collected, and 252 usable questionnaires after list-wise deletion were used in the analysis, representing a valid response rate was 80.4%.Based on the responses of these 252 employees, a sample descriptive analysis showed that 55.2% of the participants were female, with an average age of 30.4 years (SD = 6.8).A majority of the subjects (57.1%) had a four-year university education, just over 33.7% had at least a two-year vocational college education, 6.7% had a graduate school education, and only 0.8% had a high school education.On average, participants had almost 6.0 years of work experience (SD = 5.6).

Measurement Scales
Following Brislin's [42] suggestion, we translated the original version of the measurement items into Korean and two bilingual translators independently translated them back from Korean to English.Responses were rated on 5-point Likert-type scales (see Table 1).We measured perceived CSR with three items based on Hur et al. [43] and Wagner et al. [44], with the example item, "My organization is concerned to improve the well-being of society".AOC was evaluated with four items adapted from Allen and Meyer [23] and King et al. [45], including "I am proud to be a part of the organization I work for."We used five items to measure organizational or organization-directed CWB (CWB-O), such as "I discuss confidential company information with an unauthorized person", and four items to measure interpersonal or colleague-directed CWB (CWB-I), as adapted from Bennett and Robinson [5], including "I make fun of someone at work".Customer-directed CWB (CWB-C) was evaluated with a five-item measurement used by Hunter and Penney [8], such as "I act rudely toward a customer".This study controlled for gender (female = 0, male = 1), age (in years), and work experience (in years) in analyzing the hypotheses because these were shown to be related to both AOC and CWB in the existing studies (e.g., [8,[46][47][48]).

Reliability, Validity, and Common Method Bias
In order to test the internal reliability and validity of the resulting measurement items, we relied on the commonly used validation processes.First, we evaluated the internal reliability of research constructs using coefficients of Cronbach's alpha (see Table 1).The reliability coefficients for the variables ranged from 0.85 to 0.93, indicating sufficient levels of reliability [49].We conducted confirmatory factor analysis (CFA) with M-plus 8.0 program to verify the convergent and discriminant validity of the measurement scales.
To check the fitness of the measurement model, this study used some commonly accepted goodness-of-fit indices such as the Chi-square (χ 2 ) statistic with its degree of freedom, the comparative fit index (CFI), the Tucker-Lewis index (TLI), the root-mean-square error of approximation (RMSEA), and the standardized root-mean-square residual (SRMR).The measurement model fit well with the data (χ 2 (303) = 714.78;RMSEA = 0.07; SRMR = 0.06; CFI = 0.92; TLI = 0.91).All standardized factor loadings exceeded 0.67, suggesting testimony of convergent validity among our measures.All the measurement items showed acceptable reliability, with composite reliabilities ranging from 0.85 to 0.94 (see Table 1).According to the proposition by Fornell and Larcker [50], we checked the condition for discriminant validity among constructs.Given that all average variance extracted (AVE) values were bigger than the squared correlation between a specific variables and any other variables (see Table 2), we believed that these constructs demonstrated sound measurement properties.
Since we used totally the self-reported measures, we also assessed the degree of common method bias (CMB), which might be serious in the case that a single latent variable explained the majority of the manifest variables' variance [51].According to Podsakoff et al. [51], it is possible for researchers to control CMB with procedural and statistics treatments.Accordingly, we introduced procedural treatments by improving item wording, separating the questionnaire of the predictors and consequences, ensuring respondent anonymity, and reducing evaluation apprehension.As a statistical treatment, we introduced a CFA approach to Harman's single-factor analysis.Being compared to our original measurement model, all measurement items of the goodness of fit showed a worse fit for the single-factor model (χ 2  (189) = 2021.68;RMSEA = 0.20; SRMR = 0.16; CFI = 0.52; TLI = 0.47).Therefore, we believed that CMB did not cause the serious problems we initially thought it might do.

Hypothesis Testing
To test the three mediation hypotheses, we firstly investigated all the path coefficients, ruling for gender, age, and work experience (see Table 3).Of the control variables, only gender was positively related to organization-directed CWB, indicating that males tended to engage in more organization-directed CWB than females.Nonetheless, age and work experience were not associated with the three types of CWBs as well as AOC in this sample.It was shown that perceived CSR was positively associated to AOC, and AOC was negatively related to the three types of CWBs.Next, we investigated three simple indirect effects [52].This fundamental condition of indirect [53] was examined with structural equation modeling (SEM).These specific mediation effects represented the potential of AOC to intervene the effects of perceived CSR on the three types of CWBs (i.e., organization-, colleague-, and customer-directed CWB).
The overall of model fit showed that the indirect model was reasonably well: χ 2 (227) = 480.55;RMSEA = 0.07; SRMR = 0.05; CFI = 0.94; TLI = 0.92.Overall, the suggested mediation model was quietly meaningful in explaining variance (R 2  (AOC) = 39.0%,R 2 (organization-directed CWB) = 8.2%, R 2 (colleague-directed CWB) = 9.2%, and R 2 (customer-directed CWB) = 10.1%).We examined three mediation hypotheses using bootstrapping approach (N = 5000; [54]), which is recognized as a re-sampling technique that investigates the accuracy of the estimates, coefficients and their standard errors in this study, from random samples with replacement [52].As indicated in Table 3, the intervention effect, which is of dominant interest in a simple mediation model, is estimated in terms of investigation for the seven path coefficients: (1) the direct effect of perceived CSR on AOC; and (2) the direct effect of AOC on the three types of CWBs while controlling for perceived CSR.As a result, the indirect effect of employees' perceptions of CSR on organization-directed CWB with the mediation of AOC was not so strong (b = −0.147),but the bootstrapped estimation of the indirect effect was statistically significant (95% confidential interval [CI] [−0.292: −0.057]).Nonetheless, it was found that employees' CSR perceptions did not directly influence on the level of organization-directed CWB (b = 0.074, n.s.).Hypothesis 1 was thus supported, in that the negative linkage between perceived CSR and organization-directed CWB was fully intervened by AOC.
Second, the intervention effect from perceived CSR on colleague-directed CWB through AOC was small (b = −0.118),however in statistical terms, it is still significant (95% CI [−0.250: −0.024].Furthermore, employees' perceptions of CSR didn't directly influence on colleague-directed CWB (b = −0.029,n.s.).Consequently, our study revealed that the negative linkage between CSR perceptions and colleague-directed CWB was fully mediated by AOC, and that Hypothesis 2 was confirmed.
Finally, the indirect effect of perceived CSR on customer-directed CWB with the mediation of AOC was significant (b = −0.185,95% CI [−0.344: −0.076]), whereas the direct effect between perceived CSR and customer-directed CWB was not significant (b = −0.002,n.s.).These results showed that the negative linkage between CSR perceptions of employees and customer-directed CWB was fully mediated by AOC.Therefore, Hypothesis 3, which predicted that perceived CSR would also indirectly have a negative relationship with customer-directed CWB mediated through AOC, was supported.

Discussion
The prevalence and cost of employees' counterproductive behavior at work warrant the research effort to understand why and how they engage in such behavior [37].Furthermore, it is necessary to examine what organizational activities and policies might prevent employees from exhibiting deviant behavior in the workplace.By focusing on the potential benefits that could be achieved by organizations which make efforts to fulfill their responsibilities beyond the maximization of profit, this study has provided evidence that employees' perceptions of these organizational efforts serve as a preventive against their counterproductive behavior.In particular, the results revealed that AOC is positioned as the critical mediator in the links between perceived CSR and the three types of CWBs.

Theoretical Contributions and Managerial Implications
The current study contributes to the relevant research by investigating the links between perceived CSR and CWBs.More specifically, by linking employees' perceptions of CSR to their organization-, colleague-, and customer-directed CWBs, our model uncovers the hidden roles of CSR that have been neglected so far.On the one hand, scholarly attention regarding CWBs has mainly concentrated on the reason why employees engage in CWB in the workplace (e.g., [7][8][9]).Given the explicit threat and cost of CWBs, we consider it desirable to explore possible constraints against CWBs rather than investigate its triggers and identify the potential of CSR as a preventive against the three types of CWBs.On the other hand, researchers interested in CSR have tended to focus on the extent to which employees perceive their organization to be participating in CSR activities and policies, and thereby produce a variety of constructive and productive outcomes in the workplace (e.g., [16][17][18][19]).The present study contributes to the CSR literature field in extending the CSR research scope.
This study also provides evidence regarding why and how perceived CSR has an effect on CWB by seeking to identify the bridging role of affective commitment to the organization in the CSR-CWB linkage.This effort helps to advance the CSR literature because we attempted to answer the call by previous research (e.g., [10,15]) to account for the processes through which CSR activities result in specific outcomes.We proposed that employees holding the view that their organization demonstrates a concern for social issues and is ethically responsible are unlikely to conduct deviant behavior in the workplace.By grounding in social identity and social bonding theories, we provide a sound rationale and empirical support for the argument that perceived CSR is indirectly and negatively related to the three types of CWBs mediated through AOC rather than directly related to them.The findings of this study thus indicate that the premise could be only realized by employees' affective commitment to their organization.
In addition, this study offers practical recommendations for managers in organizations.We found that employees who perceive that their organization participates in CSR activities and policies experience a strong emotional bond with their organization and consequently tend not to engage in deviant behavior toward their organization, colleagues and customers.This suggests at least two strategies for preventing counterproductive behavior at work.First, managers should exactly understand how employees are receptive responsive to organizational CSR activities and policies, and then be certain to actively communicate with their staff regarding the organization's CSR plans and practices [18].It is basically useful for an organization to develop CSR policies, but if its employees do not actually recognize that fact, their negative behavior is unlikely to be mitigated.Therefore, a proactive strategy to help enhance CSR awareness among employees would ultimately have the effect of preventing their CWBs through increased AOC.
Considering the significance of AOC as the proximal antecedent of CWB, a second strategy would be to improve the level of employees' emotional attachment to their organization so as to prevent CWB more directly.As revealed in this study, affective commitment to the organization reduces the likelihood of CWB.Hence, in addition to participation and publicity in CSR, it is desirable for an organization to establish codes of ethics and quality procedures and to share them with its members because these codes and procedures create dialogue among employees about ethical issues, improve the level of its employees' affective commitment to the organization, and encourage them to behave ethically [55].Further, if an organization adopts and applies organizational management techniques including organizational support and effective leadership, as well as human resource management approaches such as high commitment work policies and practices (e.g., [35]), the level of its members' affective commitment to the organization will increase and their CWB will decrease.

Limitations and Future Study Directions
While this study presents theoretical contributions and practical implications, there are several limitations which need to be noted and potentially addressed in the future.First, we relied on self-report scales for the variables utilized in this study.We would expect future research to avoid the potential issues associated with common method bias by surveying different sources of response.Second, we focused on employees' perceptions of CSR in order to examine antecedents of counterproductive behavior at work.Our understanding of the relationship between CSR and CWB would be extended if a future study were to apply a multi-level analytic approach to an additional research model so as to investigate the effect of actual CSR activities at the organizational level on organizational members' attitudes and behavior at the individual level.Third, there is a need for improvement in terms of the study subjects.As this study focused on the hotel sector, a future study needs to incorporate companies belonging to other sectors (e.g., manufacturing and financial sectors) in order to explore the possibility of generalization of the research results.We also examined the frontline employees of hotels to comprehensively examine counterproductive behaviors at work including customer-directed CWB; however, we believe that it would be helpful to extend this study to backline employees (e.g., human resource management, finance, and accounting employees).It is necessary to identify the working groups in the hotel (e.g., front desks agents, event coordinators, and guest service representatives) to determine whether the degree of perceived CSR and the level of CWB are different as well.We hope that future research will deal with these issues.Lastly, we focused on a single attitudinal variable as the proximal antecedent of CWBs (i.e., affective commitment to the organization).However, previous studies have suggested that the differentiation of CWB targets allows logical reasoning that their triggers might also be distinguished (e.g., [37]).Accordingly, future research needs to explore the roles of various mediating variables to gain a clearer understanding of the relationships between CSR and the three types of CWBs.

Conclusions
The social responsibility of a corporation is sometimes referred to as its corporate conscience.By generating and testing a model to reveal the hidden roles of CSR, we have found that the CSR actions and policies of an organization are effective ways of leading employees to work with a personal conscience.That is, recognizing that an organization struggles to offer social benefits above and beyond meeting its legal obligations and maximizing profits for its shareholders, organizational members become affectively immersed in the organization to which they belong and, as a result, do not engage in counterproductive behavior toward their organization, coworkers, and customers.It is thus necessary for an organization to first actively participate in CSR and share it with its members so that its members naturally avoid the harmful actions to the objects inside and outside the organization.

corporate social responsibility Affective organizational commitment Organization-directed counterproductive work behavior Colleague-directed counterproductive work behavior Customer-directed counterproductive work behavior
, we would hypothesize that employees' perceptions of CSR would enhance their levels of AOC and thereby hinder the three types of CWBs. theories

Table 1 .
Results of the CFA and reliability test.

Table 2 .
Descriptive statistics and research variables inter-correlations in this research.

Table 3 .
Path coefficients and indirect effects for models.