Since the State-Owned Assets Supervision and Administration Commission (2008) issued guidelines for corporate social responsibility in China, these have become mandatory practice for both state-owned and nonstate-owned enterprises (Kong, 2022). Corporate social responsibility (CSR) is defined as the activities of organizations aiming to improve the welfare of various stakeholders (B.-J. Kim et al., 2021). It has received considerable research attention in the field of human resource management worldwide, including in Chinese contexts, for the past decade because of claims that it contributes to human resource management outcomes (Xiao et al., 2020). Studies have reported that CSR is beneficial for building positive employee attitudes and behaviors such as work engagement, organizational trust and commitment, citizenship behavior, and creativity (Gond et al., 2017; Xiao et al., 2020).

Voice behavior, a proactive behavior emphasizing expression of constructive challenge with the intent to improve a situation (Van Dyne & LePine, 1998), is a consequence of CSR. Ilkhanizadeh and Karatepe (2017) investigated the consequences of CSR in the airline industry and revealed that CSR practices foster flight attendants’ voice behavior. Liu et al. (2021) observed that CSR plays a positive role in facilitating voice behavior, and that both positive emotions and felt obligation mediate this relationship. Wang et al. (2020) revealed there are positive associations between CSR and both prohibitive and promotive types of voice, and noted that self-focused and other-focused climates mediate these connections. Memon et al. (2021) found that CSR has an indirect positive effect on voice behavior through the mediating mechanisms of organizational trust and psychological empowerment. K. A. Kim et al. (2018) examined the effect of employees’ negative perceptions of CSR on voice behavior, and showed there is a negative relationship between the constructs.

Despite these efforts, the mechanisms behind the relationship between CSR and voice behavior remain essentially unresolved. Voice behavior is an extrarole behavior that is not described in employment contracts (Van Dyne et al., 1995). Employees usually choose to remain silent or inactive and are reluctant to make suggestions or give opinions (Van Dyne et al., 1995). In particular, Chinese employees show less proactive behavior and are less likely to exhibit voice behavior than their Western counterparts due to belonging to a high power distance culture (Li & Sun, 2015).

Psychological safety has been reported to be a critical antecedent of voice behavior (Ge, 2020). Previous research has also supported the view that psychological safety plays a mediating role between voice behavior and organizational contextual factors (e.g., leadership and work environments; Yang & Wang, 2020; Zhang et al., 2020). Ahmad et al. (2019) revealed that psychological safety mediates the impact of CSR on employee creative behavior. Thus, it is plausible to expect that psychological safety will mediate the relationship between CSR and voice behavior. Figure 1 shows our conceptual model.

Method

Participants and Procedure

This study was conducted in accordance with the ethical guidelines of our universities. We collected data from employees in China with the assistance of the Chinese Enterprise Association. First, we explained the survey objectives and assured respondents of the anonymous and confidential nature of participation. Then, we distributed 324 paper-and-pencil questionnaires. Of the 258 questionnaires that were returned, 27 were discarded for inaccurate or incomplete responding (valid rate of return = 89.53%).

There were 133 (57.58%) men and 98 (42.42%) women in the sample (M_age = 35.48 years, SD = 8.73, range = 27–52). The average organizational tenure was 11.54 years (SD = 6.27, range = 3–26). Among the respondents 128 (55.41%) held a bachelor’s degree, 44 (19.05%) held a master’s degree, and 59 (25.54%) held a junior college degree or lower level of education. Regarding industry types, 85 (36.80%) worked in manufacturing, 53 (22.94%) in retail, 45 (19.48%) in hospitality, 41 (17.75%) in services, and seven (3.03%) in other industries.

Measures

The existing measures originally developed in English were translated and then back-translated to ensure semantic equivalency by two China-based bilingual researchers. All measures were previously validated. Responses were made on a 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree).

Corporate Social Responsibility

We measured CSR using five items from Rupp et al. (2018). A sample item is “A program is in place to reduce the amount of energy and materials wasted in our business.” Cronbach’s alpha in this study was .88.

Psychological Safety

We measured psychological safety using three items from Yang and Wang (2020), who developed their scale based on that of May et al. (2004). A sample item is “I feel completely comfortable in the environment at work.” Cronbach’s alpha in this study was .92.

Voice Behavior

We used Van Dyne and LePine’s (1998) six-item scale to measure voice behavior. A sample item is “I get involved in issues that affect the quality of life in the work unit.” Cronbach’s alpha in this study was .87.

Results

The descriptive statistics and intercorrelations among study variables are shown in Table 1. We conducted a confirmatory factor analysis (CFA) to examine the fit of the measurement model, which included CSR, psychological safety, and voice behavior. The three-factor model provided a good fit to the data, chi square/degrees of freedom = 2.46, incremental fit index = .92, comparative fit index = .92, root mean square error of approximation = .08. Composite reliability values were above .70. All factor loadings were significant (p < .01) and higher than .70. Average variance extracted values were above the threshold of .50 and higher than the square root of interconstruct correlations. These findings collectively supported the discriminant and convergent validity of the measures.

To test the proposed hypotheses we conducted structural equation modeling (SEM) and bias-corrected bootstrapping with 2,000 resamples. The SEM results did not support Hypothesis 1, as we found that CSR was not significantly related to voice behavior, .12, p > .05, 95% confidence interval (CI) [–0.06, 0.27]. Conversely, Hypotheses 2 and 3 were supported, as we found that CSR was significantly and positively related to psychological safety, .44, p < .01, 95% CI [0.30, 0.57] and that psychological safety was significantly and positively related to voice behavior, .51, p < .01, 95% CI [0.35, 0.65]. We also found that the indirect effect of CSR on voice behavior through psychological safety was significant and positive, .23, p < .01, 95% CI [0.13, 0.34]. Thus, Hypothesis 4 was supported.

Discussion

We examined whether and how CSR is related to voice behavior through psychological safety. We did not find support for the direct relationship between CSR and voice behavior; however, CSR was indirectly and positively related to voice behavior through the mediating mechanism of psychological safety. Our findings have several theoretical and practical implications.

Theoretical Implications

This study extends the micro-CSR literature, which has mainly emphasized employee attitudes and behaviors (Gond et al., 2017; Xiao et al., 2020). First, we did not find a significant direct relationship of CSR with voice behavior among employees in China. This indicates that CSR alone cannot spur employees’ voice behavior, which is inconsistent with the findings of previous researchers (Ilkhanizadeh & Karatepe, 2017; Liu et al., 2021; Memon et al., 2021; Wang et al., 2020). This may be because Chinese employees seldom speak out voluntarily and are reluctant to share opinions contradictory to the status quo when facing the system of hierarchical order in organizations (Li & Sun, 2015).

Second, our results offer empirical support for the positive relationship between CSR and psychological safety, and are consistent with those of Ahmad et al. (2019). As Bauman and Skitka (2012) noted, CSR can satisfy employees’ need for safety by enhancing organizational morality. Employees trust moral organizations, therefore leading to their perception of safety (Ahmad et al., 2019).

Finally, this study sheds light on the underlying process that explains the CSR–voice behavior link. Ng et al. (2019) called for research to uncover the mediating mechanisms of micro-CSR. In response, we have incorporated psychological safety into the link and revealed that it serves as a critical mediator in this relationship. These findings give important insights into the complex nature of the relationship between CSR and voice behavior. Considering the potential risk, employees are hesitant to voice their suggestions. Using CSR practices helps to improve the perception of safety, resulting in employees reciprocating to their organizations with constructive ideas.

Practical Implications

This study offers significant implications for organizations. First, although we did not find a direct link between CSR and voice behavior, it is beneficial for organizations to implement CSR activities based on their impact on psychological safety, which is positively connected with voice behavior. Organizations should focus on a CSR communications strategy to offer the necessary background and information to enhance employees’ perception and awareness of a CSR program. Moreover, this study found that CSR is not enough to facilitate voice behavior directly. Organizations should focus on improving employees’ psychological safety to encourage them to perform voice behavior.

Limitations and Recommendations for Future Research

This study has limitations that should be addressed. First, our use of a cross-sectional research design prevents us from drawing causal inferences. Future research could use longitudinal or experimental designs to provide convincing evidence of the causality of relationships. Second, although we revealed that psychological safety functions as a mediator of the link between CSR and voice behavior, we did not investigate the boundary conditions of this relationship, such as self-efficacy. Prior research has reported that employees with higher self-efficacy are more competent in achieving the desired outcomes (Gist & Mitchell, 1992). Thus, future research could explore the boundary conditions of the connection between CSR and voice behavior. Finally, our sample had a high level of education and there were more men than women, which raises the concern of the generalizability of our results. Future research could replicate our study using diverse samples.

This paper was supported by the Jilin Province Education Science Planning Project: Construction and Innovation of Ideological and Political Teaching Mode of a Tax Law Course Serving “Three Integrity Education” (GS2109).

Pingsheng Liu, College of Finance and Commerce, Zhongshan Torch Polytechnic, No. 60, Zhongshangang Avenue, Torch Development Zone, Zhongshan, Guangdong, 528436, People’s Republic of China. Email: [email protected]