Prosocial employees represent valuable assets for organizations. Research has shown that prosocial motivations which refer to desires to help others (Batson, 1987; Grant, 2008) have positive effects on a number of work outcomes such as organizational citizenship behaviors (Grant & Mayer, 2009; Rioux & Penner, 2001) and performance (Grant & Sumanth, 2009). However, without supportive work environments, employees’ prosocial motivations may not necessarily lead to positive outcomes, and the benefits of prosocial motivations may disappear over time. Thus, it is critical to examine conditions under which prosocially motivated employees produce more positive work outcomes, to enable organizations to develop supportive work environments that capitalize on these potential benefits. The main goal of this study, therefore, is to investigate when prosocial employees react to their work environment favorably.

Despite the importance of the topic, only a limited number of studies have examined the contextual factors that may foster and facilitate the positive effects of prosocial motivations on work outcomes. Grant (2007) proposed that enriched task characteristics (e.g., task significance, task identity, and autonomy) and social characteristics (e.g., interdependence, friendship opportunities, and interactions with insiders) may enhance the positive relationship between prosocial motives and repeated participation in corporate volunteering projects. Grant and Sumanth (2009) found that prosocially motivated employees were more likely to perform effectively when they perceived their managers as trustworthy. Grant (2008) also found that intrinsic motivation moderated the relationship between prosocial motivation and persistence, performance, and productivity. Prosocial motivation had a stronger positive relationship with persistence, performance, and productivity when intrinsic motivation was high rather than low. The present study extends this line of research by investigating how perceived prosocial work environments such as perceptions of supervisors’ prosocial motivations and an organization’s corporate social responsibility (CSR) synergized employees’ prosocial motivations to enhance their organizational commitment.

We choose to examine organizational commitment as a dependent variable for two reasons. First, from a theoretical perspective, we aim to expand existing knowledge about how employees with prosocial motivations develop psychological attachment to their organizations, particularly in the Asian context (cf. Morin, Meyer, McInerney, Marsh, & Ganotice, 2015; Seong, Hong, & Park, 2012). Research has suggested that employees’ prosocial orientations may potentially help to foster their organizational commitment (Grant, Dutton, & Rosso, 2008); however, no research that we know of has explicitly examined the question of whether employee prosocial motivation predicts employees’ organizational commitment, and more importantly, under which conditions do the benefits of prosocial motivation to organizational commitment exist or increase. Therefore, we choose to examine organizational commitment as a dependent variable in order to advance the current understanding of organizational commitment from a novel perspective (i.e., prosocial motivation). Second, from a practical perspective, organizational commitment is an important work outcome that has been found to decrease employees’ absenteeism and turnover (Meyer, Stanley, Herscovitch, & Topolnytsky, 2002). Many employees nowadays hold prosocial motives and values (Grant 2007, 2008), and hence research on how prosocial motivation contributes to organizational commitment can guide organizations’ efforts in increasing employee retention, which is of practical importance.

Based on person-environment (PE) fit theory (e.g., Kristof-Brown, Zimmerman, & Johnson, 2005) and social identity theory (e.g., Ashforth & Mael, 1989; Tajfel & Turner, 1985), we hypothesize that two types of bipartite synergy effects and a tripartite synergy effect will promote prosocial employees’ organizational commitment. Specifically, this study predicts that when employees and supervisors’ prosocial motivations are aligned at a high level (compared to when they are misaligned or aligned at a low level), employees’ organizational commitment will be stronger. Further, when employees’ prosocial motivations are aligned with their perception of their organizations’ CSR, employees will be more committed to their organizations. In addition, we predict that employees’ organizational commitment will be the strongest when employees and supervisors’ prosocial motivations and their perception of their organizations’ CSR are aligned at a high level.

This study contributes to the literature on prosocial motivations, PE fit, and CSR in several ways. First, it enriches the current models of prosocial motivation by examining the boundary conditions under which prosocial motivation is related to organizational commitment. Prosocial motivation is regaining popularity in the past few years and has been introduced to organizational studies (e.g., Grant, 2007, 2008; Grant & Berg, 2010; Grant & Sumanth, 2009); however, our understanding of the topic is still limited (for a review, see Grant & Berg, 2010). In particular, as many employees hold prosocial motives in organizations (Grant, 2007, 2008), understanding how their organizational commitment can be developed and sustained becomes important, both theoretically and practically. Although extant literature on prosocial motivation has examined benefits of prosocial motivation to organizations, such as persistence, performance, and productivity (e.g., Grant, 2008; Grant & Sumanth, 2009), no research has investigated whether and how prosocial motivation relates to organizational commitment. The present study identifies perceived supervisors’ prosocial motivation and CSR as facilitating factors that can predict the organizational commitment of employees with prosocial motivation. In doing so, our study enriches the literature on prosocial motivation, and advances the understanding of contextual factors that create and sustain the benefits of prosocial motivation to organizational commitment.

Second, PE fit theory has been linked with the literature on prosocial motivation. Different types of person-supervisor (PS) fit—for example, PS fit on personality (Schaubroeck & Lam, 2002; Zhang, Wang, & Shi, 2012), values (Ashkanasy & O’Connor, 1997; Kim & Kim, 2013), and goals (Witt, 1998)—have been shown to be associated with favorable employee outcomes. However, very few studies have specifically examined the prosocial dimension of PS fit (for an exception, see Cha, Chang, & Kim, 2014). Thus, this study enhances understanding of the effects of PS fit on employee outcomes by examining a new and specific content dimension of PS fit and exploring how it relates to organizational commitment.

Third, this study contributes to the growing literature on the internal impact of CSR and thus advances understanding on how employees perceive and react to their organization’s CSR. Previous research has focused on the external impact of CSR, but given little consideration to the link between employees’ perceptions of CSR and their work behaviors and attitudes (Brammer, Millington, & Rayton, 2007). The present study seeks to address this gap in the literature by examining how prosocial employees’ perceptions of their organizations’ CSR influence their organizational commitment. In doing so, insight is gained into the important relationships between employees’ perceptions of CSR and their work outcomes (Hofman & Newman, 2014).

Finally, this study expands the literature on PE fit by introducing a tripartite synergy effect in the prosocial orientations of employees, supervisors, and organizations. Previous research in the PE fit literature has focused either on PS fit (e.g., Ashkanasy & O’Connor, 1997; Schaubroeck & Lam, 2002; Zhang et al., 2012) or person-organization (PO) fit (Bretz & Judge, 1994; Cable & Judge, 1996; Cha et al., 2014), but failed to address the congruence among multiple aspects. This study highlights PE fit as a system concerning multiple aspects and levels, and suggests that the alignment of prosocial orientations among employees, supervisors, and organizations at a high level should have additional, positive impacts on employees’ organizational commitment beyond dyadic congruence either between employees and supervisors, or between employees and their organization. Thus, the present study advances the PE fit literature by highlighting not just the bipartite synergy (e.g., PS fit), but also tripartite synergy among employees, supervisors, and organizations.

Literature review and hypotheses development

Prosocial motivation is gaining popularity in organizational studies in recent years, and increasing research has been dedicated to examining the potential benefits of prosocial motivation to organizations (Grant, 2007, 2008; Grant & Berg, 2010; Grant & Sumanth, 2009). While prosocial motivation has been shown to increase task effort, persistence, and organizational citizenship behaviors (e.g., Grant, 2007; Rioux & Penner, 2001), the realization and sustaining of these benefits require that employees are committed to and want to stay in the organization in the first place. However, little research that we know of has investigated how to develop and sustain the organizational commitment of employees with prosocial motivation. Grant et al. (2008) examined how employee support programs may strengthen commitment by fulfilling employees’ other-interested motives to give, and suggest that employees who are prosocially motivated (i.e., those give to a support program) are likely to develop organizational commitment through a sensemaking process. However, the direct effect of prosocial motivation on organizational commitment is yet to be validated, and more importantly, researchers have begun to challenge the assumption of a direct relationship between prosocial motivations and positive work outcomes (Grant & Berg, 2010).

Consistent with recent studies examining contingencies that moderate the relationship between prosocial motivation and positive work outcomes (Grant, 2008; Grant & Mayer, 2009; Grant & Sumanth, 2009), the present study aims to identify factors that help predict the organizational commitment of employees with high prosocial motivation. In the following sections, this study contends the prosocial orientation of employees’ supervisors and organizations play important roles in the creation of synergies to enhance the organizational commitment of prosocial employees.

Bipartite synergy between employees and supervisors’ prosocial motivations

The present study draws on the PE fit literature to examine how the fit between employees and supervisors’ prosocial motivation could affect the organizational commitment of prosocial employees. PS fit is a specific type of PE fit that includes personality, attitude, and value fit and can produce positive outcomes for employees and organizations such as enhancing the quality of leader-member exchanges (e.g., Ashkanasy & O’Connor, 1997; Kristof-Brown et al., 2005; Phillips & Bedeian, 1994; Zhang et al., 2012), and improving behavioral integration, promotion (Schaubroeck & Lam, 2002) and performance (Strauss, Barrick, & Connerley, 2001). The present study examines PS fit in relation to prosocial motivation, a factor not examined by any previous PS fit studies. Similar to studies that examined PS fit in relation to personality or values (Ashkanasy & O’Connor, 1997; Zhang et al., 2012), this study defines PS fit as the compatibility between employees and their supervisors in relation to their prosocial motivation levels. Prosocial motivation has been the subject of extensive research; however, to date, no research appears to have examined how fit and misfit between employees and supervisors’ prosocial motivations affect work outcomes such as organizational commitment.

This study conjectures that in relation to prosocial motivation, PS fit will be positively related to organizational commitment. According to the PE fit literature (Edwards & Cable, 2009; Kristof-Brown et al., 2005), the alignment of important attributes (e.g., values, personality, and attitude) between employees and their supervisors may foster various positive employee outcomes. Zhang et al. (2012) found that when the proactive personalities of leaders and followers are congruent, followers have better exchanges with their leaders, higher job satisfaction, and stronger affective commitment to their organizations than when the personalities of leaders and followers are incongruent. Similarly, when the prosocial motivation of employees and their supervisors are congruent, positive outcomes such as strong organizational commitment are more likely to be fostered. This is because, when employees perceive that their prosocial motivation is aligned with their supervisors’ prosocial motivation, they are likely to believe that they have similar goals with their supervisors. This belief of goal congruence will make it easier for employees and supervisors to set up mutually agreed upon role expectations in the workplace, reduce possible conflicts in work processes, and enhance the quality and efficiency of interactions (Edwards & Cable, 2009; Zhang et al., 2012). In addition, when employees perceive that they have similar levels of prosocial motivation with their supervisors, they are more likely to feel that they are supported by their supervisors, because people tend to help those who are similar to them (Penner, Dovidio, Piliavin, & Schroeder, 2005). Accordingly, together, the benefits associated with the perceived fit between employees and supervisors’ prosocial motivation, such as perceived goal congruence, reduced conflicts, quality relationship and perceived supervisory support, should lead to employees’ high levels of organizational commitment (e.g., Kuvaas & Dysvik, 2010; Zhang et al., 2012).

Conversely, when there is a perceived misfit of prosocial motivations between employees and supervisors, employees may not feel strongly committed to their organizations; for example, if employees perceive that they have higher levels of prosocial motivation than their supervisors, employees may feel that their prosocial needs may not be met and their goals may not be recognized as they expected. Additionally, they may feel that the environment is insecure due to the incongruence between motivations and goals (Edwards & Cable, 2009), and thus may be less committed to working at their organizations. On the other hand, when employees perceive that they have lower levels of prosocial motivation than their supervisors, they may feel pressured to get involved in prosocial activities that are not consistent with their aspirations and goals. Consequently, when the prosocial motivations of employees and their supervisors are perceived to be incongruent, employees may not feel strongly attached to their organizations.

In addition, PS fit may occur at either high or low levels. It is expected that when employees and supervisors’ prosocial motivations are perceived to be both high, organizational commitment will be higher than when their motivations are perceived to be both low. Prosocial motivation is closely associated with other-directed qualities, including benevolence (i.e., concern for other’s well-being; Meglino & Ravlin, 1998), and integrity (i.e., social justice and sincerity; Marcus, Lee, & Ashton, 2007). Thus, when employees with high prosocial motivations work with supervisors with high prosocial motivations, the perceived alignment between their prosocial goals and values should foster the development of better mutual understandings and relationships and encourage them to take care of others and their organizations (Dirks & Ferrin, 2002; Gerstner & Day, 1997; Zhang et al., 2012). Conversely, if employees perceive that they and their supervisors have low levels of prosocial motivation, employees may still benefit from goal congruence; however, because both parties are less motivated to benefit each other and other colleagues due to their low level of prosocial motivation, employees may feel that the relationships between employees and supervisors, and between employees and their coworkers, will be characterized as being self-interested rather than other-directed, compared to when both employees and supervisors are high in prosocial motivation. Research has shown that an ethical climate of benevolence is positively related to organizational commitment, whereas an egoistic climate is negatively related to organizational commitment (Cullen, Parboteeah, & Victor, 2003). Therefore, it is likely that these employees who perceive that their prosocial motivation is convergent with their supervisors’ prosocial motivation at a low level will be less committed to their organizations.

Hypothesis 1a

PS fit in relation to prosocial motivation is positively related to organizational commitment.

Hypothesis 1b

Organizational commitment is higher when both employees and perceived supervisors’ prosocial motivations are high than when both are low.

Bipartite synergy between employees’ prosocial motivations and their perceptions of corporate social responsibility

Perceived CSR refers to the prosocial orientation of an organization and can critically affect the organizational commitment of employees with strong prosocial motivations. CSR covers a broad range of factors, including all prosocial organizational activities and outcomes other than those pursued to maximize profits (Houghton, Gabel, & Williams, 2008). CSR has been conceptualized and measured in a variety of ways (cf. Garriga & Mele, 2004). In the present study we define CSR as an expressed commitment to solving problems in a community and society as a whole (Goll & Zeitz, 1991). Thus, this definition encompasses philanthropy and community contributions, and reflects the stances of organizations toward the physical environment, consumers, and other external stakeholders (Brammer et al., 2007; Carroll, 1979). The idea of CSR on the organizational level is also consistent with prosocial motivation on the individual level—although prosocial motivation concerns individuals’ prosocial orientation, and CSR concerns organizations’ prosocial orientation, both are other-directed features (Grant, 2008; Bénabou & Tirole, 2010; Zhu, Sun, & Leung, 2014).

To date, most CSR research has focused on the effect CSR on external stakeholders; however, “few studies have investigated the impact of external CSR strategies on internal stakeholders and in particular work attitudes” (Brammer et al., 2007: 1703). Emerging research has examined the psychology of CSR and focused on how employees perceive and subsequently react to CSR. In general, existing literature has provided evidence that perceived CSR is positively related employees’ job satisfaction (Valentine & Fleischman, 2008), and organizational citizenship behaviors (Lin, Lyau, Tsai, Chen, & Chiu, 2010; Rupp, Shao, Thornton, & Skarlicki, 2013). However, research has yielded conflicting findings regarding how perceived CSR relates to employees’ organizational commitment. One the one hand, some research has shown a strong positive relationship between perceived CSR and employees’ organizational commitment (e.g., Maignan, Ferrell, & Hult, 1999; Peterson, 2004; Turker, 2009); On the other hand, some studies did not find or only found marginally significant relationship between perceived CSR and employees’ organizational commitment (e.g., Hofman & Newman, 2014; Kim, Lee, Lee, & Kim, 2010). This suggests that perceived CSR may not produce organizational commitment in all employees. Whether perceived CSR leads to organizational commitment depends on employees’ personal characteristics. In the present study, we propose that those employees who have a high level of prosocial motivation may develop strong organizational commitment when they perceive their organizations to be high in CSR.

Social identity theory (Ashforth & Mael, 1989; Tajfel & Turner, 1985) states that individuals see themselves as members of social categories (e.g., members of the organizations with which they are affiliated). Employees often identify with and commit to organizations that have positive organizational values and reputation (Kim et al., 2010; Peterson 2004) to establish or enhance positive self-concepts (Ashforth & Mael, 1989; Tajfel & Turner, 1985). It is predicted that this tendency will be particularly strong among employees with strong prosocial motivations. Specifically, it is expected that prosocially motivated employees will have stronger organizational commitment when they perceive their organizations as having high levels of CSR. This is because, individuals “choose activities congruent with salient aspects of their identities, and support the institutions embodying those identities” (Ashforth & Mael, 1989: 25). Thus, when organizations engage in high levels of CSR, prosocially motivated employees should be more committed to their organizations, as these CSR activities should align with their identities and motivations. Conversely, when prosocially motivated employees work in organizations that are not socially responsible, they are more likely to disengage from organizational activities and thus, be less committed to their organizations. In support of this view, Peterson (2004) found that perceived corporate citizenship and organizational commitment is stronger among employees who believe strongly in the importance of the social responsibility of their organizations.

Hypothesis 2

Employees’ perceptions of CSR moderate the relationship between employees’ prosocial motivations and their organizational commitment such that positive relationships between employees’ prosocial motivations and organizational commitment will be stronger when CSR is perceived as high rather than low.Footnote 1

Tripartite synergy among employees and supervisors’ prosocial motivations and perceptions of corporate social responsibility

In addition to both bipartite synergies, congruency between the three factors may produce additional benefits. Previous PE research has largely focused on either PS fit or PO fit; however, it has overlooked the “tripartite synergy” that can occur when there is congruence among employees, supervisors, and organizations. The present study conjectures that employees will have high levels of organizational commitment when there are high levels of congruence among employees and supervisors’ prosocial motivations, and organizational CSR (i.e., at high rather than low level of CSR, PS fit in relation to prosocial motivation should have a stronger, positive effect on employees’ organizational commitment). Previous research on domain centrality (e.g., Edwards & Rothbard, 1999) has shown that the degree to which one domain is considered important, may influence the strength of fit in the domain and its outcomes. In other words, the more central and important one domain is in an organization, the more benefits will be produced by the congruence between employees and organizations’ orientations in that domain. For example, Edwards and Rothbard (1999) found that as the importance of work (or family) to an employee increased (i.e., as centrality of work or family increased), the relationship between work (or family) supply-value fit and well-being generally became stronger. Similarly, if an organization is perceived as high in CSR, employees will likely believe that prosocial values and orientations are considered as central and important (i.e., high in centrality) to the organization. Consequently, employees will feel that their convergent prosocial orientation (i.e., PS fit) with their supervisors is recognized and supported. Therefore, the benefits involved in PS fit on prosocial motivation, such as goal congruence, similar expectations, low conflict, quality interactions between employees and supervisors, will be increased due to the prosocial organizational environment, resulting in employees’ strong organizational commitment. Conversely, in organizations in which employee perceive that prosocial orientation is peripheral rather than central (i.e., low centrality) to the organization, the prosocial motivation fit between employees and their supervisors will not be recognized and supported by the organization. Therefore, employees will feel that even their prosocial motivation is congruent with their supervisors’ prosocial motivation, this congruence as well as any prosocial activities resulted from this congruence may not receive any credit from the organization. Consequently, PS fit on prosocial motivation is less likely to lead to employees’ organizational commitment. Thus, we hypothesize that:

Hypothesis 3

CSR moderates the positive impact of PS fit in relation to prosocial motivation on organizational commitment so that the impact is stronger at high rather than low CSR.

Methods

Sample

A professional survey platform in China (www.sojump.com) was used to collect data from 216 full-time Chinese employees. Online data collection has been shown to be as effective as traditional methods of social science research (e.g., Adam & Shirako, 2013; Jacquart & Antonakis, 2015; Trau, Härtel, & Härtel, 2013). A Chinese sample was used to test the four hypotheses, as retaining talent in emerging markets such as China has become a significant challenge (Ready, Hill, & Conger, 2008). The participants in the present study were relatively young (63.5% were aged 21–30). Of the participants, 57.4% were male, 31% had 6–10 years work experience, and 17.6% had 3–5 years work experience.

Measures

Unless otherwise noted, all multi-item scales were measured using a 5-point Likert scale (where 1 = “Strongly disagree” and 5 = “Strongly agree”). All measures were translated from English to Chinese.

Employees’ prosocial motivations

A 4-item scale was used to measure prosocial motivation (Grant, 2008). The participants were first asked the following an introductory question: “Why are you motivated to do your work?”. An example item of the scale is “Because I care about benefiting others through my work.” In the present study, the reliability of the scale was α = .80.

Supervisors’ prosocial motivations

Supervisors’ prosocial motivations were assessed using employees’ perceptions (Caplan, 1987; Edwards & Cable, 2009; Endler & Magnusson, 1976). The same 4-item scale was used as the scale for employees’ prosocial motivations with the exception that the referent in the items referred to “my supervisor.” The reliability of the scale in the present study was α = .86.

Perceived corporate social responsibility

A 3-item scale was used to measure CSR (Goll & Zeitz 1991). An example scale item is: “My company believes in performing in a manner consistent with the philanthropic and charitable expectations of society.” In the present study, the reliability of the scale was α = .80.

Organizational commitment

The Organizational Commitment Questionnaire (Mowday, Steers, & Porter, 1979) was used to measure organizational commitment. This 9-item scale has been used extensively in previous studies (e.g., Bishop & Scott, 2000). An example item of the sale is: “I am willing to put in a great deal of effort beyond that normally expected in order for my company to be successful.” In the present study, the reliability of the scale was α = .92.

Control variables

The following variables were controlled: age, industry, and work tenure. Adopting the approach of Farh, Hackett, and Liang (2007), the age of participants was divided into eight categories: 15 years and under, 16–20 years, 21–25 years, 26–30 years, 31–40 years, 41–50 years, 51–60 years, and over 60 years. Six categories were used to measure work tenure: 1 year or less, 2–5 years, 6–10 years, 11–20 years, and over 20 years. Industry was coded as follows: zero indicated a “manufacturing industry,” and one indicated a “service industry.”

Participants’ value orientations for materialism (i.e., a set of centrally held beliefs about the importance of possessions in one’s life) were also controlled for (Richins & Dawson, 1992). Materialists are self-centered and unconcerned about others (Richins & Dawson, 1992). Therefore, materialism is at odds with prosocial motivations which represent a desire to benefit others (Batson, 1987; Grant, 2008). Materialism may affect how employees perceive their supervisors’ prosocial motivations and the organization’s CSR. For example, Kolodinsky, Madden, Zisk, and Henkel (2010) found that those participants who had higher levels of materialistic values had a negative attitude about business having a social responsibility beyond profit maximization. Materialism was also found to negatively affect people’s environmental concerns and environmentally responsible behaviors (Kilbourne & Pickett, 2008; Hurst, Dittmar, Bond, & Kasser, 2013), suggesting that individuals who score higher on materialism are less likely to be responsive to supervisor’s prosocial motivation and CSR which may involve environmental concerns that benefits other people. Therefore, it is possible that the joint effects of employee prosocial motivations and supervisor prosocial motivations, or the interactive effects of employee prosocial motivation and CSR perception, on organizational commitment can be explained by the (negative) interactive effects of employee materialism with supervisor prosocial motivation or with CSR perception. Accordingly, to rule out this alternative explanation, materialism was measured using an 18-item scale (Richins & Dawson, 1992; α = .82), and was controlled for in subsequent analyses.Footnote 2

Analysis

A polynomial regression analysis was performed to test the relationships between PS fit and organizational commitment (Edwards & Parry, 1993; Shanock, Baran, Gentry, Pattison, & Heggestad, 2010). Polynomial regression analyses have been frequently used in research on PE fit (e.g., Edwards & Cable, 2009; Van Vianen, De Pater, Kristof-Brown, & Johnson, 2004), and have a great advantage over the method of difference scores used in earlier organizational studies of fit (cf. Edwards & Parry, 1993). The following equation was formulated to test the effects of PS fit on organizational commitment:

$$ \mathrm{Organizational}\ \mathrm{commitment} = {\mathrm{b}}_0 + {\mathrm{b}}_1\mathrm{P} + {\mathrm{b}}_2\mathrm{S} + {\mathrm{b}}_3{\mathrm{P}}^2 + {\mathrm{b}}_4\mathrm{PS} + {\mathrm{b}}_5{\mathrm{S}}^2+\mathrm{e} $$
(1)

Where, P and S represent employee and supervisors’ prosocial motivations, respectively.Footnote 3

The following equation (that substituted an incongruence line (i.e., S = −P) into Eq. (1) was formulated to test Hypothesis 1a:

$$ \begin{array}{c}\hfill \mathrm{Organizational}\ \mathrm{commitment} = {\mathrm{b}}_0 + {\mathrm{b}}_1\mathrm{P}\hbox{--}\ {\mathrm{b}}_2\mathrm{P} + {\mathrm{b}}_3{\mathrm{P}}^2\hbox{--}\ {\mathrm{b}}_4{\mathrm{P}}^2 + {\mathrm{b}}_5{\mathrm{P}}^2+\mathrm{e}\hfill \\ {}\hfill = {\mathrm{b}}_0+\left({\mathrm{b}}_1\hbox{--}\ {\mathrm{b}}_2\right)\mathrm{P}+\left({\mathrm{b}}_3\hbox{--}\ {\mathrm{b}}_4 + {\mathrm{b}}_5\right){\mathrm{P}}^2+\mathrm{e}\hfill \end{array} $$
(2)

According to Edwards and Parry (1993), if Hypothesis 1a was supported, (b3− b4 + b5) should be negative, and statistically significant.

Hypothesis 1b predicted that organizational commitment was higher when both employees and supervisors’ prosocial motivations were higher than when they are both were low. This hypothesis was tested by substituting congruence line (i.e., S = P) into Eq. (1) (Edwards & Parry, 1993). Thus:

$$ \begin{array}{c}\hfill \mathrm{Organizational}\ \mathrm{commitment} = {\mathrm{b}}_0 + {\mathrm{b}}_1\mathrm{P}+{\mathrm{b}}_2\mathrm{P} + {\mathrm{b}}_3{\mathrm{P}}^2+{\mathrm{b}}_4{\mathrm{P}}^2 + {\mathrm{b}}_5{\mathrm{P}}^2+\mathrm{e}\hfill \\ {}\hfill = {\mathrm{b}}_0+\left({\mathrm{b}}_1+{\mathrm{b}}_2\right)\mathrm{P}+\left({\mathrm{b}}_3+{\mathrm{b}}_4 + {\mathrm{b}}_5\right){\mathrm{P}}^2+\mathrm{e}\hfill \end{array} $$
(3)

If Hypothesis 1b was supported, the results should show that (b1 + b2) were positive and statistically significant, and (b3 + b4 + b5) should not differ from 0 (Edwards & Rothbard, 1999; Zhang et al., 2012).

In relation to Hypothesis 2, a hierarchical regression analysis was undertaken to test the moderating effect of perceived CSR on the linkages between employees’ prosocial motivations and organizational commitment (Cohen & Cohen, 1983).

Finally, a hierarchical regression analysis was undertaken to test the moderating effects of perceived CSR on the relationship between PS fit and organizational commitment (see Hypothesis 3). Different from linear regressions where the statistical significance of the coefficient for a three-way interaction should be assessed to establish a three-way moderating effect, in polynomial regressions, the increment in R 2 after adding a moderator and products of the moderator with each of the original terms should be assessed to establish the moderating effect (Edwards & Rothbard, 1999).Footnote 4 Specifically, the following equation was formulated to test Hypothesis 3:

$$ \begin{array}{l}\mathrm{Organizational}\ \mathrm{commitment} = {\mathrm{b}}_0 + {\mathrm{b}}_1\mathrm{P} + {\mathrm{b}}_2\mathrm{S} + {\mathrm{b}}_3{\mathrm{P}}^2 + {\mathrm{b}}_4\mathrm{PS} + {\mathrm{b}}_5{\mathrm{S}}^2 + {\mathrm{b}}_6\mathrm{CSR}+\\ {}{\mathrm{b}}_7\mathrm{CSR}\ast \mathrm{P} + {\mathrm{b}}_8\mathrm{CSR}\ast \mathrm{S} + {\mathrm{b}}_9\mathrm{CSR}\ast {\mathrm{P}}^2 + {\mathrm{b}}_{10}\mathrm{CSR}\ast \mathrm{P}\ast \mathrm{S} + {\mathrm{b}}_{11}\mathrm{CSR}\ast {\mathrm{S}}^2+\mathrm{e}\end{array} $$
(4)

The following equation was formulated as a result of rewriting Eq. (4):

$$ \begin{array}{l}\mathrm{Organizational}\ \mathrm{commitment}=\left({\mathrm{b}}_0 + {\mathrm{b}}_6\mathrm{CSR}\right)+\left({\mathrm{b}}_1 + {\mathrm{b}}_7\mathrm{CSR}\right)\ast \mathrm{P}+\left({\mathrm{b}}_2 + {\mathrm{b}}_8\mathrm{CSR}\right)\ast \mathrm{S}+\\ {}\left({\mathrm{b}}_3 + {\mathrm{b}}_9\mathrm{CSR}\right)\ast {\mathrm{P}}^2+\left({\mathrm{b}}_4 + {\mathrm{b}}_{10}\mathrm{CSR}\right)\ast \mathrm{P}\ast \mathrm{S}+\left({\mathrm{b}}_5 + {\mathrm{b}}_{11}\mathrm{CSR}\right)\ast {\mathrm{S}}^2+\mathrm{e}\end{array} $$
(5)

From Eq. (5), the moderation effect of CSR should depend on the joint effect of b7, b8, b9, b10, and b11. If the increment in R 2 is statistically significant, further interpretation can be made by looking at the shape of the response surfaces at different levels of the moderator. Similar to Eq. (2), the negative (i.e., downward) curvature along the S = −P line can be tested at different levels of CSR.

Results

Discriminant validity of focal constructs

Confirmatory factor analyses were conducted to assess the discriminant validity of the variables used in this study (i.e., employees’ prosocial motivations, supervisors’ prosocial motivations, CSR, and organizational commitment). The results indicated that the four-factor model (χ 2 = 310.66, df = 164, root-mean square error of approximation [RMSEA] = .06, comparative fit index [CFI] = .94, and Tucker-Lewis index [TLI] = .93) fit the data well and better than the best-fitting three-factor model that treated employees’ prosocial motivations and supervisors’ prosocial motivations as one factor (χ 2 = 401.16, df = 167, Δχ 2 = 90.50, p < .01; RMSEA = .08; CFI = .91; TLI = .90), the best-fitting three-factor model that treated supervisors’ prosocial motivations and perceived CSR as one factor (χ 2 = 406.47, df = 167, Δχ 2 = 95.81, p < .01; RMSEA = .08; CFI = .91; TLI = .89), the best-fitting two-factor model that treated employees and supervisors’ prosocial motivations as one factor and CSR and organizational commitment as one factor (χ 2 = 518.21, df = 169, Δχ 2 = 207.55, p < .01; RMSEA = .10; CFI = .86; TLI = .85), and the one-factor model (χ 2 = 638.85, df = 170, Δχ 2 = 328.19, p < .01; RMSEA = .11; CFI = .82; TLI = .80). These results provided support for the discriminant validity of the constructs used in this study.

Addressing the issue of common method biases

We used two techniques to address the concern of common method biases potentially produced by the same raters (i.e., employees). The first technique was Harman’s single-factor test which is one of the most widely used techniques in the literature (Podsakoff, MacKenzie, Lee, & Podsakoff, 2003). All items from focal constructs in the study were loaded on one factor to determine whether the one-factor model fits the data well. Based on the results from assessing the discriminant validity of the variables used in the study, the one-factor model (χ 2 = 638.85, df = 170; RMSEA = .11; CFI = .82; TLI = .80) was statistically worse than the hypothesized four-factor model (χ 2 = 310.66, df = 164, RMSEA = .06, CFI = .94; TLI = .93), Δχ 2 = 328.19, p < .01. This provides evidence that common method biases did not pose a serious threat to the testing of our hypotheses (cf. Dulac, Coyle-Shapiro, Henderson, & Wayne, 2008; Iverson & Maguire, 2000; Mossholder, Bennett, Kemery, & Wesolowski, 1998).

However, Harman’s single-factor has its limitations. Therefore, we also used a second method by adding a latent method factor to the hypothesized four-factor model, recommended by Podsakoff et al. (2003). In this approach, items are allowed to load on their theoretical constructs, as well as on a latent common methods variance factor. This method model allowed us to explore the potential increase in model fit due to the addition of the method factor, as well as the variance extracted by this factor. We specified that the method factor was not correlated with any of the four hypothesized factors (employee prosocial motivation, supervisor prosocial motivation, CSR, and organizational commitment). This model fit the data slightly better than the hypothesized four-factor model (χ 2 = 303.78, d f = 164, RMSEA = .06, CFI = .95; TLI = .94). However, the average variance extracted by the common methods factor was only .35, falling below the .50 cut-off point that was proposed as indicating the presence of a substantive methods factor (Fornell & Larcker, 1981), and similar to the average variance extracted by the common method factor in other published works (e.g., Baron, Franklin, & Hmieleski, 2013; Dulac et al., 2008). Therefore, although the common method variance seemed to be present in our data, it did not appear to pose a serious threat to the testing of our hypotheses.

Descriptive statistics

Table 1 sets out the descriptive statistics, reliability estimates, and correlations for all measures used in this study. The zero-order correlations between employees’ prosocial motivations, supervisors’ prosocial motivation, CSR and organizational commitment were all significant.

Table 1 Means, standard deviations, correlations, and reliabilities for all variables
Full size table

Hypothesis testing

Table 2 sets out the results for the polynomial regressions and hierarchical regression analyses in relation to the effects of employees’ prosocial motivations, supervisors’ prosocial motivations, and employees’ perceptions of CSR on the employees’ organizational commitment. Hypothesis 1a proposed a fit effect that suggested that organizational commitment would increase as employees’ prosocial motivations increased toward supervisors’ prosocial motivations, and would decrease as employees’ prosocial motivations exceeded supervisors’ prosocial motivations (i.e., a curvilinear relationship). The curvature along the incongruence line (P = −S) for organizational commitment was lower than zero statistically (b3− b4 + b5 = −.35, p < .05) (see Table 2). The response surface curved downward along the incongruence line (i.e., the diagonal from the upper-left corner to the lower-right corner on the bottom surface) (see Fig. 1). Thus, Hypothesis 1a was supported.

Table 2 Results of polynomial regression analyses for prosocial motivation fit and organizational commitment and moderation effects of corporate social responsibility
Full size table
Fig. 1
figure 1

Response surface depicting hypothesized relationship between PS fit on prosocial motivation and organizational commitment

Full size image

Hypothesis 1b stated that organizational commitment would be higher when employees and supervisors’ prosocial motivations were both high (as compared to when they were both low). The coefficient for (b1 + b2) was positive and statistically significant (b1 + b2 = .77, p < .01), whereas the coefficient for (b3 + b4 + b5) was not different from zero statistically (b3 + b4 + b5 = .10, p > .05) (see Table 2). Further, the slope was upward along the P = S line (i.e., the diagonal from the lower-left corner to the upper-right corner on the bottom surface), showing that organizational commitment increased as both employees and supervisors’ prosocial motivations increased from low to high (see Fig. 1). Thus, Hypothesis 1b was supported.

Hypothesis 2 stated that perceptions of CSR would moderate the relationship between employees’ prosocial motivations and their organizational commitment, and, more specifically, that the relationship would become stronger when CSR was high rather than low. The results from hierarchical regression analyses showed that the effect of interaction terms was significant (β = .16, p < .01) (see Table 2). Tests of the simple slopes showed that the positive relationship between employees’ prosocial motivations and organizational commitment was weaker when CSR was low (β = .31, p < .01) than when CSR was high (β = .53, p < .01). To further interpret the nature of the interaction, the simple slopes of the relationship between employees’ prosocial motivations and organizational commitment were plotted at one standard deviation below and above the mean of CSR perceptions (see Fig. 2). The slope was flatter at low CSR than at high CSR (see Fig. 2). Thus, Hypothesis 2 was supported.

Fig. 2
figure 2

Simple slopes of employee prosocial motivation on organizational commitment at levels of CSR

Full size image

Hypothesis 3 stated that as perceptions of CSR increased, the positive relationship between PS fit and organizational commitment would become stronger. The third column in Table 2 shows that R 2 increased by adding the interaction terms between the moderator and each of the five terms in the original polynomial regression equation (ΔR 2 = .06, p < .01). Figures 3a and b demonstrate the response surfaces when perceptions of CSR were high and low (one standard deviation above and below the mean, respectively). Specifically, results from the simple surface tests reveal that a negative (i.e., downward) curvature along the P = −S line was steeper when CSR was high (estimate of curvature = −.81, p < .05) than when it was low (estimate of curvature = −.24, p > .10), indicating that the effects of PS fit on organizational commitment were stronger when CSR was high rather than low. Thus, Hypothesis 2 was supported.

Fig. 3
figure 3

Response surface of hypothesized relationship between PS fit and organizational commitment when CSR was high (a) and low (b)

Full size image

Discussion

Recently, prosocial motivations have been the subject of increasing attention both in research and practice (e.g., Baston, Ahmad, Powell, & Stocks, 2008; Grant & Mayer, 2009; Grant & Sumanth, 2009). The present study examined the factors that strengthen the relationship between prosocial motivations and organizational commitment. First, the results suggested that supervisors’ prosocial motivations may have synergized with employees’ prosocial motivations to foster employees’ organizational commitment. Second, organizational prosocial orientation (i.e., CSR) also may have synergized with employees’ prosocial motivation to enhance the organizational commitment. Third, the alignment among employees’ prosocial motivations, supervisors’ prosocial motivations, and organizational CSR created a tripartite synergistic effect on employees’ organizational commitment. The finding of this study makes contributions to both theories and practice.

Theoretical and practical implications

This study made several theoretical contributions to the current literature. To begin with, the present study was the first one that explicitly examined the relationship between prosocial motivation and organizational commitment, and factors that may influence this relationship. The study identified supervisors’ prosocial motivations as a facilitating factor that predicted the organizational commitment of employees with prosocial motivations. Specifically, it found that the higher the PS fit in relation to prosocial motivation, the stronger an employee’s organizational commitment. It also showed that levels of congruence on prosocial motivation were consequential (i.e., when the alignment of employees and supervisors’ levels of prosocial motivation was high rather than low, their organizational commitment was stronger). In summary, this study drew on the PE fit literature to examine prosocial motivations, and provided a new perspective by studying the contextual factors that predict the organizational commitment of employees with strong prosocial motivations.

Second, the findings showed that an organization’s prosocial orientation resonated strongly among prosocially inclined employees. Grant and Berg (2010) called for research that investigated how organizations can spark and sustain employees’ prosocial motivations. This study found a moderating effect of perceived organizational CSR on the relationship between employees’ prosocial motivations and their organizational commitment. Thus, when organizations have a prosocial orientation, employees with high levels of prosocial motivation are likely to commit to their organizations. These findings are consistent with research on PO fit (e.g., Cha et al., 2014; O’Reilly, Chatman, & Caldwell, 1991; Sekiguchi, 2006) that emphasized the importance of compatibility of certain values between employees and their organizations. This finding also added to the literature on prosocial motivation by identifying a new contextual factor—perception of CSR—(in addition to task characteristics [Grant, 2007] and manager trustworthiness [Grant & Sumanth, 2009]) at the organizational level.

Importantly, the findings enrich the CSR literature by examining the micro-level impact of CSR, and addressing Peterson’s (2004) concern that the benefits of CSR in relation to internal stakeholders have been understudied. In particular, among the limited number of studies, findings regarding whether CSR benefits employees’ organizational commitment were conflicting and inclusive. Consistent with this line of research, we found that the perception of an organization’s external CSR has a positive relationship with the organizational commitment of employees with prosocial motivation. It extends previous research in the way that perceived CSR does not necessarily lead to organizational commitment (Hofman & Newman, 2014), but instead, whether perceived CSR benefits organizational commitment depends on employees’ personal characteristics, such as their prosocial motivation. For example, Hofman and Newman (2014) also conducted a study with a sample of Chinese employees, and found that employees’ perceptions of external CSR had a non-significant or marginally significant relationship with organizational commitment. However, these results do not contradict the present study’s findings, as Hofman and Newman (2014) examined the main effect of employees’ CSR perception on their organizational commitment, whereas the present study investigated the moderating effect of employees’ CSR perception on the relationship between their prosocial motivation and organizational commitment. Consequently, this study complements and extends Hofman and Newman’s findings, as it suggests that while overall, employees’ perceptions of external CSR have less impact on employees’ organizational commitment, the CSR perceptions are important in predicting the organizational commitment of those employees who have strong prosocial motivation.

Finally, this study showed a tripartite synergistic effect among employees, supervisors, and organizations in relation to their prosocial orientations, in addition to bipartite synergy effects between employees and supervisors, and employees and organizations. The findings showed that optimal outcomes (e.g., improving employees’ organizational commitment) can be predicted when employees, supervisors, and their organizations are all high in their prosocial orientations. This finding extends previous research that focused on bipartite fits between employees and their supervisors (Ashkanasy & O’Connor, 1997; Kim & Kim, 2013; Schaubroeck & Lam, 2002; Zhang et al., 2012) and employees and their organizations (Bretz & Judge, 1994; Cha et al., 2014; O’Reilly et al., 1991), and highlights congruence among all three aspects (i.e., employees, supervisors, and organizations).

The results of the present study also have a number of practical implications. First, organizations may wish to develop their prosocial cultures by actively engaging in socially responsible activities, while simultaneously training and educating their managers and employees to be more attentive to the prosocial aspects of their work to create a synergistic commitment effects. Additionally, the results indicate that organizations should implement strategies that raise employees’ awareness of CSR activities. The findings suggested that perceptions of CSR increased employees’ organizational commitment. Thus, organizations may wish to increase employees’ perceptions of their prosocial cultures (e.g., by providing employees with information on how the organization contributes to the community and how external stakeholders benefit from these activities).

Finally, the results also have implications for human resource managers in both China and other similar emerging markets in the Asia Pacific region. Employees in these markets “often for the first time in their lives, have numerous options and high expectations” (Ready et al., 2008: 62) due to their societal changes. Using a sample of Chinese participants, the present study showed the importance of fostering prosocial work environments for employees with prosocial motivations. Such information could assist organizations seeking to acquire and retain talent in these emerging markets. The present study’s findings are also consistent with conclusions from a meta-analysis study of Hou, Liu, Fan, and Wei (2016) that found that CSR practices create advantages in the East Asian context.

Limitations and future directions

Despite its important theoretical and practical implications, this study had several limitations. First, all the data were self-reported and collected at a single time. Thus, the issue of common method variance arises (Podsakoff et al., 2003). Although common method variance is less likely to have an impact on nonlinear relationships (Crampton & Wagner, 1994), and more importantly, we provided evidence that common method biases did not pose a threat to our findings, the results should still be interpreted with caution. On the other hand, when testing the effect of PS fit on organizational commitment, using same raters should not be viewed as a flaw because, consistent with previous studies (Chan, 2009; Edwards & Cable, 2009; Hofman & Newman, 2014), self-report measures were appropriate for this theoretical model, as the effects of congruence on employees’ own psychological outcomes (i.e., organizational commitment) can be more effectively predicted by employees’ own perceptions (Caplan, 1987; Endler & Magnusson, 1976). If prosocial motivation data were obtained from different sources (i.e., employees’ prosocial motivation rated by employees, and supervisors’ prosocial motivation rated by supervisors), the fit between the two does not necessarily lead to positive outcomes such as organizational commitment, because employees may not have perceived that there is a fit. Thus, while the use of a common source could potentially increase the relationship strength between PS fit and organizational commitment, it could also reflect the reality rather than an artificial bias (Kristof-Brown et al., 2005).

Second, employees’ perceptions of CSR were examined rather than objective CSR at an organizational level. Actual organizational involvement in CSR activities was not measured; thus, there could be a gap between employees’ perceptions and reality. However, as previous research in the PE fit literature (e.g., Kristof-Brown et al., 2005) and CSR literature (e.g., Turker, 2009) has shown, subjective perceptions might be a more important determinant of the consequences of CSR than objective measures of organizational social performance, regardless of the accuracy of employees’ perceptions.

Third, this study focused on one single outcome; that is, organizational commitment. Future research should examine other outcomes such as employees’ performances or organizational citizenship behaviors (OCB). Employees’ prosocial motivations have been found to be associated with citizenship behaviors (Grant & Mayer, 2009). Thus, it is possible that this relationship will be stronger if supervisors also have high levels of prosocial motivation (i.e., fit at high level of prosocial motivation). Additionally, organizational CSR may provide a supportive environment for prosocially motivated employees to engage in more OCB. Future research should expand on the current study by testing these relationships.

Finally, cross-sectional data was used; thus, causal relationships could not be verified. Rather than congruence between prosocial motivations and employees and supervisors leading to organizational commitment, it is possible that employees who are committed to their organizations are likely to develop similar prosocial motivations to their supervisors. However, we see the reversed causal relationship as less plausible: while the congruence on a specific dimension (i.e., prosocial motivation) between employees and their supervisors likely leads to employees’ general attitude toward their organizations (i.e., organizational commitment), the general attitude toward the organization may results in employees’ congruence with their supervisors on a variety of dimensions, not necessarily on prosocial motivation. Nevertheless, future research should use experimental or longitudinal research methods to establish the causal relationships among this study’s variables.

Conclusion

This study provided insight into the interplay between employees’ prosocial motivations, supervisors’ prosocial motivations, and organizations’ prosocial orientations. The findings shed light on how to foster and sustain the organizational commitment of those employees with strong prosocial motivations. Prosocial employees could be valuable assets to organizations; thus, organizations should provide prosocial work environments for these employees to capitalize on these benefits.