Introduction

Corporate social responsibility (CSR) can be viewed as an organization’s initiatives, beyond those that are required by law, aimed at benefiting its stakeholders (McWilliams & Siegel, 2001). When implemented well, CSR initiatives contribute to the attainment of an organization’s triple bottom-line objectives, specifically goals related to economic, social, and environmental sustainability (Schulz & Flanigan, 2016). Carroll (1991) frames CSR as the economic, legal, ethical, and philanthropic or discretionary expectations a society has of an organization. In the current climate of the early twenty-first century, these societal demands placed on organizations are steadily increasing. Due to this, more organizations are considering CSR as a means to achieve sustainable growth and performance while simultaneously addressing societal issues (Maignan & Ferrell, 2004).

Beyond this, there is a recognition that social and economic changes in the twenty-first century have posed significant challenges for career counselors with a greater emphasis being placed on social responsibility, ethics, and lifelong guidance as opposed to the previously more narrowly defined remit of their roles (Amundson, 2006). Significantly, in a post-Covid-19 world, counselors may lack the wherewithal to help individuals deal with a raft of heretofore unexperienced work-, and indeed life-related worries (Drosos et al., 2021). At the same time, CSR has been seen to contribute to the attraction and retention of talent in a competitive environment (Ohlrich, 2015). Thus, by placing greater emphasis on career counseling within a CSR framework, organizations and counselors can satisfy the broader demands of modern career counseling while contributing to employee and societal needs. As such, career counseling can represent some of an organization’s CSR activities, alleviate individuals’ work- and life-related stresses, increase an employee’s capacity and confidence to fulfill their roles, and present an opportunity for organizations to grow.

Dual-research streams have emerged in relation to CSR, whereby internal CSR is concerned with the psychological and physiological well-being of an organization’s employees, and external CSR is directed at stakeholders outside the organizations’ boundaries (Ferreira & Real de Oliveira, 2014). Originally, much of the discourse around CSR was conducted at an institutional level about the impact organizations had on society, i.e., external CSR (Lee, 2008, p. 200). At the micro-level, interest in internal CSR developed later. This is not to suggest that internal CSR is lesser, however, as it has been posited that measures at this level may be more appropriate for determining how embedded CSR is in an organization (Trivellas et al., 2019). This has led to a recognition among CSR scholars of the need for greater analysis of CSR at the individual level in addition to at the organizational level both within and outside organizations (Hofman & Newman, 2014; Ng et al., 2019).

Alternatively, Basu and Palazzo (2008) identified three dominant streams of inquiry within CSR research: stakeholder driven, performance driven, and motivation driven. Of these, the performance-driven approach is concerned with assessing the impact of CSR on organization profitability and performance metrics (Aupperle et al., 1985; Basu & Palazzo, 2008). From this perspective, CSR has been highlighted as a tool available to organizations for potentially satisfying their performance goals and links have been identified between CSR and organizational financial performance (Miller et al., 2020; Weber, 2008). Relationships have also been uncovered between organizations with CSR policies and positive overall organizational performance, which is comprised of HR-related, organizational, and financial outcomes (Bučiūnienė & Kazlauskaitė, 2012). Indeed, in 2021, Velte conducted a review of the extant meta-analyses related to CSR. The review discovered that the vast majority of studies focused on the impact of CSR on financial outcomes. This is problematic as the impact of CSR on financial performance has been brought into question due to the equivocal nature surrounding the measurement of the relationship between the two concepts (Galant & Cadez, 2017), while attention has been drawn to the existence of CSR-related activities which do not directly contribute to financial performance (Wright & Ferris, 1997).

Other research has shown improvements in HR-related concepts such as individual task performance, and organizational-level outcomes such as relative competitive performance as a result of CSR (Story & Neves, 2015; Zhao & Zhang, 2020). Velte’s (2021) study, however, points to a dearth of meta-analyses examining non-financial outcomes of the relationship. By including heretofore under-represented groupings related to employees’ behaviors and attitudes, as well as groupings reflecting disparate countries, regions, and industries, we have attempted to address some of the shortcomings in the existing literature.

Despite the existence of positive connotations of CSR, conflicting negative results have also been uncovered. In a literature review examining the results of 76 empirical papers published over the forty-year period between 1972 and 2012, the muddiness surrounding the relationship between CSR and performance was highlighted and can be attributed to the fact the variables being analyzed were still being conceptualized in their nascent form (Raza et al., 2012). In that study, the relationship between CSR and performance was variously shown as being positive, negative, mixed, and non-existent. In other research, the ownership type of an organization was shown to be important to the relationship between CSR and performance with state-owned enterprises demonstrating a negative relationship and non-state-owned enterprises a positive effect (Kao et al., 2018). Finally, the sincerity with which employees perceive CSR initiatives being undertaken by their organization, and the degree to which the CSR activities align with individual personal values, affect individual performance. When an employee trusts that the CSR activities are intrinsically motivated, there is a positive effect. A negative effect is found when an employee does not entirely trust the motives of their organization (Donia et al., 2019). This differs somewhat with the results of Story and Neves (2015) who found that an organization’s CSR practices can influence employees to exert extra effort but only if the employees view the practices as being both intrinsically and extrinsically motivated.

The previous paragraphs highlight some of the potential benefits to pursuing CSR initiatives for organizations. They also draw attention to some of the discrepancies in the results of research surrounding CSR. Perhaps due to these discrepancies, many organizations are hesitant to participate in CSR initiatives. Furthermore, organizations are often reluctant to invest in CSR due to uncertainty around the direct value it provides to performance metrics of organizational success (Chung et al., 2018). This is compounded when organizations view CSR and performance as an “either/or” choice with CSR initiatives being undermined by, and discontinued due to, competing commercial and financial interests (Glover, 2007). Indeed, large global companies such as Ford have been guilty of placing commercial interests ahead of CSR concerns despite significant public opposition (Schwartz & Saiia, 2012). Additional criticisms have been leveled at organizations pursuing ultra-capitalist agendas to the detriment of society (Ragodoo, 2009). Therefore, it can be seen that there is a potential disconnect between the ever-increasing public demands that organizations participate in CSR activities (Fenwick, 2014), and the organizations’ reluctance to spend money on activities which do not have a clear, measurable, positive impact on the organization or its employees (Chung et al., 2018). In practice, CSR is perhaps not prioritized as highly as it is in theory.

Based on the previous research and perspectives outlined in the above paragraphs, there remains a lack of clarity surrounding CSR research and its relationship with performance in the workplace. There is a clear need to continue the research in this area and develop a greater understanding of the links between the concepts. Therefore, the purpose of this research is to investigate the total effect size of the relationship between CSR and performance-related outcomes in order to facilitate the development of CSR-related education programs in the career counseling and development field. Further criteria to be examined include differences in the relationship based on region and industry among others. In order to do this, a meta-analysis of 17 studies examining CSR and performance in organizations across a variety of industries and locations was conducted and the following research questions were formulated:

  1. 1.

    What is the total effect size of CSR on performance?

  2. 2.

    What other performance-related groupings emerge from the data and how does the effect size of CSR differ based on these groupings?

Literature review

Corporate social responsibility (CSR)

Corporate social responsibility has been defined as “policies and practices that organizations engage regarding creating positive social, including environmental, changes aimed at different stakeholders. It is important to note that this is a voluntary approach” (Ferreira & Real de Oliveira, 2014, p. 234). Elsewhere, the Commission of the European Communities (2001) viewed CSR as “a concept whereby companies integrate social and environmental concerns in their business operations and in their interaction with their stakeholders on a voluntary basis” (p. 4). At first glance, these definitions seem similar. In both definitions, and indeed further definitions, stakeholders are prominent either by implication, or by being explicitly mentioned (Aguilera et al., 2007; Aguinis, 2011; Commission of the European Committees, 2001; Ferreira & Real de Oliveira, 2014).

A subtle difference exists between the definitions though. In the first case, CSR is seen as an adjunct to more traditional, narrow economic objectives of organizations. The second definition, however, views CSR as a complement to these objectives. Although CSR-related activities are still seen as being voluntary in this definition, they are more closely integrated into the operations of organizations. Collier and Esteban (2007) hold a more extreme view as they assert that CSR is not an optional extra. Rather, they see organizations as being “responsible not only for the ethical conduct of their operations but also for their impact on all their stakeholders now and in generations to come” (p. 30).

A differentiation has been made between CSR activities or behaviors, and perceived CSR behaviors. Yoon et al. (2006) found that CSR activities improve an organization's image only when their motivation is perceived as being pure. Contrary to this, if stakeholders view organizational motives as being ambiguous or insincere, there is no effect, or a negative effect on the organization’s image respectively.

This is significant as greenwashing, which refers to an over-stating of an organization’s CSR activities or intentions, can actually be counter-productive in terms of creating a positive image among stakeholders (Delmas & Burbano, 2011; Nyilasy et al., 2014). As such, perceptions of CSR activities play a large role in whether these activities achieve their goals, or not. These findings support those of Weaveret al. (1999) who found that pursuing CSR activities due to imposed requirements rather than commitment to positive change, leads to decoupled CSR activities which are disconnected from an organization’s core business and have fewer positive impacts than would otherwise be the case.

External and internal CSR

CSR is concerned with organizations providing positive changes for the benefit of their stakeholders (Aguilera et al., 2007; Aguinis, 2011; Commission of the European Committees, 2001; Ferreira & Real de Oliveira, 2014). These stakeholders exist internally within the organization and externally to the organization, and the organization’s CSR activities can be differentiated on this basis (El Akremi et al., 2018).

Activities which aim to satisfy the social needs of local communities (Boehm, 2005), advocacy groups (Davidson, 2009), or other members of civil society (van Marrewijk, 2003) can be viewed as external CSR activities. Examples of external CSR include volunteerism (Perrini et al., 2007), corporate charitable contributions (Chen et al., 2008), or support of social and environmental issues (Babiak & Trendafilova, 2011).

Contrary to this, an organization’s activities which are concerned with the psychological and physiological well-being of its members can be viewed as relating to internal CSR (Hameed et al., 2016). Examples of internal CSR include diversity and equality in the workplace (Stewart et al., 2011), health and safety standards for all employees (Montero et al., 2009), training opportunities (Muller & Kolk, 2009), and manageable workloads and work-life balance (Khallash & Kruse, 2012). Employee perceptions of both external and internal CSR activities have positive relationships with employee attachment towards the organization, job performance, and job satisfaction (Colquitt et al., 2001; Lee et al., 2013; Rupp et al., 2006). It has become clear, however that the impacts of internal and external CSR on individuals and organizations are not uniform. Employees react in different ways to CSR activity which they perceive as benefiting people within their organization as opposed to society as a whole. As such, CSR represents a concept worthy of closer examination related to its impact on individuals, organizations, and societies, particularly in regard to the direction in which an organization’s CSR activities are focused.

Performance in the workplace

Organizations have numerous and differing objectives. Various metrics for interpreting and measuring performance have emerged over the years. This reflects the disparate goals of different sized organizations within diverse industries and societies (Richard et al., 2009). Although financial performance was perhaps the first method employed to determine an organization’s success or failure, it has long been recognized that as a metric of performance it is limited in scope, particularly in work environments that are becoming more socially conscious in the twenty-first century. This is perhaps due to the diverse approaches in the quantification of financial performance which makes it difficult to definitively measure as an outcome of CSR (Galant & Cadez, 2017). As such, alternative, non-financial indicators can add value when interpreting organizational outcomes while simultaneously addressing a humanistic approach to management (Chalofsky, 2014). Indeed, Reverte et al. (2016) advocated for a more comprehensive, less-rigidly financially focused approach when interpreting performance. From this perspective, HR-related and organizational-level outcomes can be viewed as good indicators of performance (Bučiūnienė & Kazlauskaitė, 2012). This section briefly outlines the HR-related and organizational-level ways of measuring performance which are adopted in this research.

Firstly, aggregated individual job performance has been viewed as a barometer of an organization’s relative success (Motowidlo & Kell, 2012). Outside of financial metrics, this represents perhaps the most intuitive method of measuring an organization’s performance and refers to the accumulated impact of the results of individual in-role tasks and behaviors on organizational outcomes. Alternatively, when individual employees contribute to the effectiveness of their organization as related to the expectations associated with their role, they can be said to be performing well (Zablah et al., 2012). Although in-role job behaviors may be self-rated, they are often measured from the perspective of the supervisor who judges how effectively they deem their subordinates to be fulfilling their roles. From this point of view, job performance may encompass a range of behavior-based concepts which individuals undertake as core aspects of their work (Cascio & Aguinis, 2018).

In addition to in-role behaviors, extra-role behaviors have also been used as indicators of performance in the workplace and have drawn attention since the 1980s (Bateman & Organ, 1983). They have been defined as “those behaviors that go beyond specified role requirements, and are directed towards the individual, the group, or the organization as a unit, in order to promote organizational goals” (Somech & Drach-Zahavy, 2000, p. 650). Among these extra-role behaviors, it has been posited that individual workplace proactivity results in positive change at the individual and organizational level. When employees are proactive, their personal well-being is enhanced, and they contribute more to organizational effectiveness (Strauss & Parker, 2014). Similar positive effects on performance have been uncovered through innovative behaviors (Aryee et al., 2012), customer-oriented behaviors (Grizzle et al., 2009), and helping behaviors (Ng & Van Dyne, 2005). As such, measurements of employees’ behavioral tendencies could be described as an accurate interpretation of performance. Indeed, Gilbert (2013) went so far as to say that behavior is the means of performance.

Aside from their behavior, individuals’ affective attitudes towards their organizations have been shown to be a good measure of job-related performance. Of these attitudes, employee engagement, job satisfaction, organizational commitment, and turnover intention are among the most prominent (Memon et al., 2018; Shuck et al., 2013). Although there are areas of overlap among these concepts, it has been argued that they are discrete constructs which contribute to workplace and organizational performance to varying degrees via differing theoretical mechanisms (Memon et al., 2018; Shuck et al., 2013; Vogel et al., 2016). Indeed, positive effects on job performance have been found in relation to engagement (Bakker & Bal, 2010), job satisfaction (Judge et al., 2001), and organizational commitment (Meyer et al., 1989), as well as a negative relationship with turnover intention (Hui et al., 2007). Although, significant relationships between these job-related attitudes and performance have been demonstrated, within a holistic, humanistic framework it can be argued that these concepts can be used as measures of performance themselves, as opposed to being antecedents to task-related performance.

Finally, in the context of the current research, and in an increasingly socially conscious global work environment, CSR-related activities have a positive impact on a variety of individual, organizational, community, and society-based issues (Maignan & Ferrell, 2004). As work and life are becoming increasingly entwined, many organizations can no longer afford to view CSR activities as discretionary (Szczanowicz & Saniuk, 2016). This is particularly true in the post-Covid-19 era where individual employees are more likely to be working from home. This means they may be placed under increased stress and find it difficult to switch off when their workday is finished (Gómez et al., 2020). As CSR activities are increasingly recognized as being integral to organizational operations, and they occurred as dependent variables in several of the papers included in this research, they have been included as a grouping for analysis in this study (Collier & Esteban, 2007).

The relationship between CSR and performance

As has been outlined in the previous section, there are numerous options available to organizations when deciding on which basis to judge performance. CSR has different impacts on each of these performance outcomes. First, CSR activities affect individual job-related performance. He et al. (2019) found that CSR had a positive effect on task performance when mediated by organizational identification. Furthermore, perceived CSR has a positive impact on corporate performance, as well as contributing to employees’ attitudes towards their work and their organizations (Lee et al., 2013; Rupp et al., 2006; Valentine & Fleischman, 2008).

Similarly, Rahman et al. (2016) found a positive relationship between CSR and several employee attitudes including job satisfaction, affective organizational commitment, and continuance organizational commitment. The same positive relationship did not extend to normative organizational commitment. The results may be brought into question, however, as Bouraoui et al. (2020) found that the relationship between CSR and affective commitment was largely moderated by the need to belong and mediated by person-organization fit. This suggests that CSR may have a reduced impact on individualistic employees and those with differing ideologies to their organization. Elsewhere, positive relationships between CSR and organizational citizenship behavior (OCB), and work engagement were uncovered when mediated through organizational justice (Farid et al., 2019), although other research contradicted the relationship with OCB (Azim, 2016). Additionally, a negative relationship has been highlighted between highly ethical climates in the workplace, and turnover intention among employees (Stewart et al., 2011).

Finally, CSR affects employee behavior in several ways. It has been shown to have a positive effect on extra-role helping behavior when mediated by organizational identification (Shen & Benson, 2016). CSR also mitigates individuals’ tendencies to participate in customer-directed counter-productive work behavior (Hur et al., 2018). Another study points out the dearth of research on the relationship between CSR and employee behavior (Zulfiqar et al., 2019). The research, however, employs work engagement and organizational identification as proxies for employee behavior which are conceptualized as employee attitudes in the current study. This draws attention to the still uncertain nature of the research surrounding CSR and emphasizes the need for the current study.

Groupings for analysis

Among the articles examined in this meta-analysis, several recurring themes and trends emerged. By grouping the variables in the articles according to these trends, a greater depth of analysis was facilitated. As the focus of this research is the outcomes of CSR, CSR-related variables were included. Further groupings are representative of HR-related, and organizational-level outcomes as they are good metrics of performance in the workplace (Galant & Cadez, 2017). The variables were grouped based on their conceptual similarity to other dependent variables as they emerged through the literature review. For example, variables such as job satisfaction, commitment, and involvement, among others display a considerable degree of conceptual overlap and collectively represent the individual’s feelings towards their company (Shuck et al., 2013). As such, they were grouped together as “employee attitudes towards the organization”. On this basis, the groups CSR-related variables, employee attitudes towards the organization, in-role behaviors, extra-role behaviors, and organizational-level outcomes were formed. CSR/ethics related variables naturally occurred in both the dependent and independent groupings as they were included in our original search terms. Employee attitudes towards the organization include individual affective attitudes or states such as work engagement and turnover intention which have been shown to impact employees’ performances in the workplace (Bakker & Bal, 2010; Saeed et al., 2014). In-role behaviors include related concepts such as in-role job performance and task performance and relates to those behaviors which are deemed to be essential to the undertaking of an employee’s job, while extra-role behaviors, including helping behaviors and innovative behaviors, focused on individual behaviors which are not central to an employee’s job but still contribute to the attainment of organizational objectives. Finally, organizational-level outcomes, including firm performance and relative competitive performance, were grouped together.

Next, groupings were organized on the basis of country, and also region. These groupings were made as individuals from similar national or cultural milieus have been shown to prioritize similar aspects of their work-lives, espouse similar value systems, and respond to similar motivational factors (Jo & Park, 2020; Sortheix & Schwartz, 2017). Industry-type, too, has been shown to affect the degree to which organizations are likely to participate in CSR-related activities (Brammer & Millington, 2003). The status or level of participant was also included as a group of analysis as every paper derived from the original search articulated whether the participants were employees, managers, or dyads. This analysis may provide insight into who CSR-related behaviors provide more value to in the workplace. Finally, CSR perspective was added as a grouping due to the fact that internal and external CSR activities have provided differing outcomes in relation to employee behaviors and attitudes (Hameed et al., 2016; Jia et al., 2019).

Methods

Data collection

In order to conduct a meta-analysis on CSR, this study examined academic papers published in scientific journals from 2009 to 2020. These dates represent the ten-year period prior to our initial search and were chosen as they allowed us to examine CSR as it is currently conceptualized. This meant we could exclude historical conceptualizations which may not adequately reflect how CSR is viewed today. It has been posited that, when compared with older studies, recent primary studies may be more adequately powered in terms of extracting results for secondary meta-analysis (Turner et al., 2013). By limiting our search to the years since 2009, we could focus on papers employing modern statistical methods for the measurement of psychometrics which were sufficiently powered for interpretation via meta-analysis.

In September 2020, a search of the ProQuest online database was conducted using the key search terms 'Corporate Social Responsibility', or 'Corporate Ethics', or 'Company Ethics', or 'CSR' as the concepts are thematically similar and have a large degree of overlap (Visser et al., 2010). These terms were searched in conjunction with 'job performance' as we are interested in the impact of CSR-related activities on individuals in the workplace and how they perform their roles. ProQuest is a composite database comprising 27 sub-databases. The number of journals included in these databases ensures a broad return on searches and somewhat mitigates against the possibility of location and reporting bias.

The initial search returned 395 papers of which duplicates, non-peer-reviewed research, and articles not focusing on our key variables were removed leaving a total of 150 remaining papers. Both authors read the abstracts of these papers and determined their suitability for inclusion based on the following criteria.

First, because quantitative meta-analysis requires the calculation of the effect size, data for meta-analysis must provide relevant quantitative values. Therefore, prior research based on literature reviews and qualitative analysis were excluded from this study. Only quantitative studies with statistics including the mean, standard deviation, sample size, correlation coefficient, t value, and F value were selected. Second, when an article comprising the same content was published in multiple different formats, the same data may be duplicated and lead to increased incidences of error being incorporated into subsequent research. In these cases, only one iteration of the article was included in the study. Finally, differing measures for CSR-related and other psychometric variables were employed in the papers returned in the original search. To ensure quality and depth of data, we elected to exclude research which had not utilized validated representations of the variables in question.

Through this process, a total of 17 papers were finally selected for inclusion in the current research. Through independent analysis of these papers by both authors, clear areas of overlap between variables began to emerge. Both authors’ analysis was then compared and consensus was formed on the groupings to be included for analysis in the final study. Statistics were entered into sub-groups for the 17 papers, and the total number of subgroup statistics used for this study was 70. The characteristics of the thesis to be studied are as follows in Table 1.

Table 1 Meta-analysis articles list
Full size table

Data coding

Coding of the selected 17 articles for the meta-analysis was conducted by the author's name, article title, publication year, sample size, dependent variables, sub-categories of the dependent variable, country of research, global region, industry of the company participating in the research, and responders’ status (employee or manager-level). For studies in which several statistical values ​​of the dependent variables were presented in one study, such as when multiple sub-categories of the dependent variables were included in a single study, the values ​​for each were used for analysis. Thus, in this study, a total of 70 sub-groups were derived from the 17 research articles. More specifically, the dependent variables were coded for each category, and the dependent variables were classified into CSR-related variables, employee attitudes, in-role behavior, extra-role behavior, and organizational-level outcomes. Initial coding of the collected data was conducted by two researchers using the Microsoft Excel 2016 program. Afterwards, one doctor of education participated in the coding process and conducted cross-checking of the coding results.

Data analysis

The research model in meta-analysis is divided into a random effect model and a fixed effect model (Hong et al., 2016). The random effect model assumes that the effect size may vary depending on the study, and the fixed effect model assumes that the effect size of all studies is the same (Hwang, 2014). While heterogeneity between studies is not considered in the fixed effect model, it is assumed that heterogeneity exists between studies in the random effect model (Borenstein et al., 2021). In this study, since different weights were applied according to the size of the sample numbers, a random effect model was applied to estimate the variances between studies.

In this research, the CMA (Comprehensive Meta-Analysis) program was used for data analysis. The analysis was conducted by calculating the effect size for CSR, confirming related variables, and reviewing statistical significance. For calculating the effect size, Hedges’ g method which is the corrected standardized mean difference was used. When conducting a meta-analysis, it is important to figure out if each article’s data shows its representativeness in the final study. In general, since studies with a large effect size or large sample size are more likely to be published than studies with a small effect size or sample size, publication bias may occur (Hwang, 2014). In this study, a funnel plot was used to visually grasp the effect size of each study and its standard error, and the analysis of publication bias was carried out using the Trim-and-Fill method.

Results

Total effect size of CSR

The results of calculating the total effect size for 17 papers examining CSR and performance are as follows. First, as a result of the homogeneity test, we rejected the null hypothesis with Q = 502.861 (p = .000), revealing that the effect sizes of individual studies were heterogeneous. In addition, I2 value was used for the heterogeneity test (Borenstein, et al., 2021), and it was found that I2 was 96.818, showing a high level of heterogeneity of about 97%. As shown in Table 2, in 17 research papers, the value of Hedges' g, which is the standardized average effect size in the random effect model, was 0.841. The effect size can be classified as a small effect size if the Hedges' g value is about 0.2, a medium effect size if it is about 0.5, and a large effect size if it is 0.8 or more (Cohen, 1988). In the random effect model, the 95% confidence interval of the effect size value showed significant results as the lower limit was 0.631 and the upper limit was 1.060 (p = .000).

Table 2 Total effect size of 17 articles
Full size table

Table 3 shows the results of examining the overall effect size using 70 analysis groups from 17 studies. The standardized average effect size in the random effect model based on 70 analysis groups showed a result of 0.838, which can be judged as a large effect size. The 95% confidence interval of the effect size value in the random effect model shows a significant result with a lower limit of 0.689 and an upper limit of 0.987 (p = .000). In this model, I2 was 93.855, indicating a high level of heterogeneity.

Table 3 Total effect size of 70 groups of analysis in 17 articles
Full size table

Effect size by the dependent variable groups

To discern the dependent variables’ effect size within the 70 groups of analysis, the effect size of dependent variables was calculated. As shown in Table 4, the dependent variable representing CSR studies was divided into CSR-Related Variables, Employee Attitudes, Extra-Role Behavior, In-Role Behavior, and Organizational-Level Outcomes. The results show that Organizational-Level Outcomes had the highest effect size of 1.263 (p = .000). Next, CSR-Related Variables showed a high effect size of 1.025 (p = .000), Employee Attitude’s effect size was 0.786 (p = .000), Extra-Role Behavior’s was 0.785 (p = .000) and finally, In-Role Behavior showed a medium-sized effect of 0.559 (p = .007).

Table 4 Effect size by the dependent variable groups
Full size table

As a result of the heterogeneity test between the dependent variable categories, the Q value was 9.041 (df = 3, p = .029), which revealed that there was a significant difference in the effect size between each group. Also, it was found that the effect size of CSR-Related Variables was larger when they were measured as dependent variables as opposed to independent variables.

Effect size by country

In this study, the effect size was calculated by country to examine the factors that influence CSR. The effect sizes for each country are shown in Table 5. Taiwan showed the largest effect size of 1.469, and Germany also showed a large effect size of 1.453. Vietnam (1.446) and Portugal (1.105) followed next. The country with the smallest effect size was Greece (0.663). Egypt did not show significant statistical results.

Table 5 Effect size by country
Full size table

As a result of the heterogeneity test between countries, the Q value was 19.557 (df = 10, p = .034), indicating that there was a significant difference in the effect size for each country.

Effect size by region

Next, effect size by each region was calculated. As shown in Table 6, Europe had the highest effect size of 0.933 (p = .000). After that, Asia showed a high effect size of 0.909 (p = .000), America also had a high effect size of 0.828 (p = .000) and MENA showed a medium–high level of effect size, 0.746 (p = .000).

Table 6 Effect size by region
Full size table

Effect size by industry

Next, the effect size was calculated by industry to which each company belonged. As shown in Table 7, when a company is included in various industries, it showed the largest effect size at 0.982, followed by Finance with a large effect size of 0.837. Logistics was 0.663, indicating a medium effect size. The statistical results of Manufacturing were found to be insignificant.

Table 7 Effect size by industry
Full size table

Effect size by perspective on CSR

The perspective on CSR was divided into both Internal and External, and External Only. Using this data, the effect size was calculated. As portrayed in Table 8, when a perspective on CSR was both internal and external, the effect size was 0.931 and showed higher value than an only external perspective (0.656).

Table 8 Effect size by perspective on CSR
Full size table

Effect size by survey participant group

Last, the effect size by survey participant group—manager level, employee level, and no division of level were calculated. As portrayed in Table 9, in the manager-level group, the effect size of CSR was the largest at 1.115. At the employee level, there was a large effect size of 0.870. The no division group indicated a medium effect size of 0.546.

Table 9 Effect size by survey participant group
Full size table

Publication bias

Publication bias analysis was performed to confirm the reliability of this study. Publication bias occurs when primary research results are not published due to their strength or direction. This can mean that a disproportionate number of studies display significant positive results, whereas negative or insignificant results may not be published. Accordingly, the aggregate of published research results does not represent the totality of actual research conducted (Hwang, 2014).

In a meta-analysis, it is helpful to show that this risk has been minimized. In order to do this, the relationship between sample size and effect size can be shown through a funnel plot. In the funnel plot, if the research is arranged symmetrically around the effect size, it can be considered that there is no publication bias. If left–right symmetry is not achieved, there is a possibility of publication bias (Hyun, 2014). The trim-and-fill method corrects an asymmetric funnel plot to be symmetric and analyzes the effect of missing data on the overall effect (Hwang, 2014).

In this study, a funnel plot with the effect size on the x-axis and the standard error on the y-axis for the 17 studies included in the meta-analysis was calculated.

According to the funnel plot in Figure 1, the distribution of the average fixed effect size of the 17 studies displays a high degree of symmetry (Hedges’ g = 0.760).

Figure 1
figure 1

Funnel plot

Full size image

Next, a trim-and-fill test was performed, and the corrected effect size results and funnel plot are presented as shown in Table 10. As a result of the analysis, 3 studies were added and the corrected effect size was 0.682, which is lower than that before correction. In conclusion, there is no reason to doubt the validity of the study results in relation to publication bias.

Table 10 Trim-and-fill analysis results
Full size table

Discussion

This research was designed to uncover general conclusions about CSR and workplace performance and discover relevant variables which have an impact on the relationship. In order to do this, 17 previous studies were selected via the online database ProQuest. A meta-analysis of 17 studies with 70 statistical sub-groups examining CSR and performance in organizations across a variety of industries and locations was conducted. The discussion based on the results of this study is presented as follows.

First, the overall effect size of CSR on performance was 0.841, indicating that CSR has a large effect on performance. The effect size in the random model based on 70 statistics was 0.838, which can also be interpreted as a large effect size. In addition, as a result of examining the relationship between the sample size and the effect size in this study through a funnel plot, the effect size distribution of 17 research articles was symmetrical based on the average fixed effect size (Hedges' g = 0.760), and through the trim-and-fill method, the corrected effect size was similar to the original effect size. As we could not uncover any evidence of publication bias in our research, the validity of the results of this study were verified and it was confirmed that the effect size of CSR on performance was quite large. These results are in line with previous research which uncovered a relationship between the two concepts (Tang et al., 2012) and adds further weight to the argument of those who view CSR as an integral aspect of modern business practice (Collier & Esteban, 2007).

Second, the dependent variables related to CSR were divided into the groups CSR-related variables, employee attitudes, extra-role behavior, in-role behavior, and organizational-level outcomes. Organizational-level outcomes showed the highest effect size. This suggests that CSR activities do indeed contribute to performance in organizations in a manner that adds value and can be measured. This, in turn, should go some way to putting the minds of upper managers who are reluctant to participate in CSR activities due to uncertainty around its worth at ease. The next-highest effect sizes were attributed to CSR-related variables, employee attitudes, extra-role behaviors, and in-role behaviors in that order. Employee attitudes, extra-role behaviors, and in-role behaviors all showed a medium effect size. This can be seen as partially statistically supporting the various results shown in previous studies on the relationship between CSR and employee attitudes and behaviors (Azim et al., 2014; Shen & Benson, 2016). These results highlight a range of areas within organizations to which CSR activities can contribute and should encourage increased participation in these activities in the future to the benefit of stakeholders.

Third, among the countries included in this study, Taiwan and Germany showed a large effect size between CSR and performance. Europe and Asia showed a high effect size based on region. According to these results, practitioners in the countries and regions included in this research may choose to appropriately funnel larger or smaller amounts of resources into CSR-related activities. These results run somewhat counter to previous research which found that the Asia–Pacific region was the worst in terms of participation in CSR activities (Fisher et al., 2016). Our results serve to support the position that CSR-related activities are highly valued in Asia and there may be a growing appetite among stakeholders for organizations to participate more visibly in these activities. This places the onus on businesses to practice CSR in a global capacity as opposed to merely in the Western context which had, until recently, been the overwhelming point of focus in the CSR literature (Ramasamy et al., 2013). This may have implications for Western organizations which have been accused of operating in some parts of Asia in order to take advantage of lax local legal infrastructure and participate in unethical business practices for financial gain (Clelland, 2014; Ngai, 2005). One potential confounding issue relates to within-country cultural differences which may exist and were beyond the scope of this study. Future researchers may consider including this subject for analysis in their studies.

Fourth, related to industry, the finance industry showed a larger effect size than the logistics industry, while the results for the manufacturing industry were not significant. This could be due to the fact that in the wake of the global financial crisis, increased external pressure on organizations within the finance industry to behave in a socially responsible manner has borne fruit (Weber et al., 2014). However, a caveat must be recognized in order to maintain this progress. The prevalence of amoral, or merely profit-based, justifications for CSR within the finance industry means, in the long-term, organizations may be at risk of participating in decoupled CSR activities, or greenwashing (Norberg, 2018). If this is recognized by stakeholders, the benefits associated with these CSR activities may be diminished. An alternative view on this could be that the stringent rules and regulations the finance industry is subject to may play a role. Stronger rule of law can facilitate greater degrees of trade in financial services so the same may be the case for participation in CSR-related activities (Gani & Clemes, 2016). Added to this, research has shown that stock-market prices of organizations in the finance industry increase more due to positive CSR-related news in the media than the prices of organizations in other industries (Pérez et al., 2020). All of this reinforces the importance of CSR activities within the finance industry. Whatever the underlying rationale, it seems financial institutions should persevere with these activities.

Fifth, the effect size on managers was greater than that on employees. This points to the role of managers or employees in upper levels of organizational hierarchies in creating shared values and organizational visions (Høvring, 2017; Valentino, 2004). If organizations wish to attend to the values of their customer-base, other stakeholders, and society as a whole, managers will play a key role in communicating these values to their subordinates. This is especially salient when environmental and social concerns are becoming bleaker, and emphasis is being placed on CSR to become more proactive and transformative (Visser, 2016). As such, by approaching their role through a career-counseling lens, managers are empowered to help employees take control of their narratives, find meaning in their work, and actualize overarching objectives (Savickas, 2011). This can contribute to the formation of fruitful, positive partnerships between individuals and their employers and ultimately lead to a career development culture based on shared values which benefits both employees and managers (Neault, 2000). Organizations, too, will benefit from the associated improved attitudes and behaviors of their employees if managers can inculcate a pro-CSR culture in the workplace.

Finally, when organizations participate in CSR activities related to both its internal and external aspects, there is a larger effect size than when they only participate in external CSR activities. This suggests that employees are concerned about societal responsibilities, but also ethical issues which impact their functioning in the workplace such as work-life balance, and equitable access to opportunities (Khallash & Kruse, 2012; Muller & Kolk, 2009). From this perspective, career counseling and development represent an opportunity for organizations to fulfill their internal CSR obligations. They also play a formative role in an employee’s capacity to manage their various personal and work-related responsibilities insofar as counselors can help individuals develop the psychological tools to deal with the stresses of their daily lives and work (Hobson et al., 2001). Further, in the post-Covid-19 world, as work increasingly encroaches on family life, and vice versa, the distinctions between the two may become somewhat blurred further highlighting the need for employees to receive guidance in how to take ownership of their career direction through counseling and development opportunities (Hobson et al., 2001; Vyas, 2022). Thus, if organizations are to receive the full performance-related benefits of CSR, they should endeavor to integrate both internal and external programs into their strategies. They may also find it worthwhile to consider implementing programs from a bottom-up perspective as guidance in this form can provide employees with a sense of agency and control over their personal and professional situations (Kwon et al., 2020).

These discussion points highlight several implications of our study for both researchers and practitioners in the future. With regard to future studies, the meta-analytical nature of our paper will facilitate the formulation of robust hypotheses and provides multiple bases upon which researchers may choose to build their research. For example, they could elect to focus on and expand a single aspect of our results such as the impact of CSR in specific territories. Alternatively, they may wish to further explore a combination of our results to satisfy an, as yet, under-represented area in the CSR literature. One approach could be to examine the impact of CSR on performance in the workplace across similar industries in different regions in a single study with a common method. This would provide further clarity to the existing literature.

For practitioners, our study will help in the design of interventions and other leadership approaches aimed at improving perceptions of their organizations’ CSR activities and their impact on career counseling and development, and workplace performance. Practitioners in specific countries, regions, or industries, for example, may choose to elevate or reduce the prominence of CSR-related activities or interventions based on the results of our study. By recognizing that workers in different areas and fields have different attitudes towards and reactions to CSR, practitioners can alter their own strategic plans and shape their employees’ behaviors. As the effect size of CSR is comparatively larger for managers than employees, another option may be to concentrate CSR-related interventions at the management level first. If managers are brought on board, they may be more inclined to direct their energies to workplace activities such as constructing healthy work-climates. This could be done by implementing counseling programs designed to help individuals tackle challenges specific to the current changing social and economic situations. By offering these programs, organizations address some of their obligations with regard to CSR, but also contribute to the development of an empowered and robust workforce. This, in-turn, could have a positive trickle-down effect on subordinate employees and contribute to improved performance in the workplace.

Conclusions

The purpose of this study was to investigate the total effect size of the relationship between CSR and performance-related metrics and provide some insight to the benefits career counseling and development can provide to organizations in this context. Consumers, governments, and other stakeholders are becoming increasingly aware of the need for robust environmental, social, and governance (ESG) systems within organizations (Friede et al., 2015). The appetite for CSR activities among the general populace is growing, yet this has not always been reflected in business practice due to uncertainty around the returns organizations can receive on CSR-related investment (Chung et al., 2018). Through our research, we have added to the literature surrounding the ever-increasing push for organizations to behave in a more socially responsible manner. By analyzing a wide variety of areas including countries, industries, and levels within organizations, we have helped to paint a fuller picture of the importance of CSR in a socially conscious era which will benefit future researchers. This will also benefit practitioners within organizations who will be able to discern in which areas to focus their CSR resources with particular regard to career counseling and development opportunities.