The Fit between Employees’ Perception and the Organization’s Behavior in Terms of Corporate Social Responsibility

: Corporate social responsibility (CSR) refers to the set of behaviors that businesses ought to or are expected to perform in a society. Many companies expect to increase proﬁts through CSR behavior. However, it is a reasonable question to consider whether employees commit to an organization when they do not agree with the organizations’ CSR behavior. Therefore, this study explores the effect of ﬁt between employees’ perceptions of CSR and organizations’ CSR behavior on organization commitment. This study found that the ﬁt between employees’ perceptions and organizations’ CSR behavior has a positive effect on commitment. Among the four CSR dimensions of economic, legal, ethical and philanthropic social responsibility, all dimensions excluding philanthropic responsibility are positively related to organizational commitment. Finally, comparing the effects of ﬁt in Korea and China the effect of ﬁt for ethical social responsibility differed between the two countries. The congruence between need for ethical C SR (NETC) and behavior of ethical CSR (BETC) and its effect on organizational commitment (OC).


Introduction
Corporate social responsibility (CSR) refers to the set of behaviors businesses ought to or are expected to perform in a society [1]. Many companies expect to increase profits through CSR action, and in fact, Fortune 500 companies spend in excess of $15 billion annually on CSR activities [2]. CSR behavior build positive brand image and business reputation, which leads to higher sales and investment [3]. The positive effects of CSR also apply to consumers. The factors that consumers consider in their purchase decisions are not only the functional values of products, but also the brand values of the business. When they purchase products from businesses known for their CSR behavior, consumers feel that they have contributed to fulfilling their social responsibilities.
The research interests of recent CSR studies have been extended to internal stakeholders including employees [4,5]. Initial studies have examined if and how CSR influences the attitudes or behaviors of employees. In addition, they have attempted to explore if CSR behavior have the same effects on internal stakeholders as on external stakeholders. Despite the contributions of previous studies, the following four questions have risen.
First, CSR studies at the individual level remain lacking, even though studies on the relationships between employees and CSR have been conducted. A CSR review paper [6] in 2014 indicated that only 4% of total CSR studies focused on the individual-level. It is very important to understand how internal members of an organization perceive the activities and performances of the organization because the perceptions of the employees about the organization will influence their attitudes and behavior, which will ultimately have an impact on both personal and organizational performance [7]. The same can be applied to CSR. Employees' perceptions of CSR behavior will influence their attitudes and behaviors regarding the business, which will in turn influence the performance of the business [4]. D'Aprile and Talo [5] viewed the relationships between CSR and employees' commitment to the organization as a psychological process, and emphasized the need for in-depth study of this process. Barrena-Martínez et al. [8] recently pointed out the need for socially responsible human resource management to increase CSR, employee commitment, and performance. Therefore, it is needed to study further about CSR at the individual level and therefore socially responsible human resource management, as well.
Second, the question is whether CSR behavior has the same positive effects on all employees. In recent years, the concept of fit or congruence has been presented in the CSR field. For example, the fit between the business' CSR behavior and the fairness of internal management of the business [9], and the fit between the business' CSR behavior and ethical level of individuals in the business, will have different effects on employees' attitudes and behaviors [10]. Sen & Bhattacharya [11] explained that customers were not always supportive of businesses which engaged in CSR actively, and rather, they were favorable only when the products/service of the company were associated with the CSR behavior of the business. Rupp et al. [9] illustrates that employees may perceive injustice when the businesses' CSR behavior towards external stakeholders, including consumers, are not consistent with the businesses' policies towards internal stakeholders.
Third, the question of whether each dimension of economic, legal, ethical, and philanthropic CSR has the same positive effect on employees is an important consideration. According to the pyramid of CSR presented in Caroll [12], there is a difference in required social level for each dimension. Among the dimensions of CSR, economic and legal responsibilities are two basic required responsibilities that must be achieved. On the other hand, ethical and philanthropic responsibilities are considered optional (not mandatory). Thus, the difference in the necessity and importance of CSR can be applied equally to employees, and employees may perceive the importance and necessity of each dimension differently.
Fourth, we consider the generalization of the positive effects of CSR found in the previous research. Since CSR is literally a corporate activity required by society, there will be a difference in CSR standards depending on which society the business belongs. This is due to the fact that the scope or standard applied to CSR is associated with the values or moral norms of the society or the country where the businesses are located [13]. So far the CSR studies have been conducted on businesses in Europe and America, and therefore, the empirical studies in the context of other countries will be critical to generalize the results of these CSR studies [13].
Based on these research questions drawn from the literature review, this study attempts to explore the relationships between organizational commitment and the fit between perceived needs of CSR of individual employees and CSR behavior in Korea and China. To obtain a better understanding of the effect of fit between employees' CSR needs and organizations' CSR behavior on employees' organizational commitment, this study adopts the person-organization fit theory (P-O fit theory).

Corporate Social Responsibility
Discussion regarding the relationship between society and business have continued for a long time, and these discussions have extended to the topics of CSR. In the 1960s and 1970s, the main question for researchers was "what should management do for society?" This has been developed further by Carrol (1979) [1]. Carroll offered the following definition: "The social responsibility of business encompasses the economic, legal, ethical, and discretionary expectation that society has of organizations at a given point in time" [1] (p. 500).
CSR is viewed as containing four dimensions: economic responsibility, legal responsibility, ethical responsibility, and philanthropic responsibility. Economic responsibility refers to the responsibility that business, as a main agent of economic activity, ought to produce products and services and to conduct business honestly. The legal responsibility means that business should obey the laws and regulations related to the business, while ethical responsibility means that the business as a member of a society must conform to ethical standards. Lastly, philanthropic responsibility is defined that the business must conduct social altruistic activity. Among the four responsibilities, economic and legal responsibilities are required and mandatory, ethical responsibility is expected of businesses and philanthropic responsibility is desired.
CSR has both mandatory (obligation) and voluntary (expectation) characteristics. The four dimensions of CSR (economic, legal, ethical, and philanthropic) are categorized as either mandatory or voluntary based on the ethical standards of the society. In other words, if ethical consciousness is spread, ethical responsibility as well as economic and legal responsibility can be recognized as a mandatory level.
As well, various theoretical perspectives (e.g., contingency, stakeholder theory, institutional theory) of CSR have been integrated recently. The formation of the CSR need of the stakeholder (external, internal stakeholder) of the enterprise is related to the social situation and social awareness, which can be related to stakeholder theory and institutional approach. Martínez et al. [14] argued for an extension of the CSR perspective through a combination of stakeholder theory and institutional approach to CSR. The differences in the pyramid of CSR can be determined by contingency, and this contingency is related to the recognition of the stakeholders constituting the society and the institutional characteristics that affect their perception [15]. According to the institutional approach, adopting CSR policies in the organization itself is supposed to increase employees' commitment or organizational performance. However, as in stakeholder theory, it is important to have a good fit between CSR policies of an organization and its employees' perception of CSR to increase organizational commitment and performance. Therefore, not only the policies, but also the HRM for social responsibility of an organization have received more attention.
Studies of the effects of CSR are divided into the following three categories. The first category includes the studies about the effects of CSR on external stakeholders such as customers and investors. Empirical studies have continuously shown that consumers pay willingly for the incentives of CSR activities from businesses [11,16]. CSR behavior provides not only motivation to purchase a product from a certain company, but also positively affects the consumers' evaluations of the business [11].
The second category of CSR studies concerns the relationship between CSR and financial performance. Many studies have reported the positive relationship between CSR and financial performance, but some have found a negative relationship or no correlation [17]. These inconsistent findings lead to the idea of the existence of a black box between CSR and financial performance, which has to be explained further.
The third category includes the recent studies focusing on the effects of CSR on internal members of organizations [18]. These studies demonstrated that CSR leads to positive attitudes and behaviors of internal members of organizations. Researchers assume that employee commitment is influenced by CSR because CSR leads to more enjoyable work behaviors and also employees' greater pride in the organization. Glavas and Kelley [19] highlighted the importance of perception of internal members due to the fact that internal members' perception of the CSR activities of organizations will make differences in the effect of CSR behavior of organizations on the individuals' attitudes and behaviors. Thus, among the research areas of CSR, the studies on the effects of CSR on individuals require further study.

Fit and Person-Organization Fit Theory
Generally, "fit" suggests 'being in harmony with', or 'being in agreement with' something else [20]. As a social science research term, fit implies an association of variables 'being congruent with' each other [21]. To validate this point, one study adopting an early approach to fit found that rather than investigating the effects of two variables upon an outcome separately, the fit between those two variables could predict the outcome more precisely [22]. The concept of fit has been applied to various research contexts representing organizational behavior. The most well-known theory about fit is person-organization fit proposed by Kristof [23]. It is conceptualized as "the fit (congruence) between individual and organizational values, in motivating an employee's preference for and commitment to organization" [23]. The fit between person-organization can be one of the determinants to predict a person's attitude and behavior. According to Edwards and Cable [24], employees' attitudes and behaviors related to their jobs are likely to be more positive when the organizational value and employees' perceived value are congruent compared to when they are not. The P-O fit theory illustrated that when full congruence between the values of employees and organization occurs, employees' attitudes toward the organization become more positive.
Greening and Turban [25] explained CSR behavior and perception by using the concept of person-organization fit. They found CSR behavior of organizations are not always accompanied with individuals' positive attitudes. Rather, individuals' characteristics also influence their attitudes toward CSR behavior of organizations. Zhang and Gowan [26] argued that individuals who place a higher value on ethical behaviors should be attracted to more socially responsible businesses that espouse higher ethical standards than businesses that do not.
To date, only three studies [25][26][27] have examined this relationship (between CSR and employee or consumer) from the perspective of person-organization (P-O) fit. Therefore, the current study extends previous research by examining the influence of CSR perception on employee attitude and behavior from a P-O fit perspective.

Fit between Need and Behavior of CSR and Organization Commitment
Empirical studies concerning the effect of employees' perceived CSR on employees' commitment to the organization have been conducted. Previous findings regarding the relationships between CSR and organizational commitment are divided into two categories. First, the relationship between CSR and organizational commitment depended on the target of CSR practices. CSR towards social and nonsocial stakeholders, employees, and customers were found to be significant determinants of organizational commitment while CSR toward government were not [28]. Employees' perceptions of CSR practices targeting internal stakeholders were also found to be significantly related to their organizational commitments, while the effects of employees' perceptions of CSR towards external stakeholders on organizational commitments were found to be either non-significant or marginally significant [29]. Second, there are studies concerning variables meditating CSR and organizational commitment. Farooq et al. [30] verified that organizational trust and organizational identification fully mediated CSR and organizational commitment. D'Aprile and Talo [5] illustrated the psychosocial process in which organizational sense of community mediated the perceptions of CSR and organizational commitment.
From the two categories of CSR studies, a concept of fit has emerged as a critical factor as well as a common theme. It is noticeable that the fit between CSR practice and employees' perceptions has a positive effect on organizational commitment. It is not the CSR practice itself but the congruence between CSR practice and employees' needs for CSR that can strengthen employees' commitment to the organization. Recent studies [27] have also proposed a conceptual model regarding the fit between CSR behavior and employees' perceptions of CSR affect employees' attitudes and behaviors.
Therefore, it is expected that not the organization's CSR behavior, but the fit between CSR behavior and employees' needs of CSR will have an effect on organizational commitment. In the case where an organization is actively engaged in CSR behavior and concurrently, employees of the organization strongly perceive the need of CSR, which can be viewed as a full fit between CSR behavior and employees' needs of CSR, employees tend to maintain their relationships with the organization and to report higher levels of organizational commitment [31,32]. According to P-O fit theory, employee devotion and commitment are enhanced when employees perceive that their values and needs match those of the organization. Therefore, this study predicts that fit between employees' perceived needs of CSR and organization's CSR behavior could positively influence organizationational commitment. The following hypothesis was developed: Hypothesis 1 (H1). The fit between employees' perceived needs of CSR and organization's CSR behavior will be related to organizational commitment.

Pyramid of CSR and Organizational Commitment
The pyramid of CSR provided by Carroll [12] illustrated the four dimensions of CSR. The basic level at the bottom of the pyramid represents economic performance. As well, the businesses are expected to obey the laws of society, which usually concerns what you should and should not do therein. The upper level of the pyramid represents the ethical responsibility of businesses. Unethical, however, does not mean illegal. The highest level of the pyramid is philanthropic performance in which businesses are expected to be good members of society.
Carroll [12] noted that these four dimensions of CSR (economic, legal, ethical, and philanthropic) form a pyramid system. Therefore, without fulfilling the lower levels of responsibility, higher responsibilities were not able to be fully achieved. That is, members of a society would doubt the motivation of philanthropic behaviors if the business did not achieve the economic, legal, and ethical responsibilities, which would lead to no positive effect of the CSR behavior. Therefore, Amalric and Hauser [33] argued that it was necessary to decide priorities when the businesses engaged in CSR behavior.
However, it has been argued that "responsibility" is not a proper term for philanthropic behaviors, unlike the other three CSR responsibilities, because of its voluntary or discretionary nature. Others also agree that philanthropy cannot be considered a responsibility in itself [34]. Therefore, we expect that there are differences in the effects of the four dimensions of CSR. Economic, legal, and ethical responsibilities are perceived as fundamental "responsibilities" of the businesses. Therefore, for these three dimensions, the fits are expected to have positive effects on organizational commitment. For the philanthropic dimension of CSR, the fit is expected to have no effect. Therefore, we hypothesize: Hypothesis 2 (H2). The fits for economic, legal, and ethical responsibilities of CSR will have positive effects on organizational commitment.

Hypothesis 3 (H3).
The fit for philanthropic responsibility of CSR will have no significant effect on organizational commitment.

Difference of China and Korea in Effect of CSR on Organizational Commitment
CSR differs depending on the situation of the country where the businesses belong [35] because not only the ethical or moral norms that people accept, but also economic and legal regulations differ by country [36]. Therefore, the standards that apply to CSR and people's perceptions regarding CSR differ for each country. So far most studies concerning CSR have been conducted in USA or the UK, and limited studies have considered countries with different social and cultural environments [37]. Therefore, it is valuable to study different countries for the generalization of findings of CSR studies [38].
The study of Maignan and Ferrell [39] was the earliest CSR study comparing different countries. It analyzed differences in consumer perceptions of CSR in USA, France, and Germany. For Americans, economic responsibility was proposed to be the most important because American consumers were believed to perceive business as economic subjects that sell products and services in order to satisfy consumers at the right prices, grow their businesses, increase employment, and make enough profits to satisfy investors.
These differences in perception concerning CSR are found between Korea and China, both of which are similar in the aspects of collectivism and Confucianist culture. The differences are thought to be caused by the different economic policies in the two countries. Despite rapid economic development, China has faced difficulty regarding CSR because the awareness of concept is relatively low, even though the need for CSR is increasing due to environmental pollution and human rights concerns in relation to labor [35]. In Korea, the perception of CSR has spread since the 1990s; relatively earlier than in China.
The need for CSR is widely recognized by the public, and a positive evaluation for the CSR companies has led to a rise in corporate value, as well. Therefore, CSR seems to be perceived as mandatory in Korea [35] while in a Chinese business environment, where not many companies are engaged in CSR, CSR is perceived as voluntary rather than as mandatory. Thus, Chinese employees will evaluate the voluntary CSR behavior of the company positively [36]. In Korea and China, the economic growth and legal regulations differ and therefore, perceptions of CSR in the two countries are expected to differ. Kolk et al. [37] has studied Chinese consumer perceptions of CSR. They recognized the importance of studies on perceptions of CSR in non-Western markets since most previous studies were in the context of Western countries. Originally the four dimensions of CSR as well as the CSR construct seemed generalizable to China, but findings showed that consumers in China perceived only two dimensions of CSR, and not four dimensions. In their view, economic and legal responsibilities were combined into one (labelled "required CSR"), and the other dimension combined ethical and philanthropic responsibilities (called "expected CSR"). Chinese consumers expected that Chinese businesses engaged in required CSR more than foreign businesses. Another study on CSR in China was conducted by Anthony Wong and Hong Gao [40]. They found that CSR activities focused on external stakeholders did not have significant effect on organizational commitment of employees. However, in their study of CSR in Korea, Kim, Song and Lee [41] found that employees' perceptions of CSR are related positively to organizational commitment and negatively to turnover intention.
Previous research noted that the relationship between perceptions of CSR and organizational commitment in Korea differs from that in China. These differences between Korea and China resulted from the social norms of each country which influences employees' perceptions of CSR [19]. Therefore, this study expects that the relationship between the fit of perception of CSR and need of CSR and organizational commitment in Korea differs from that in China. Therefore, we propose: Hypothesis 4 (H4). The relationships between the fits of the four dimensions of CSR and organizational commitment will differ by country.

Data Collection and Sampling
This study surveyed employees working in Chinese and Korean companies from March to April 2017. In order to compare Korea to China, data were collected from the small to mid-sized companies with 100~200 employees in manufacturing sector (60%) and service sector (40%) in both countries. Before the questionnaire was distributed, the purpose of the study was explained to members of each company. In China, 256 survey questionnaires were distributed and 209 responses were collected. In Korea, 234 surveys were distributed and 202 responses were collected. Of the 411 collected questionnaires, a total of 400 were completed useable. Eleven responses were not used due to inappropriate answers. Respondents included Korean (49.3%) and Chinese (64.7%), and 216 (54%) were male and 184 (46%) were female. The age group of respondents was composed of 260 persons (65%) aged 20-39, and 111 persons (27.8%) aged 40-49. There were 138 persons (34.5%) who had work experience of less than three years, followed by 97 persons (24.3%) who had between 5 and 10 years of experience.

Measure
The study's measurement items were developed based upon a comprehensive literature review. Employee's need of CSR and behavior of CSR were used as independent variables, and organizational commitment was set as a dependent variable. The need and behavior of CSR, independent variables, were divided into economic, legal, ethical, and philanthropic responsibilities based on the work of Carroll [12] and Maignan and Ferrell [42]. Each variable was measured with four items. In CSR, economic responsibility refers to awareness of the company's sound and sincere economic activities, and legal responsibility means the degree of compliance with relevant laws and regulations. In addition, the recognition of corporate ethical responsibility refers to the degree of behavior and activity expected of a company as a member of society, though it cannot be prescribed by law. Finally, philanthropic responsibility refers to the recognition of activities belonging to a voluntary area, such as social donation behavior, which is a responsibility entrusted to decisions from individuals or choices of corporations. This study also measured organization commitment, which is a dependent variable, based on the results of Moon et al. [43]'s study. Organizational commitment, which means strong trust and attachment to organizational values and goals, was measured using three items. All measurements were modified for the purposes of this study and measured using a five-point Likert scale.

Data Analysis: Polynomial Regression with a Response Surface Analysis
This study followed the data analyses used in previous studies [44]. Thus, this study employed a nonlinear approach representing polynomial regression with a response surface analysis. Polynomial regression models with a response surface analysis were developed to investigate the congruence between Need of CSR (NEC, NLC, NETC, and NPC) and Behavior of CSR (BEC, BLC, BETC, and BPC) and the effect of their congruence upon Organizational Commitment (OC).
The following equation is a form of polynomial regression for testing nonlinear relationships: Z = b0 + b1X + b2Y + b3X 2 + b4XY + b5Y 2 + e In polynomial regression models, the dependent variable (Z: OC) is regressed on two independent variables (X: Need of CSR and Y: Behavior of CSR), two quadratic terms (X 2 and Y 2 ) for each independent variable, and one interaction term between the two independent variables (XY). For the four types of Need of CSR dimensions, and four types of Behavior of CSR, four polynomial regression models (1, 2, 3, and 4) were developed and tested separately.
To interpret the significance of the quadratic terms, the coefficients were used to graph the results of the polynomial regression analysis in a three-dimensional space, called a "response surface analysis" [45]. A response surface analysis (RSA) is an advanced and powerful methodological tool that shows interplay between X and Y value combinations and outcomes in a three-dimensional space [46].
This study centered all variables using a midpoint (4 on the 7-point Likert scale) and inputted these values into polynomial regression models to minimize any multicollinearity and to accurately interpret the three-dimensional graphs [46]. Data analyses for polynomial regression and response surface analyses were conducted using SYSTAT 13 and SPSS 21 statistical package programs.

Validity and Reliability of Measurements
Confirmatory factor analysis (CFA) was conducted to validate internal and external consistency of constructs. Cronbach's α coefficients were ranged from 0.722 to 0.899, indicating high internal consistency among the variables within each factor [47]. Results of the CFA are presented in Table 1. All factor loadings (>0.50), composite construct reliability (>0.70) and average variance extracted (>0.50) were considered acceptable, and satisfied the recommended values [48]. Therefore, convergent validity and reliability of constructs used in this study were found to be acceptable.  Table 2 presents the means, standard deviations and correlation coefficients for the nine constructs. All values of the square root of the AVE were greater than the correlations among the constructs, and this result supports the discriminant validity of the study's measures [49]. This section may be divided by subheadings. It should provide a concise and precise description of the experimental results, their interpretation, as well as the experimental conclusions that can be drawn.

Polynomial Regression Analysis
To perform polynomial regression analyses for Hypothesis 1, a second-step procedure was employed in which two independent variables (Need of CSR (X) and Behavior of CSR (Y)) were entered in step 1, and subsequently, two quadratic terms (X 2 and Y 2 ) and one interaction term (XY) were entered in step 2. Consistent with the studies of Edwards [50] and Edwards and Parry [46], this study assessed the validity of the quadratic terms by examining the significance of changes in variance explained by adding a set of nonlinear terms to the linear terms (see Table 3). Table 3. Polynomial regression analyses of H1.

Variables
Step 1 Step 2 3.729 *** + p < 0.1, * p < 0.05, ** p < 0.01, *** p < 0.001; ∆R 2 indicates an increase in variance explained by adding the set of nonlinear terms (X 2 , XY and Y 2 ) above the linear terms (X and Y). Table 3 shows that the value of R2 significantly increased in step 2 for the polynomial regression model (0.021, ∆F = 3.729 ***). This implies that the quadratic regression model is more appropriate to investigate how the congruence (or incongruence) between Need of CSR and Behavior of CSR affects OC. The interaction term in step 2 for the polynomial regression model is significant. Therefore, H1 is supported.
Next, the subdimensions of CSR were analyzed. Table 4 shows that the values of R 2 significantly increased in step 2 for all three polynomial regression models: model 1 (0.050, ∆F = 8.628***), model 2 (0.078, ∆F = 14.169***), and model 3 (0.089, ∆F = 15.714***). The value of R 2 did not significantly increase for model 4. This implies that the quadratic regression models of model 1, model 2, and model 3 are more appropriate to investigate how the congruence (or incongruence) between Need of CSR and Behavior of CSR affects OC. Therefore, the five polynomial regression coefficients (b1, b2, b3, b4 and b5) in models 1, 2, and 3 were used to test response surface graphs showing the relationships in a three-dimensional space [46].

Variables
Step 1 Step 2 Variables Step 1 Step 2 Variables Step 1 Step 2 Variables Step 1 Step  Notes: NEC (need of economic CSR); BEC (behavior of economic CSR); NLC (need of legal CSR); BLC (behavior of legal CSR); NETC (need of ethical CSR); BETC (behavior of ethical CSR); NPC (need of philanthropic CSR); BPC (behavior of philanthropic CSR); * p < 0.05, ** p < 0.01, *** p < 0.001; ∆R 2 indicates an increase in variance explained by adding the set of nonlinear terms (X 2 , XY and Y 2 ) above the linear terms (X and Y).

Response Surface Analysis
Response surfaces were plotted using a 3D space in which the three values of Need of CSR (NEC, NLC, NETC) and Behavior of CSR (BEC, BLC, BETC) were perpendicular horizontal axes and OC was a vertical axis. As evident in Table 5, the Y = X line represented the perfect congruence between Need of CSR and Behavior of CSR, and Y = −X represented the line of incongruence, indicating the case that Need of CSR and Behavior of CSR levels are incongruent. The slope of the perfect congruence line (Y = X) was estimated by the equation (b1 + b2) and the curvature along the Y = X line was assessed by the equation (b3 + b4 + b5). When the slope value (b1 + b2) was significant, but a curvature value (b3 + b4 + b5) was not, this indicates the existence of a linear relationship along the perfect congruence line (Y = X) with an outcome (Z), [51]. Results of Need of CSR showed that the slopes (b1 + b2) of model 1 (0.353, t = 6.993 ***), model 2 (0.375, t = 7.980 ***), and model 3 (0.428, t = 8.099 ***) along with the Y = X line were significant and positive. On the other hand, the curvature (b3 + b4 + b5) of all three models were not significant (see Table 5).  Regarding economic, legal, and ethical aspects, results indicated that the congruence between Need of CSR and Behavior of CSR was positively and significantly associated with OC. Figures 1-3 showed positive relationships between the X and Y congruence values and Z value. Especially, the height of the surface (OC levels) appears to vary depending on congruence levels. For instance, when Need of CSR and Behavior of CSR are congruent at the combination of (X = 3, Y = 3), OC is greater than when Need of CSR and Behavior of CSR are congruent at the combination (X = −3, Y = −3), implying that a higher congruence level leads to better OC compared to a lower congruence level. As shown in Figures 1-3, OC increases as both X and Y increase, with a concave shape. Thus, OC is higher when both Need of CSR and Behavior of CSR are high than when both factors are low. Therefore, findings showed that Need of CSR and Behavior of CSR have a significant effect on OC in economic, legal, and ethical aspects.

Response Surface Analysis
Response surfaces were plotted using a 3D space in which the three values of Need of CSR (NEC, NLC, NETC) and Behavior of CSR (BEC, BLC, BETC) were perpendicular horizontal axes and OC was a vertical axis. As evident in Table 5, the Y = X line represented the perfect congruence between Need of CSR and Behavior of CSR, and Y = −X represented the line of incongruence, indicating the case that Need of CSR and Behavior of CSR levels are incongruent. The slope of the perfect congruence line (Y = X) was estimated by the equation (b1 + b2) and the curvature along the Y = X line was assessed by the equation (b3 + b4 + b5). When the slope value (b1 + b2) was significant, but a curvature value (b3 + b4 + b5) was not, this indicates the existence of a linear relationship along the perfect congruence line (Y = X) with an outcome (Z), [51]. Results of Need of CSR showed that the slopes (b1 + b2) of model 1 (0.353, t = 6.993***), model 2 (0.375, t = 7.980***), and model 3 (0.428, t = 8.099***) along with the Y = X line were significant and positive. On the other hand, the curvature (b3 + b4 + b5) of all three models were not significant (see Table 5). Table 5. Response surface analysis.
Regarding economic, legal, and ethical aspects, results indicated that the congruence between Need of CSR and Behavior of CSR was positively and significantly associated with OC. Figures 1-3 showed positive relationships between the X and Y congruence values and Z value. Especially, the height of the surface (OC levels) appears to vary depending on congruence levels. For instance, when Need of CSR and Behavior of CSR are congruent at the combination of (X = 3, Y = 3), OC is greater than when Need of CSR and Behavior of CSR are congruent at the combination (X = −3, Y = −3), implying that a higher congruence level leads to better OC compared to a lower congruence level. As shown in Figures 1-3, OC increases as both X and Y increase, with a concave shape. Thus, OC is higher when both Need of CSR and Behavior of CSR are high than when both factors are low. Therefore, findings showed that Need of CSR and Behavior of CSR have a significant effect on OC in economic, legal, and ethical aspects.

Moderating Effects of Nations: Difference of China and Korea
This study hypothesized that the relationship between the congruence of Need of CSR and Behavior of CSR and OC would be affected by the difference in nation (i.e., China or Korea). In doing so, a three-step procedure was employed for each of the three polynomial regression models [52]. In step 1, the five variables (X, Y, X 2 , XY and Y 2 ) were entered. In step 2, the moderating variable (W: nation), which was dummy-coded with China as a reference group, was entered. In step 3, three interaction terms (XW, YW and XYW) were inputted. The following polynomial regression model was developed: Z = b0 + b1X + b2Y + b3X 2 + b4XY + b5Y 2 + b6W + b7XW + b8YW + b9XYW + e (2) Likewise, an increment in R 2 yielded by three additional interaction terms (XW, YW and XYW) with the original set of independent variables and moderating variable was assessed to test the validity of the polynomial regression Equation (2). Table 6 provides evidence that the R 2 in step 3 significantly increased in model 3 (0.018, ΔF = 3.550**), whereas the increment in the R 2 of model 1 (0.012, ΔF = 1.583) and model 2 (0.004, ΔF = 0.782) were not significant. These findings imply that the effects of the congruence between NETC and BETC on OC were significantly influenced by whether

Moderating Effects of Nations: Difference of China and Korea
This study hypothesized that the relationship between the congruence of Need of CSR and Behavior of CSR and OC would be affected by the difference in nation (i.e., China or Korea). In doing so, a three-step procedure was employed for each of the three polynomial regression models [52]. In step 1, the five variables (X, Y, X 2 , XY and Y 2 ) were entered. In step 2, the moderating variable (W: nation), which was dummy-coded with China as a reference group, was entered. In step 3, three interaction terms (XW, YW and XYW) were inputted. The following polynomial regression model was developed: Z = b0 + b1X + b2Y + b3X 2 + b4XY + b5Y 2 + b6W + b7XW + b8YW + b9XYW + e (2) Likewise, an increment in R 2 yielded by three additional interaction terms (XW, YW and XYW) with the original set of independent variables and moderating variable was assessed to test the validity of the polynomial regression Equation (2). Table 6 provides evidence that the R 2 in step 3 significantly increased in model 3 (0.018, ΔF = 3.550**), whereas the increment in the R 2 of model 1 (0.012, ΔF = 1.583) and model 2 (0.004, ΔF = 0.782) were not significant. These findings imply that the effects of the congruence between NETC and BETC on OC were significantly influenced by whether

Moderating Effects of Nations: Difference of China and Korea
This study hypothesized that the relationship between the congruence of Need of CSR and Behavior of CSR and OC would be affected by the difference in nation (i.e., China or Korea). In doing so, a three-step procedure was employed for each of the three polynomial regression models [52]. In step 1, the five variables (X, Y, X 2 , XY and Y 2 ) were entered. In step 2, the moderating variable (W: nation), which was dummy-coded with China as a reference group, was entered. In step 3, three interaction terms (XW, YW and XYW) were inputted. The following polynomial regression model was developed: Z = b0 + b1X + b2Y + b3X 2 + b4XY + b5Y 2 + b6W + b7XW + b8YW + b9XYW + e (2) Likewise, an increment in R 2 yielded by three additional interaction terms (XW, YW and XYW) with the original set of independent variables and moderating variable was assessed to test the validity of the polynomial regression Equation (2). Table 6 provides evidence that the R 2 in step 3 significantly increased in model 3 (0.018, ∆F = 3.550 **), whereas the increment in the R 2 of model 1 (0.012, ∆F = 1.583) and model 2 (0.004, ∆F = 0.782) were not significant. These findings imply that the effects of the congruence between NETC and BETC on OC were significantly influenced by whether the nation is Korea or China, but the effect of congruence between NEC and BEC, and also the congruence between NLC and BLC on OC, were not significantly influenced by whether the nation is Korea or China.
Response surface analyses were conducted for only model 3 to provide a more sophisticated interpretation of Korea and China. Figure 4 presents that when the surface runs along the perfect congruence line (Y = X), the congruence between NETC and BETC for Korea influences OC more positively than it does for China. The OC for both Korea and China were maximized with the highest congruence level (X = 3, Y= 3), and particularly, OC for Korea was higher than for China. Regarding ethical CSR, this result implies the congruence between perceived need of CSR and behavior of CSR would be a more significant and important factor for Korea than China. the nation is Korea or China, but the effect of congruence between NEC and BEC, and also the congruence between NLC and BLC on OC, were not significantly influenced by whether the nation is Korea or China. Response surface analyses were conducted for only model 3 to provide a more sophisticated interpretation of Korea and China. Figure 4 presents that when the surface runs along the perfect congruence line (Y = X), the congruence between NETC and BETC for Korea influences OC more positively than it does for China. The OC for both Korea and China were maximized with the highest congruence level (X = 3, Y= 3), and particularly, OC for Korea was higher than for China. Regarding ethical CSR, this result implies the congruence between perceived need of CSR and behavior of CSR would be a more significant and important factor for Korea than China.  Figure 4 also indicates that when the surface runs along the incongruence line (Y= −X), OC for both Korea and China was found to be lowest at the combination point (X = 3, Y= −3). The OC for Korea was found to be much lower than for China. This result revealed that when both nations face a high level of perceived need of ethical CSR and are not performing ethical CSR as much as needed, its impact might be a more serious impediment for decreasing performance in Korea than it would be in China. Therefore, the OC for Korea was significantly and negatively influenced by the incongruence between perceived need of CSR and behavior of CSR with respect to ethics. In summary, hypothesis 4 was supported.  Figure 4 also indicates that when the surface runs along the incongruence line (Y= −X), OC for both Korea and China was found to be lowest at the combination point (X = 3, Y= −3). The OC for Korea was found to be much lower than for China. This result revealed that when both nations face a high level of perceived need of ethical CSR and are not performing ethical CSR as much as needed, its impact might be a more serious impediment for decreasing performance in Korea than it would be in China. Therefore, the OC for Korea was significantly and negatively influenced by the incongruence between perceived need of CSR and behavior of CSR with respect to ethics. In summary, Hypothesis 4 was supported. Step 1 Step 2 Step 3 Step 1 Step 2 Step 3 Step 1 Step 2 Step 3 Constant  6. Discussion

Conclusions
CSR is no longer optional for companies, but is a necessary factor for long-term growth. Previous empirical studies have found that CSR positively affects external stakeholders such as customers and shareholders. On the other hand, studies on the influence of CSR on internal members have been lacking. In this study, the influence of CSR on organizational members was examined in detail. The results of this study are as follows.
First, this study found that CSR activities affect organizational members according to the fit between employee's CSR needs and CSR behavior. Previous studies have examined the two-dimensional relationship between CSR and employee's commitment. In this study, we examined the three-dimensional relationship between NEED, CSR activities and employee's commitment to CSR. Haski-Leventhal et al. [27] emphasized that the impact of CSR activities on employees' attitudes and behaviors can be clearly identified through the FIT between CSR behavior and employees' perception of CSR. The results of this study shed light on the deeper understanding of the effect of CSR on employees by verifying that the CSR activities themselves do not increase the commitment of the organization members, but rather the fit to the employees' need for CSR results in increased commitment.
In the case that the need for CSR is larger than behavior of CSR, employees' satisfaction is lower because employees may perceive that the company does not perform CSR well enough, which lead to lower organizational commitment. If behavior of CSR is larger than the need of CSR, employees' satisfaction becomes lower as well because employees may perceive that the company wastes the resources, which leads to lower organizational commitment. Therefore, the fit between need and behavior of CSR becomes strongly influential.
Second, in this study, we examine the relationship between CSR and attitude of its employees based on the person-organization fit. As suggested by Rupp et al. [9], individual employees take certain roles for the organization's evolving social consciousness since they are members of the organization. They are concerned with, contribute to, and respond to it. In addition, it can be seen that the positive behavior of the organization positively affects the organizational members through the 'trickle-down effect'.
Third, this study verified the different effects of CSR dimensions. Carroll [12]'s CSR pyramid theory has proposed different effects of each dimension of CSR, but studies validating this theory are limited. Results of this study found that companies should prioritize economic and legal social responsibility. In addition, the results of this study show that ethical responsibility is also recognized as fundamental as economic and legal responsibility are. Results of this study suggest that companies need to prioritize social responsibility activities. It is expected that fulfilling charitable responsibilities should be based on the fulfillment of other basic responsibilities, and for companies with limited resources, it is expected that prioritizing CSR activities will result in maximizing the positive effects derived from CSR.
Fourth, in this study, there was a statistically significant difference between Korea and China in terms of the effect of CSR on organizational commitment. For both countries, there was no difference in the effect of need and conformity between the economic and legal CSR on organizational commitment, but there was a significant difference in ethical responsibility. These results may imply the difference in CSR perception of each country. In China and Korea, congruence between the needs of employees and the organization's behavior regarding the economic and legal responsibility has a significant positive effect on organizational commitment. On the other hand, in terms of ethical responsibility, incongruence between CSR needs and CSR behavior results in a significantly more negative reaction to organizational commitment in Korea than in China. This indicates that negative attitudes toward the organization are more strongly formed in the case of Korea in which the CSR needs of employees do not fit CSR behavior. Ethical responsibility is recognized as essential and fundamental in Korea. Thus, compared to China, in Korea, economic, legal, and ethical responsibilities of CSR are perceived as "obligation" of organizations that exceed "responsibility". These findings can be explained in relation to CSR described in social stakeholder theory and institutional theory. The CSR effectiveness of a company can be different in relation to the understanding of stakeholders, and the level of CSR in which social justification is secured can be applied differently depending on the social situation. In other words, it was understood that the interest group could affect the socially responsible behavior of the company.

Limitation and Further Research
The limitations and future research of this study are as follows. First, subsequent relationships among the four dimensions of CSR may be considered. Some dimensions of CSR may take precedence over the others. Results of this study suggest that the effect of philanthropic responsibility differs from those of the other social responsibilities, so more detailed research on each dimension of CSR is needed. Second, we can consider various control effects in the relationship between CSR and organizational commitment. Therefore, future studies also need to take into account control variables such as psychological ownership, perceived corporate image or corporate reputation that strengthen or weaken the relationship between CSR awareness and organizational commitment. Third, in this study, differences of CSR effects between Korea and China were analyzed, but results are likely to be interpreted as a result of reflecting the collective cultures of these countries. Therefore, in the future, it will contribute to the generalization of CSR effects by analyzing CSR effects among various countries. Fourth, it will be necessary to study socially responsible human resource management practices that will link corporate-level CSR with employees' attitude and behavior. Finally, even though CSR is viewed as essential for long-term survival, few studies has been done on long-term period. Therefore it will be needed to conduct longitudinal studies.