Job Satisfaction as a Factor That Moderates the Relationship Between Internal Social Responsibility and Organizational Commitment: A Structural Equation Analysis

Abstract

Research on the relationship between employees’ perceptions of internal social responsibility and organizational commitment is a widely considered topic in the literature. The objective of this paper is to analyze whether job satisfaction moderates this relationship. A random sample was chosen of 419 workers residing in the same country (Uruguay). The participants answered a questionnaire that they received through Facebook. The perception of internal social responsibility and job satisfaction were operationalized through indicators selected by the research team based on the literature. For the organizational commitment, a shortened version of the Organizational Commitment Questionnaire was used. This questionnaire was constructed and validated in 1974 by professors Porter, Steers, Mowday and Boulian. The relationship between internal social responsibility, organizational commitment and job satisfaction was studied through correlations and structural equations. It was found that there exists a correlation between internal social responsibility and organizational commitment, and that job satisfaction moderates that relationship. In particular, it was found that the relationship between internal social responsibility and organizational commitment is stronger in the segments with less job satisfaction. From a practical point of view, these results suggest that the strategies of internal social responsibility allow an increase in the commitment of employees, but that when applying them, the state of their job satisfaction must be considered. For those same reasons, this study contributes with arguments for those who promote internal social responsibility as a strategy that improves the performance of employees.

1. Introduction

A huge number of studies have found that Corporate Social Responsibility (CSR), and particularly the one geared towards employees, is an important antecedent of organizational behavior. Yassin and Beckmann [1] revised a significant number of scientific papers on the positive impacts of CSR inside companies, identifying the following effects: desirable employee outcomes (organizational pride, organizational identification, organizational commitment, organizational attractiveness, and job pursuit intentions, employee engagement), job satisfaction and organizational citizenship behavior. According to [1], the organizational commitment is the effect which has the greatest number of academic publications. Despite that, research on the matter has not yet arrived at conclusive results. There is yet to be a consensus regarding the definition of both concepts and the scales that are utilized to measure them. Research base themselves mainly in interviews with employees or experts, but there is a lack of papers based on data provided by the companies. In general, the results are partial and for specific contexts.

There are three gaps in research on the relationship between RSI and CO. Firstly, it has centered around aspects of external CSR, while the internal dimension of it has rarely been considered [2,3]. This can be proved just by seeing the title of most publications: they talk about CSR and not internal social responsibility (ISR). Secondly, in most published studies only the employees of a sole organization or a small group of them are considered; thus, there is a lack of studies that consider all workers of a country or of an activity sector. These approaches limit the ability to understand the effect of factors that are constant for all employees in an organization but vary across organizations (such as human resources policies) [2]. Thirdly, the relationship between both variables is studied in isolation, without considering the effects that other variables have on that relationship (job satisfaction, organizational citizenship behavior, motivation, etc.) [1]. Overall, these studies tend to focus only on contrasting the existence of the statistical relationship between CSR and OC, without considering the mediating effect that other variables could have [2]. Thus, it has been pointed out that research only examines the direct effect of CSR on OC but does not contribute to the explanation of what are the processes that lead to that, particularly the question of causality [4]. In particular, few papers have studied the mediating role of job satisfaction, even though many studies have found that this variable correlates strongly with the other two. This study was designed to address these three gaps.

This paper aims to study the impact of the application of CSR practices on organizational commitment (OC) and to analyze if job satisfaction (JS) mediates in the relationship between both variables. More specifically, this paper aims to answer the following research questions: (1) Do ISR practices have an impact on the development of employees’ organizational commitment? (2) Is this relationship mediated by job satisfaction? (3) Is this relationship stronger among employees who are more satisfied with their jobs or among those who are less satisfied? It is important to note that this paper does not propose or validate an explanatory model for the relationship between the variables. The objective of this paper is limited to studying the existence of relationships between them.

The research was conducted in Uruguay. Some characteristics of this country constitute the specific context in which the research was conducted. Uruguay is a small country (3.2 million inhabitants) located in South America. It is classified as a developing country, with a GDP of USD 59.3 billion. Uruguay has the lowest GDP in South America, but it also has the highest GDP per capita (USD 23,900). Nearly a third of its inhabitants (33.6%) are employed: 25.2% in the private sector, 7.9% in the public sector, and 0.5% in the social sector. The average household income is approximately USD 1800, and the minimum wage is approximately USD 500. In 2023, there were 215,293 private businesses in Uruguay, 97% of which are micro or small. Only 900 businesses have more than 100 employees [5] There is no previous research in this country on the relationship between RSI and CO based on employee surveys.

A representative sample of employed people in Uruguay was selected. A total of 419 employees participated in this study. They were asked to answer a questionnaire on their perception of CSR in the organizations they work at (public, private or of the third sector), their organizational commitment and their job satisfaction. A structured questionnaire was utilized, which was sent to the target audience through Facebook. The relationship between the variables was analyzed using structural equations. The approach of this study allows the avoidance of some of the problems mentioned before. Firstly, this paper focuses on the perception of ISR, leaving aside CSR practices towards other stakeholders or the environment. Secondly, the analyzed population is not limited to employees of one or a few organizations, as a sample of employees residing in the same country is considered. And, thirdly, it avoids the problem of studying the impact of CSR on organizational commitment in isolation.

The results align themselves with those obtained in a large part of previous research. It was found that there exists a strong correlation between ISR and organizational commitment, and that said relationship is moderated by job satisfaction. Particularly, it was found that the impact of ISR is greater when employees feel a lower level of satisfaction. This finding has relevant practical implications because the word demonstrates that incorporate ISR practices could boost organizational commitment in contexts of low job satisfaction.

Section 2 of this paper reviews the literature. This section: (1) provides a definition of the three concepts (OC, ISR, LS), (2) describes the bibliographic background that supports the hypotheses, and (3) formulates the hypotheses. Section 3 describes the methodological aspects of the work. Section 4 presents the statistical results. Finally, Section 5 presents the conclusions of the work.

2. Literature Review

2.1. Definition of Internal Social Responsibility

ISR is the part or dimension of Corporate Social Responsibility (CSR) that focuses specifically on the responsibility of enterprises with their employees. Or, as Yassin and Beckmann ([1], p. 598) say, ISR “refers to the activities, actions, and policies of the organization that are directed to the internal stakeholders of the organization, such as employers”. Consequently, its definition is closely tied to the concept of CSR. Given that there is no consensus on the definition of CSR [6,7,8], there is also no consensus for ISR. Particularly, it highlights the fact that the idea of internal responsibility does not make sense in some ways of interpreting CSR. This is the case in theoretical approaches that assimilate or reduce CSR to the idea of Corporate Citizenship (where the focus is on the social commitment of the company) or to the concept of sustainability (where the focus is on matters of governance, and social and environmental sustainability). On the other hand, the concept of ISR makes sense in the context of theoretical approaches of CSR that interpret it as: (1) incorporating the needs and demands of stakeholders into company management [9,10] and (2) responsible management of the impacts (externalities) of the enterprise’s operation on stakeholders, society and the environment [8,11,12].

In the context of the first of these two approaches, Johnson [13] affirmed that the socially responsible enterprise balances its interest with those of its employees. To do so it is necessary that enterprises identify, involve themselves and respond to the expectations of their employees and the syndicates that represent them. Following this approach, ISR has been defined as the “part of corporate social responsibility that manages a company’s human resources in a social, ethical, humanistic, and supportive manner” ([14], p. 58). Additionally, it has been noted that ISR refers to actions of CSR related to all internal operations of the enterprise carried out with the dual objective of improving employee well-being and increasing their productivity [15]. Chatzopoulou, Manolopoulos and Agapitou [16] associate ISR with human resources practices. Van, Lang, Ngo and Ferreira [17] point out that the main difference between external and internal CSR is that the former focuses on matters related to sustainability, the protection of the environment, corporate volunteering and philanthropy, while the internal CSR focuses on issues related to the physical and psychological environment of employees. This approach is shared by many authors [18,19].

In the context of the second approach on CSR, the range of responsibilities of the enterprise towards their employees is extended, as, in addition to answering their needs and demands, ISR includes minimizing the negative externalities of the operations over them. Reder [20] proposed that CSR includes the responsibility for the internal impacts of CSR on the labor force; Gaete ([21], p. 43) proposed that companies “should pay special attention to mitigating behaviors that are related to socially irresponsible forms of expression” towards their employees. On the other hand, the ISO 26000 Guide [12] used the term Labor Practices to define a wide range of impacts of the operation of the organizations on their employees. More recently, Ofenhejm and Queiroz [22] defined ISR as sustainable work systems that minimize human and societal damages in work relations.

This study follows a definition of ISR aligned with the approach that interprets CSR as the responsible management of externalities. This definition was proposed by Licandro ([23], p. 63), who defines it as:

The responsible management of the impacts that are generated or could be generated by the operations of the enterprise over their employees, which implies minimizing the negative impacts and generating positive impacts that they are not forced to do so due to contracts or the law. These impacts occur or may occur in the context of day-to-day relations inside the company and involve all the human resource processes.

The management of ISR is performed through a group of specific practices that, in general, could be catalogued within what is called Human Resources Management (HRM) [24]. An important part of the literature on ISR includes partial lists of those practices, but it is difficult to find a somewhat exhaustive classification in homogeneous categories. Some authors include lists of ISR practices in the section of literature review [10,21,24,25,26]. Other authors include ISR practices in the indicators they use to measure ISR [18,27,28,29,30]. Other authors obtain their lists from explorative qualitative research [31]. It is also possible to find lists of ISR practices in documents of organizations that promote CSR [12,32,33,34]. In this context, some few efforts to make a classification as exhaustive as possible are noteworthy. An example is [23], who proposed a classification using 20 categories. The tool developed in our research to operationalize ISR is based on this classification, but only utilizing ten of those categories.

2.2. Definition of Organizational Commitment

Throughout the years commitment in general, and organizational commitment in particular, have been defined in different ways [35], depending on the background of each author [36]. Thus, different tools have been utilized to measure them [36]. Sheldon ([37], p. 143) defines the concept of commitment as “an orientation toward the organization which links or attaches the identity of the person to the organization. For Bouraui, Bensemmane, Ohana and Russo ([38], p. 154), commitment is a psychological state which “capturing the extent to which an individual is attached to his/her organization”. Buchanan ([39], p. 533) argued that commitment “is viewed as a partisan, affective attachment to the goals and values of an organization to one’s role in relation to goals and values, and to the organization for its own sake, apart from its purely instrumental worth”. Closon, Leys and Hellemans ([40], p. 40) pointed out that commitment is an internal process, “whose investment object may vary and is characterized by antecedents, correlation and consequences”. On the other hand, Choi and Yu ([41], p. 354) claim that organizational commitment “is defined as the extent to which an individual identifies with and is involved in an organization”. Hernández Chávez, Jaramillo Villanueva and Hernández Chávez ([42], p. 106) claim that employees develop OC when “they demonstrate belonging and participation in compliance or respect for the organization, showing gratitude and moral responsibility”. For Putri, Tanuwijaya, and Putra ([43], p. 180), OC “is characterized as an affective state that anchors employees to an organization, thereby diminishing their inclination to depart”.

In the cumulative research on OC, two theoretical approaches dominate, with their respective methodologies. The first one was proposed by Porter, Steers, Mowday and Boulian [44], who developed a measurement tool for OC that they named the Organizational Commitment Questionnaire (OCQ). They defined OC as “the strength of an individual’s identification with and involvement in a particular organization” ([44], p. 604). Additionally, they proposed that OC is composed of three elements, which they named: identification, involvement and membership. Identification consists in “a strong belief in and acceptance of the organization’s goals and values” ([44], 1974, p. 604). It is an element closely related to the concept of organizational identification proposed by [45], who interpreted this concept as the people’s perception that an organization shares their fundamental values. The second element, ([44], p. 604) characterized it as “a willingness to exert considerable effort on behalf of the organization”. This element relates to the concept of engagement, that happens when people go further than their established obligations, cooperate and help their employers [15]. Finally, a third element refers to the intention of remaining in the organization; this is a “desire to maintain organizational membership” ([44], p. 604). It should be noted that [44] does not consider those elements as dimensions of OC, as for them OC is a one-dimensional concept [46].

The second theoretical approach was proposed by Allen and Meyer [47], who suggested that OC is a multidimensional concept, composed of three dimensions. The first dimension (affective commitment) refers to “Affective or emotional attachment to the organization such that the strongly committed individual identifies with, participates in, and enjoys membership in the organization” ([47], p. 2). The second dimension (continuance commitment or perceived costs) is based on the idea that OC is also associated with the recognition of additional costs related to the interruption of an activity or the abandonment of a place. The third dimension (normative commitment or obligation) relates to the commitment with “a belief about one’s responsibility to the organization” ([47], p. 3), which has a strong moral content. Based on this framework they constructed a model and a scale, which they named the Three-Component Model (TCM).

This study adopts the definition of Porter, Steers, Mowday and Boulian [44] because a unidimensional focus is more appropriate than a multidimensional one.

2.3. Relationship Between Internal Social Responsibility and Organizational Commitment

2.3.1. Theories That Establish the Relationship Between ISR and OC

Different theories have been utilized to explain, from a theoretical point of view, the relationship between CSR (particularly, ISR) and OC. One of the most used ones is the theory of social identity. Based on this theory, some authors claim that ISR encourages employees identify with the company’s values and, therefore, commit to it [48]. Chatzopoulou, Manolopoulos and Agapitou [16] say that the actions of CSR will increase the commitment of the employees, as it generates a greater identification with the company and strengthens their sense of belonging in it, all which reinforces their self-esteem. On the other hand, Bouraoui, Bensemmane, Ohana and Russo [38] establish the relationship between ISR and OC in the theory of social exchange, which proposes that employees tend to behave according to the perception they have of how they are treated and valued by their organizations. A basic principle in this theory is reciprocity: when employees feel that the company is socially responsible with them, they respond reciprocally with a greater organizational commitment [4].

Shen and Zhu [49] establish the relationship between ISR and OC in the theory of social attribution [50]. This theory proposes that employee behavior responds to how they perceive the policies that are applied in the workplace [49]. According to this theory, OC rises when employees attribute an honest attitude behind the CSR policies of the company. The theory of organizational justice has also been utilized, which combines three concepts: distributive justice inside the company (in other words, how resources are distributed), procedural justice (in other words, how the procedures are applied) and interactional justice (in other words, the way in which employees are treated in the company) [51]. Based on this theory, Hofman and Newman [52] propose that, when employees have a positive perception of CSR of their enterprises, they also have a perception of organizational justice in them, which increases affective and moral commitment.

2.3.2. Empirical Evidence on the Relationship Between ISR and OC

Many studies have found a positive correlation between CSR (or ISR) and OC. A review of some academic articles that achieved this result are presented in Table 1. It is possible to see there are some elements that strengthen the validity of this relationship. In first place, geographical diversity suggests that the relationship between CSR/ISR and OC was registered in different cultures and economies. Secondly, this relationship was found in different segments of companies, determined by activity (banking, industry, food, telecommunications, education, etc.), the size (big, SMEs, etc.) and the type of property (private and state). And, thirdly, the studies were applied to different segments, determined by their roles in the organization (directors, executives, employees, etc.).

On the other hand, it can be observed, in Table 1, that there is a problem that negatively affects the comparability of results and their extrapolation: both CSR/ISR and OC were operationalized and measured with different tools (indicators and scales). In the case of CSR, some authors used the tool developed by Maignan and Ferrell [53] that is based on the definition of CSR proposed by Archie Carroll [5,16,40,53,54,55,56,57]. Other authors utilized a tool designed by Turker [18], that supports the interpretation of CSR as satisfying the needs and demands of stakeholders [3,38,52,58,59]. Finally, a group of authors utilized indicators designed by other researchers [48,55,60,61] or their own indicators [19,43,51,62].

In the case of OC, most authors utilized the tool designed by Allen and Meyer [47] or one of its previous versions. Some authors considered the three types of OC (affective, normative and of continuity) [41,49,63], others considered two of them [16,52]; but most of the studies only took into account the affective commitment [5,38,40,41,56,59,64]. The OCQ was utilized by a small group of authors [18,57,58]. Additionally, very few authors utilized their own indicators or ones adapted from previous works [43,48,61]. Despite how widespread the tool designed by Allen and Meyer [47] is, this study opted to utilize a shortened version of the OCQ.

The combination of theories and the reviewed research establish the basis for the first hypothesis of this study. The former sets a theoretical basis, while the latter sets an empirical one, for the hypothesis that establishes:

Table 1. Synthetized information on scientific articles that found a positive relationship between ISR and OC.

Article	Theories	Sample	ISR (CSR) Measurement	OC Measurement—Utilized Indicators
[53]	Does not mention	Sample 1: marketing executives (n = 210) and students of MBA in the United States (n = 154)	Own scale	Indicators of other authors
[55]	Social identity	Business professionals of varied functional and organizational areas, in the USA (n = 278)	Maignan and Ferrell [53]	Indicators of other authors
[48]	Social identity	Retail banking employees of the United Kingdom (n = 4712)	Own scale	Indicators of other authors
[18]	Social identity	Business professionals from Turkey (n = 269)	Turker’s indicators [18]	Shortened OCQ
[58]	Does not mention	Employees of various companies in Pakistan (n = 371)	Turker’s indicators [18]	Shortened OCQ
[19]	Social exchange	Banking employees in Jordan (n = 336)	Own scale that measures give dimensions of ISR	Indicators from the TCM
[49]	Attribution	Employees and managers of two aluminum companies in China (n = 784)	Own indicators, taken from various authors	Indicators from the TCM
[64]	Does not mention	Employees of a German multinational that reside in 17 countries (n = 1084)	Own scale	Indicators from affective commitment, extracted from the OCQ and TCM
[52]	Social identity Organizational justice	Employees of five Chinese SMEs (n = 280)	Turker’s indicators [18]	Indicators from the TCM on Affective and Normative
[4]	Social exchange Social identity	Employees of 11 companies in Pakistan that disseminate CSR activities (n = 392)	Turker’s indicators [18] plus an indicator of Maignan and Ferrell [53]	Shortened version of the TCM
[40]	Social identity	Employees of enterprises with more than 300 employees in Belgium (n = 621)	Own scale, adapted from Maignan and Ferrell [53]	Revised version of the TCM
[51]	Social identity, Signaling Social exchange Organizational justice	Employees of 18 food enterprises in the USA (n = 827)	Own scale with eight indicators	Indicators from affective commitment from the TCM
[41]	Social identity Vision based on resources	Directors in Chinese enterprises (n = 175)	Own indicators on CSR (environment, ethic, philanthropy, relationship with stakeholders)	Five indicators of the TCM, that use the three types of commitment
[56]	Social identity Social exchange	Employees of enterprises located in three Chinese provinces (n = 700)	Scale based on Maignan and Ferrell [53]	Indicators from affective commitment from the TCM
[57]	Social identity	Teacher and employees of eight universities in Pakistan (n = 245)	Scale based on Maignan and Ferrell [53]	Shortened OCQ
[38]	Social identity Social exchange Deontic justice Social justice	Employees of 19 public and private enterprises, in Tunisia (n = 161)	Turker’s indicators [18]	Indicators from affective commitment from the TCM
[59]	Social identity Social exchange	Employees of five telecommunication enterprises in Pakistan (n = 229)	Turker’s indicators [18]	Indicators adapted from the TCM
[16]	Social identity	Employees of different hierarchical levels in Greece (n = 189)	Maignan and Ferrell scale [53] Differences between ISR from CSR	Indicators from affective and continuity commitment from the TCM
[61]	Social identity	Employees of six multinationals in Pakistan (n = 216)	Own scale inspired by Carroll [54]	Indicators of other authors
[62]	Does not mention	Workers in Ecuador (n = 318)	Own indicators, based on various previous works	Own indicators based on the OCQ, the TCM and others
[2]	Social identity Social exchange	Employees of manufacturing companies in Italy (n = 263)	Items on Turker’s scale [18] on ISR	Indicators from affective commitment from the TCM
[43]	Does not mention	Employees of universities in Indonesia (n = 254)	Own indicators on ISR and external CSR.	Indicators on affective, normative and continuity commitment.
[65]	Social exchange	Internal auditors of 15 banks in Jordan (n = 148)	Own scale	Three indicators from the TCM

Source: own elaboration.

Table 1. Synthetized information on scientific articles that found a positive relationship between ISR and OC.

Article	Theories	Sample	ISR (CSR) Measurement	OC Measurement—Utilized Indicators
[53]	Does not mention	Sample 1: marketing executives (n = 210) and students of MBA in the United States (n = 154)	Own scale	Indicators of other authors
[55]	Social identity	Business professionals of varied functional and organizational areas, in the USA (n = 278)	Maignan and Ferrell [53]	Indicators of other authors
[48]	Social identity	Retail banking employees of the United Kingdom (n = 4712)	Own scale	Indicators of other authors
[18]	Social identity	Business professionals from Turkey (n = 269)	Turker’s indicators [18]	Shortened OCQ
[58]	Does not mention	Employees of various companies in Pakistan (n = 371)	Turker’s indicators [18]	Shortened OCQ
[19]	Social exchange	Banking employees in Jordan (n = 336)	Own scale that measures give dimensions of ISR	Indicators from the TCM
[49]	Attribution	Employees and managers of two aluminum companies in China (n = 784)	Own indicators, taken from various authors	Indicators from the TCM
[64]	Does not mention	Employees of a German multinational that reside in 17 countries (n = 1084)	Own scale	Indicators from affective commitment, extracted from the OCQ and TCM
[52]	Social identity Organizational justice	Employees of five Chinese SMEs (n = 280)	Turker’s indicators [18]	Indicators from the TCM on Affective and Normative
[4]	Social exchange Social identity	Employees of 11 companies in Pakistan that disseminate CSR activities (n = 392)	Turker’s indicators [18] plus an indicator of Maignan and Ferrell [53]	Shortened version of the TCM
[40]	Social identity	Employees of enterprises with more than 300 employees in Belgium (n = 621)	Own scale, adapted from Maignan and Ferrell [53]	Revised version of the TCM
[51]	Social identity, Signaling Social exchange Organizational justice	Employees of 18 food enterprises in the USA (n = 827)	Own scale with eight indicators	Indicators from affective commitment from the TCM
[41]	Social identity Vision based on resources	Directors in Chinese enterprises (n = 175)	Own indicators on CSR (environment, ethic, philanthropy, relationship with stakeholders)	Five indicators of the TCM, that use the three types of commitment
[56]	Social identity Social exchange	Employees of enterprises located in three Chinese provinces (n = 700)	Scale based on Maignan and Ferrell [53]	Indicators from affective commitment from the TCM
[57]	Social identity	Teacher and employees of eight universities in Pakistan (n = 245)	Scale based on Maignan and Ferrell [53]	Shortened OCQ
[38]	Social identity Social exchange Deontic justice Social justice	Employees of 19 public and private enterprises, in Tunisia (n = 161)	Turker’s indicators [18]	Indicators from affective commitment from the TCM
[59]	Social identity Social exchange	Employees of five telecommunication enterprises in Pakistan (n = 229)	Turker’s indicators [18]	Indicators adapted from the TCM
[16]	Social identity	Employees of different hierarchical levels in Greece (n = 189)	Maignan and Ferrell scale [53] Differences between ISR from CSR	Indicators from affective and continuity commitment from the TCM
[61]	Social identity	Employees of six multinationals in Pakistan (n = 216)	Own scale inspired by Carroll [54]	Indicators of other authors
[62]	Does not mention	Workers in Ecuador (n = 318)	Own indicators, based on various previous works	Own indicators based on the OCQ, the TCM and others
[2]	Social identity Social exchange	Employees of manufacturing companies in Italy (n = 263)	Items on Turker’s scale [18] on ISR	Indicators from affective commitment from the TCM
[43]	Does not mention	Employees of universities in Indonesia (n = 254)	Own indicators on ISR and external CSR.	Indicators on affective, normative and continuity commitment.
[65]	Social exchange	Internal auditors of 15 banks in Jordan (n = 148)	Own scale	Three indicators from the TCM

Source: own elaboration.

Hypothesis 1.

There exists a positive relationship between ISR practices and organizational commitment of employees.

2.4. Relationship Between Job Satisfaction and Internal Social Responsibility with Organizational Commitment

2.4.1. Definition of Job Satisfaction

The interest in studying job satisfaction has been going on for decades, and it has been motivated by its influence on matters of commitment, rotation and absenteeism [66]. This concept has been defined by a wide range of scientists from numerous perspectives. Cherrington [67] proposed that job satisfaction can be conceptualized from two perspectives: as an attitude towards things and as something that responds to each individual. But, in general, job satisfaction is conceptualized as a feeling, something emotional. Churchill, Ford and Walker ([68], p. 327) defined the satisfaction of employees as “his affective or attitudes feelings toward his job, his organization and his work environment”. For Odom and Sharda ([69], p. 159), this concept refers to “the extent to which a worker feels positively or negatively about his or her job”. Currivan ([70], p. 497) defined it as the degree “to which an employee has positive emotions toward the work role”, while for Llovet and Fito ([71], p. 1071) job satisfaction is “the degree of positive emotions which a member of the organization has in connection with his/her employment”. On the other hand, Brammer et al. ([48], p. 1706) defined it as “the degree to which people like their Jobs”, while Eliyana, and Pradana ([72], p. 432) proposed that job satisfaction “is a feeling of pleasure or a positive emotional state which is the result of the quality of one’s work and work experience”. Locke [73] proposes that satisfaction and dissatisfaction constitute complex reactions to the job, and states that both “are a function of the perceived relationship between what one wants from one’s job and what one perceives it as offering or entailing” ([73], p. 316).

Lum, Kervin, Clark, Reid, and Sirola [74] proposes that job satisfaction is affective response to specific aspects of the job and is not a response to the organization as a whole. This sentiment comes from an evaluative judgement (positive or negative) that employees have of their job or on their work situation [40]. This evaluation can be realized both on the job in general as well as specific parts or elements of it [75]. For [76], job satisfaction is a feeling that responds to working conditions and relationships with colleagues and superiors. Amongst these elements, recognition and compensation received by the employees are accounted for, as well as the way they are supervised, the work conditions, work challenges, and the relationship with co-workers [77]. In this study, the elements that give employees satisfaction or dissatisfaction are divided into three groups. Firstly, structural aspects, that is, characteristics of the environment in which they work: the work environment, the work conditions, the salary conditions, the work relationships, the norms, etc. In second place, aspects related to the function they perform: the type of tasks they do, the responsibilities assigned to them, the possibilities of professional development, etc. Finally, the interpersonal relationships: the relationships with their co-workers, the one with their supervisors, etc.

2.4.2. Relationship Between Internal Social Responsibility and Job Satisfaction

ISR practices help employees to manage their personal relationships with their professional ones, reducing their stress and improving their job satisfaction. For Glavas and Kelly [51], CSR can contribute a meaning and purpose to employees, which impacts their job satisfaction. For Barakat, Isabella, Boaventura and Mazzon [78], CSR can generate positive evaluations on employees regarding their own job. Similarly, behaviors opposed to CSR can generate a negative evaluation. In the first case this evaluation can translate into higher job satisfaction, while in the second, the opposite happens. Based on the theory of cognitive dissonance, Valentine and Fleishmann [79] propose something similar. They highlight the ethical behavior of companies and the development of social responsibility strategies towards stakeholders that generate a higher job satisfaction, while behaviors contrary to that ethic and of social irresponsibility could generate major dissatisfaction.

Asrar-ul-Haq, Kuchinke and Iqbal [57] claimed that there has been little research on the relationship between these two variables, while [78] pointed out that there are few studies that show that social responsibility influences job satisfaction. Amongst these studies, some noteworthy ones are: [2,16,40,48,51,56,57,65,80,81,82]. Based on these precedents, this document evaluates the following hypothesis:

Hypothesis 2.

There exists a positive relationship between ISR practices and job satisfaction.

2.4.3. Relationship Between Job Satisfaction and Organizational Commitment

Many decades ago, academics suggested that job satisfaction is a antecedent of organizational commitment [83]. More recently, Hefny [84] claimed that the relationship between both variables is one of the most studied aspects in the literature of management. According to Brammer, Millington and Rayton [48] theoretical analysis and empirical research agree that a higher level of satisfaction is related to a higher level of commitment. This is because, in general, more immediate emotions (such as satisfaction) generate more lasting affiliative attitudes (as is the case with OC) [71]. Despite OC and JS (job satisfaction) being affective responses, Mowday, Steers and Porter [85] propose that there exists at least two differences between those concepts: (1) OC refers to a general response to the organization, while JS reflects a response to the personal work experience towards specific situations; (2) OC is more stable across time than JS, because OC is developed in a slow but constant manner, while JS is conditioned by everyday events.

Across the last couple of decades, numerous studies, in different contexts, found a positive correlation between job satisfaction and OC [3,16,17,43,48,56,57,80,86,87,88,89,90]. The study by Meyer and Allen (1987) [86] is noteworthy, it verified the correlation between JS and OC with measurements in three different moments: one month, five months, and nine months after new workers joined.

Despite these results, the existence of a causal relationship between two variables is still up for debate. The most common perspective in academia is that there is a causal relationship between JS on OC. This point of view is based on the implicit assumption that the orientations of employees towards a particular work experience precedes the orientations towards the organization as a whole [70]. However, some academics state the opposite relationship: that JS is a consequence of OC [91]. This point of view has been formulated by Bateman and Strasser [92] and Vandenberg and Lance [93]. This debate is of a conceptual nature as most research studies the correlation between both variables, but do not do it strictly on the causality. A different case is [70], who throughout a longitudinal study evaluated four causal models between OC and JS. It was found that three models which propose the existence of causality between both variables (OC on JS, JS on OC and reciprocity) were not valid, and the only model that had some validity is the one that established no causal relationship between JS and OC.

Based on the previous considerations, the following hypothesis will be evaluated:

Hypothesis 3.

There exists a positive relationship between job satisfaction and organizational commitment.

2.4.4. Job Satisfaction as a Mediator in the Relationship Between ISR and OC

When analyzing previous Works on the perception of CSR and OC, Peterson [55] observed that the mechanisms ingrained in that relationship had not been studied, that is, the possible mediating role of other variables. Glavas [94] found that, out of 181 articles on the relationship between CSR and OC, only 16 studied the mechanisms that intervene in that relationship; and amongst those, only three articles considered factors related to attitudes and behaviors of individuals. Bouraoui, Bensemmane, Ohana and Russo [38] pointed out that it is known that there is a relationship between the perception of CSR and OC, but there is no sufficient knowledge on “why and how” this relationship functions. The lack of studies on the mediation of other variables has led to the assumption that the application of CSR would automatically produce a higher OC [2]. On this line, Chatzopoulou, Manolopoulos and Agapitou [16] claim that previous research has not explored adequately the precise nature of the link between CSR and organizational commitment. Therefore, Ashan and Khalid [2] stated that knowledge about how CSR impacts CO is still a significant gap in the literature.

The few works that studied the existence of the underlying mechanisms in the relationship between CSR and OC have considered the following variables: organizational identification [5,59], organizational trust [5,62], organizational climate [23], intrinsic motivation [62], perception of receiving organizational support [51], finding meaning in the job [51], satisfaction with HR policies [56], organizational ability [95] or cultural aspects [52]. Job satisfaction has been considered a mediating variable in [3,16,23,43,64,96]. Jawaad, Amir, Bashir and Hasan [97] pointed out that job satisfaction mediates the relationship between Human Resources practices and organizational commitment. Nambudiri and Tweari [98] proposed that this variable is probably a mediator in the relationship between CSR perception and CO. For Hossen, Chan and Hasan [96], job satisfaction is an underlying mechanism that connects ISR practices with organizational commitment. In the study, it was found that JS mediates the relationship between some dimensions of ISR (Empowerment and Employee stability) and OC, but not others (working environment, training education). Chatzopoulou, Manolopoulos and Agapitou [16] established this idea using the argument that commitment is a consequence of satisfaction across time.

Overall, research on this matter validates the hypothesis that job satisfaction mediates the relationship between ISR (or CSR) and organizational commitment [3,16,23,43,96]. Additionally, Ting [99] found that job satisfaction mediates the relationship between internal marketing (which shares some practices with ISR) and organizational commitment; even demonstrating that the direct effect of internal marketing on OC is lower than the indirect effect mediated by JS. On the other hand, Mueller, Hattrup, Spiess and Lin-Hi [64] found that the mediation does not exist. Consequently, the cumulative of these studies allows the assumption that job satisfaction could mediate in some way the relationship between CSR and OC. In accordance with that, the following hypothesis will be tested:

Hypothesis 4.

The relationship between ISR and organizational commitment is mediated through job satisfaction.

Social exchange theory indicates that the implementation of SRI practices generates reciprocal behaviors in employees. Among these behaviors is commitment. In this context, job satisfaction can act as a trigger for reciprocal behaviors. Job satisfaction, as a positive feeling, activates employee commitment as a way of returning to the company the benefits they receive from it. Therefore, hypothesis 4 can be supported by social exchange theory. Social identification theory also serves to support this hypothesis, but it does so more indirectly. According to this theory, SRI practices foster employee identification with the company, which is one of the components of organizational commitment. More satisfied employees are likely more willing to perceive the benefits derived from SRI practices. In this sense, satisfaction creates better conditions for SRI to positively impact OC.

Figure 1 illustrates the relationships between the variables proposed in the hypotheses.

3. Methodology

A quantitative study was designed to evaluate the hypotheses. This study is based on a structured questionnaire administered to a random sample of employees residing in Uruguay. The data collected were analyzed using various statistical techniques. These methodological aspects are described below.

3.1. Population and Sample

The concept of CSR was introduced in Uruguay in the early 2000s. Little is known about the implementation of CSR in Uruguay because little research has been conducted on this topic. In particular, there is no diagnosis of the current state of CSR in this country. The few studies have been based on the study of very small samples of companies that are part of organizations that promote CSR. These studies found that: (1) CSR has a positive impact on financial performance [100], (2) companies that publish sustainability reports have more developed CSR [101], (3) CSR is more developed in companies that have managers in charge of this issue [102], (4) the implementation of CSR is not influenced by the gender of the CEO [103] and (5) there is a strong relationship between internal social responsibility and organizational commitment [22]. However, these results cannot be generalized to the universe of companies operating in Uruguay. Unlike previous works, this work addresses the study of CSR from the perspective of citizens.

The research was applied to all people living in Uruguay who work in private companies, state organizations or the third sector. It was chosen to work with this population in order to avoid some of the research gaps in this field of study, which were already described in Section 1. Most research in this field is applied to a more bounded population. In general, it is applied to employees of a single organization. Consequently, in a large part of the research on employees’ perceptions are conditioned by the same organizational context, that can have some influence on them. On the other hand, in this study the perceptions of employees are not conditioned by the same organizational context: each employee is from a different organization.

A structured questionnaire was utilized, which was sent through Facebook. The sampling procedure was as follows: (1) Facebook was contracted; (2) Facebook was informed of the profiles of the people to be surveyed; (3) Facebook applied its algorithms to randomly send the questionnaire to people with that profile; (4) the responses were deposited in an Excel database. The sample size was determined based on the funds available to contract the Facebook service. We were informed that with the available funds, a sample of approximately 400 cases would be obtained. The final sample was 419. Given how this social network uses its algorithms to identify the target population and to invite them to respond to the questionnaire, the randomness of the sample is affected by numerous biases: (1) excludes employees that do not use Facebook; (2) over-represents the segment of people that tend to answer surveys through social media.

The sample is heterogeneous regarding the type of organization that the employees work at, their size and geographical location within Uruguay. The sample includes three types of organizations: companies (285), public or state organizations (99) and organizations of the third sector (35). It is also heterogeneous regarding the size of the organizations in which they work: with up to 20 employees (133), between 21 and 100 employees (85), between 101 and 500 employees (84), more than 500 employees (117). As for geographical location of the companies, the sample is divided by employees residing on the capital of Uruguay (170) and the rest of the country (249). Additionally, the sample is heterogeneous in the type of tasks performed by the employees: management (33), technical (93), administrative or commercial (113), operating tools, machines or vehicles (54) and others (125).

3.2. Questionnaire and Variables

A structured questionnaire was utilized, which was sent through Facebook. The surveys distributed through Facebook cannot be extensive. Thus, few indicators were used to operationalize the three main variables. For internal social responsibility, 10 of the 20 SRI dimensions proposed by [23] were selected. Nine of these dimensions were operationalized using a single indicator. Only the “Health and job safety” dimension was operationalized with two indicators. Each indicator was arbitrarily determined by the research team (see

Table 2. Indicators utilized to operationalize each variable.

Variable	Dimension	Code	Text
Internal Social Responsibility	Discrimination/diversity	ISR1	Does everything possible so people do not feel discriminated by their age, gender, race, religion, handicap, etc.
	Protection of human rights	ISR2	Handles disciplinary measures respectfully
	Handling of information	ISR3	Informs the employees on relevant matters that affect or could affect them (changes in tasks or place of work, etc.)
	Dialogue with employees	ISR4	Listen to employees when they express dissatisfaction, raise concerns, or file a complaint
	Labor climate	ISR5	Promotes a good work and relationship environment within the company
	Health and job security	ISR6	Prevents accidents at work and cares for the security of employees
		ISR7	Prevents work illnesses (stress, tendinitis, back issues, etc.) and brings support to the employees when they get them
	Promote professional development	ISR8	Promotes and facilitates employee training and professional development
	Promote personal development	ISR9	Promotes the participation and initiative of their employees
	Social benefits for employees and their families	ISR10	Provides support to employees facing personal or family problems (health, financial, addiction, etc.)
	Act in an ethical manner	ISR11	Promotes that bosses act in an ethical manner and treat their subordinates in a just way
Organizational Commitment	Identification	OC1	I think that my values and those of this company are very similar
		OC2	I have pride in telling others I am part of this company/organization
		OC3	For me, this is the best company/enterprise possible to work at
	Effort	OC4	I am willing to make much more effort than what is normally expected of me, in order to help the success of this company/organization
		OC5	This company/organization really brings out the best of me regarding my performance at work
		OC6	I would be willing to do almost any task in order to continue working in this company/organization
	Membership	OC7	I feel a lot of loyalty towards this company/organization
		OC8	For me to quit this company/organization drastic changes in my current circumstances would have to occur (work or personal)
		OC9	Having chosen to work in this company/organization has been a good decision on my part
Job Satisfaction	Structural aspects	JS1	I am satisfied with the work environment in this company
		JS2	I am satisfied with the conditions of my job
	Function	JS3	I am satisfied with the tasks I do
		JS4	I am satisfied with the responsibilities I am given
	Interpersonal relationships	JS5	I am satisfied with the understanding I have with my peers
		JS6	I am satisfied with the understanding I have with my direct superior

Source: Own elaboration.

). Each indicator consists in a phrase that describes an ISR practice. The respondents were asked to answer the question: What is your opinion on the actions of the directives of the organization/public company in which you work, in each of the following situations? To answer, the respondents utilized a Likert scale in accordance with (and with semantical support), five values, with the following equivalences: one (disagree), two (somewhat disagree), three (neither disagree nor agree), four (somewhat agree), five (agree).

For organizational behavior, a shortened version of the OCQ was utilized. Each one of the 18 indicators was translated to Spanish, the contents of each one, were analyzed and, finally, nine were selected (three for each one of the components). In each case small changes were introduced to the text, in order to facilitate their comprehension. In particular, the negative affirmations were avoided. The ICQ has been validated in multiple previous studies conducted in various contexts. Consequently, it was decided not to conduct a new validation in the Uruguayan context. Table 2 includes the selected indicators. The respondents were asked to indicate the level of agreement they had with the contents of each affirmation, utilizing the same Likert scale. The choice of the OCQ rather than Allen and Meyer’s tool was argued in Section 3. For the answers, the same scale as the one for ISR was utilized.

Job satisfaction is something difficult to measure as it is based on general evaluative judgements on a wide array of affective experiences at work [40]. Thus, none of the set of indicators utilized to operationalize this variable has been able to encompass this diversity of experiences. This has occurred both in studies designed to include many indicators (for example: surveys carried out throughout a personal interview), and studies designed to only utilize short questionnaires, as is the case in this study. Three types of factors were chosen to determine employee satisfaction or dissatisfaction: structural, function-related, and interpersonal relationship-related. Two indicators were chosen to operationalize each of these (see Table 2). These indicators were chosen arbitrarily by the research team. The indicators were not subjected to a pretest nor were their psychometric properties evaluated. For the answers, the same scale as the previous two variables were utilized.

The survey was conducted in the last semester of 2023. It is unknown what the answering rate was because the social network did not give out that information. The processing of the information and the statistical analysis were realized through SPSS 22.0 and Smart PLS 4.

4. Results

4.1. Common Method Bias

The common method variance bias refers to the possible alterations to the true correlations between observed variables due to the use of a common scale in the measures derived from a sole source of data [104,105]. In order to evaluate the bias of the common method variance the VIF indicator of collinearity was used on Smart PLS throughout the complete evaluation of collinearity [106], for that all the variables were connected to just one. There is bias in the method if the VIF is higher than 3.3 at a factor level [106]. The resulting values of the variance’s inflation factor (VIF) were lower than 3.3. These results suggest that it is not likely that the CMB represents a significant worry in this study.

4.2. Estimation of the Measurement Model

Following the recommendations of Hair, Risher, Sarstedt and Ringle [107], the estimation of the proposed conceptual model is realized in two stages: first the measurement model is analyzed, to verify that the indicators associated with each construct are valid and sufficiently solid to represent adequately the latent variables included; and secondly, the structural model is evaluated, with the aim of analyzing the causal relationships between latent constructs and to determine the degree fit and the predictive validity of the model. Regarding the measurement model, the indicators’ factorial loads were calculated, as well as the composite reliability, Cronbach’s alpha, the average variance extracted (AVE) and the discriminant validity of the variables considered in the model.

Regarding the reliability indicators of the first order reflective variables, those gave adequate results as Cronbach’s alpha and the composite reliability (CC) were above the recommended value of 0.70 [108]. In regard to the additional reliability indicator of Rho of Dillon-Goldsteins, it was confirmed that all the values of Rho were higher than 0.70 [109]. To evaluate the convergent validity the values of Average Variance Extracted (AVE) were examined, confirming that the values are greater than the minimum value of 0.5 [110]. Additionally, the individual reliability of the indicator variables was evaluated through the factorial loads, corroborating that their factorial loads were superior to 0.70 [107]. Considering what was previously presented, a good degree of internal consistency of the constructs was confirmed, as well as an adequate convergent validity (see

Table 3. Measurement model.

Variable	Mean	Standard Deviation	Loading	t Student	p	Cronbach’s Alpha	Rho A	CR	AVE
Internal Social Responsibility						0.976	0.977	0.979	0.811
ISR1	3.413	1.613	0.799	29.266	0.000
ISR2	3.060	1.626	0.907	43.521	0.000
ISR3	3.181	1.580	0.917	50.385	0.000
ISR4	3.198	1.621	0.903	45.823	0.000
ISR5	3.232	1.635	0.901	47.714	0.000
ISR6	2.838	1.573	0.918	49.070	0.000
ISR7	2.936	1.591	0.925	53.497	0.000
ISR8	3.033	1.611	0.937	51.854	0.000
ISR9	3.136	1.581	0.874	45.019	0.000
ISR10	3.024	1.592	0.933	47.860	0.000
ISR11	3.053	1.569	0.884	40.983	0.000
Job Satisfaction						0.976	0.977	0.979	0.811
JS1	3.184	1.591	0.839	55.630	0.000
JS2	3.737	1.465	0.891	66.174	0.000
JS3	3.267	1.454	0.873	63.921	0.000
JS4	3.726	1.376	0.838	39.983	0.000
JS5	3.689	1.462	0.905	85.438	0.000
JS6	3.606	1.490	0.870	61.104	0.000
Organizational Commitment						0.951	0.955	0.959	0.722
OC1	3.043	1.552	0.815	40.575	0.000
OC3	3.422	1.554	0.901	87.792	0.000
OC3	2.862	1.514	0.879	79.550	0.000
OC4	3.532	1.487	0.843	46.735	0.000
OC5	3.143	1.535	0.908	88.167	0.000
OC6	2.690	1.485	0.743	29.524	0.000
OC7	3.403	1.553	0.878	64.099	0.000
OC8	3.470	1.539	0.797	36.693	0.000
OC9	3.492	1.486	0.869	35.772	0.000

Source: Own elaboration.

).

To confirm the discriminant validity, the criterion of Fornell and Larcker (1981) [111] and the method of HTMT relation [105,107] were utilized. On the criterion of Fornell and Larcker [111], the square root of AVE must be higher than the correlations with other variables. The study showed that all the square roots represented in the main diagonal, are higher than the correlations with other variables of the model. On the other hand, Hair, Risher, Sarstedt and Ringle [107] determined that to corroborate the discriminant validity the values of HTMT cannot be higher than 0.90.

Table 4. Discriminant validity.

	Internal Social Responsibility	Job Satisfaction	Organizational Commitment
Internal Social Responsibility	0.900	0.772	0.777
Job Satisfaction	0.743	0.870	0.858
Organizational Commitment	0.751	0.813	0.850

Note: The diagonal elements (in cursive) are the square roots of AVE. Above the diagonal the proportions of HTMT are found. The elements below the diagonal are the correlations of the constructs.

shows the obtained results.

Finally, before evaluating the proposed structural relationship, the non-existence of collinearity between the variables of the model had to be proved [107]. The collinearity can be estimated through the variance inflation factor (VIF). The ideal values of VIF should be close or lower than 3 [107].

Table 5. Result of the structural model.

Hypothesis	Paths	β	VIF	t-Student	p-Value	Remarks
H1	Internal Social Responsibility → Job Satisfaction	0.743 ***	1.000	29.906	0.000	Accepted
H2	Internal Social Responsibility → Organizational Commitment	0.328 ***	2.230	6.538	0.000	Accepted
H3	Job Satisfaction → Organizational Commitment	0.570 ***	2.230	11.823	0.000	Accepted
H4	Internal Social Responsibility → Job Satisfaction → Organizational Commitment	0.423 ***		11.149	0.000	Accepted

R² (job satisfaction) = 0.552; R² (organizational commitment) = 0.710. SRMR = 0.049. *** p < 0.001.

shows that all values of VIF of the group of predictive variables in the model are lower than 3, which ensures the inexistence of multicollinearity problems.

4.3. Modelling of Structural Equations and Testing of the Hypothesis

Once the measurement model was validated, the structural model was examined in global and local ways. For the global evaluation, the SRMR adjustment criterion was utilized, which indicates a good fit of the model with values lower than 0.08 [109]. The proposed model has an estimated value of 0.049 (see Table 5), which indicates that the expected frequencies are like those observed, showing a good fit. The local model was estimated by analyzing the standardized trajectory coefficients (β) which were obtained through the resampling bootstrap method [108], on over 5000 subsamples. Additionally, the evaluation of the model was made through the variance of dependent latent variables explained by the constructs that predict them (R2). The values of R2 obtained with bootstrap indicate the capacity of the model to explain the variability of the variables [112], the higher the R2 values are, the more their predictive capacity is, with the minimum value being 0.10. The results of Smart PLS showed that all the direct impacts of ISR are positive and significant: job satisfaction (β = 0.743, t-value = 29.906, p < 0.001); organizational commitment (β = 0.328, t-value = 6.538, p < 0.001), which means that hypothesis H1 and hypothesis H2 are supported, respectively.

On the other hand, the results support the positive impact of job satisfaction in organizational commitment (β = 0.570, t-value = 11.823, p = 0.000), so hypothesis H3 is supported. The results also give data on the specific indirect effect to prove the mediating effect of job satisfaction in the relationship between internal social responsibility and organizational commitment. This indirect effect resulted significant and positive (β = 0.423, t-value = 11.149, p = 0.000), which supports the mediating effect of job satisfaction in the relationship between internal social responsibility and organizational commitment, and thus, hypothesis H4 is supported. All the results are shown in

Table 6. Indexes of ISR, organizational behavior and job satisfaction: descriptive statistics.

Indexes	Mean	Median	Standard Deviation	Percentiles
				33%	67%
ISR Index	3.13	3.27	1.48647	1.91	4.36
Job Satisfaction Index	3.47	3.83	1.36374	2.67	4.50
Organizational Commitment Index	3.20	3.44	1.35390	2.44	4.11

.

Additionally, robustness checks were performed to reinforce the credibility of the findings. The common method bias was examined using the full collinearity assessment (VIF), and all values were below 3.3, suggesting no significant threat of common variance bias. Likewise, the multicollinearity diagnostics confirmed that all VIF values were well below the accepted threshold of 3. These results, already reported in Section 4.1 and Section 4.2, strengthen the statistical robustness of this study. Nevertheless, as with any self-reported survey, the possibility of biases such as social desirability cannot be completely ruled out; therefore, strategies such as anonymity and the use of validated scales were applied to mitigate them.

4.4. Correlations in Segments Determined by the Degree of Job Satisfaction

Complementarily, the correlation between ISR and OC was studied within three segments, determined by the degree of job satisfaction. To that end, indexes were constructed for each of the three variables. In Table 6, descriptive statistics of the three indexes are shown. In the case of job satisfaction, the limits of the percentiles were used to divide the sample into three segments determined by the degree of satisfaction. Given the small sample size, it was decided to use a small number of segments. Percentiles were chosen so that the three segments included the same number of cases. A satisfaction level below 2.67 was called low JS; a satisfaction level between 2.67 and 4.5 was called medium JS; and a satisfaction level above 4.5 was called high JS. In

Table 7. Simple correlation between the ISR index and organizational behavior in segments based on job satisfaction.

Job Satisfaction Level	Correlation Coefficient
Low satisfaction (<2.67)	0.504 **
Average satisfaction (2.67 to 4.50)	0.516 **
High satisfaction (>4.50)	0.389 **

** The correlation is significant at a level of 0.01 (bilateral).

, it can be observed that the correlation between ISR and OC is lower in the segment with employees with a higher degree of satisfaction. Thus, the mediation of JS consists in moderating the relationship between ISR and OC. This indicates that the introduction of ISR practices has a bigger impact on OC in cases of low or medium JS, than when JS is high.

5. Conclusions

5.1. Theoretical Implications, Contribution to the Field of Study, and New Lines of Research

The results of this work contribute to theoretical development in the field of CSR. CSR theory argues that the implementation of ISR practices generates positive impacts for companies. In the case of CSR practices, these impacts are related to employee attitudes and behaviors: increased organizational commitment, reduced turnover, increased productivity, etc. This work adds to the long list of previous studies that have shown that the implementation of CSR generates greater organizational commitment. Therefore, this work reinforces the value of CSR theory.

Another important contribution of this study is that it focuses exclusively on the impact of ISR on OC. In most previous studies, the authors utilized general indexes of CSR or indexes of external CSR, but did not account for the specific effects of ISR on OC. The research based on external CSR can be based on the theory of social identification (CSR influences OC because employees tend to identify themselves more with organizations that act responsibly with society, the clients or the environment) and the theory of social attribution (CSR influences OC because employees can verify the consistency between what the organization says and what they do). But external CSR cannot be based adequately on the theory of social exchange nor the theory of organizational justice. The former, because external CSR does not benefit employees, and the latter because external CSR does not say anything about how the organization behaves with them. On the other hand, ISR can base itself on four theories, as the socially responsible behavior of the organization with its employees facilitates that they identify themselves with it and its values (theory of identification), stimulates the reciprocity behaviors of the employees (theory of social exchange), allows employees to verify the consistency between what the organization says and what they do (theory of social attribution) and because an important part of ISR practices involve behaviors of procedural and interactional justice (theory of organizational justice).

Another contribution is related to the population studied and the sample used. As opposed to most previous research, this work uses a representative sample of the employees of different types of organizations (companies, state organizations and social organizations), different size and different geographical locations. Consequently, its results are much more generalizable than other studies that included employees of only one or a few organizations. In particular, this study confirms in a more solid way the results from them.

Despite the existence of a wide variety of publications that demonstrate that there exists a positive relationship between CSR and OC, there are few works that study the mechanisms that mediate this relationship. In this research, it was found that job satisfaction moderates this relationship, that is, that the impact of ISR on OC is not direct. In particular, it was found that this impact is higher in contexts of low satisfaction. These results contribute to the field of study because it asks new research questions. How does job satisfaction moderate the relationship between ISR and OC? Which dimensions of job satisfaction have the greatest influence (task satisfaction, supervisor satisfaction, coworker satisfaction, etc.)? Why does ISR have a higher impact on OC in contexts of low JS than in high JS? Which ISR practices have the greatest impact on OC in low-satisfaction contexts? This work was conducted in the specific context of a small developing country. It is worth asking whether the moderating effect of CSR operates equally in small and large economies, in developing and developed economies. These questions may lead to new lines of research.

5.2. Practical Implications

Also, this paper contributes to the management of ISR. The introduction of new practices of ISR in organizations is made with the aim of generating positive impacts on the performance of employees (amongst them, increasing OC). The results of this work suggest to managers the need to evaluate what the starting situation is in terms of job satisfaction, before introducing those practices. These results suggest that managers can expect a strong impact of these practices in contexts of low job satisfaction. Consequently, ISR is a good tool to increase commitment of employees in organizations where they realize tasks that are not very satisfactory, receive low salaries and have low chances of personal development. Inversely, managers should not expect ISR to increase substantially the commitment of employees in organizations in which there already exists an important degree of job satisfaction. In summary, this paper informs managers of the importance of investigating employee satisfaction before introducing SRI practices.

5.3. Limitations

Finally, some limitations of this study must be pointed out. Firstly, the sample is not fully random, due to the biases that Facebook’s algorithm introduces. This indicates that the assumptions of randomness are not fulfilled in their entirety, which is required to apply the statistical techniques used. Secondly, a small number of indicators were used to operationalize ISR and JS. Therefore, this study does not allow us to address the multidimensionality of both concepts. Thirdly, the indicators used to operationalize SRI, OC, and SJ were arbitrarily selected and were not evaluated through a pretest. Fourthly, the research was applied to only one country. In consequence, the results obtained must be considered as provisional, until other studies confirm or invalidate them with greater force. In particular, new similar studies are needed that: (1) utilize samples selected through strictly random methods, (2) include a set of indicators that enable to better encompass the multidimensionality of the three concepts and (3) are realized in other countries or include a variety of countries with different characteristics.