Effects of corporate social responsibility on employee organizational identification: Authenticity or fallacy

Greenwashing implies the predominance of the fallacious application of the corporate social responsibility (Symbolic CSR) over its authentic practice (Substantive CSR). It represents, by definition, a conceptual conflict, worth it to observe in light of its impact on the organizational identification of employees. This work examines this impact by carrying out a cross-sectional investigation and analyzing it through structural equation modeling. The study is applied to a sample of two hundred and twenty professional employees in Colombia. Results obtained are mainly in line with the theory; they show a positive effect of Substantive CSR on employees’ organizational identification, as well as a futile character of Symbolic CSR on this same attitude. However, the study also finds a direct correlation between Symbolic CSR and Substantive CSR, which suggests a possible tacit acceptance of greenwashing as a valid organizational practice. This work contributes to enhance the theory, given the fact that a recently proposed conceptual model was academically tested for the first time ever. Likewise, at a practical level, results suggested that the decision-making process when related with CSR is under continuous scrutiny by employees, in function of its sense of authenticity and therefore with consequences that may affect their behavior.


Introduction
Greenwashing, as an antagonistic idea of the true meaning of corporate social responsibility (CSR), has become an emerging topic whose means and ends have captivated academics and professionals who seek to understand how it works and at the same time mitigate its spread (Lyon & Montgomery, 2015). One of the many ways to do this could be through an analysis of the effect of its adoption on employees and their behaviors in the workplace (Donia & Tretault-Sirsly, 2016). Mainly, organizational identification (OID) has been consolidated in the literature as a construct that allows the connection between a series of organizational stimuli with some behavioral results at the individual level (Lee, Park & Koo, 2015). Therefore, considering that this behavior has been widely analyzed in different manifestations at the organizational level, an evaluation of the way in which it could be affected by the adoption (or not) of practices that coincide with greenwashing could be pertinent. For the purpose of this work, it is determined by the perception of the authenticity (or not) of CSR.
In this way, on the one hand, it is clear that the theoretical conception of greenwashing, interpreted as an exercise of conveying untrue messages regarding CSR, can be a valid antecedent of adverse effects both for organizations and for societies in general (Lyon & Montgomery, 2015). On the other hand, some organizational results such as more significant commitment and motivation, less absenteeism and higher job performance are in turn driven by adequate levels of employee OID (Jones & Volpe, 2011). Nevertheless, further and more refined knowledge is still needed regarding the connection that may exist between these two concepts (Glavas & Godwin, 2013). In particular, this need is tangible within contexts where aspects such as the environmental and social impact generated by the dynamics of organizations are critical to the construction of the future development of their societies (Jamali & Karam, 2016;Rodrigo, Duran & Arenas, 2016).
Thus, understanding greenwashing as the difference between false and authentic actions related to CSR (Walker & Wan, 2012), and OID as the self-perception of an employee expressed as an approximation of their sense of belonging to the organization of which they are a member (Jones & Volpe, 2011), this work addresses the impact that the former can have on the latter. In other words, the present paper intends to identify how much the authentic or false nature of the CSR of an organization influences the sense of belonging of its employees. In this vein, the framework used integrates some contributions of the theory of organizational legitimacy (represented by the notion of greenwashing) with the foundation on CSR at the individual level, or micro CSR (represented by the OID).
Methodologically, an exploratory cross-sectional research is carried out to empirically evaluate a theoretical model in which the OID takes on the role of a dependent variable and interrelates with CSR. In turn, this latter variable is analyzed regarding its dual lens; this is through both substantial (practical and authentic) and symbolic (rhetorical or false) perceptions. Specifically, we develop a multivariate study covering a sample of two hundred and twenty professional employees with jobs of responsibility in Colombia. Thus, three hypotheses are tested using a model of structural equations based on exploratory and confirmatory factor analyses.
The structure of the manuscript consists of five additional chapters. In the first instance, a theoretical framework in which the constructs to analyzed are presented together with scientific antecedents where we demonstrated a more profound relevance of their theoretical association. Subsequently, there is address the methodology, where we develop the working hypotheses, the design of the instrument, and the respective structural equation model formulation. The fourth chapter includes the results obtained in its various stages. The fifth chapter contains the discussion generated from the respective interpretation of the results, their implications, and possible future lines of research. Finally, the paper closes with the conclusion reached.

Theoretical framework
Within the body of knowledge of organizational theory, there has been an exciting development with respect to the effects generated by the different ways in which CSR and its multiple associated concepts impact employees at the level of their perceptions and behaviors (such as motivation, commitment, satisfaction, intention to resign and identification, among others). This field of study has been called Micro Corporate Social Responsibility (Micro CSR), which is a derivation of the field of study of organizational behavior at the individual level (Rupp & Mallory, 2015). In the same manner, organizational legitimacy has played an essential role within this discipline, and despite a series of fascinating contributions that shed light on this relation, there is a continuous warning of the existence of countless opportunities for the creation of more and better knowledge in this regard (Aguinis & Glavas, 2012;Delmas & Pekovic, 2013;Evans & Davis, 2014;Evans, Davis & Frink, 2011;Lamm, Tosti-Kharas & King, 2015). The following theoretical framework is structured from the general to the particular, starting with CSR as a macro-field of study linked to the concept of legitimacy and ending with some research background that has somehow involved them with the OID. Finally, we will note the specific literature gap that this work intends to address.

Corporate Social Responsibility (CSR)
Since the decision-makers in the organizations understood that the final objectives should not be only financial but also social and environmental (Elkington, 1998), the CSR concept has been linked to the efforts of searching this type of results in its environments. In other words, the so-called sustainable development has gained an essential space in management (both practical and theoretical), implying that the emergence of CSR has become a very relevant issue in management (Aguilera, Rupp, Williams & Ganapathi, 2007). The Sustainable development is universally defined as: "development that meets the needs of the present without compromising the ability of future generations to meet their own needs" (United Nations World Commission on Environment and Development, 1987, p. 3). Moreover, in fact, such as the influence of this concept that some authors cataloged it as the new dominant paradigm in terms of evaluating organizational and personal performance (Boudreau & Ramstad, 2005).
Research in the field of CSR has shown a recent trend towards assessing the authenticity of its adoption by organizations (De Roeck & Maon, 2016). In this sense, the precise definition of CSR explicitly excludes from its domains aspects such as compliance with the law (including environmental or labor legislation itself). The payment of taxes and royalties, the attempt to compensate irresponsible action with other benevolent actions (of a quid pro quo type), the concealment or manipulation of the truth, conspiracies by specific sectors to increase profits at the expense of their clients and suppliers, and direct and indirect participation in industries of indiscriminate exploitation of natural resources (Berrone, 2016).
In a recent literature review on organizational psychology linked to CSR, Professors Jones & Rupp (2017) argue that social responsibility has commonly been related to feelings, expressions, and actions. Therefore, "CSR is often conceptualized, described, and implemented as a set of socially oriented practices, policies, and initiatives" (idem; p. 4). At the same time, it should be noted that CSR has a stable link with stakeholder theory (Freeman & McVea, 2006), which in turn focuses on actions taken into consideration of "those that could be affected by, or affect the achievement of, the objectives of the organization" (idem; p. 46). In that same line, given that employees effectively represent a group of stakeholders that may affect and (or) are affected by the achievement of the objectives of the organizations to which they belong, they must undoubtedly be considered as internal stakeholders.

Legitimacy and greenwashing
The theoretical foundations of the concept of legitimacy go back to strong philosophical roots that reach themes anchored in institutional theory. Its unfolding dates (back to the 4th century BC), from the definition of Aristotle's famous "good act" (Koehn, 1995) up to the emergence of tensions between philanthropy and the creation of value for Godfrey's stakeholders (2005); through the analysis of the dissociation (or better known by its original term: decoupling, due to its difficult polysemy) between real organizational structures and those that one wishes to present to society (Meyer & Rowan, 1977). In this sense, this pressure to legitimize a responsible image can lead companies to adopt CSR practices only in order to comply with a particular formalism and end up manipulating the truth (Ramus & Montiel, 2005). In this manner, at the organizational level, the greenwashing effectively implies a dissociation between the real commitment to CSR and the pressure (legal, social, or economic) to apply CSR. That reason is, perhaps the explication of greenwashing as a phenomenon that is continuously associated with the disclosure and management of information intentionally manipulated in order to show a responsible public image (Phyper & MacLean, 2009). Suchman (1995) suggests that legitimacy is based on consistency between the values stated by the organization, its practices, and behavior standards. In other words, it is when a company seeks to fit appropriately into society in order to seek a kind of social license (moral license) to operate. Therefore, according to this definition, greenwashing implies a kind of absence of unaccepted legitimacy (Milne & Patten, 2002). As such, greenwashing could also see as a legitimizing strategy that occurs when organizations voluntarily promote an impression of social and environmental values even when they do not perform strongly in these aspects (Mahoney, Thorne, Cecil & LaGore, 2013).
A definition of Greenwashing is the gap between substantial and symbolic corporate social actions (CSAs) (Walker & Wan, 2012). Symbolic CSAs refer to what the company does to demonstrate socially responsible behavior to society, while substantial CSAs relate to real intentions to engage in socially responsible behavior. A mismatch between them (i.e., the intention of the symbolic CSAs prevails over the intention of the substantial CSAs) means the emergence of greenwashing. The most common way of evidencing such an appearance is by comparing rhetoric or symbols, with action or substance (Lyon & Montgomery, 2015). And although most of the time this phenomenon is perceived as a form of calculated deception by organizations, the evidence also shows some cases in which, approaching the precepts of cognitive dissonance (Festinger, 1962), an aspirational rhetoric could be the initial step-voluntary or involuntary-for the subsequent implementation of genuine CSR actions (Christensen, Morsing & Thyssen, 2013).

Symbolic (Symbolic CSR) and substantial (Substantial CSR) corporate social responsibility
In recent research that evolved favorably from a presentation at the 2015 Academy of Management Conference, the works of Donia & Tretault-Sirsly (2016) and Donia, Tetrault Sirsly & Ronen (2017). The works were able to develop a concrete proposal for the theoretical cut and operationalization of constructs, which suggests their link to the development of organizational behaviors based on the authenticity of CSR. It is a logic base on the tension between two sides of the same coin: Symbolic CSR and Substantial CSR.
Their contributions derive from the concepts previously defined by Walker & Wang (2012), concerning symbolic and substantial CSAs. Thus, it is argued that while Symbolic CSR involves those practices that represent any action related to CSR issues that a company develops to show a kind of ceremonial formalism, Substantial CSR involves real changes at the operational level, usually involving tangible and measurable sacrifices, which in turn require a significant mobilization of resources from the organization (Weaver, Trevino & Cochran, 1999;Zott & Huy, 2007). Therefore, the simplified approach of this two-dimensional construct is based on the difference between the authenticity or satisfaction of the real needs of society from the action of the organization in question (Substantial CSR), and the fallacy (or importance) given to the appearance of meeting those needs and thus it finds benefits in a better reputation in society (Symbolic CSR).

Organizational Identification (OID)
With deep conceptual roots anchored in the realms of organizational behavior, OID is treated initially as a topic, defined and developed by Mael & Ashforth (1992) in an academic exercise conducted with graduates of an educational institution. There, it is possible to interpret this behavior as a form of social identification, where the identity of the individual is a derivation from its classification into social categories or groups. Thus, OID was initially defined as "the perception of oneself concerning the sense of belonging to an organization" (Mael & Ashforth, 1992;p. 105). In other words, OID may be what defines an individual to the organization of which they belong. This construct is considered necessary in the literature because-among other more humanistic arguments-it has been catalogued as a positive activator of other organizational behaviors (Lee et al., 2015), such as commitment at work (Riketta, 2005), job satisfaction (Van Dick et al., 2004), and the low intention of job rotation (Cole & Bruch, 2006).
It must be acknowledged that from the contributions of Mael & Ashforth, OID became an academic and practical benchmark in organizational studies. Additionally, it began to be treated as a work behavior related to the affective commitment of employees and their potential to promote tangible and collective behavior in organizations. In fact, the academic importance of the concept has widely highlighted in both research spaces. From the select forum on it, in conjunction with the concept of organizational identity, in Volume 25 of the American Management Review introduced by Albert, Ashford & Dutton (2000) to the recent contributions of Miscenko & Day (2016) an exciting effort is made to compile history and literature concerning this and other related concepts.

Association between CSR and OID
One of the visible common elements among the studies published to date has been the determination of the role of OID as an effect or as a mediator due to the different forms of presenting the CSR topic (Lee et al., 2015;Riketta, 2005). In this sense, it is worth highlighting the impact generated by authors such as Carmeli, Gilat & Waldman (2007); Grant, Dutton & Rosso (2008); and Kim, Lee, Lee & Kim (2010), who laid the groundwork for future contributions that have further expanded the corresponding knowledge. In general terms, all the contributions have coincided in the validation of a positive association between these two significant concepts. However, although OID has basically been governed by its original guidelines (Mael & Ashforth, 1992), the treatment given to CSR has entailed a great concept of diversity. Some examples of the above may be: CSR perceptions (Brammer, Millington & Rayton, 2007), internal vs. external CSR perception (De Roeck & Maon, 2016;Glavas & Godwin, 2013), CSR activities or practices (Farooq, Merunka & Valette-Florence, 2013;Grant et al., 2008;Peloza & Papania, 2008), CSR associations and participation in CSR (Kim et al., 2010;Mozes, Josman & Yaniv, 2011), corporate responsibility on stakeholders and perception of organizational support on the environment (Lamm et al., 2015;Shen, Dumont, Deng & Deng, 2016), among others. Therefore, the versatility of CSR allows the exploration of more evolved concepts in line with the real problems of the organizational field. Also, in this vein, as mentioned above, this work highlights the importance of exploring the tension between the authentic and false facets of CSR in terms of its impact on OID.

Methodology
Some mentions founded concerning the involvement of the authenticity of CSR in its specific relation to individual employee behaviors and which call for further research development in this regard (Brammer, He & Mellahi, 2015;Glavas & Godwin, 2013;Rupp, Skarlicki & Shao, 2013). Similarly, it is also stated that studies are needed to help examine the dynamics of stakeholder perceptions of the authentic and inauthentic nature of CSR. However, it is necessary to recognize that the work of Donia et al. (2017) has been the only work to make a concrete proposal to date on the subject. In fact, in their manuscripts, they explicitly invite the continuation of the particular line of research based on the empirical testing of their proposals, and the possibility of their study in new contexts.
Specifically, considering the impact of those concepts on organizational behavior, the theoretical propositions of these authors are two. First, the attributions of substantial CSR will be positively related to favorable behaviors on the part of employees; and Second, the attributions of symbolic CSR will have no-relation, or this relation will be negative with favorable behaviors on the part of employees (Donia et al., 2017;Donia & Tretault-Sirsly, 2016). In this order of ideas, and taking into account the predictive character of modeling through structural equations developed in this study, it could be said that taking into account gender, age and profession as control variables, the first hypotheses that adjust to the problem posed here would be as follows: H1: The attributions of substantial CSR are positively associated with the level of employee OID.
H2: The attributions of symbolic CSR are not associated, or their association is negative on the level of employee OID.
Likewise, the interaction between the independent variables of the model and the-what we could call, logical-the result of the null or negative interdependence between them found in Donia et al. (2017) is taken into consideration. In this manner, the aim is to validate that the concepts of symbolic CSR and substantial CSR are indeed contrary to, or at least independent of, each other, meaning that authentic CSR practices are incompatible with greenwashing practices. Consequently, this article tests the following hypothesis: H3: The correlation between symbolic CSR and substantial CSR is negative or non-existent. Based on the above hypotheses, we applied a cross-sectional study in order to carry out the present investigation using the application of a self-administered physical questionnaire (applied in September 2016). Later, we analyzed the data collected using structural equation modeling (Wright, 1921) carried out in R language version 3.2.2 (a free and open-source language), which allows evaluating the relationship between one or more independent variables. The Structural Equation Model (SEM) is a statistical methodology that departs from the proposal of the theoretical interaction between variables from a path diagram, finally determining the existence of this relationship through its robustness and eventual level of significance. Specifically, the SEM uses a confirmatory factorial analysis or multiple linear regression approach, which involves determining the linear relation between latent (e.g., theoretical constructs) and observable variables (e.g., each component or item of a theoretical construct), in conjunction with dependency or covariance relations among themselves (Byrne, 1998). The method starts from the development of a measurement model in order to identify the observable variables that affect the latent variables; in such a way that-and after cleansing-it is possible to proceed with the determination of their interaction and the subsequent obtaining of results (Manzanares & Pérez, 2015). It is worth highlighting that the main reasons for justifying the selection of the methodology used are the capacity of the SEM to support a possible non-normality in the data. As well, the methodology can estimate through a principal components analysis loads of model (always guaranteeing the same load value), where there are as many latent variables as observed. (Kline, 1994).

Sample and procedure
The corresponding instrument originally was applied to two hundred and seventy-six professionals working in Colombia, particularly in the cities of Bogotá, Bucaramanga, Cúcuta, and Barrancabermeja. The data collection was physically useful, taking advantage of the fact that these professionals were also participants in different graduate programs in management areas of the School of Industrial and Business Studies of the Industrial University of Santander. Once the surveys were completed and collected; we carry out a primary data cleaning, eliminating forty-three answers due that they were incomplete or belonged to individuals who had no direct employment relationship with any company (i.e., many of them are independent professionals and (or) contractors), and the respective foliation of the others was carried out. Subsequently, after the data tabulation, we conduct a second cleansing with the elimination of three surveys that presented inconsistent information (such as more than one answer in the same question or a high number of blank answers).
Finally, we consolidated a homogeneous sample of two hundred and twenty professionals (they work in different types of companies with diverse responsibility roles). All of them were in the preparation stage for the management exercise. Therefore, taking into account their own academic or experiential perspective, it is assumed that they have at least a basic knowledge of issues related to organizational dynamics such as CSR and human talent management. A brief description of the sample is a group of young adults with a marked engineering profile, where the majority are under 32 years of age (64.5%), without dissimilarly among genders. Figure 1 shows the main characteristics of the sample based on gender, age, and profession; variables of interest for the formation of socio-demographic profiles.

Measurements and instrument
In addition to the control variables previously defined: gender, age and primary professions (which did not present statistical differences), the survey measured three latent variables or constructs: substantial CSR, symbolic CSR and OID, the observable variables or items of which were rated by the respondents on a Likert scale of 1 to 5, with 1: completely disagreeing and 5: completely agreeing. In order to measure OID, we used what could be called the 'golden standard' in the calibration of this construct: a six-item scale proposed by Mael & Ashforth (1992), highly cited and validated in the relevant literature. Likewise, we considerate the work of Donia et al. (2017) for the measurements of symbolic CSR and substantial CSR, using a six-item scale for the former, and an eight-item scale for the latter.
In this manner, the corresponding instrument included a total of twenty observable variables structured in the form of a sentence or statement. Of the codification described in Annex 7.2, the set of statements in 2.1.1 to 2.1.8 measure the degree of substantial CSR, in questions 2.2.1 to 2.2.6 the level of symbolic CSR, and questions 3.1 to 3.6 the OID. The order of these items was randomly altered to reduce the perception of monotony in the responses by the respondents. (See Annex 7.1 for the relation between the variables, and Annex 7.2 for the statements), the summary of the variables is in Table 1. We highlight that before the official application of the instrument, both the English and Spanish versions were subject to external review by five academic experts in France, Spain, and Colombia. Their areas of expertise ranged from CSR and organizational behavior to research questionnaire design. In this way, the final version applied collected their comments. Once the survey was applied, and before initiating an exploratory factor analysis (EFA), the reliability of the instrument was verified using Cronbach's Alpha coefficient. The co-variability characteristics of the data were also found to be relevant for a dimension reduction analysis; for these purposes, this work applies the Kaiser-Meyer-Olkin (KMO) sample adequacy measure and Bartlett's Test of Sphericity (BtS). Table 2 compares the study values with the compliance parameters of the proposed statistical tests to verify the relevance of performing an EFA.

Results
We have done the EFA choosing three factors using the Kaiser criterion (1960) through the principal components method, without finding a significant statistical difference between the control variables, thus explaining 60.7% of the data variance. From the original dataset, questions 2.2.3, 2.2.6, and 3.6 (two questions related to symbolic CSR and one to OID) are not considered because their loads did not fit the proposed three-factor model (their value was not high enough: to influence the assigned factor), decision supported by a confirmatory factor analysis or structural measurement model. Table 3 shows the relations between observed variables and constructs (loads). As a highlight, we noted that depending on the value of the loads, factor 2 (OID) presents negative but high values (absolute value greater than 0.5). Later, we evaluated the proposed structural equation through the sem and lavaan statistical packages of the R software. Getting a high statistical significance (goodness of fit index of 0.929), thereupon, the organizational identification depends on the substantial perception of CSR developed in the companies (the value of the slope of the Substantial CSR is 8.805). In addition, although the true and false CSR perceived by professionals varies (the p-value of the influence of Symbolic CSR is significant, and there is a correlation between the variables in Annex 7.1), there is not enough statistical information to affirm a significant degree of relation (neither positive or negative) between the latter and OID. In the same way, there is a direct and statistically significant correlation between substantial CSR and symbolic CSR (estimated value of the correlation between the constructs is 0.178), which contradicts H3. Table 4 summarizes the main results. Similarly, the model represented in Figure 2 presents a graphic approximation of the findings. Table 4 records the fitness indices or parameters of the corresponding model, finding a value that falls in the lower setting range only in the Chi-square test; however, this behavior may be due to the fact that said test is sensitive to the sample size, therefore, when increasing the latter, the probability of rejection of the fitness hypothesis is increased (Bagozzi & Yi, 1988). Therefore, other estimation indices are analyzed (considering degrees of freedom and complexity of the model) that show an adequate adjustment of the proposed model. Besides, we calculate the Root Mean Square Error of Approximation (RMSEA), with the interpretation as the mean approximation error per degree of freedom. Therefore, we conclude a proper fit of the model. Source: Own elaboration.

Discussion
Previous studies have focused on the relation between CSR (and its different manifestations) with different employee work behaviors. Predominately, they have found some positive relations between the constructs analyzed. In fact, some scholars consolidated the idea of a positive effect generated by CSR on OID in the work environment (Glavas & Godwin, 2013;Jones & Rupp, 2017). The literature has been commissioned, then, to present sufficient evidence to affirm that CSR practices effectively improve the way an employee identifies with their organization. However, these nuances have not taken into account the search for the legitimacy of such organizational practices. This study was concerned with the latter questioning, in such a way that it assessed both the authentic (i.e., substantial) and false (i.e., symbolic) nature of CSR and the effect they have on the OIDs of the employees. We obtained results varied in nature. On the one hand, there was a strong and significant positive dependence between substantial CSR and OID, as well as a causal disconnection between symbolic CSR and OID (supporting hypotheses 1 and 2). On the other hand, we found a weak positive correlation between substantial CSR and symbolic CSR (rejecting hypothesis 3). The analysis has two fundamental assertions. The first is the initial ratification of two of the proposals to the theoretical model proposed by Donia & Tretault-Sirsly (2016), which later served as input to the validation and operationalization of constructs (Donia et al., 2017). The second is an unexpected contradiction observed, with two opposite concepts having a direct correlation, especially when in the initial validation tests of these authors, the correlation found between them was-logically in appearance-of an inverse or null nature. As for the former, it coincides with the theory. Therefore, it could be thought that an employee is indeed capable of perceiving and processing signals derived from CSR practices emitted by their organization and differentiating them between authentic (substantial) and false (symbolic) practices, as well as independently appropriating them for shaping or not shaping their behaviors (in this case, their OID). About the later, we observed a sort of consent to the organizational adoption of false CSR practices by employees in a context of self-definition as members of an organization.
In consequence, the theoretical and practical implications of the study are diverse. In terms of theory, we made progress in the knowledge of both the field of study of micro CSR and the conceptual aspects surrounding the theory of legitimacy over employees, especially when this legitimacy is defended through deception (which is part of greenwashing). Additionally, according to the knowledge and understanding of the authors, this paper is the first empirical partial validation of a recently proposed theoretical model. The fact that the research proposed by Donia & Tretault-Sirsly (2016) was carried out in the context of a Latin American country (with non-conventional results partly different to the original theoretical proposal), opens the door to identifying nuances in the formulated theory that might make one think of context as an explanatory element of the results (Jamali & Karam, 2016;Rodrigo et al., 2016). However, it is essential to acknowledge that the latter requires further research to arrive at more definite conclusions.
From a practical standpoint, it is possible to capitalize on these results in the light of organizational decisions related to CSR, together with their dissemination to different sectors of society. The evidence obtained suggests that employees appear to be no strangers to these signals and can process them to shape their behaviors (perceptions, intentions, and behaviors) in the work environment. Therefore, if the decision-makers can manipulate the genuine concept of CSR, the consequences may not be so convenient for their workforce and, therefore, have repercussions in terms of the sustainability of society in general, and organizations in particular.
One of the apparent limitations of this study relates to the almost exclusive focus of the instrument used on the social issues of CSR, which could leave aside the environmental aspects of CSR. The preceding, (despite to the fact that we no included the contributions obtained), is previously justified by Donia et al. (2016), since they explicitly propose the equivalence of social consideration with environmental protection on the part of organizations, in light of the perceptions of their employees. However, an exercise that could project more specific research fronts would be the clear delineation of the social, environmental, and even economic dimensions of CSR. Likewise, as previously done by Amato, Busraschi & Peretti (2016), the fact that we differently approached CSR and sustainability is something that could add more value to the discussion in the present study, given that the latter concept is sometimes perceived as more critical than CSR itself.
Likewise, another point of discussion relates with the cross-sectional nature of our data, which would not allow affirming to suggest causal relations. However, we mitigated it with the use of structural equation modeling as well as strong theoretical support that in part allows predicting such relations in the conceptual model proposed. In the same way, we recognized that it is impossible to guarantee the generalization (or external validity) of the results obtained, taking into account the characteristics of the sample. Nevertheless, the balance generated by the use of similar samples in previous studies (Brammer et al., 2015;Carmeli et al., 2007;Kim et al., 2010) together with the homogeneity in the profile of the members of the sample examined here, are elements that provide some reassurance in order to reduce potential biases in the study.
Despite the above, some controversy may also arise regarding the context in which it was presented. In this sense, having collected data in Colombia may make the results relevant in different ways in other cultural contexts. However, if that is the case, in addition to contributing to a better understanding of the specific process in that country, this study also opens the door to future contrast and comparison of its results in other contexts-similar or not-in order to produce a more precise and comprehensive knowledge on the subject. Both CSR and the nature of the individuals who make up the organizations are concepts with territorial nuances, and their analysis cannot ignore those variables that separate them from the rest. For example, Colombia, like many other Latin American countries, is a biodiverse and developing country. However, it presents profound social problems that make the results of this type of study a real challenge in terms of its interpretation and implementation. In this sense, corporate deception about CSR practices is something that should not be tolerated in these types of regions, as the implications of doing so can be costly for their present and future well-being.
In addition to the refinement of the above limitations, future research could focus on the validation of the relations between substantial CSR and symbolic CSR with other behavior constructs specific to the field of organizational behavior at the individual level, including, if deemed appropriate, OID through its mediatory character. Likewise, the possibility of involving moderating variables and (or) more mediating variables is an opportunity that will always be present to achieve more progress in the knowledge gathered. It also opens the door for qualitative research projects, which could provide more profound clues that add to the definition of greenwashing in a context similar to those typically studied in the literature. In this sense, case studies, content analysis, or ethnographic work in emerging countries could represent exceptional examples that contribute even more to the comprehensive understanding of employee behavior in organizational environments.

Conclusions
The evidence gathered in this study suggests that employees may perceive substantial CSR practices positively and be indifferent to symbolic CSR practices. We observed this evidence in the evaluation of their effect on the identification perceived by them as members of their organizations. However, considering previous studies, it could also be inferred that other behavioral outcomes such as job satisfaction, commitment, and even effectiveness may depend on such perceptions.
On the other hand, a disturbing finding that challenges the initial assumptions of the theoretical propositions used (Donia & Tretault-Sirsly, 2016) is that in some ways these employees also vindicate the use of symbols, ceremonies, and formalisms to legitimize the (often authentic) CSR practices of their organizations. In other words, there is a kind of validation of the positive, aspirational discourse (Christensen et al., 2013) where it would seem that there would not be much controversy. It must be accepted that organizations have the right to make use of various mechanisms to disseminate the fulfillment of a purpose-oriented to CSR when it exercises a genuine presence in organizational aspects (policies, processes, operations, and goals, among others), in addition to making use of them to reinforce their positions and define objectives (Berrone, 2016).
Nevertheless, the results could also be a clear sign of concern where the concept of CSR could have alterations according to past academic or work experience. Therefore, it is cause for concern that some empty practices such as the validation of discourse without action, environmental certifications seeking only the fulfillment of a requirement, donations seeking tax benefits or even claiming having contributed to the development of society through compliance with the law, are recognized as CSR. They all represent forms of greenwashing through the use of subtle mechanisms that attempt to create an altered image of reality (Mahoney et al., 2013;Milne & Patten, 2002). The fact is that this eventual situation would imply an enormous challenge at the educational level to redirect efforts to make clear the real definition of CSR (Berrone, 2016). Moreover, while the latter is a comprehensive topic at all levels and for all areas of knowledge, it is particularly sensitive at the management training level, as the participants are the ones who, subsequently make decisions in the organizations.

A1.
Correlation matrix between variables observed for the structural model Source: Own elaboration.

Code
Var. Phrase -Statement P2.1.1 CSR_Sub1 I think my organization is interested in CSR because it really cares about the community in which it operates.

P2.1.2 CSR_Sub2
My organization is interested in CSR issues because it cares about those actors with whom it interacts and (or) does business.

P2.1.3 CSR_Sub3
My organization is interested in CSR programs and activities because it wants to help solve community problems.
P2.1.4 CSR_Sub4 I feel that my organization has a genuine interest in the well-being of those outside individuals who are affected by its operation or business. P2.1.5 CSR_Sub5 My organization considers it essential to help those in need.

P2.1.6 CSR_Sub6
My organization is involved in CSR programs because it really wants to help those with whom it interacts and (or) does business.
P2.1.7 CSR_Sub7 I consider that having the opportunity to interact with the community is something that is highly valued in my company. One of the interests of my company in sympathizing with CSR is to avoid the criticism (bad press) by some media or the actors with whom it interacts and (or) does business.
P2.2.5 CSR_Sym5 I believe that my organization takes advantage of CSR to look like an ethical company to society.

P2.2.6 CSR_Sym6
My organization adopts CSR with the aim of making its employees understand that there is a concern for those external actors who, in one way or another, are related to its operation. P3.3 OID3 When I refer to the organization of which I am a member, I prefer to speak of "we" rather than "them." P3.4 OID4 The success of the organization is my own.
P3.5 OID5 When somebody says something positive about my organization, I take it as a personal compliment.

P3.6 OID6
If I saw any real news that criticized or mentioned anything negative about my organization, I would feel ashamed.