The KAC-CSR Model in the Tourism Sector

: This article carries out, for the ﬁrst time in the scientiﬁc literature, an integrated analysis of the variables of knowledge, application, and communication of CSR in an economic sector (in this case, tourism), through the application of a novel model called KAC-CSR (knowledge, application, and communication of corporate social responsibility); this model interrelates the three concepts and the possible factors causing them. The objective is to identify the possible causal relationships among these CSR variables and their respective determining factors. In practice, this implies carrying out a ﬁrst empirical veriﬁcation of the proposed model, seeking to determine its viability as an explanatory tool. Applying the structural equation modeling (SEM) approach, using partial least squares (PLS), to a sample of 224 hotels in the Colombian Caribbean region, this analysis veriﬁes the proposed model and speciﬁes its characteristics. This study ﬁnds that a greater degree of advanced CSR knowledge lends itself to a greater degree of CSR application. This, in turn, positively inﬂuences CSR communication. Likewise, motivations have a positive inﬂuence on CSR application, particularly in economic and social activities, while obstacles have a negative inﬂuence on CSR application. Moreover, different characteristics of the manager and the hotels determine the levels of knowledge, application, and communication of CSR. The paper also provides evidence on the determining factors inﬂuencing the ‘knowledge–application–communication’ sequence of CSR, an aspect not studied until now. Future research should consider more stakeholders and replicate the KAC-CSR model in other economic sectors and geographical areas.


Introduction
Today, it is clear that many scholars agree that corporate social responsibility (CSR) is a social construct and that companies, in this matter, have the responsibility to ensure that all aspects of the business have a positive impact on society [1].In this sense, it is interesting to carry out a detailed analysis of CSR within the field of tourism, since it is one of the most important sectors of the world economy, considered key to development, prosperity, and wellbeing.
However, the social and environmental potential negative impacts of tourist activity on the destinations are also well known; therefore, there is a need to implement CSR initiatives at the international, national, sectoral, and business levels that allow for the mitigation of these potential negative externalities [2].
The concept of CSR has been thoroughly debated in the tourism and hospitality literature due to its importance to sustainable tourism development [3].Authors such as Paskova and Zelenka [4] investigated how the CSR concept is implemented in tourism and analyzed how it contributes to its sustainability, evidencing the practical importance of CSR for tourism sustainability.Under this reality, CSR has become a guiding tool that allows progress in the debate and practice of tourism sustainability [5].Mihalic [6] went further and made a new proposal integrating the terms sustainable and responsible in tourism under the word 'responsustable', demonstrating that responsible tourism (CSR in tourism) is a practical means to achieve sustainable tourism.
Clearly, the tourism sector was one of the most affected by the COVID-19 pandemic [7], wreaking havoc like never before [8], but also creating new challenges [9] that became an opportunity to redirect the industry toward the true sustainability path.In 2019, travel and tourism represented one of the world's largest sectors, accounting for 10.4% of global GDP (USD 9.2 trillion) and 10.6% of all jobs (334 million), and it was responsible for creating one in four of all new jobs across the world.Moreover, international visitor spending amounted to USD 1.7 trillion in 2019 (6.8% of total exports and 27.4% of global services exports).As a result of COVID-19 and the ongoing restrictions on international mobility, the travel and tourism sector recorded its worst year in 2020, with international arrivals falling by 74%.It suffered losses of almost USD 4.5 trillion, with its global contribution to GDP declining by 49.1% compared to that in 2019, reaching only USD 4.7 trillion in 2020, going from 10.4% to 5.5%.A total of 62 million jobs were lost, leaving just 272 million employed across the sector globally.This 18.5% decrease was felt across the entire travel and tourism ecosystem, with small and medium-sized enterprises (SMEs), which make up 80% of all global businesses in the sector, being particularly affected [10].
The unexpected event of COVID-19 confronted the hospitality industry with an unprecedented challenge, which has led to a great opportunity to carry out studies [11][12][13][14][15][16] with a significant theoretical and practical contribution to the hospitality industry on this topic, thereby allowing a better understanding of the situation and the proposal of recovery strategies for tourism companies.
The tourism sector has been scrutinized regarding its CSR performance [17]; in this regard, CSR implementation within tourism and hotel organizations has become a useful and effective tool that can increase sustainability and minimize such negative impacts, while improving the company's performance [18] and contributing directly to the development of the host communities [19].Regarding CSR in the tourism sector, the scientific literature between 2001 and 2020 was composed of four research topics: (1) CSR and financial performance; (2) CSR practices, drivers, and inhibitors; (3) reporting and communication of CSR; (4) CSR and stakeholder behavior [20].Regarding CSR in the hotel sector, the scientific production between 2011 and 2020 gradually paid more attention to the external or holistic impact, although the focus on the internal impact was maintained, including the influence on employees and clients, who are crucial stakeholders in the long-term development of hotel establishments [21].Moreover, research on CSR implementation in emerging economies has been relatively scarce, with mainstream research being concentrated in developed economies.
Currently, there remain limited contributions that focused on CSR strategic models within the hospitality industry, and this research aims to fill this gap in the field [22].This research aims to make a concrete contribution to the business and academic world, as well as to the CSR debate, through the application of the KAC-CSR (knowledge, application, and communication of corporate social responsibility) model, a novel proposal that consists of conducting, for the first time in the scientific literature, an integrated analysis regarding the knowledge, application, and communication of CSR in an economic sector (in this case, tourism).It is worth mentioning that, in previous studies, the variables of knowledge, application, and communication of CSR were studied individually, or, at best, a combined study of two variables was performed.However, the three aspects have not previously been integrated, as proposed in this work.In alternative words, previous research focused on analyzing the vertical links in Figure 1, i.e., the factors influencing the level of knowledge of CSR among managers in different industries and the antecedents or determinants of its application and communication.However, the horizontal links among these three variables have received little previous research attention.How and to what extent one variable influences the others is a question that remains little studied across the different industries, particularly in the field of tourism.
what extent one variable influences the others is a question that remains little studied across the different industries, particularly in the field of tourism.
In essence, the novelty of this paper lies in the simultaneous analysis of the relationships indicated in Figure 1, with the analysis of the relationships represented horizontally in Figure 1 being especially novel.Therefore, the results of this study may have important theoretical and practical implications in the area of CSR, both for academia and tourism and hospitality firms.Although research on CSR in tourism has seen increased attention [23], specifically the literature on CSR in the travel, tourism, and hospitality industries [24], even more so in the last decade for tourism in general [1] and hospitality in particular [25], no work integrating the three variables mentioned has been reported, probably resulting in the notorious difficulty of transferring CSR to the real world of the tourism and hotel industry.
The remainder of the article is structured into five sections.The next section revises the determinant factors identified by academic research as influencing the knowledge, application, and communication of CSR in different industries (i.e., the vertical links in Figure 1).Results from the scarce research dealing with some of the horizontal links in Figure 1 are also analyzed, and some hypotheses are proposed.Section 3 describes the methodological aspects of the field research carried out, whose results are highlighted in Section 4. Lastly, Sections 5 and 6 provide a discussion of the results and extract conclusions, including theoretical and practical implications.

Literature Review
Interest in the study of CSR has remained strong, given, among other things, the practical complexity of its interpretation.Not in vain, Votaw and Sethi [26] defined it as a brilliant term that means something but is not always the same for everyone.Carroll [5] identified the year 1953 as the starting point of the CSR construct in the scientific literature, with the publication of the book Social Responsibilities of the Businessman by Howard Bowen, who defined CSR as the obligation of the employer to pursue these policies, take those decisions, or follow those lines of action that are desirable in terms of the objectives and values of our society.For Carroll [5], CSR represents the economic, legal, ethical, and philanthropic responsibilities of companies; perhaps the most accepted definition is that of the European Commission [5], which defines it as the responsibility of companies for their impact on society.
CSR, in addition to being indeterminate, dimensional, regional, and cultural [27], is a dynamic phenomenon with an ever-greater demand by socially conscious stakeholders [28] that can change over time and be different according to the place and economic sector where it is conducted, which implies a very probable difference in its implementation depending on the characteristics of the countries and companies.
It is also important to comment on the close relationship that exists between CSR and sustainability, and authors such as Mostepaniuk et al. [29] demonstrated this fact by proposing a model for organizations in different sectors of the economy, including a variety of determinants that were divided into two groups depending on their relevance to an In essence, the novelty of this paper lies in the simultaneous analysis of the relationships indicated in Figure 1, with the analysis of the relationships represented horizontally in Figure 1 being especially novel.
Therefore, the results of this study may have important theoretical and practical implications in the area of CSR, both for academia and tourism and hospitality firms.Although research on CSR in tourism has seen increased attention [23], specifically the literature on CSR in the travel, tourism, and hospitality industries [24], even more so in the last decade for tourism in general [1] and hospitality in particular [25], no work integrating the three variables mentioned has been reported, probably resulting in the notorious difficulty of transferring CSR to the real world of the tourism and hotel industry.
The remainder of the article is structured into five sections.The next section revises the determinant factors identified by academic research as influencing the knowledge, application, and communication of CSR in different industries (i.e., the vertical links in Figure 1).Results from the scarce research dealing with some of the horizontal links in Figure 1 are also analyzed, and some hypotheses are proposed.Section 3 describes the methodological aspects of the field research carried out, whose results are highlighted in Section 4. Lastly, Sections 5 and 6 provide a discussion of the results and extract conclusions, including theoretical and practical implications.

Literature Review
Interest in the study of CSR has remained strong, given, among other things, the practical complexity of its interpretation.Not in vain, Votaw and Sethi [26] defined it as a brilliant term that means something but is not always the same for everyone.Carroll [5] identified the year 1953 as the starting point of the CSR construct in the scientific literature, with the publication of the book Social Responsibilities of the Businessman by Howard Bowen, who defined CSR as the obligation of the employer to pursue these policies, take those decisions, or follow those lines of action that are desirable in terms of the objectives and values of our society.For Carroll [5], CSR represents the economic, legal, ethical, and philanthropic responsibilities of companies; perhaps the most accepted definition is that of the European Commission [5], which defines it as the responsibility of companies for their impact on society.
CSR, in addition to being indeterminate, dimensional, regional, and cultural [27], is a dynamic phenomenon with an ever-greater demand by socially conscious stakeholders [28] that can change over time and be different according to the place and economic sector where it is conducted, which implies a very probable difference in its implementation depending on the characteristics of the countries and companies.
It is also important to comment on the close relationship that exists between CSR and sustainability, and authors such as Mostepaniuk et al. [29] demonstrated this fact by proposing a model for organizations in different sectors of the economy, including a variety of determinants that were divided into two groups depending on their relevance to an organization: the components of human and social capital, the technical characteristics of an organization and financial dimensions, and the outside business environment, which is determined by the political system and the level of corruption.
For all of the above, CSR has become a topic of growing interest in countless areas, and the number of studies focused on researching different CSR aspects is increasing.Although this type of research within the hospitality industry is less developed compared to other fields, the academic interest in this sector is growing [30].
There are various studies that tried to identify CSR theories; in this sense, the work of Garriga and Melé [31] stands out for being one of the most rigorous and most widely accepted in the scientific literature when classifying CSR theories related to benefits (instrumental theories), political action (political theories), social demands (integrative theories), and ethical values (ethical theories).The present study was identified with ethical theories, since CSR, in the first place, is an ethical concept [32] that, in a broad sense, leads to sustainability.
Companies that base their CSR on ethical theories seek to do the right thing, and they contribute to the common good, as well as to building a better society and a more supportive, just, and equitable world for all, through the implementation of four subcategories of theories based on the normative theory of stakeholders, human rights, sustainable development, and the common good approach.In this sense, the theory of sustainable development based on ethics was used for this research, framed specifically in hotel companies, since these were the main object of the study.
CSR decisions are not made in a vacuum, but instead through an informed understanding of the benefits gained and the costs incurred; In terms of costs, activities associated with CSR can lead to both increased cash outlays and reductions in cash inflows, which could affect organizational financial performance [33].Regarding the benefits, authors such as Kusik and Lozano [34] argued that they are manifested in the improvement of the image and reputation, as well as in the improvement of the competitiveness of the company through human talent (improvement of the work environment, satisfaction, and retention and attraction of employees), via innovation (processes and products/services), via productivity (operational efficiency throughout the business value chain), and via the market (customer loyalty and attraction and entry into new markets).Moreover, Franco et al. [35] analyzed how corporate financial performance (CFP) is affected by CSR in the hospitality industry, finding that the impact of CSR on CFP has a U-shaped form, where CSR is a cost that translates into higher benefits only when it generates solid relationships between firms and their stakeholders.Another benefit of CSR was shown in the study of Srivastava and Singh [36], which suggests the strong mediating effect of a hotel's reputation on customer retention when engaging in CSR.Nazir and Islam [37] revealed CSR's positive effect on employee engagement in hotels.Another benefit of CSR in hotels was revealed in the study of Nazir et al. [38], which concluded that employees' CSR participation positively affects their sense of purpose and their experienced meaningfulness, which consequently affects employee engagement.
Taking into account the above, CSR is particularly important in the tourism industry, generating more benefits than costs, since it is evident that its well-directed actions contribute more effectively to environmental conservation and the social wellbeing of all stakeholders in the sector, as well as to the improvement of the competitiveness of companies, which ultimately generates greater benefits and value in the medium and long term, both for them and for the societies that host them.Something similar occurs in the hotel sector, and it has been shown that the benefits are related to the improvement of profitability, product quality, the natural and social environment, and the satisfaction of its clients and employees, contributing positively to the competitiveness and sustainability of the sector, as well as the tourist destination where they operate [5].
Academic research concerning the determinant factors influencing the knowledge, application, and communication of CSR in different economic sectors is quite rich, albeit less abundant in the tourism industry.According to previous research findings [39][40][41][42], the degree of CSR knowledge within the hotel sector is influenced by the size, age, and category of the hotel establishment, as well as the type of contract, financial performance, and the level of investment in innovation.The gender, age, level of education, and degree of autonomy for CSR decision making of the manager also influence this variable.CSR application is influenced by the same set of variables mentioned above, in addition to the motivations and barriers perceived or felt by the managers [24,34,[43][44][45][46][47][48][49][50].Concerning the degree of CSR communication, this is influenced by the size and age of the hotel establishment, the level of education of the manager, and the external pressures from stakeholders, particularly customers and the local community, followed by business partners/suppliers, competitors, and governments [51][52][53][54].
However, as mentioned before, researchers have not focused on the possible influential links among the variables of knowledge, application, and communication of CSR.No studies have been identified in the tourism field analyzing the potential relationships among these three variables.Only three exploratory studies, all of them out of the tourism field, have partially analyzed the relationships between two of these three variables, but not all three together.A report by the European Commission suggested that a greater CSR application has been evidenced within companies with better knowledge of the concept, which are usually the larger ones [55].Jeppesen et al. [56], focusing on SMEs, concluded that companies with greater CSR knowledge have greater CSR communication, and Wijesinghe [57], analyzing a sample of 75 large companies in Sri Lanka, suggested that greater CSR application can lead to greater CSR communication.
Meanwhile, Peña et al. [58], when analyzing the determinants of CSR application in a hotel context, applied six principal component analyses (henceforth PCAs) for different variables of the study to facilitate the interpretation of the data with the least possible information loss.One PCA corresponded to CSR knowledge, three PCAs corresponded to CSR application (one for each group of activities: economic, social, and environmental), and two PCAs corresponded to motivations and obstacles.Regarding CSR knowledge, the PCA analysis revealed the existence of two factors, one associated with the highest levels of the concept of CSR (advanced CSR knowledge), in which CSR is involved with the sustainable development of society, and the other identified with the lower levels of the concept of CSR (basic CSR knowledge), where the main social responsibility of the company revolves around compliance with the economic requirements stipulated by the shareholders.
Regarding CSR application (economic, social, and environmental activities), the PCA of the economic activities evidenced the existence of two main factors: one associated with activities aimed at marketing and competitiveness (ECO-ACT M/C) and the other associated with the activities aimed at the CSR strategy (ECO-ACT STR).The PCA of the social activities also showed two main factors: one associated with the activities targeting work conditions and respect for the local community (SOC-ACT LAB) and the other associated with activities aimed at policy, as well as the social and disability plan (SOC-ACT P/S).On the other hand, the PCA of the environmental activities revealed the existence of a principal factor associated with activities targeting the environment actions (ENV-ACT EA).
In the same way, the PCA carried out for 'motivations and obstacles' showed that there was a principal factor for each.In the first case, the principal factor was associated with the motivations related to the values and the management style, the competitiveness and the image of the hotel, and the pressure from the stakeholders.In the second case, the principal factor was associated with the obstacles related to management attitude and style, lack of knowledge, and shortage of resources (financial, time, and human) and government support.
Drawing on the results of these previous investigations and the variables used in this study (Table 1), some causal relationships are proposed on the basis of previous studies [55][56][57][58] and of an exploratory approach seeking to expand the options of causal relationships analyzed.CSR knowledge was divided into basic and advanced and CSR application in various types of economic, social, and environmental activities, which allows a more detailed understanding of the implications of the model.The hypotheses implicit in the causal model initially proposed generally state that CSR knowledge has a direct positive effect on CSR application [55] and CSR communication [56].Similarly, the CSR application positively influences CSR communication [57], whereas CSR motivations and obstacles affect the CSR application [24,34,[43][44][45][46][47][48][49][50].The causal model created and the corresponding development in the form of a hypothesis are highly complex when seeking to simultaneously analyze the interaction among all variables with a possible significant effect, as well as the concepts discussed in detail.The hypotheses are detailed below.

Hypothesis 1 (H1).
Basic CSR knowledge has a direct effect on CSR application.

Hypothesis 1.1 (H1.1).
Basic CSR knowledge has a direct effect on CSR economic activities related to market/competitiveness.

Hypothesis 1.2 (H1.2).
Basic CSR knowledge has a direct effect on CSR economic activities related to CSR strategy.

Hypothesis 1.3 (H1.3).
Basic CSR knowledge has a direct effect on CSR social activities related to labor conditions and respect for the local community.

Hypothesis 1.4 (H1.4).
Basic CSR knowledge has a direct effect on CSR social activities related to policy, social action plans, and disability.

Hypothesis 1.5 (H1.5).
Basic CSR knowledge has a direct effect on CSR environmental activities.

Hypothesis 2 (H2).
Basic CSR knowledge has a direct effect on CSR communication.

Hypothesis 3 (H3).
Advanced CSR knowledge has a direct effect on CSR application.

Hypothesis 3.1 (H3.1).
Advanced CSR knowledge has a direct effect on CSR economic activities related to market/competitiveness.

Hypothesis 3.2 (H3.2).
Advanced CSR knowledge has a direct effect on CSR economic activities related to CSR strategy.

Hypothesis 3.3 (H3.3).
Advanced CSR knowledge has a direct effect on CSR social activities related to labor conditions and respect for the local community.

Hypothesis 3.4 (H3.4).
Advanced CSR knowledge has a direct effect on CSR social activities related to policy, social action plans, and disability.

Hypothesis 3.5 (H3.5).
Advanced CSR knowledge has a direct effect on CSR environmental activities.

Hypothesis 4 (H4).
Advanced CSR knowledge has a direct effect on CSR communication.
Hypothesis 5 (H5).CSR motivations have a direct effect on CSR application.Hypothesis 7.4 (H7.4).CSR social activities related to policy, social action plan, and disability have a direct effect on CSR communication.
Hypothesis 7.5 (H7.5).CSR environmental activities have a direct effect on CSR communication.

Methodology
This study consisted of the application of a structured questionnaire on CSR that was designed from the existing literature.The questionnaire was applied in September 2021 and consisted of 82 questions distributed as follows: 23 questions about general issues of the hotel and its director, seven questions related to the knowledge that directors have about CSR, 43 issues related to the CSR practices of hotel establishments, and 9 questions to measure their CSR communication.The reference literature used for the design of the questions was Peña et al. [58], and a five-point Likert scale was used for the responses to items.Table 1 provides a description of all variables involved in the study.

Sampling and Data Collection
The population studied consisted of hotels in the Colombian Caribbean region (with Barranquilla, Santa Marta, and Cartagena as the main cities), made up of 506 hotels, of which 116 operate in Barranquilla (22.93%), 168 operate in Santa Marta (33.20%), and 222 operate in Cartagena (43.87%).From this population of 506 establishments, the sample size was calculated, whose value was 224 hotels, and the most important statistical information is shown in detail in Table 2.

Population (N) 506 Hotels
Geographical scope Colombian Caribbean region (Barranquilla, Santa Marta, and Cartagena) Confidence interval, as well as error and proportions (p and q) predefined to calculate the sample size 95.5-5% (with p = q = 0.5)

Respondent Hotel managers
Response rate 99.11% (222 hotels) Sampling error obtained from the response rate with a confidence interval and predefined proportions (p and q) 95.5-0.64%(with p = q = 0.5) Source: own elaboration.
Specifically, the CSR questionnaire was applied to the sample of 224 hotels, of which, geographically, 58 were in Barranquilla (25.90%), 72 were in Santa Marta (32.14%), and 94 were in Cartagena (41.96%), with this sample having approximately the same proportion of hotels by city compared to the total population (506 hotels).Of the 224 hotel managers, a total of 222 responded correctly to the entire questionnaire, assuming an effective response rate of 99.1%.Three means were used to collect the data: e-mail, telephone, and personal interviews, with telephone interviews the most used means (77.93%), followed by personal interviews (14.86%) and e-mail (7.21%).Concerning the selection of the 224 hotels, a nonprobabilistic sample was made for convenience (hotels affiliated with the Hotel Association of Colombia, COTELCO), due to the great disposition of the latter to support the research.As a result, 121 hotels (54% of the sample) were obtained; for the remaining 103 hotels (46% of the sample), snowball sampling was used, which is a modality of convenience sampling.In practice, the 121 hotel managers affiliated with COTELCO were asked to recommend hotel managers not affiliated with COTELCO who might be interested in participating in the study, and they agreed.
One interesting aspect of the sample is that it was composed almost entirely (95%) of micro-sized, small-, or medium-sized hotels (average of 53 rooms), with the highest proportion in the micro-sized and small categories (85%).Similarly, the majority did not belong to international hotel chains or groups (92%), and a large part (89%) were privately owned and managed, with only 11% being franchised or with a lease or management contract.

Data Analysis
After collecting the information, data analysis was carried out using structural equation modeling (SEM), a technique increasingly applied in the social sciences [59] and for some time in research in the area of business management [60]; it has also been used in tourism research [61,62], as well as in CSR research [63][64][65], specifically in the tourism sector [18,48,66].
The fundamental advantage of SEM lies in its ability to analyze more than one relationship of dependence between variables, addressing complex phenomena and, thus, allowing it to go from exploratory to confirmatory analysis [67].SEM proposes the type and direction of the relationships expected to be found between the different variables and estimates the parameters specified by the relationships proposed at the theoretical level [68].It was interesting to use SEM to identify the possible relationships among the three main variables of the study (knowledge, application, and communication of CSR), as well as among their determining factors.
The statistical method used for the analysis was partial least squares (PLS) regression, using the SmartPLS software [69].The application of PLS-SEM is relevant for this study since it allows working with very small samples [70,71] and is quite robust [72].Although the results do not differ significantly among the different alternative weighting systems, the most recommended (used in this case) is path weighting, since it provides the highest R 2 value for the latent variables [73].
The SEM analysis began with a model with all the hypotheses raised and all the possible determining factors of the three actions taken into consideration (knowledge, application, and communication).Subsequently, the SEM analysis was repeated, eliminating the nonsignificant causal relationships and the variables without any significant explanatory capacity.This last causal model without insignificant causal relationships is detailed in Section 4.

Results
Various model measurement checks were performed, in terms of reliability, convergent validity, and discriminant validity.Individual reliability requires that items have loads of ≥0.707 [74], and items that do not comply must be eliminated.
The construct reliability measures the consistency of the items or, in other words, whether the variables are measuring the same concept or a latent variable.Cronbach's alpha [75] or individual reliability and composite reliability [76] are the two most widely used reliability indicators.According to Nunnally [77], as a reliability criterion, the values of both indicators should be greater than 0.7, although it is better if they are greater than 0.8.The analysis of Cronbach's alpha and composite reliability for the constructs used showed very high values for both measurements (Table 3), indicating the reliability of the constructs used.The convergent validity [78] implies that a set of indicators represents a single underlying construct, and this can be demonstrated through their unidimensionality [73].It is usually evaluated through the average extracted variance (AVE); in this sense, Baggozi and Yi [79] recommended 0.5 as a minimum value, indicating that more than 50% of the variance of the construct is due to its indicators.As shown in Table 3, the constructs used showed an AVE higher than 0.700, a value much higher than the minimum of 0.5 required by the literature; thus, it can be said that the level of convergent validity was high.
Lastly, discriminant validity implies that each construct must be different from the remaining constructs, with two methods of evaluation [78].The first explains that factorial loads must be higher than factorial cross loads, which means that items must be more correlated with their own construct than with the others.The second explains that the square root of the AVE of the construct must be greater than the correlation between that construct and all the others [80].When analyzing the cross-loading table (Table 4) and comparing the correlations with the AVE square root (Table 5), the discriminant validity in the used constructs was fulfilled.
After assessing that the measurement model met the reliability and validity criteria (convergent and discriminant), it was time to evaluate the relationships among the constructs, i.e., the structural model itself.According to the analysis of R 2 , the amount of variance explained by the model was important and exceeded 0.600 in all constructs (Table 3), except for basic CSR knowledge.In the case of CSR communication, the value of R 2 was 0.700 (CSR communication was measured using a quantitative variable).
Table 6 shows the items and loadings of the reflective variables used in the analysis, once refined on the basis of the previously mentioned reliability and validity analyses.Advanced CSR Knowledge: CSR refers to integrating the triple economic-social-environmental sphere into the company's strategy.
ADV-KNOW/1 0.966 CSR aims to contribute to the wellbeing and improvement of the quality of life in society.
ADV-KNOW/2 0.949 CSR refers to the tactical practices carried out by the company above the legal regulations.
ADV-KNOW/3 0.847 Basic CSR Knowledge: The main social responsibility of the company is the financial requirements stipulated by the shareholders.
BAS-KNOW/1 0.915 CSR deals with philanthropic activities that are not related to the company's business.
BAS-KNOW/2 0.914 CSR is identified with compliance with all legal regulations that affect the company.BAS-KNOW/3 0.809 However, to test whether the regression coefficients are significant, PLS is based on a nonparametric bootstrapping procedure [81].Bootstrapping is a resampling process in which samples are randomly generated from the original sample, through substitution with replacement.Hair et al. [82] stated that parameter estimates generated from the subsamples are used to determine the standard errors of the estimates.To determine the critical values of the Student's t-test, a two-tailed Student's t-distribution with 4.999 degrees of freedom was used.Specifically, in Tables 7 and 8, the results of the analysis of the significance of the different causal relationships conserved in the final causal model are detailed.On the basis of the results of Table 7, it can be indicated that a greater degree of advanced CSR knowledge lends itself to a greater degree of CSR application (H3), mainly in the economic (H3.1 and H3.2) and social (H3.3 and H3.4) activities.On the other hand, if the degree of basic CSR knowledge increases, there will be a decrease in the degree of CSR application (H1), mainly in the economic (H1.1 and H1.2) and social (H1.3 and H1.4) activities.Similarly, a greater degree of CSR application, through its economic (H7.2), social (H7.3 and H7.4), and environmental (H7.5) activities, has the direct effect of increasing the degree of CSR communication (H7).Likewise, motivations have a positive influence on the CSR application (H5), specifically in economic (H5.1 and H5.2) and social (H5.3) activities.On the other hand, obstacles have a negative influence on the CSR application (H6).
In summary, the results show that CSR knowledge has a direct effect on CSR application (H1 and H3), but not on CSR communication (H2 and H4).Likewise, CSR application directly influences CSR communication (H7); lastly, the motivations (H5) and obstacles (H6) directly influence CSR application.
As a complement to the above, concerning the determining factors of each of the three variables (Table 8), it can be said that the size of the hotel, within the range of <10 workers, 10-50 workers, and 51-200 workers, has a direct negative effect on the degree of CSR application (economic and social activities).Likewise, hotels with <10 workers have a direct negative effect on the degree of CSR communication.The age of the hotel within the range of <40 years has a direct positive effect on the degree of basic CSR knowledge and a negative effect on the degree of CSR application (economic and social activities).
Similarly, the low categories of hotels (0 and 2 stars) have a direct negative effect on the degree of CSR application (environmental activities); on the other hand, the high categories of hotels (5 stars) have a positive effect on social activities (policy and social plan).The type of contract in which the hotel is leased has a direct negative effect on the degree of basic CSR knowledge.Franchising the hotel has a direct positive effect on the degree of CSR application (economic and social activities).The hotel's regular financial performance has a direct negative effect on the degree of CSR application (economic and social activities) and advanced CSR knowledge.
The regular level of hotel investment has a direct negative effect on the degree of CSR application.The bad level of hotel investment has a direct negative effect on the degree of advanced CSR knowledge and CSR application.The female gender of managers has a negative effect on the degree of basic CSR knowledge and a positive effect on the degree of CSR application.Younger (<40 years) managers have a direct positive effect on the degree of CSR application (economic activities).Similarly, the educational level of the manager has a direct positive effect on the degree of advanced CSR knowledge and CSR application (economic activities).The postgraduate educational level of the manager has a direct negative effect on the degree of basic CSR knowledge and a positive on the degree of advanced CSR knowledge and CSR application (social and environmental activities).The director's degree of autonomy in CSR decision making has a direct positive effect on the degree of CSR application (economic activities).Lastly, pressure from external stakeholders (government) has a positive effect on the degree of CSR communication.

Discussion
This study analyzed, in a comprehensive way, all the links illustrated in Figure 1, both vertical and horizontal, i.e., the factors influencing knowledge, application, and communication of the CSR, as well as the links among these three constructs.The main contribution of this paper is the simultaneous analysis of all these variables, especially knowledge, application, and communication, and their interrelationships.The results obtained regarding the vertical links are in line with those obtained in previous studies analyzing these links in an isolated way, thus validating the method used.The causal relationships among knowledge, application, and communication turned out to be significant, with this study being the first time that the interrelationships among the three concepts were analyzed simultaneously.In addition, these concepts were broken down into other more specific and concrete concepts, providing an added value of nuance to the study carried out.
With regard to the factors influencing the degree of CSR application, our results are coincident with a series of previous investigations outside the hotel industry (e.g., [24,[43][44][45][46][47][48][49][50]) and within the hotel industry [83,84].Melissen et al. [84] evidenced the influence of the type of contract in the application of CSR activities that lead to sustainability, arguing that the separation of ownership and operation generates obstacles.Camisón et al. [83] argued that hotels operating within the chain model without capital (franchising) enjoy advantages for activities (particularly economic ones); therefore, cooperation and association are important for independent establishments.
On the other hand, the factors determining the degree of CSR communication are in line with those evidenced in several previous studies (e.g., [51][52][53][54]).Concerning this variable, it is interesting to emphasize its importance for CSR, which is why, in the area of tourism, more actions are required from the marketing of sustainability [85] to prevent companies from falling into the double error of greenwashing or greenhushing, showing with their communications a moral silence in the first case and moral hypocrisy in the second [86].Therefore, care must be taken with communicating more than what has been achieved or with the extreme case of not communicating anything of what has been achieved.
Regarding the causal horizontal relationships among the variables of knowledge, application, and communication, the results of this study coincide with those shown in the works of the European Commission [55] and Wijesinghe [57].However, this study analyzed the links among the three variables in a comprehensive and combined manner, concluding that, to reinforce the 'knowledge-application-communication' sequence and boost the process, it is of paramount importance to assure an advanced level of CSR knowledge among managers.If this objective is achieved, then the process will flow smoothly, and the levels of CSR implementation and communication will be intensified.
In the case of Colombian Caribbean hotels, advanced knowledge of CSR correlates to higher levels of education among top managers.Therefore, the key for the sequence to work optimally is the combination of a top manager holding higher levels of education (preferably postgraduate), a company with strong financial resources, and a top manager's values favorable to CSR.The top manager being a woman is also positive for CSR implementation in environmental, social, and economic activities.The main obstacle is the shortage of financial resources.
On the other hand, CSR represents a highly publicized notion with important strategic implications for tourism companies [87]; in terms of managerial implications, the findings from this study could help hotel managers make the best decisions to ensure that their companies understand, apply, and communicate CSR more appropriately.In particular, the following can be highlighted: (i) The characteristics of hotels related to size, age, category, type of contract, financial performance, and the level of investment in innovation influence the levels of CSR knowledge and CSR application.In the case of CSR communication, only size has an influence.In this respect, managers will have to bear in mind that the levels of knowledge and application will increase as the hotel increases its size (application), acquires experience (knowledge and application), improves its category (application), and importantly always maintains good levels of financial performance (knowledge and application) and investment in innovation (knowledge and application).Likewise, the types of hotel contracts that are leased or franchised would improve the degree of knowledge and application.In turn, the levels of communication will increase as the size of the hotel increases.(ii) With regard to the characteristics related to the manager, decisions aimed at having higher levels of CSR knowledge and CSR application will involve having a relatively young person in the management (application), preferably a woman (knowledge and application), with university studies (knowledge and application), ideally with a postgraduate degree (knowledge and application) and with autonomy for decision making in the area of CSR (application).It is important to remember that these characteristics are not influential in CSR communication.
Similarly, to contribute to a higher level of CSR application, incentives should be provided for managers: internal motivations related to values, management style, and improved competitiveness, and external motivations related to the image/reputation of the hotel and pressure from its internal and external stakeholders (workers, clients, local community, society in general, investors, business partners, suppliers, competitors, and public administrations).On the other hand, efforts should be made to counteract the obstacles that become barriers to the CSR application, which are of internal origin, concerning attitude, management style, and lack of knowledge on the subject, or of external origin, related to the lack of resources (financial, time, and human) and government support.Therefore, more training in CSR should be encouraged, as should greater availability of resources for its application and more decisive intervention in the issue by governments, through support or incentives (fiscal and/or subsidies) for its proper implementation.
Without a doubt, greater participation and involvement of the government through public policies for the promotion of CSR are fundamental, given that, in addition to what was mentioned above, this contributes to improving the levels of CSR communication.It became evident in this study that public administration is the main stakeholder capable of persuading businesspeople to communicate more (quantity) and better (quality) the CSR activities and actions that they carry out within and outside their organizations.
In a comprehensive approach, the results of this study seek to counteract the deficiencies in the area of CSR implementation in tourism, as well as contribute to its better application and communication by hotel firms (in this case, in the Colombian Caribbean region), providing a strategic tool that could enable them to obtain benefits associated with improved profitability [49,88], product quality [89], the physical and social environment [45,90], and the satisfaction of their customers and employees [40,91], all of which was ratified by Rhou and Singal's [92] recent literature review study on the business case for CSR in the tourism industry, which showed the positive impact that CSR generates on the environment, long-term profitability, customers, and the product.
Therefore, these findings become a practical and useful information tool for hotel managers in the Colombian Caribbean, which could positively contribute to the competitiveness and sustainability of both their establishments and the tourist destination.

Conclusions
Analyzing the progress of the sustainable tourism agenda, through the lens of CSR in tourism, discussions on the subject have been dominated by theoretical presentations, rather than promoting a more proactive approach to CSR in the tourism industry.However, on the basis of advances in the relationship between CSR and sustainable development within the tourism sector, it can be argued that long-term sustainability can only be achieved if industry professionals successfully address their social and community deficits, which can undoubtedly become a reality through CSR practice [93].In this sense, the results of this study attempt to promote, in a practical way, CSR implementation in the tourism industry and, therefore, the sustainable development of the entire sector.In other words, the study also contributes to the promotion of responsible tourism that benefits local communities and minimizes negative social and environmental impacts [94].
Specifically, the main contribution of this work is that an integrated analysis of the variables of knowledge, application, and communication of CSR was performed for the first time through the application of the KAC-CSR model.The most relevant managerial implication is that, according to the results shown, hotel managers in the Colombian Caribbean will have useful information enabling them, in practical terms, to improve the levels of knowledge, application, and communication in their hotels, thanks to the identification, on the one hand, of its determining factors and, on the other, of the causal relations that exist between them.The fundamental decisions of managers should revolve around improving CSR knowledge itself since this has a direct positive effect on implementation.Likewise, emphasis must be placed on raising the levels of CSR application, since this increase has a positive influence on the levels of CSR communication, in terms of both quantity and quality of information.Moreover, efforts must be made to maintain motivations and mitigate obstacles, since it is known that the former has a positive influence on the CSR application, while the latter has a negative influence on it.
Lastly, it is important to add that this article makes a relevant contribution to the existing literature on CSR in several ways, given that (a) it provides evidence on the determining factors influencing the 'knowledge → application → communication' sequence of the CSR, by conducting an integrated analysis of these variables, an aspect that, as stated before, has not been studied in this way in any economic sector until now, (b) it helps to continue understanding how CSR can be implemented, in a better way, in the hotel industry, since it is known that research on CSR has focused more on other industries with apparently more evident negative social and environmental impacts than those in the tourism and hotel sector, and (c) it provides more information on CSR management in developing countries, as is the case in Colombia and specifically in its hotel sector, being the first sectoral study that integrally analyzes the knowledge, application, and communication of CSR in the country.
With regard to the limitations of the study and suggestions for future research, although the percentage of the 224 hotels in the sample was very similar to the proportion of the 506 hotels in the population, thus ensuring certain representativeness in terms of geographical distribution, the use of nonprobability convenience sampling, especially snowball sampling, makes the conclusions representative of the sample but not of the population.Therefore, the use of other types of statistical sampling is suggested for the selection of cases to be studied.It would also be interesting to repeat the analysis with much larger samples since this would allow the detection of small but significant effects.With the sample used in this case, only the most important quantitative effects could be confirmed.It is also important to highlight the importance of repeating this analysis in other tourist regions in order to determine if the conclusions presented here are generalizable.
On the other hand, authors such as Farmaki and Farmakis [95] argued that, if a holistic and inclusive approach is to be promoted in the multistakeholder tourism industry and if CSR is to contribute to the overall sustainability of tourism, a wide range of stakeholders must be considered.In this sense, it should be noted that the analysis of this investigation was based exclusively on the perception/opinion of the hotel's top manager (an internal stakeholder); therefore, it is recommended to carry out studies taking into account the opinion on CSR of more stakeholders, such as the employees (and other internal stakeholders).Likewise, further research should be promoted focusing on identifying the role that the host communities, ecosystems, suppliers, NGOs, and the government play as external stakeholders, an issue that has been little studied in the existing literature [3], which would provide a more complete overview of the subject.Similarly, given the novelty of the KAC-CSR model, it is suggested to replicate it in other economic sectors and geographical areas, to continue evaluating its usefulness, reliability, and validity.

Figure 1 .
Figure 1.Horizontal links among knowledge → application → communication of CSR and vertical links with their determinant factors.

Figure 1 .
Figure 1.Horizontal links among knowledge → application → communication of CSR and vertical links with their determinant factors.

(
00-10) YEARS Time of operation of the hotel <10 years (10-20) YEARS Time of operation of the hotel 10-20 years (21-40) YEARS Time of operation of the hotel 21-40 years (>40) YEARS Time of operation of the hotel >40 years 0-STARS-H Hotel without stars

Table 1 .
Description of the study variables.

Table 2 .
Statistical summary of the fieldwork.

Table 3 .
Reliability and convergent validity of the constructions used.

Table 6 .
Loadings of the items (multi-item constructs).