Abstract:

This study aims to examine intangible assets as mechanisms underlying the direct corporate social responsibility (CSR)-corporate financial performance (CFP) relationship. Based on signalling theory, we developed an integrated model taking the corporate image (CIM) and corporate identity (CID) as double mediators between the relationship of CSR and CFP. Using the convenience sampling procedure, we dispersed an anonymous survey to employees and customers in 33 multi-national companies working in Pakistan. The structural equational modelling using SmartPLS 3.0 on 456 valid responses revealed the best fit for the study measurement and structural models. Further, the mediation analyses of intangible assets in the simple form identified both CIM and CID have partial indirect effects on the CSR-CFP relationship. Our overall mediation model taking both mediators simultaneously, revealed the full mediation effect of the CSR-CFP connection. Specifically, we found that corporate image has higher relative significance than the CID in the mediation model. However, both simultaneously result in greater CFP for the firms.

Key Words:

Corporate Social Responsibility, Corporate Identity, Corporate Image, Corporate Financial Performance

Introduction

Corporate social responsibility (CSR) has been a highly discussed issue equally in the business world and academic settings in recent decades (D’Amato & Falivena, 2020; Ye et al., 2021). According to Wheelen and Hunger (2012), corporations have social responsibilities that go beyond profitability and their legal obligations. CSR is generally defined as initiatives that make corporations responsible citizens that serve the well-being of the community beyond their own business operations and so is said to be “doing good” for firms (Ye et al., 2021). Though the concept of CSR has added significance as 90 per cent of Fortune 500 companies participate in CSR initiatives for the firms, and the bulk of the research was carried out (McKinsey and Company, 2009; Kelley et al., 2019; Yang & Basile, 2019), yet the results of CSR investment, particularly in terms of corporate financial performance (CFP) are still unclear. Thereby,  the question of whether it is feasible to translate CSR into positive CFP has not been fully resolved (Liu et al., 2021).

This is because the recent research on the topic found inconclusive and even contradictory results. For example, positive, negative and no relationships between CSR and CFP were found previously (Muhammad et al., 2020; Nguyen et al., 2020; Shabbir, 2016). It is argued in favour of a negative relationship that investment in CSR initiatives is considered a financial cost that may reduce a portion of the profit for a firm. Whilst, the majority of the studies found a positive link between the direct link of CSR and CFP (Ali et al., 2020), arguing that CSR initiatives for societal and the environment produce positive signals in the context of operating firms that result in attracting human capital, investors, suppliers, and costumers towards firms and its products. In this way, they are able to enhance their revenues and profits, i.e., CFP.  

Nevertheless, recent research has been devoted to answering such inclusive results between the CSR-CFP link. In this vein, some experts suggested that these contradictory results might be the result of incorrectly measuring the constructs, given that many studies used different scales to measure the same construct, while others used actual financial secondary data with different proxies, especially for CFP (Fourati & Dammak, 2021). Some authors proposed that institutional factors, including a country’s socio-economic status, laws and culture, may differently impact the relationship (Javed et al., 2020; Kelley et al., 2019). Apart from these reasons, researchers also suggested studying intervening mechanisms that perhaps play an important role in answering the dilemma of inclusive results of the CSR-CFP relationship (Ali et al., 2020; Ye et al., 2021). Though, a considerable stream of research also focused on such underlying intervening factors, yet with simple mediation analyses in which only one mediator was studied and tested (Fourati & Dammak, 2021). However, such analyses are prone to specification errors and thus may produce inflated or deflated associations between the variables (ul-Hassan et al., 2021).   

This study set out to fill the aforementioned gap in the relevant literature and contribute to the literature in four ways. First, in previous research, CSR was measured reflectively either unidimensional or its different dimensions were separately studied with different variables. To the best of our knowledge, CSR was never measured as a higher order encompassing its four dimensions for a holistic view of the concept. In this vein, we study the concept of CSR representing four dimensions in a proper way of measurement as reflective-formative higher order, given that each dimension represents a different aspect of the concept that was previously omitted in the management research (Javed et al., 2020; Saeidi et al., 2015). Second, we assess the quality and broaden the generatability of different measurement scales used in the study in a newly developing country context, namely Pakistani. Third, in the previous research regarding the direct relationship among the study variables in the developed world, we try to replicate these direct relationships in a new country and multi-organisational data that is rare if not absent in the literature. Finally, we extend prior research to examine underlying novel intangible assets of corporate image (CIM) and corporate identity (CID) to better understand the relationship between CSR and CFP. 

Literature Review and Hypotheses Development CSR and CFP

After first appearing in the corporate world in the 1950s, the concept of CSR is now heard loud and has become a buzzword in the business context (Javed et al., 2020). Initially, in the conceptual underpinning period, it was considered unidimensional but later on recognised as a multidimensional concept. Carroll (1979) distinguished economic, legal, ethical, and discretionary expectations/philanthropy as four different dimensions of CSR. Economic responsibility is the basic social responsibility of corporations since they are the basic units of the general societal creation of commodities and services. The responsibility of corporations to undertake economic missions in accordance with legal regulations is called legal responsibility and to comply with laws and regulations imposed by the government. Ethical responsibility, which has received a lot of consideration recently, means that corporations are required to behave in manners beyond just what is required by law, and lastly, discretionary/ philanthropic responsibilities perform activities on a voluntary basis to increase the quality of life of the community and human welfare.

The existing evidence on the association between CSR and CFP is mixed, as stated in the previous section. However, the majority of the study conducted around the world, for example, in Australia, Iran, UAE, Pakistan and Taiwan etc., found a positive direct relationship between the two constructs (Javed et al., 2020; Saeidi et al., 2015). Further, the meta-analysis studies also endorsed the positive association between CSR and CFP (Margolis et al., 2009; Orlitzky et al., 2003). In accordance with the previous research, we develop the following direct relationship hypothesis:

H1: There is a positive relationship between CSR and CFP.

The Mediating role of Intangible Assets CIM and CID

The customers’ overall impression or judgemental feeling about a company is referred to as CIM (Nguyen & Leblanc, 2001). CIM of a firm plays a significant role in developing customers' purchase decisions (Horng et al., 2018). It influences not only customers' attitudes and perceptions about the corporation but also customers’ retention likelihood (Ball et al.,2006), whereas it is very difficult for the competitors to imitate intangible assets like the CIM of a firm (Kant et al., 2017. We believe, based on signalling theory, that CSR translates into CFP indirectly via CIM influence. Signalling theory (Boulding & Kirmani, 1993) posits that stakeholders interpret the available information as positive or negative signals about a firm as usually they do not have complete information about the operations of a firm. This is reasonable to argue that the engagement of firms in CSR initiatives transmits positive signals towards their CIM that influence the internal and external stakeholders’ investment and purchase decision in the organisation and its product, respectively. These actions of stakeholders turn into superior CPF. 

Previous empirical research also supported this notation that corporations with a strong sense of CSR and are seen as having positive CIM by their external and internal stakeholders are better at generating healthy financial outcomes in comparison to companies with negative CIM (Bhattacharya & Sen, 2004; Mohr & Webb, 2005). Similarly, recent studies in different organisational settings found a positive relationship between CSR and CIM (Kim et al., 2017; Liu et al., 2019), whereas CIM has a direct positive impact on CFP Ghaderi et al., (2019). Evidence of mediation of CIM between CSR-CFP is also substantiated. In this line of research, we develop the following set of hypotheses:

H2: There is a positive direct relationship between CSR and CIM. 

H2: There is a positive direct relationship between CIM and CFP. 

H3: CIM mediates the relationship between CSR and CFP.

Further, the next mediator, CID, is referred to as “the self-presentation of an organisation, rooted in the behaviour of individual, organisational members, expressing the organisation's ''sameness over time'' or continuity, ''distinctiveness,'' and ''centrality” (Van Riel, 1997, p. 290). Simply, regarding CID, there should be compatibility between what firms are and what they say (Fukukawa et al., 2007). Dowling (2004) differentiated the CID from CIM in these words that CID is the set of qualities used to characterise a firm, whereas CIM takes the beliefs held by the public about a firm. Thereby CIM is concerned with addressing the question, "what do people think about you”? Whilst CID addresses the question of “Who are you”?

This argument suggests that, at first, firms need to develop their CID through CSR that further improves their CIM in the eyes of external stakeholders. Given that CSR engagement of firms again disperses such positive signals that may also influence the CID. These initiatives influence the internal stakeholders, i.e., employees, to behave and communicate in a way that improves the CID of the firms to external audiences. Arendt and Brettel (2010) advocated in favour of this belief that CSR is particularly beneficial in creating CID due to its potential to improve CIM appeal and consequently boost competitive advantage as well as CFP. Similarly, multiple studies found positive relationships between CID and best financial outcomes (Margulies, 1977; Brown, 1995; Bharadwaj & Menon, 1993).

The preceding paragraphs lead us to conclude that CID being the outcome of CSR and antecedent of CFP may mediate the relationship between CSR and CFP. Further, when firms actively engage in CSR first improve their CID, which turns into improved CIM of the firms and both CID and CIM work as double mediation between the CSR-CFP link. Accordingly, we formulate the following set of hypotheses and portray all our preceding discussion in the below given conceptual model of the study in Figure 1:

H4: There is a positive direct relationship between CSR and CID 

H4: There is a positive direct relationship between CID and CFP.

H5: CID mediates the relationship between CSR and CFP.

H6: CID and CIM serial mediate the CSR and CFP relationship. 

Method Sample and procedure

This cross-sectional study was conducted in multi-national companies working in Pakistan. For data collection, we used both online as well as traditional paper and pen methods. An anonymous questionnaire was distributed among both internal and external stakeholders, i.e., employees and customers, respectively. A cover letter explaining the study's goal, requesting voluntary participation and assuring the participants' confidentiality of their responses, and we stated that their responses would only be used for research purposes. To avoid common method variance, we put a set of items of different constructs randomly in the questionnaire (Podsakoff et al., 2003, 2012). Resultantly, we collected a total of 473 responses representing 32 different companies. However, the screening process excluded 17 invalid questionnaires. Thus, we had 456 valid questionnaires for data analyses and testing of our study models. 

Measures

The dependant variable, i.e., CFP, was measured with five items adapted from a balance score card scale  (Kaplan & Norton, 1992) that has four different quadrants. However, we used indices that were relevant to CFP only that were previously used in multiple studies (Ali et al., 2020; Saeidi et al., 2015). For CFP, two items represented growth determinants, while 3 items were related to monetary accounting performance. 

The scale of Maignan and Ferrell (2000) that was derived from Carroll's (1979) conceptualisation was adopted to measure CSR comprising a total of 18 items. A set of 4 items captured economic and legal components, whereas 5 items each were for ethical and philanthropy dimensions of CSR. 

Five items scale by Bayol et al. (2000) was used to tape the concept of CIM, while (Arendt & Brettel, 2010) 7 items scale originally taken from Bhattacharya et al. (2003) was borrowed to measure the CID. For all the measurement instruments, we took responses using the Likert scale, wherein 7 indicated strongly agree, and 1 represented strongly disagree. 

Analysis and Results

Figure 1 shows the path coefficients of the study’s model that were estimated using a structural equation modelling with partial least squares approach. The SmartPLS version 3.0 was utilised to investigate the theoretical model because Hair et al. (2017) proposed that this statistical technique and software have the potential to analyse a complicated model with a series of mediation comprising the formative second order constructs. Since we measured both CSR and CFP in the fashion of second order reflective-formative manner and first order formative way, respectively, given that both these constructs either have various dimensions representing different aspects or each item represents a different proxy index.  

The data was analysed using a three-step process in this research. First, the measurement model was necessary to run in order to verify the validity and reliability in a newer Pakistani organisational context for our adopted or adapted scales. In the second step, we estimated the structural model to check whether the data fit the overall model of the study. With this estimation, we examine both direct and indirect connections between variables in the measurement model before moving on to step two. Prior to investigating the theoretical model's structural connections, this statistical analysis approach was utilised to verify the validity and reliability (Hair et al., 2017).

Measurement Model Results

In order to validate the study, reflective measurement scales that were used in different national cultures, we assessed them with the coefficients of reliability and validity, whereas for the formative scale, a different method of assessment was used, discussed next paragraph (see, e.g., Hair at el., (2017). For the reliability and validity of reflective scales, two yardsticks were used as proposed by Hair et el., (2017). The scores of composite reliabilities and Cronbach Alpha are shown in Table 1, which all surpass the recommended value of 0.70. Thus, endorse no issue with regard to the reliability of the study reflective measurement instruments. 

For convergent validity, the factor loading of each reflective scale also exceeds the threshold value of 0.70, except for one item each of the first-order components of ethical and philanthropy of CSR that were 0.426 and 0.477, respectively, as shown in Table 1. Since the reflective scales can be interchangeable and thus deleting one or more items does not affect the results (see, e.g., Hair et el., (2017), therefore, both these items with lower factor loading were deleted in the further analyses. Besides, another index for convergent validity, i.e., average variance extracted (AVE), was also estimated, and results were found over the threshold value of 0.5 for each construct, as shown in Table 1. Further, to estimate the discriminant validity, we used a newer and more reliable procedure of hetero- trait-monotrait ratio of correlations (HTMT: Hair et al., 2017). The values of HTMT are all below 0.90 and range between 0.627-798 for each pair of the study’s reflective constructs that satisfy the condition of discriminant validity, whereas 1st order reflective dimension of CSR was also within the range of critical value, i.e., minimum 2.49 to maximum 5.65 (see, e.g., Hair at el., 2017). 

For formative scales, the CSR was measured in a higher order reflective-formative way, and CFP was formative. The experts of the methods contend that reliability is irrelevant for formative scales while evaluating them (Hair et al., 2017). However, the quality of the formative scale is determined by the significance of outer weight, keeping in view the minimum level of multi-collinearity among the different items/attributes of a construct. In Table 1, we see no issue with multi-collinearity as VIF values are between a minimum of 1.12 and a maximum of 1.48 that below the criteria of 5 (Mason & Perreault, 1991). Further, 2nd order formative scales that represent four dimensions of CSR have all statistically significant, and 1st order reflective scale representing each dimension of CSR is evaluated with similar criteria to the reflective scale as explained in the above paragraph and found satisfactory. Moreover, the outer weights of all five items of CFP are also statistically significant, as shown in Table 1, validating them as satisfactory measurement scales. 

Table 1. Results of the measurement model

Construct		Item
Reflective Measures			Loading	Composite Reliability	AVE	Cronbach’s Alpha
CIM		CIM1	0.801**	0.857	0.600	0.778
		CIM2	0.790**
		CIM3	0.760**
		CIM4	0.768**
		CIM5	0.765**
CID		CID1	0.791**	0.884	0.604	0.836
		CID2	0.826**
		CID3	0.764**
		CID4	0.714**
Formative Measure			Outer Weight	VIF
	2nd order
	CSR	Economic	0.267**	1.119
		Legal	0.398**	1.488
		Ethical	0.373**	1.353
		Philanthropy	0.345**	1.363
CFP		CFP1	0.226**	1.416
		CFP2	0.396**	1.249
		CFP3	0.442**	1.306
		CFP4	0.236**	1.327
		CFP5	0.125**	1.365
lower-order (Reflective) Measures		Item	Loading	Composite Reliability	AVE	Cronbach’s Alpha
Economic		CSRE1	0.817**	0.898	0.688	0.849
		CSRE2	0.808**
		CSRE3	0.822**
		CSRE4	0.869**
Legal		CSRL1	0.853**	0.893	0.676	0.840
		CSRL2	0.863**
		CSRL3	0.864**
		CSRL4	0.777**
Ethical		CSRET1	0.814**	0.905	0.705	0.860
		CSRET2	0.844**
		CSRET3	0.786**
		CSRET4	0.842**
		CSRET5	0.426d
Philanthropy		CSRP1	0.813**	0.885	0.658	0.827
		CSRP2	0.861**
		CSRP3	0.784**
		CSRP4	0.785**
		CSRP5	0.474d

Structural model Results

To check the in-sample explanatory power, predictive relevance fit, and direct relationship among different constructs of the study, we used the criteria of R2, f2, Q2 and path coefficients, respectively (Hair et al., 2017). We found a moderate level of R2 values in our model that indicate good predictive power explaining 28%, 51% and 66% of the variance of CID, CIM and CFP, respectively, as presented in Table 2. This value of R2 represents that all the model’s variables explain a total of 66% variance in the CFP. We found statistically significant path coefficients that are a direct relationship between the pair of two variables. 

For instance, we found a positive relationship between CSR and CIM (?= 0.34; p <0.001) as well as CIM and CFP (?= 0.36; p < 0.001). These results substantiated both our H2a and H2b. The next pair of direct relationship hypotheses H4a and H4b pertaining to the link of CSR?CID (?= 0.53; p < 0.001) and CID?CFP (?= 0.46; p < 0.001) were also supported. Further, the effect size f2 that is representing the relative impact of each construct in the model (Hair et al., 2017). Table 2 shows that in the full model, CID has a high impact on CFP, followed by CIM and CSR. In addition, the in-sample model predictive relevance fit was determined through the blindfolding method and values of Q2>0, as shown in Table 2, indicated a good level of the model predictive relevance (Hair, 2020).

Table 2. Full model results of R2, Effect Size, and Q2

Constructs	R2	f²	Q2
CID	0.277	0.332-	0.163
CIM	0.505	0.188	0.299
CFP	0.658	-	0.296
CSR	-	0.02	0.344

Mediation Analyses

To test the indirect effects, with the recommendation of Klarner et al. (2013), we performed a direct relationship model and three different mediation models, two for simple mediation and one as an overall study serial mediation model. For these analyses, first, we run a baseline model of the direct only relationship between CSR and CFP excluding both mediators and found a positive relationship between the constructs as shown in Table 3 (? = 0.565, p < 0.001) that conforms to the study hypothesis H1. In the next step, we performed two simple mediation models considering each mediator separately in the model. As a result, we found a partial mediation effect of each mediator of CIM and CID between the relationship of CSR and CFP, given that variance accounted for (VAF) of 58% and 63% that between the range of 20%? VAF ? 80% partial mediation (Hair et al., 2017). The simple mediation models’ results suggest partial support for our H3 and H5. 

Finally, we executed the model of the study, taking both mediators simultaneously in a sequential way as double mediation of both CIM and CID between the CSR-CFP relationship in a way represented in Table 3. The finding of this indirect relationship model with double mediation suggests a full mediation as VAF exceeded 80%, and also the direct relationship of CSR and CFP in the full model reduced substantially but still remained significant (? = 0.105, p < .01) that is considered as complementary mediation (Hair et al., 2017). Hence, fully support our study’s hypothesis H6.

Table 3. Results of mediation Models

Model Description	Direct effect (DE)	Indirect effect (IE)	Total effect	VAF	Mediation?
1. CSR®CFP Direct relationship of     excluding both mediators	0.565*
3. DE(CSR®CFP), IE(CSR®CID®CFP)	0.234**	0.328**	0.562**	58.4%	Partial
2. DE(CSR®CFP), IE(CSR®CIM®CFP)	0.200**	0.354**	0.554**	63.4%	Partial
4.DE(CSR®CFP), CSR®CID®CFP  IE(CSR®CIM®CFP IE(CSR®CID®CIM®CFP)	0.105*	0.452**	0.557**	81.1%	Full

Discussion and Conclusion

Over the last several decades, organisational experts have continued to pay close attention to the topic of CSR. A bulk of studies have focused on CSR's antecedents and outcomes, but little has been done to investigate how CSR impacts CFP via intangible assets (Saeidi et al., 2015). Most of this research is rooted in the stakeholder theory (Freeman, 1984) and resource-based view (Amit & Schoemaker, 1993; Wernerfelt, 1984). We proposed signalling theory (Boulding & Kirmani, 1993) as an alternative theoretical lens to explain the relationship between CSR and CFP via underlying mechanisms of intangible assets. Using signalling theory, we hypothesised that CIM and CID as two sequentially intervening mechanisms through which CSR is related to CFP that were a novel contribution of this study and never examined previously to the best knowledge of the authors. Overall, we found full support for our main hypothesis H6. In this way, we estimated our hypothesised model adopting a path analytic procedure, in which simultaneously all hypothesised propositions of the model were estimated. Resultantly, the issues caused by testing in a piecemeal and causative step manner were much reduced in the current research, which was seldom tested before (Xie et al., 2015). 

Besides this contribution, we replicated a direct relationship in line with the previous studies in a newer context of a developing country. For example, our hypotheses H2a,b and H4a,b were substantiated on multi-national companies' perceptual data and we found a positive relationship between CSR and CIM and CID (Al Mubarak et al., 2019; Arendt & Brettel, 2010; Kim et al., 2020). We also determined a positive relationship between both mediators and CFP (Arendt & Brettel, 2010; Ghaderi et al. (2019); Margulies, 1977; Brown, 1995; Bharadwaj and Menon, 1993). Further, we tested simple mediation models in order to answer the hypotheses H3 and H5. For both hypotheses, we found partial support for hypothesis H3, that is, CSR effect indirectly via CIM to CFP that is congruent with the study of Ali et al. (2020). We believe that the partial median effect of CID is the novel contribution of this study that was previously omitted as a mediator between the CSR and CFP connection. 

Another prominent feature of this research is the measurement approach, especially tapping the concept of CSR that was previously measured either as unidimensional or reflectively measured in the different dimensions of CSR (Ali et al., 2020; Saeidi et al., 2015). Such an approach of measurement was seriously criticised equally by the management experts and statisticians (Lee & Cadogan, 2013; Podsakoff et al., 2012) when each dimension/item of a concept represents a unique aspect like CSR or CFP. Our study overcomes such weakness and provides a holistic view of properly measuring the concept of CSR as reflective-formative higher order and CFP as a formative way of measurement. Thereby, we are sure that our study’s results are reliable enough given that properly measuring the concept of  CSR with its various dimensions as well as CFP. In this vein, our study also validated the measurement instruments in a newer developing country context that were framed in developed countries.

Limitations and Future Research

There are obviously few limitations apart from the contributions of this study that hint at possible research avenues in the future. Since this research is cross-sectional in nature, thereby we are unable to make causal conclusions in terms of the relationship among the constructs studied in this research. Future researchers may replicate the serial mediation model of this study using time lag or longitudinal research design. We could not totally rule out the occurrence of common method variance since the questionnaires were self-rated, though we collected data from employees as well as customers. Pakistani multi-national provided the data for this research. 

To generalise our results, it is important to keep in mind the setting and culture of our investigation. Although the amount of data gathered was substantial as taken from different types of firms, the results' generalizability to other industries and nations should be examined with caution. Further, this research adds to the literature by analysing a robust double serial mediation model where two variables, namely CIM and CID, were utilised as mediators, and future researchers need to work on the boundary conditions of this serial mediation model for a better understanding of in which situation this mediation path becomes the contingent on one or more moderators. Finally, we attract the attention of future researchers to replicate the study’s model using secondary financial data of firms as this study tested the model on perceptual data of employees and customers of various firms. 

References

• Al Mubarak, Z., Ben Hamed, A., & Al Mubarak, M. (2019). Impact of corporate social responsibility on bank’s corporate image . Social Responsibility Journal, 15(5), 710– 722.
• Ali, H. Y., Danish, R. Q., & Asrar-ul-Haq, M. (2020). How corporate social responsibility boosts firm financial performance: The mediating role of corporate image and customer satisfaction. Corporate Social Responsibility and Environmental Management, 27(1), 166–177.
• Amit, R., & Schoemaker, P. J. H. (1993), `Strategic Assets and Organisational Rents' , Strategic Management Journal, 14. 33-46.
• Arendt, S., & Brettel, M. (2010). Understanding the influence of corporate social responsibility on corporate identity, image, and firm performance. Management Decision, 48(10), 1469– 1492.
• Bayol, M. P., de la Foye, A., Tellier, C., & Tenenhaus, M. (2000). Use of PLS path modelling to estimate the European consumer satisfaction index (ECSI) model . Statistical Applicate, 12(3), 361– 375.
• Bharadwaj, S. G., & Menon, A. (1993). Determinants of success in service industries? Journal of Services Marketing, 7(4): 19-44.
• Bhattacharya, C. B., & Sen, S. (2003). Consumer–company identification: A framework for understanding consumers’ relationships with companies. Journal of marketing, 67(2), 76-88.
• Boulding, W., & Kirmani, A. (1993). A consumer-side experimental examination of signalling theory: do consumers perceive warranties as signals of quality? Journal of consumer research, 20(1), 111-123.
• Brown, S. L., & Eisenhardt, K. M. (1995). Product development: Past research, present findings, and futuredirections. Academy of management review, 20(2), 343-378.
• Carroll, A. B. (1979). A three-dimensional conceptual model of corporate performance. Academy of management review, 4(4), 497-505.
• D’Amato, A., & Falivena, C. (2020). Corporate social responsibility and firm value: Do firm size and age matter? Empirical evidence from European listed companies. Corporate Social Responsibility and Environmental Management, 27(2), 909–924.
• Education, M. (2009). Shaping the future: How good education systems can become great in the decade ahead. McKinsey & Company, Singapore .
• Fourati, Y. M., & Dammak, M. (2021). Corporate social responsibility and financial performance: International evidence of the mediating role of reputation. Corporate Social Responsibility and Environmental Management, 28(6), 1749-1759.
• Freeman, R. B., & Medoff, J. L. (1984). What do unions do. Industrial and Labor Relations Review, 38, 244.
• Ghaderi, Z., Mirzapour, M., Henderson, J. C., & Richardson, S. (2019). Corporate social responsibility and hotel performance: A view from Tehran, Iran. Tourism Management Perspectives, 29, 41-47.
• Hair, J. F., Hult, G. T. M., Ringle, C. M., and Sarstedt, M. (2017). A primer on partial least squares structural equation modeling (PLS-SEM). 2nd Ed. Los Angeles: SAGE Publications, Inc.
• Hair, J.F. (2020). Next-generation prediction metrics for composite-based PLS-SEM. Industrial Management and Data Systems. 121(1), 5-11.
• Horng, J. S., Hsu, H., & Tsai, C. Y. (2018). An assessment model of corporate social responsibility practice in the tourism industry. Journal of Sustainable Tourism, 26(7), 1085-1104.
• Javed, M., Rashid, M. A., Hussain, G., & Ali, H. Y. (2020). The effects of corporate socialresponsibility on corporate reputation and firm financial performance: Moderating role of responsible leadership. Corporate Social Responsibility and Environmental Management, 27(3), 1395-1409.
• Kant, R., & Jaiswal, D. (2017). The impact of perceived service quality dimensions on customer satisfaction: An empirical study on public sector banks in India. International Journal of Bank Marketing. 35(3), 411-430
• Kelley, K. J., Hemphill, T. A., & Thams, Y. (2019). Corporate social responsibility, country reputation and corporate reputation: A perspective on the creation of shared value in emerging markets. Multinational Business Review, 27(2), 178–197.
• Kim, Y. (2017). Consumer responses to the food industry’s proactive and passive environmental CSR, factoring in price as CSR trade off. Journal of business ethics, 140(2), 307-321.
• Klarner, P., Sarstedt, M., Hoeck, M., & Ringle, C. M. (2013). Disentangling the effects of team competences, team adaptability, and client communication on the performance of management consulting teams. Long Range Planning, 46(3), 258- 286.
• Liu, Y., Liu, B., & Huan, T. C. T. (2019). Renewal or not? Consumer response to a renewed corporate social responsibility strategy: Evidence from the coffee shop industry. Tourism Management, 72, 170- 179.
• Maignan, I., & Ferrell, O. C. (2000). Measuring corporate citizenship in two countries: The case of the United States and France. Journal of Business Ethics, 23, 283–297.
• Margolis, J. D., Elfenbein, H. A., & Walsh, J. P. (2009). Does it pay to be good? A meta- analysis and redirection of research on the relationship between corporate social and financial performance.
• Mason, C. H., & Perreault Jr, W. D. (1991). Collinearity, power, and interpretation ofmultiple regression analysis. Journal of marketing research, 28(3), 268-280.
• Miyazaki, A. D., Grewal, D., & Goodstein, R. C. (2005). The effect of multiple extrinsic cues on quality perceptions: A matter of consistency. Journal of consumer research, 32(1), 146-153.
• Mohammed, A., & Al-Swidi, A. (2020). The mediating role of affective commitment between corporate social responsibility and eWOM in the hospitality industry. Journal of Sustainable Tourism, 29(4), 570-594.
• Mohr, L. A., & Webb, D. J. (2005). The effects of corporate social responsibility and price on consumer responses. Journal of consumer affairs, 39(1), 121-147.
• Nguyen, N., & Leblanc, G. (2001). Corporate image and corporate reputation in customers’ retention decisions in services. Journal of retailing and Consumer Services, 8(4), 227-236.
• Nguyen, T., Pham, T., Le, Q., & Bui, T. (2020). Impact of corporate social responsibility on organisational commitment through organisational trust and organisational identification. Management Science Letters, 10(14), 3453-3462.
• Orlitzky, M., Schmidt, F. L., & Rynes, S. L. (2003). Corporate social and financial performance: A meta- analysis . Organisation studies, 24(3), 403-441.
• Podsakoff, P. M., MacKenzie, S. B., Lee, J. Y., & Podsakoff, N. P. (2003). Common method biases in behavioral research: a critical review of the literature and recommended remedies. Journal of applied psychology, 88(5), 879.
• Podsakoff, P. M., MacKenzie, S. B., and Podsakoff, N. P. (2012). Sources of method bias in social science research and recommendations on how to control it, Annual Review of Psychology, 63, 539– 569.
• Shabbir, A., Malik, S. A., & Malik, S. A. (2016). Measuring patients’ healthcare service quality perceptions, satisfaction, and loyalty in public and private sectorhospitals in Pakistan. International Journal of Quality & Reliability Management, 33(5) 538 – 557.
• Shafer, W. E., Fukukawa, K., & Lee, G. M. (2007). Values and the perceived importance of ethics and social responsibility: The US versus China. Journal of Business Ethics, 70(3), 265-284.
• Van Riel, C. B., & Balmer, J. M. (1997). Corporate identity: the concept, its measurement and management. European journal of marketing, 31(5/6), 340-355.
• Wernerfelt, B. (1984). A resource-based view of the firm. Strategic management journal, 5(2), 171-180.
• ul-Hassan, F. S., Ikramullah, M., Khan, H., & Shah, H. A. (2021). Linking Role Clarity and Organizational Commitment of Social Workers through Job Involvement and Job Satisfaction: A Test of Serial Multiple Mediation Model. Human Service Organizations Management, Leadership and Governance, 45(4), 337– 351.
• Wheelen, T. L., & Hunger, J. D. (2012 ). Strategic management and business policy: concepts and cases.
• Xie, X. Y., Ling, C. D., Mo, S. J., & Luan, K. (2015). Linking colleague support to employees’ promotive voice: A moderated mediation model. PLoS ONE, 10(7), 1–15.
• Yang, J., & Basile, K. (2019). The impact of corporate social responsibility on brand equity. Marketing Intelligence and Planning, 37(1), 2–17.
• Ye, M., Wang, H., & Lu, W. (2021). Opening the “black box” between corporate social responsibility and financial performance: From a critical review on moderators and mediators to an integrated framework. Journal of Cleaner Production, 313, 127919.