Impact of corporate social responsibility on customer loyalty: Evidence from the Vietnamese jewellery industry

Abstract Corporate social responsibility takes an increasingly important role for businesses and receives the attention of many people. The relationship between Corporate social responsibility and customer loyalty has been tested in many business fields. So, the study aims to determine and estimate the influence of corporate social responsibility on customer loyalty in the jewellery industry. To achieve this objective, the study uses AMOS software that supports confirmatory factor analysis and structural equation modelling (SEM) to clarify whether or how Corporate social responsibility impacts customer loyalty. The inputs to the analyses were obtained through answers from 300 respondents, which completed the survey questionnaire with a total of 22 items is utilised to collect data from customers of top jewellery companies in Vietnam. The research results emphasize the corporate social responsibility is not only useful for jewellery companies but also enhances the customers’ perception that relates to some factors trust, satisfaction, and loyalty.

ABOUT THE AUTHOR Ngoc Tu TRAN Tran Ngoc Tu is currently a lecturer of the Faculty of Business Administration, Saigon University, Vietnam. She is interested in researching the business administration, marketing, and so on.

PUBLIC INTEREST STATEMENT
Corporate social responsibility (CSR) is the key to a company's success. CSR is the goal of most organisations to provide not only benefits for the community but also value for the enterprise. CSR application studies in the hospitality, manufacturing, banking, insurance, and transportation industries exist. CSR takes an increasingly important role for businesses and receives the attention of many people. The relationship between CSR and customer loyalty has been tested in many business fields. So, the study aims to determine and estimate the influence of corporate social responsibility on customer loyalty in Vietnamese jewellery industry. To achieve that, the study uses AMOS software that supports confirmatory factor analysis and structural equation modelling (SEM) to clarify whether or how CSR impacts customer loyalty.

Introduction
Corporate social responsibility (CSR) is the key to a company's success (Maqbool & Zameer, 2018;Paek et al., 2013) and has concerned in academic research and management practice (Youn, Lee and Lee, 2018;Boulouta & Pitelis, 2014;Tingchi Liu et al., 2014). Thus, CSR investment is a competitive advantage rather than an expense, a limitation, or an activity separate from the business operation (El-Garaihy et al., 2014;Porter & Kramer, 2006). Large companies invest in CSR (Luo & Bhattacharya, 2006) mainly because such an investment is considered as a competitive advantage and provides benefits to companies such as improving their reputation, reducing advertising and operating costs, and attracting talents and investors (Fu et al., 2014;Li et al., 2015). Hence, CSR provides various advantages to companies, such as enhancing brand loyalty, trust, and positive attitudes towards a brand and countering negative publicity (McWilliams & Siegel, 2001;Sen, 2006;Sen & Bhattacharya, 2001).
Consumers consider a company's CSR image in their purchase and consumption decisions (S. Choi & Ng, 2011). Although this statement does not apply to all product categories or industry segments (Peloza & Shang, 2011), Luo and Bhattacharya (2006) notes that in the current competitive service market, CSR represents a high-level concept of strategic importance to many companies. Bhattacharya et al. (2009) identified the positive benefits of CSR programme implementation for consumers, such as enhanced awareness-beliefs, attitudes and loyalty during a period of financial crisis. Even though, Foreh and Grier and Forehand (2003) claim that a company's profit-driven CSR activities can damage the company's reputation, espeically from the customers' perception. Some studies still confirmed that CSR improves a company's reputation and financial performance and society. In detail, recent reviews of theories related to CSR showed that well-executed CSR initiatives can improve consumer loyalty (Asatryan, 2012;Yusof et al., 2015). Business strategy creation bases on the concept of CSR is an important opportunity for differentiation, as it considers the interests of different stakeholders and is dedicated to the achievement of social goals and environmental protection (Stanisavljevíc, 2017). It can be clarified that many companies are realizing the importance of CSR and its impact on social benefits and increasing customer loyalty, bringing about operational efficiency for the business itself. The rating of CSR is quite necessary and set in line with the business performance. So, some companies add CSR categories to their disclosed reports to achieve CSR ratings to announce their CSR scores. This rating system is monitored by independent supervisory bodies, such as the Stock Exchange of Malaysia (Lim, 2011). This redefines CSR's relationship with existing business and management.
However, CSR activities become harmful when customers perceive an exaggeration compared with reality to promote business. Customer attitude towards CSR is complex, and customers may react negatively to CSR activities if viewed as a company's primary motive for profit. It claims that the perception of CSR hype is a way to discourage customers from buying products related to CSR activities. Similarly, Foreh & Grier and Forehand (2003) claims that a company's CSR activities can damage a company's reputation when the company just focuses on their own profits. As a result, determining whether the CSR theory is truly an incentive for businesses to consider implementing it in order to ensure the development of society or simplify a business tool to generate profits. Recently, Osakwe & Yusuf (2020) suggested that customers are likely to support companies through their positive actions, regardless of the motivation behind CSR commitments.
CSR is the goal of most organisations to provide not only benefits for the community but also value for the enterprise. CSR application studies in the hospitality, manufacturing, banking, insurance and transportation industries exist. However, within the scope of this research, empirical studies have yet to examine the impact of CSR on consumer loyalty in the jewellery industry. In Vietnam, CSR has become a considerable issue requiring attention from the government, society, and businesses. Therefore, the primary objective of this study is to examine the influence of CSR on customer loyalty in the jewellery industry in Vietnam. To achieve this objective, this study aims to answer the question, "How does CSR affect customer loyalty in the Vietnamese jewellery industry?".
After providing the research gap, the paper is structured as follows. First section mentions about the literature review of basic concept on CSR and its relation to the stakeholder theory. Besides, this section discusses about other factors to create the hypothesis development. Then, the author outlines the research procedures, includes samples, and proposed model. Based on research method (SEM) to estimate the effect of CSR on customer loyalty in Vietnamese jewellery industry and discuss the results. Final section is the conclusion of CSR's role in being an effective strategy to secure customer loyalty. In the last section, the author points out the research's limitation and proposed the future study.

CSR
Stakeholder theory emphasizes the relationships between a business and groups that are among related parties, such as customers, suppliers, employees, communities, banks, investors, and managers in activities that make up their benefits (Freeman, 1984;Walsh, 2005). Based on that theory, Ivanović-Đukić (2011), Srbljinović (2012), Weimer & Pape (1999 claim that CSR supports the company to cover its activities that focused on the interests of all stakeholders. Therefore, Davis (1960) asserted that certain socially responsible business decisions can be justified through a long, complicated process of reasoning as having the potential to reap long-term economic gains for the firm, gaining benefits in return for its socially responsible outlook. In addition, McGuire (1963) defines CSR as a company's economic obligation, law enforcement, and other specific responsibilities towards society. Davis (1975) asserts "Social responsibility is the obligation of decision-makers to take actions which protect and improve the welfare of society along with their interests".
Obviously, CSR is enterprises' commitment to business ethics and contribution to sustainable economic development and the improvement of the quality of life of their employees and their families, local communities, and society. Carroll (1979Carroll ( , 1991 generalised CSR into four groups, namely, economic, legal, ethical, and philanthropic responsibility. In another aspect, Johnson (1971) presented a second view of CSR, stating that "Social responsibility states that businesses carry out social programs to add profits to their organization". In this view, social responsibility is perceived as long-term profit maximisation.
Generally, CSR is understood as policies and practices employed by companies to ensure the consideration and protection of society and stakeholders in their strategies and operations.

Trust
Trust signifies reliance based on the willingness of partners to exchange from the confidence between them (Morgan & Hunt, 1994). The review identified trust as a prerequisite for the creation and maintenance of long-term relationships between companies and consumers that are interested in the direct effects of trust and satisfaction as components of customer loyalty (Morgan & Hunt, 1994), especially in the service market context. The present study.
Trust plays a significant role in service organisations' management of relational exchanges (Grönroos, 1984;Gummesson, 1994). Trust from customers is crucial and minimises perceived risks of service outcomes (Berry, 2000;Darby & Karni, 1973;Laroche et al., 2004). According to Patrick (2002), customer trust manifests as thoughts, feelings, emotions, or behaviours when customers believe that a supplier can be trusted to behave in their best interest when they relinquish direct control. During the last few decades, the idea of "trust" was recognised as a significant mediator in the relationship between client satisfaction and loyalty (Roland & Kunz, 2010). Bachelet (1995) and Zeithaml and Bitner (2000) also argued that the concept of customer satisfaction is an assessment measured against a product or service. Kotler (2003) defined customer satisfaction as the level of an individual's sensory state from comparing results obtained from the consumption of a product/service meeting the individual's expectations. Customer satisfaction involves the comparison of expectations before and after the purchase of a product or service, whereas customer expectations are formed from shopping experiences, friends and colleagues and seller and competitor information. Moreover, Tran Phi (2018) mentioned understanding customer satisfaction levels, enterprises can make decisions about products and services to change their activities to improve customer satisfaction. Measuring customer satisfaction can help enterprises engage in advanced competition with their rivals to satisfy customers' needs.

Word of mouth (WOM)
WOM is defined as a form of informal trade through the person-to-person dissemination of information about a brand, product, organisation or service in noncommercial communication (Harrison-Walker, 2001). According to Bristor (1990), customers can obtain information about a product or service from previous customers based on the WOM process. Receivers (consumers) consider this communication form more significant in influencing their decisions than other forms of communication (Murray, 1991). In addition, Kotler (2000) emphasised that WOM is a low-cost marketing tool.
WOM can be considered as cultural exchange, the continuous flow of accurate information and effective communication or dialogue between two individuals. Haywood (1989) considered WOM as a type of formal conversation. Despite measuring WOM, some researchers agree that WOM is merely an informal and non-commercial type of communication that cannot be authenticated or controlled (Paridon, 2008).

Customer loyalty
In service or product case studies, customer loyalty is understood as loyalty to a product or service provider as well as to a supplier's brand. Oliver (1997) suggested that loyalty stages represent a process highlighting the relationship between attitudes and behaviours. Firstly, Oliver stated that loyalty is developed from three components, namely, cognitive, affective, and conative components. The three stages are consistent with the general definition of attitude (Bagozzi, 1978;Breckler, 1984). Subsequently, researchers vertically developed four stages of the cognitive verification of loyalty theory, sentiments, trends and behaviours (Evanschitzky & Wunderlich, 2006;Yuksel et al., 2010).
Customer loyalty indicates customers' intention or inclination to repurchase goods or services from the same organisation (Edvardsson et al., 2000). Businesses are interested in customer loyalty factors leading to increased profits. Thus, building a customer loyalty foundation became a major goal for many companies and marketing programmes (Kotler et al., 2007).

Empirical studies and hypothesis development
Empirical studies revealed the positive relationship between CSR and customer loyalty. Andrea et al. (2011) determined that CSR awareness positively affects customer loyalty. In addition, Mar Garcia-De et al. (2005) examined the influence of customer perception of CSR on customer loyalty and found that CSR has a direct influence on the total value of the received service, which in turn has a direct effect on loyalty. Especially, the service industry has a close relationship with the environment; thus, the role of CSR is considered to be an important factor. For example, the hospitality industry can significantly influence natural and social environments and local communities (Chung & Parker, 2010). Currently, some hotels are implementing CSR-related activities to demonstrate their commitment to their customers (Danuta, 2012; Li et al., 2015), as such activities contribute to social development that attracts new customers. Therefore, CSR activities can serve as an effective brand for companies (Raman et al., 2012), especially in the tourism industry, which is moving towards sustainability. Meanwhile, Balqiah et al. (2011) evaluated the relationship between CSR and customer loyalty and revealed that CSR activities are cost-consuming activities and can create a reputation for a company. In the long term, such activities can create customer loyalty, thereby contributing to the business financially. Nguyen Van Anh and Nguyen Thi Phuong Thao (2021) focused on the relationship between CSR and customer loyalty that proves the factors of social responsibility, customer engagement, trust, and satisfaction have interdependent and positive effects on customer loyalty.

Relationship between CSR and trust
Shared corporate and consumer values are incredibly influential in consumer trust (Morgan & Hunt, 1994). According to Brown and Dacin (1997) and Martinez and del Bosque (2013), CSR-related initiatives can provide helpful information for enhancing overall trust in organisations. Pivato et al. (2008) pointed out that trust is the primary outcome of CSR activities and results directly from certain CSR practices. Hansen et al. (2011) revealed that trust plays a mediating role in linking CSR to employee attitudes and behaviours (Yu and Choi, 2014). Thus, in this relationship, trust is consistent with a firm's expectations owing to its reputation (Ganesan, 1994). This phenomenon puts CSR at the forefront based on the extent of the CSR activities of a company. Such activities can enhance trust in a company and its image (Servera-France and Arteaga-Moreno, 2015). Andrea et al. (2011) proposed that perceived CSR positively impacts trust in a company. Moreover, Salmones, Crespo, Bosque et al. (2005b) demonstrated that consumers' perception of a company's CSR activities could directly affect their beliefs about the company's products and services. When a company has social responsibility commitments, customers will consider and evaluate the implementation of such commitments, which will directly affect customer trust. Hosmer (1994) argued that by incorporating principles of responsibility and ethics into corporate policies, decision-making strategies can strengthen shareholders' confidence in a company, including that of customers.
Hypothesis 1: CSR positively impacts trust in the jewellery industry in Vietnam.

Relationship between CSR and customer satisfaction
Numerous studies found that some customers are loyal to and willing to pay high prices for the products and services of socially responsible companies (McWilliams & Siegel, 2001). Luo and Bhattacharya (2006) identified CSR as the driving force behind satisfaction in large companies with increasing market value (Rhou et al., 2016). CSR can influence customer satisfaction (He & Li, 2011;Luo & Bhattacharya, 2006;Martínez & Del Bosque, 2013). Three attitudes can explain the influence of CSR on satisfaction (Luo & Bhattacharya, 2006). Firstly, perceived value has a significant influence on satisfaction (Athanassopoulos, 2000). Secondly, social and organisational identity theory indicates that CSR initiatives cause a high level of consumer identity as a form of corporate identity (Bhattacharya & Sen, 2003) that companies create to satisfy customers' needs (Luo & Bhattacharya, 2006). Thirdly, companies must have a balanced attitude towards the interests of all stakeholders to manage their needs (El-Garaihy et al., 2014). In details, according to Rashid et al. (2013), CSR refers to activities that generate benefits for customers. Such activities include product/service quality assurance, affordable prices, complaint distribution, rights to privacy, and other concerns, such as charitable activities, environmental protection activities, and CSR encompassing strategies for improving customer satisfaction.
Hypothesis 2: CSR positively affects customer satisfaction in the jewellery industry in Vietnam. Tran, Cogent Business & Management (2022)

Relationship between CSR and WOM
Social identity theory was first introduced by Bhattacharya and Sen (2003), who asserted that customers identify with a company to satisfy their need for self-identification. A company that acts socially responsible will attract recognisable customers (Deng & Xu, 2017). Once customers identify with a company in the buying process, they exhibit behaviours such as loyalty, positive WOM and civic behaviours (Ahearne et al., 2005).
Positive WOM involves customers' satisfaction with and recommendation of goods and services through informal networks and personal communication (Bansal & Voyer, 2000). Bhattacharya and Sen (2003) concluded that one of the primary behavioural outcomes of CSR activities is consumers' willingness to positively convey the image and reputation of companies that often engage in socially responsible activities.
Several studies examined the impact of interactive CSR on customer WOM. Du et al. (2007) found that CSR positively impacts customer trust, which is linked with advocacy behaviours, such as positive WOM. Mandhachitara and Poolthong (2011) demonstrated that CSR is linked with bank customer loyalty. Thus, CSR activities produce immediate results that are the positive impact of WOM on customer perceptions and attitudes.
Hypothesis 3: CSR has a positive impact on WOM in the jewellery industry in Vietnam.

Relationship between CSR and customer loyalty
When customers are satisfied with their chosen brands, they will gradually repurchase services and products and eventually become loyal customers (Martínez and Bosque, 2013). Moreover, CSR can affect loyalty (Maignan & Ferrell, 2001;Sureshchandar et al., 2002). Several studies revealed the positive impact of CSR on customer retention and loyalty (Aurier & N'goala, 2010; Leverin & Liljander, 2006). Apart from aspects related to strategic commercial prospects, CSR can affect loyalty directly or indirectly (De Los B. Choi & La, 2013;Shin & Thai, 2014. Furthermore, signalling theory suggests that accountability can influence customers to perceive a company as trustworthy through its fairness and responsibility signals (Bustamante, 2018).
In addition, Mar Garcia et al. (2005) observed the influence of customers' perception of CSR on customer loyalty and showed that CSR directly influences the total value of received services. Furthermore, Andrea et al. (2011) andBalqiah et al. (2011) supported the argument of the existence of a positive correlation between CSR and customer loyalty.
Hypothesis 4: CSR positively impacts customer loyalty in the jewellery industry in Vietnam.

Relationship between trust and WOM
The establishment of a business-to-customer relationship can help customers and businesses connect with comfort and trust (Gwinner, Gremler and Bitner, 1998). This relationship can increase customer referrals and positive WOM. Satisfaction and trust have a strong relationship but are conceptually different and thus have different effects on customer retention (Szymanski and Henard, 2001). The establishment of a relationship of mutual trust between a business and customer can increase the parties' commitment and create real customers, that is, loyalty (Ranaweera and Prabhu, 2003). Thus, a high level of customer trust can help convey a positive WOM effect.
Consumer trust plays a central role in helping consumers make purchasing decisions in the growing e-commerce landscape (Gefen, 2000;Jones & Leonard, 2008;McKnight et al., 2002). An increasing level of trust is likely to encourage consumer exploration and influence intention to purchase products from a supplier (Hajli, 2013;Gefen, 2000). In addition, trust can positively affect transfer from a supplier to a customer through new product development performance (Zhao & Lavin, 2012). Trust can help consumers overcome the perceived risks of making a purchase. Thus, trust is a powerful motivation for suppliers to interact with consumers to promote their goods.
In addition, several studies quantified the influence of WOM and found that WOM is nine times more effective than advertising in converting negative sentiments into positive attitudes (Day, 1971). Therefore, when consumers have confidence in a business, the business brand will spread.
Hypothesis 5: Trust positively affects WOM in the jewellery industry in Vietnam. Chiou and Pan (2009) argued that trust precedes satisfaction. Thus, trust is a solid foundation for increasing customer satisfaction and loyalty (Schoorman, Mayer and Davis, 2007). According to Schoorman et al. (2007), trust reflects a customer's willingness to purchase a provider's services/ products based on his/her positive expectations of the quality of the products/services provided. Contacting customers in good faith to build trust is important for organisations.

Relationship between trust and customer satisfaction
Hence, trust in an organisation generates a positive response from the organisation, which can increase customer satisfaction. For instance, customers' trust in banks is the basis for the continuous use of banking services, because trust encourages customers to build long-term relationships with banks (Ofori et al., 2018). The more a bank can win customers' trust, the more satisfied the customers, which will reduce the probability of customers switching to other banks.
Hypothesis 6: Trust positively influences customer satisfaction in the jewellery industry in Vietnam.
Consuming a product or service, product/service failure can lead to the breakdown of the customer's confidence and loyalty (B. Choi & La, 2013). Trust plays an essential role in developing customer loyalty (Kaur et al., 2013;Zeithaml et al., 1996). Morgan and Hunt (1994) argued that trust is necessary for creating and maintaining long-term relationships between businesses and consumers, especially in the service market context. In details, service quality affects trust and behavioural outcomes, and trust significantly affects consumer loyalty (Vlachos et al., 2009;Nguyen et al. (2008). For instance, a satisfied customer will be loyal if a high level of trust exists (Dimitriadis et al., 2011). Reichheld and Schefer (2000) concluded that companies must first gain customers' trust before gaining their loyalty. In the service industry, trust is the key to building relationships. Hikkerova (2011) and Andrea et al. (2011) emphasised the role of trust in building customer loyalty.
Hypothesis 7: Trust has a positive impact on customer loyalty in the jewellery industry in Vietnam. Tran, Cogent Business & Management (2022)

Relationship between customer satisfaction and WOM
Studies found customer satisfaction to be the leading factor determining loyalty (Anderson et al., 1994;Rust & Zahorik, 1993). Similarly, customer satisfaction was identified as a crucial factor generating positive WOM (File et al., 1994;Yi, 1990). Based on these views, the following hypothesis (Hypothesis 8) is proposed: Customer satisfaction has a positive impact on WOM in the jewellery industry in Vietnam.

Relationship between customer satisfaction and customer loyalty
Customer satisfaction and loyalty are positively related to a firm's profitability and market share (Anderson et al., 1994;Flint et al., 2011;Reichheld, 1993). B. Choi and La (2013) and Martínez and Del Bosque (2013) suggested that customer satisfaction will contribute to loyalty, positive WOM and a positive perception of CSR. Several studies, such as Liljander and Roos (2002) and Aldlaigan and Buttle (2005), determined that loyal customers will exhibit a more favorable attitude towards a company's reputation than its competitors. Thus, a strong relationship exists between customer loyalty and satisfaction (Hallowell, 1996;Hur et al., 2013). Customer satisfaction is the core of loyalty (Bloemer et al., 1998;Shin and Thai, 2014). Moreover, satisfaction has a positive effect on consumer loyalty (J. Lee et al., 2001;Andrea et al., 2011;Fang-Yuan et al., 2012). The relationship between satisfaction and loyalty was examined in different fields (Homburg & Giering, 2001;Lam et al., 2004;J. Lee et al., 2001). K-Y. Lee et al. (2007) suggested that a high level of satisfaction can lead to increased committed associations between companies and consumers.
Hypothesis 9: Satisfaction positively impacts customer loyalty in the jewellery industry in Vietnam.

Relationship between WOM and customer loyalty
The influence of WOM on consumers' decision to use services is stronger than its influence on consumers' decision to purchase physical products (Bristor, 1990;Murray & Schlacter, 1990). Bansal and Voyer (2000) argued that this difference arises from the intangibility of services, which can create difficulties in making prepurchase comparisons. Moreover, the extent of the influence of WOM on consumer decisions varies across services (Still et al., 1984).
Hypothesis 10: WOM has a positive impact on customer loyalty in the jewellery industry in Vietnam.

Sample
Jewellery, together with watches, makes up a segment of the market known as "hard luxury," as opposed to fashion, accessories, and cosmetics, which are comprised of "soft" materials. Especially, gold jewellery remains the most popular, accounting for roughly 70 percent to 80 percent of the income structure of jewellery firms in Vietnam. In 2021, the Luxury Watches & Jewellery segment will generate the US $224 million in revenue. The market is predicted to expand by 8.13 percent every year (2021-2025). According to experts in the jewellery industry, Vietnamese jewellery market turnover will be around USD 126 billion in 2023, more than doubling from 2013 and growing at a Compounded Annual Growth Rate of around 6% from 2019 to 2023. However, the jewellery market in Vietnam is highly fragmented, with minor retailers accounting for roughly 73 percent of the market. Only 20% of the remaining shares belong to big companies like Phu Nhuan Jewellery Joint Stock Company (PNJ), Saigon Gemstone Jewellery Limited Company (SJC), Bao Tin Minh Chau Gemstone Jewellery Company, Doji Group . . . The data of local retail stores is difficult to determine due to the highly fragmented market. Because of the fierce pricing rivalry, big brands' business activities have been hampered. However, Circular 22/2013/TTBKHCN on ensuring the quality of jewellery sold has progressively allowed major businesses to take market share from small businesses. Since Circular 22 came into effect, PNJ and SJC's jewellery retail revenue has soared.
The information collection process was conducted from November 2020 to May 2021. A questionnaire was sent to 350 individuals who bought jewellery and accessories from are reputable jewellery enterprises, such as Saigon Jewellery Company Limited (SJC), Phu Nhuan Jewellery (PNJ), the DOJI Gold & Gems Group (DOJI), Diamond World and Vinh Cara-Diamond Jewellery, in the Vietnamese jewellery industry. There are 340 of them were returned (with 97,1%) (Table 1). However, the number of collected valid questionnaires was 300, accounting for 85.7% of the total number of distributed questionnaires. As a rule of thumb, researchers suggest a relatively large sample size (n > 200) for the SEM (J.F.J. Hair et al., 1998).

Proposed model
Based on a combination of different studies with a number of research models on the effect of CSR on customer loyalty, the research model was proposed, which is illustrated in Figure 1.

Research methods and procedures
The research process to determine the effect of CSR on customer loyalty in the jewellery industry is structural equation modelling (SEM) that was used in this study to analyse the proposed model because it is a multivariate data analysis technique commonly used in the social sciences (González et al., 20092008). AMOS, a powerful statistical tool for SEM (Nachtigall et al., 2003), that was employed to examine the proposed model. In the structural model, the relationships between the constructs were examined, and the coefficients were estimated to test the hypotheses.

Descriptive statistics
The table in Appendix 2 presents the demographic information of the 300 respondents. According to this table, 184 customers were female (61.3%), and the rest were male. According to data from the World Gold Council, in 2020-2021, 57% of jewellery shoppers were female, and gold jewellery consumption was high. Thus, data match between the poll and reality was observed.
Moreover, the customers between the ages of 25 and 34 years accounted for the highest proportion (51.7%), followed by those between the ages of 35 and 44 years (21%) and those between the ages of 18 and 24 years (17.3%). The second lowest proportion was the customers between the ages of 45 and 60 years (8.3%) (Appendix 2).
The number of customers with a vocational education or training centre qualification was 122, accounting for 40.7% of the total number of customers, followed by the customers with an over undergraduate qualification, which were 121, accounting for 40.3% of the total number of customers (Appendix 2).
The sample distribution by occupation is explained as follows. The customers in the business group accounted for the highest proportion, which was 46.3% of the total number of surveyed customers. In addition, the students accounted for the lowest proportion, which was 4.0% of the total number of surveyed customers (Appendix 2).

Survey instrument
As discussed in the previous section, the survey items were adopted from the relevant literature and grouped into 22 observed variables for measuring the five constructs in the research model (Appendix 1). To ensure the clarity of each survey item, the author conducts a pilot test on the survey questionnaire. In addition, the expressions of the survey items were adjusted to fit the context of the jewellry industry under the test from experts. The survey items were rated on a five-point Likert scale ranging from 1 ("strongly disagree") to 5 ("strongly agree").

Evaluation of scales with the cronbach's alpha reliability coefficient
The scales were analysed using the Cronbach's alpha reliability coefficient. Variables with a total correlation coefficient (item-total correlation) of less than 0.3 were excluded, and the scales were accepted for the three-step analysis if their Cronbach's alpha coefficient was 0.6 or higher (Nunnally and Burnstein, 1994).
Summary of the reliability of the scales: After the reliability of the scales was determined, the scales were used to measure aspects related to CSR, WOM, trust, customer satisfaction and customer loyalty. A summary of the scales used for the subsequent analysis is presented in Table 2. Table 3 shows that a reliability analysis using Cronbach's alpha was conducted on the survey data. The Cronbach's alpha values of the five constructs were 0.924, 0.939, 0.957, 0.885 and 0.911. This result confirmed the internal consistency of the constructs, as all the Cronbach's alpha values were over 0.7 (Jr. J. F. Hair et al., 2010). Because the index RMR equals 0,000 which is less than 0.05, it is confirmed that the model is appropriate (Table 4).Moreover, the average variance extracted (AVE) is higher than 0.5 which indicates the adequate convergence (Hair et al. 2010). Table 5 shows that all AVE values figured for each construct are both larger 0.5, it means that there is an adequate convergence.

Evaluation of scales with exploratory factor analysis (EFA)
After the Cronbach's alpha reliability test was performed, EFA was conducted on the scales. However, before EFA was conducted, considering whether the data are suitable for the factor analysis method by performing Bartlett's test of sphericity was necessary. Bartlett's test of sphericity tests the null hypothesis, that is, "H0: The variables in the population correlation matrix are uncorrelated". If this hypothesis is not rejected, then factor analysis is likely unsuitable.  The method of extracting the main components (principal components) and performing Varimax rotation to simplify the interpretation of the variables and factors was used in this study. The results of Bartlett's test of sphericity are presented in Table 3. The sig. = 0.000 result allowed the rejection of the hypothesis stating that the variables are not correlated with one other in the population, and KMO index = 0.900 > 0.5; thus, the analytical method factor could be used for the data analysis.
According to Total variance explained (Appendix 3), with five groups of factors drawn, 74.67% of the data variability can be explained, which exceeds the acceptable threshold for forming new factors (50%). Variables can be grouped by factor by assessing the sample matrix. In each row of each variable, if the factor loading of the attributes is the largest for any factor and satisfies the condition of > 0.4, then it will belong to that factor. Thus, the variables in the row with the largest load factor of less than 0.4 were excluded, as they had no measurement significance for any factor (Appendix 4: Rotated component matrix).

CFA
As shown in Figure 2, to assess the overall model fit without being affected by the sample size, alternative standalone fit indices that are less sensitive to sample size were used. According to Jr. J. F. Hair et al. (2010), such indices include the GOF index (GFI), adjusted GFI (AGFI), comparative fit index (CFI) and the root mean square error of approximation (RMSEA), which must satisfy various conditions for a good model fit. Specifically, the GFI should be close to 0.90, the AGFI should be more than 0.80, the CFI should be more than 0.9, the RMSR should be less than 0.05 and the RMSEA should be less than 0.10 (Jr. J. F. Hair et al., 2010).
The validity of the structural model was examined, with acceptable values for the GOF indices, and the structural relationship parameters were investigated. The examination of the structural relationships involved the investigation of the standardised regression weights and the assessment of the explained variance in the main dependent variable (Jr. J. F. Hair et al., 2010). The GOF values (χ2 = 480.679, p < 0.000, df = 192, χ2/df = 2.504, TLI = 0.946, CFI = 0.955, RMSEA = 0.071) indicated that the structural model fit the data well. Carmines and McIver (1981) stated that the relative chi-square of an acceptable model should be within the range of 2:1 or 3:1. However, Kline (1998) argued that 3 or less is acceptable. Some researchers accepted a value as high as 5 to consider a model adequately fit (Schumacker & Lomax, 2004). Besides, according to Jr. J. F. Hair et al. (2010), the RMSEA should be less than 0.10.
The results showed that the chi-square/df = 2.504 ≤ 3, and the RMSEA = 0.071 < 0.1. Thus, the model was suitable.

SEM and hypotheses
After the compatibility test results of the entire model were obtained, SEM was implemented for the entire model. Thus, the model presented for analysis was completely satisfactory, with the following results (Figure 3): As the RMSR index was 0.000, which is less than 0.05, the model was confirmed to be appropriate. Moreover, the AVE is higher than 0.5, which indicated adequate convergence (Jr. J. F. Hair et al., 2010). Table 5 shows that the AVE value for each construct is larger 0.5, which indicated adequate convergence.

* Hypothesis testing
After the appropriate test model was confirmed, the results of the hypothesis testing are presented in Tables 6 and 7. According to Table 8 This study determined that CSR significantly influenced customers' trust, satisfaction, and loyalty. Various jewellery product providers in the Vietnamese market have a highly competitive market share, quality products and customer policies. Maintaining customer loyalty is difficult and challenging for any company. Hence, jewellery firms should implement and invest in CSR activities, as such practices can effectively increase customer satisfaction and trust. Besides, CSR activities can create not only benefits for communities but also value for a company. As a result, maintaining customer loyalty can help companies sustainably.

Discussion
As the p-values are smaller than 1%, the results of the hypothesis testing show that CSR and its interactions with other factors are consistent top previous research study. In details, CSR has a positive impact on customer loyalty (Aurier & N'goala, 2010; Leverin & Liljander, 2006;Maignan & Ferrell, 2001;Sureshchandar et al., 2002), customer trust (Brown & Dacin, 1997;Martinez and del Bosque, 2013;Andrea et al., 2011) and customer satisfaction (El-Garaihy et al., 2014;He & Li, 2011;Luo & Bhattacharya, 2006;Martínez & Del Bosque, 2013). In addition, the research shows the CSR also has an indirect influence on loyalty through customer trust (Choi, Beomjoon;La, Suna, 2013).  Table 11, the results of the regression weights showed the following evidence: -CSR affects customer trust (p = 0.000) -CSR impacts customer satisfaction (p = 0.000) -CSR affects WOM (p = 0.000) -CSR affects customer loyalty (p = 0.000) -Customer trust has an impact on customer satisfaction (p = 0.000) -Customer trust affects WOM (p = 0.000) -Customer trust affects customer loyalty (p = 0.000) -Customer satisfaction affects WOM (p = 0.000) -Customer satisfaction affects customer loyalty (p = 0.017) Moreover, the degree of impact of the independent variables on the dependent variable is presented in Table 7.
CSR is a broad category that originates from companies' activities. Hence, CSR implementation is necessary for development. The more businesses participate in social programs and sponsor humanitarian programs, the more loyal their customers are to their products/services. CSR activities help customers recognise companies that provide benefits to the community instead of focusing only on their value. Because of CSR generates benefits for the company and contributes to the sustainable development of the community, and the business operations of the jewellery industry are no exception. Therefore, enterprises must make strong CSR commitments in their business strategies to achieve customer loyalty and a sustainable competitive advantage. Besides, the implementation of practical and meaningful CSR programs that can differentiate a company from its competitors will increase consumer trust and loyalty. With continuing effort in their social responsibility activities, businesses in the jewellery industry reached various accomplishments. However, limitations remain in their social responsibility activities. To encourage enterprises to conduct social responsibility activities, the jewellery industry must involve the long-term social responsibility planning strategy for each period based on the companies' development orientation. Then, the companies establish the implementation of social responsibility activities yearly. In addition, attention should be paid to the inspection and review of social responsibility activities. When a company's development orientation changes, adjusting its long-term social responsibility strategy is necessary.

Conclusion
The essential role of perceived CSR as identified in the present research indicates that managers of jewellery companies should recognize the influence of perceived CSR and monitor service recovery encounters to ensure that the recovery process improves the customer perceptions of firm CSR. Because CSR has a direct influence on customer loyalty and indirect customer loyalty through WOM. This is also the new contribution of this paper to the jewellery industry when conducting how CSR and its interactions with customer loyalty.
Jewellery companies in Vietnam need to address CSR and customer loyalty issues to ensure their long-term development and sustainable operations. Promoting CSR would be an effective strategy to secure customer loyalty. In addition, increasing word-of-mouth via better service quality and better customer trust in association with CSR would enhance customer loyalty.
The study has limitations, such as a small sample size, because the scope of the study only surveyed enterprises within Ho Chi Minh City. Besides, the research has not mentioned the legal and ethical aspects. So legal and ethical issues should be taken seriously. I recommend that future research directions should extend to the entire country, focus on the legal and ethical aspect, and classify according to the type of ownership between the state and the private sector.

Funding
The author received no direct funding for this research.

Author details
Ngoc Tu Tran E-mail: tranngoctu1984@gmail.com Faculty of Business Administration, Saigon University, Chi Minh City, Vietnam.

Disclosure statement
No potential conflict of interest was reported by the author(s).  a. Rotation converged in six iterations.