Examining the effect of perceived quality of authentic souvenir product, perceived value, and satisfaction on customer loyalty

Abstract The aim of this research was to explore predictive relationships between perceived quality of authentic souvenir product, value, satisfaction and customer loyalty that consisted of word-of-mouth and repurchase intention. A self-administered, onsite intercept survey was deployed to gather the data to formulate the hypothesised relationships in this study. In total, 250 questionnaires were distributed to non-local Thai customers (i.e., those who live outside the area of the northeast part of Thailand) and 180 useable surveys were tested. Partial least squares structural equation modelling was used to test the proposed conceptual model and hypotheses. This research contributes to the literature by showing that the first-order components of authentic souvenir quality, service encounter quality, and physical quality constituted the second-order variable of perceived quality of authentic souvenir product. Additionally, perceived value and satisfaction mediated the effects of perceived quality of authentic souvenir product on word-of-mouth.


PUBLIC INTEREST STATEMENT
Authentic souvenirs which are made from the symbolic attributes of the destination can trigger consumers to remember a place, event, and experience, which are imbued with meaning and significance. Thai silk is considered as an authentic product which offers product that is unique, authentic, and representative of the host country or region. In addition, Thai silk has become a part of the cultural heritage of Thailand. However, prior research specifically in the context of souvenirs is still unclear as to how souvenir authenticity affects customers' behavioural intentions. In doing so, this study aims to shed new light on examining the first-order components of authentic souvenir quality, service encounter quality, and physical quality constituted the second-order variable of perceived quality of authentic souvenir product which are used to evaluate customer satisfaction, repurchase intention, and word-ofmouth intention. Therefore, this study contributes to the understanding of authentic souvenirs consumption in Thailand.

Introduction
The concept of authenticity has been examined in a variety of research areas, such as products (Algharabat et al., 2017;Keskitalo et al., 2019), branding (Fritz et al., 2017;Moulard et al., 2016), and tourism (Knudsen et al., 2016;Ram et al., 2016). In a tourism context, authenticity is defined by flagship and/or iconic objects/structures/projects that motivate many people to a location and/or to host destinations or cities (Ram et al., 2016). This definition of authenticity can also be adapted to define a variety of souvenir and tourism products. For example, Soukhathammavong and Park (2019) describe that toured objects and/or places are designed and presented by tourism suppliers. This enables visitors (here after, we use the term customer to imply a visitor or tourist) to gain a level of destination experience with some components through the lens of a vacation or trip (Cohen & Avieli, 2004). A unique tourism destination facilitates customers in addressing most of their desires and contributes positively to their overall vacation experience (Ho et al., 2020). In addition, customers have been demonstrating an increasing interest in local products and souvenirs, such as fabrics and locally grown and produced food items (Ho et al., 2020;Kozak, 2016;Li & Ryan, 2018).
According to Smith and Reid (1994), authentic souvenirs represent an image of a tourism destination, create memorable customer experiences, and can strengthen identities. Therefore, the significance of authentic souvenirs may have a different meaning for each individual customer because "they both represent the destination's image and embody tourists' experiences and are therefore related to experiential consumption" (Horodyski & Gandara, 2016, p. 884). Soukhathammavong and Park (2019) claim that authentic souvenirs include three main value components: economic, symbolic and artisanal values. More specifically, they identify authentic souvenirs as having an integrity of culture, history and place identity, product origins, handmade/ handicraft composition, skilled craftsmanship and authentic patterns and presentation (Soukhathammavong & Park, 2019). Shopping for local, authentic products is considered a major customer activity, which is an important aspect of a trip experience and represents a large element of customer spending (Li et al., 2021;Lloyd et al., 2011). Therefore, buying an authentic souvenir or product can assist residents in increasing their income, create jobs and play a crucial role in contributing a source of revenue for many tourism destinations (Lloyd et al., 2011;Wong, 2013).
This paper identifies few research gaps. First, as a matter of fact, previous research into souvenirs and their authenticity has been a minor but consistent subtheme within the marketing and tourism domains for a few decades (Trinh et al., 2014). Specifically, prior studies have examined the context of souvenirs from various perspectives, including customers' purchasing behaviour (Lin & Wang, 2012;Suhartanto, 2018), motivation for purchasing souvenirs (Kong & Chang, 2016;Swanson & Horridge, 2006), authenticity and souvenirs (Fu et al., 2018), and customer and retailer perceptions (Swanson, 2004;Wong & Cheng, 2014). The empirical research is unclear as to how souvenir authenticity affects customers' behavioural intentions. Second, this issue has rarely been investigated in countries where there is a rich architectural heritage and historical destinations. Thailand is undoubtedly one of Asia's most popular destinations in which many cultural resources are materialised for souvenir production, in part to meet the increasing demand for local features in souvenirs. Thai silk and silk production has become a part of the cultural heritage of Thailand (Graham, 2010). The identity of Thai silk not only represents at some level the images, values and cultures of the country for which they are souvenirs but may also represent the beauty of Thai traditional textiles and economic values for many (Johnsen & Riddle, 2007). Therefore, this study examines the effect of the perceived quality of authentic souvenir product on perceived value and satisfaction; customer loyalty of the perceived value and satisfaction; and the associated mechanisms and boundary conditions.

Souvenir authenticity in Thailand
An authentic souvenir shows a craftsmanship that implies its apparent culture and history. Nevertheless, the circumstances differ by one's perspective. Individuals may assess souvenirs in their own way. Somehow, the most powerful variable is the meaning that tourists have associated with an object through the process of attributing that meaning. A recent study reported that souvenir authenticity was determined by tourists' individually (e.g., pleasure-seeking tourists vs. serious tourists) and by the type of souvenirs (from high-to low-levels of value). Most importantly, this perspective differs from the traditional concept of individual authenticity standards (e.g., toured objects, settings, and services).
Thai handicrafts are created by highly skilled craftspeople for royal ceremonial and ritual purposes at palaces or temples and include gold and silversmithing products, nielloware, lacquerware, silk and brocade-woven materials (Cohen, 2000). They account for a large majority of the souvenirs sold in Thailand (Office of the National Economic and Social Development, 2009). Specifically, Thai silk has distinct characteristics that distinguish it from silk produced in other countries. The style of silk production in Thailand creates a unique fabric; every process is systematically and patiently done by hand. Most Thai silks are produced related to characteristics of each locality that represent cultural, social, authentic and heritage value. For example, the Nakha Silk Market, located in the northeast region of Thailand, has unique designs in Thai silk styles that are different when compared to silk in the southern and north parts of Thailand or other Asian countries (e.g., Laos, Cambodia and Vietnam), even though a customer can see a shared history in the products. The local silks are naturally dyed and woven in age-old patterns, and some include golden and silver threads (Pha Yok Thong). However, there are a number of concerns regarding the souvenir trade that cannot be ignored, especially in emerging countries, such as Thailand. For this reason, Thailand can serve as an excellent research site to collect data.

Perceived quality of authentic souvenir product
Quality is one component that leads to customer post-consumption evaluation such as satisfaction, value, trust, and loyalty to both products and services (Carrillat et al., 2007;Parasuraman et al., 1988). Quality is usually judged by customers who compare the expected quality and the perceived quality (Chen, 2008;Loureiro & González, 2008). The concept of quality of both products and services have conceptualised as multidimensional components (Garvin, 1984). For example, Ryu et al. (2012) studied the concept of quality within the restaurant business. These researchers mentioned that customers normally assess restaurant quality based on product, employee service and the physical environment. Also, Žabkar et al. (2010) proposed that visitors tend to measure quality at a tourist destination from the tourism product delivery and the outcome of the service, such as the food, accommodation and facilities. Product quality is used as an element to measure total quality.
The physical environment is a component that leads to customers' impressions. The store environment is a combination of tangible aspects (fixtures, decorations, colours, lighting, buildings, layout and design) and intangible aspects (music, scent and temperature) (Migdadi & Abdel-Rahman, 2020). Ryu et al. (2012) concluded that customers' perceptions of quality (product, service and physical environment) at stores can predict how they perceive value and the degree of satisfaction and thus, their behaviour intention.Customers may interpret whether a product purchase is good quality based on the actual product quality, physical environment, and service encounter (Slack et al., 2020).Also, previous studies in the service sectors have used the combination of interaction quality, physical environment and outcome quality to measure total quality (Namkung & Jang, 2008;Ryu & Han, 2010;Ryu et al., 2012). Therefore, this study aimed to test the components of perceived quality: souvenir quality, service encounter quality and physical quality. For this study, souvenir quality was identified by the variables of traditional materials, product design and product materials. Service encounter was measured by employee knowledge, willingness to serve and service quality. The physical environment consists of store cleanliness, rest areas and parking space. The following hypotheses were proposed: H1. Perceived quality contains three first-order components including: (H1a) Authentic souvenir quality, (H1b): Service encounter quality, (H1c): Physical quality aspect.

The relationships among perceived quality of authentic souvenir product, perceived value, customer satisfaction, repurchasing intention, and word-of-mouth
Creating loyal customers is a critical strategic foundation for retail shops. It is necessary for retailers to look for strategies to satisfy existing customers, which may include designing products to be more desirable and less expensive than competitors (Loureiro & González, 2008). Babakus et al. (2004) explained the concept of product quality as contributing to increased profit and market share. Customers tend to compare a product's quality to what competitors might offer and then choose the best offer. Improving product quality, such as in a product design or service process, would help attract new customers, retain existing customers and eliminate competitors who offer lower quality products. When customers perceive high-quality products, they are more likely to be satisfied and thus willing to pay more (up to a point). Therefore, the customer assessment of quality has an impact on a retail store's revenue and market growth. Ryu et al. (2012) also mentioned that customers' perceptions of product quality significantly affect a store's image. The store image also has an influence on the customers' perceptions of product value and satisfaction. The more that customers combine satisfaction with the perceptions of high values of products, the more they become loyal customers. Chen and Chen (2010) stated that the three variables of quality, perceived value and satisfaction are significant factors that predict customers' purchasing intention.
Perception is an acceptable structure regarding consumer behaviour. This structure can convey consumer behaviour, such as in the classical work on service quality by Zeithaml (1988). The value perception depends on the specific context (e.g., heritage tourism, community-based homestay visit, etc.). The marketing literature has found that inadequate products and the limits of purchased products (souvenir purchasing) are relevant to perceived value, which can become progressively higher. However, the role of purchasing souvenir behaviour has received less attention than the consumption of travel and tourism in the literature. Studies have mentioned the element of rational conception and multidimensions. There is a perception of value through the tourism context, hospitality and retailing sectors. When used in analysis, works have indicated four perceptions of value: functional value, emotional value, social value and the value of money. The value of souvenirs and spiritual value in particular have been revealed as two dimensions of the perceived value of souvenirs.
Perceived value is a concept in which customers compare the utility of products between "what is received" and "what is given" (Zeithaml, 1988). Pham et al. (2018) have stated that perceived value is actually a better independent factor than quality or satisfaction in predicting repurchasing intention. These researchers also explained the process that leads to customer repurchase intention in the context of online shopping. Customers' perceptions of product value may occur at different stages, including the pre-purchase, purchase and post-purchase stages. The more customers perceive a high value of products, the more they are satisfied and become loyal customers. The previous research also found that customer loyalty is the result of perceived value and the experience of service quality provided by retail stores (Kusumawati & Rahayu, 2020). Thus, customers' perceptions of product value as well as satisfaction influence repurchasing intention. Behavioural intention is relevant to the desired actions of consumers. Thereby, behavioural intention shows the consumer's expectation along with the setting and loyalty.
Retailers seek to increase the number of repeat customers by generating a sense of loyalty. The concept of customer loyalty can be measured by customers' attitudinal and behavioural intentions. These intentions include free advertisement to friends through word-of-mouth and repurchasing products . Previous studies have shown a significant relationship between quality, perceived value, satisfaction and loyalty within retail stores (Babakus et al., 2004), tourist destinations (Jeong & Kim, 2019;Suhartanto et al., 2020), online shopping (Pham et al., 2018) and airline service (Chen, 2008). This suggests that quality, perceived value and satisfaction are good factors predicting customer loyalty. Also, past studies showed that quality significantly influenced customers' perceived value (Chen et al., 2020;De Leon et al., 2020) and satisfaction (Loureiro & González, 2008). Moreover, perceived value and satisfaction were significant contributors to predicting customer loyalty (Pham et al., 2018). Loyal customers tend to express themselves by revisiting and spreading positive WOM (Asghar Ali et al., 2021). To have a better understanding of customers' intentions, the researchers aimed to test whether perceived value and satisfaction could mediate the relationship between customers' perceived quality and loyalty (repurchasing souvenir products and recommending souvenir products to others). Therefore, the proposed conceptual model (Figure 1) with hypotheses were suggested as follows: H2. Customers' perceived quality of authentic souvenir product has a positive effect on perceived value of souvenirs.

H3.
Customers' perceived quality of authentic souvenir product has a positive effect on satisfaction with souvenirs.

H4.
Customers' perceived quality of authentic souvenir product has a positive effect on repurchasing intention.

H5.
Customers' perceived quality of authentic souvenir product has a positive effect on word-ofmouth.
H6. Customers' perceived value of authentic souvenir product has a positive effect on customer satisfaction.
H7. Customers' perceived value of authentic souvenir product has a positive effect on repurchasing intention.
H8. Customers' perceived value of authentic souvenir product has a positive effect on word-ofmouth.
H9. Customer satisfaction of souvenirs has a positive effect on repurchasing intention. Although the direct relationship between perceived value and customer loyalty (e.g., repurchase intention and word of mouth) has been tested (Carlson et al., 2019;, perceived value has previously been found to mediate the effect of perceived quality on a range of customer loyalty and behavioral intention in the context of hospitality and tourism (e.g., Chen et al., 2020;Fu et al., 2018;Jeong & Kim, 2019). For example, the study of Meeprom and Silanoi (2020) indicated that there was an indirect correlation between the perceived quality and behavioural intention through via perceived value. Specifically, when consumers perceive a higher level of service quality offering, they may perceive more value such as monetary value and enhancing relational benefits which result in increasing customer loyalty to a certain service or product. The following hypothesis are drawn: H11a. Customers' perceived value mediates the relationship between customers' perceived quality and repurchasing intention.
H11b. Customers' perceived value mediates the relationship between customers' perceived quality and word-of-mouth.
Moreover, previous research found that perceived quality of a service or product influences customer loyalty through customer satisfaction, which means that customer's feeling of gratification when the post-consumption experience exceeds prior expectations, will improve the level of customer's willing to recommend to others and the likelihood of customer to repeat visit/purchase to the certain products or services. The importance of the customer satisfaction as a construct which mediates the relationship between both revisit intentions and positive word-of-mouth referrals and its determinants such as perceived quality has been supported by prior hospitality and tourism research (Choi & Kim, 2013;Keshavarz & Jamshidi, 2018). In line with the model and the related literature review, the following hypothesis is drawn: H12a. Customer satisfaction mediates the relationship between customers' perceived quality and repurchasing intention.
H12b. Customer satisfaction mediates the relationship between customers' perceived quality and word-of-mouth. Furthermore, perceived value plays an important role in building corporate strategies. When a consumer assesses the perception regarding a service or product what they receive compared with what they give, value contributes to higher customers-services or products long-term viability or sustainability (e.g., loyalty and retention). In service marketing literature, scholars found the indirect influence of the perceived value on customer loyalty through satisfaction Keshavarz & Jamshidi, 2018). Moreover, it was confirmed that customer satisfaction plays a mediating role between the perceived value and customer loyalty (e.g., repurchasing intention and word-of-mouth.) in various contexts such as hotel (Keshavarz & Jamshidi, 2018), tourism destination , and retail (El-Adly & Eid, 2016), the following hypothesis is drawn: H13a. Customer satisfaction mediates the relationship between customers' perceived value and repurchasing intention.
H13b. Customer satisfaction mediates the relationship between customers' perceived value and word-of-mouth.

Sampling and data collection procedures
To test the conceptual model (Figure 1), this study would ideally collect data from domestic customers. However, as the number of international customers in the Na Kha Silk Village of Udonthani, located in the northeast part of Thailand, is comparatively small, this study gathered a sample of non-local Thai customers (i.e., those who live outside the area of the northeast part of Thailand). Na Kha Silk Village is a popular silk market that is essentially composed of stores selling silk and cotton clothes. A purposive sampling via self-administrative procedure was employed to identify domestic customers above the age of 18 years old living outside the area of the northeast part of Thailand. At the beginning of the survey, the customers were approached in key hotspots of souvenir shop areas, including the shops, restaurants, and rest area. If they had purchased the products from the souvenir shops, they were asked to participate in the research. If they agreed to answer the questionnaire, they were asked to fill out a paper-and-pencil questionnaire under the guidance of the fieldwork team. After screening the 250 returned surveys, 70 surveys were excluded from analysis due to a high number of incomplete answers. At the conclusion of the data analysis, 180 useable surveys were tested, representing a 72 per cent response rate, which is more than the 50-90 per cent general response rates reported in previous studies (e.g., Bryce et al., 2015;Meeprom & Silanoi, 2020). Thus, non-response bias was not considered an issue in this study.
In total, 180 useable surveys were completed, of which 66.0% were made by female respondents (34.0% male). The age of the respondents varied: 26 to 35 years old (39.0%), followed by 18 to 25 years old (26.2%) and 36 to 45 years old (14.2%). There was a high level of respondents with educations of a bachelor's degree or above (52.4%); 37.6% had less than a bachelor's degree, such as a high school degree. Most of the respondents had a mean household income of $4001 to $8000 USD, with 13% earning $9000-12,000 USD or more per annum.

Instrument design and measures
All the modifications of the items ensured a match with the authentic souvenir product context. Three reflective items were adapted to measure each of the perceived product quality aspects. The measurement scales with construct reliability (CR) were adapted from previous studies as followed: authentic souvenir quality indicators (CR = 0.91) were adapted from Fu et al. (2018) andCarlson (2012); service encounter quality (CR = 0.74) and physical quality aspect (CR = 0.84) and were drawn from Wu and Cheng (2013) and Qin et al. (2010). The measures for perceived value (CR = 0.91) were adapted from Carlson et al. (2019) and Meeprom and Silanoi (2020). Customer satisfaction (CR = 0.78) was adapted from Suhartanto (2018); word-of-mouth (CR = 0.93) was adapted from Fu et al. (2018), and repurchase intention (CR = 0.92) was adapted from Lin and Wang (2012). A five-point Likert scale (with 1 = "strongly disagree" and 5 = "strongly agree") served to measure all the items. All the measurement items were sourced from the existing literature in English; as the sample in this research were domestic customers, a forward-andback translation procedure was required to ensure that there was no loss in meaning after the translation process. the survey was developed in the original English version, forwarded into Thai and using another translator, then translated back Thai to English. This approach can minimise misunderstanding caused by the differences of expression between languages (Meeprom & Dansiri, 2021).
To test common method variance, following Lim et al. (2017), the fit of the model in which the indicators loaded on one factor was examined. If it was largely responsible for the covariation among the measures, this would be an indication of possible common method bias. Based on this approach, a single factor using confirmatory factor analysis (CFA) was assessed to fit the data to see if the method variance was largely responsible for the covariation among the measures. The results of the CFA indicated that a one-factor model exhibited poor fit. Therefore, we included a five-factor measurement model that showed an acceptable fit, χ 2 (147) = 172.37, CFI = .95, TLI = .93, RMSEA = .04. Thus, these findings suggest that a common method variance was not a problem in our sample.

Reflective measurement models
To demonstrate the construct reliability, this study adopted a two-step analytical approach for the assessment of the measurement model, as suggested by Limayem and Cheung (2008) and Ngo and O'Cass (2009). First, the psychometric properties of the measurement model were assessed and validated. Further, the structural model was assessed using partial least squares. As suggested by Ngo and O'Cass (2009), the adequacy and significance of the reflective measurement models for the other constructs were calculated through an investigation of the range of indices that included the individual item weights and factor loadings, composite reliability, average variance explained (AVE), bootstrap t-values (critical ratio statistic), convergent validity and discriminant validity.
As shown in Table 1, the findings of the assessment of the convergent validity confirmed that the reflective measurement models and their first-order components exceeded the requirement of Nunnally (1978) criteria of reliability (> .70). The composite reliabilities of all the constructs ranged from .74 to .81 (see Table 1). Therefore, the constructs related to the reflective measurement models presented satisfactory convergent validity. Similarly, the average variance extracted (AVE), which captured the portion of the reflective indicator block's variance that the related constructs of interest could explain (Hair et al., 2017), exceeded the common threshold of .50 for all the reflective measurement models. Therefore, the AVE demonstrated an acceptable internal consistency of all the measurements (Hair et al., 2017).
To assess the discriminant validity, this study applied the Fornell-Larcker criterion ratio of correlations (Fornell & Larcker, 1981;Henseler et al., 2015). With the Fornell-Larcker criterion, the square roots of all AVEs were higher than the correlations between the corresponding constructs, supporting the existence of discriminant validity. Therefore, the discriminant validity was confirmed with the Fornell-Larcker results, see Table 2 (Jarvis et al., 2003). As such, the formative type of the hierarchical latent variable perceived product quality was substantiated through the confirmation of the discriminant validity of a first-order component (Becker et al., 2012).

Formative measurement model
Since this study examined a reflective measurement model for each of three components, which then formed the perceived product quality variable (i.e., a Type-II configuration), this study investigated the weights of first-order dimensions on the designated second-order construct (see Figure 1 and Table 3). All three first-order dimensions had significant positive relationships with perceived service quality, as hypothesised. Authentic souvenir quality had the strongest positive beta weight (β = .47; t = 14.77; p < .001) followed by the physical quality aspect and service encounter quality. In addition, this study formulated collinearity on the index by checking the variance inflation factor (VIF) values of all the predictor (exogenous) constructs because the study was anticipated to apply both formative and reflective variables. As presented in Table 3, the findings of the VIF demonstrated that the collinearity among the three dimensions of perceived product quality predictor variables was not a critical issue, as all VIF values fell below the common threshold of 5.0; they ranged from 1.11 to 1.18, which met the benchmark proposed by Hair et al. (2017). Taking this into account, the results confirmed perceived product quality was a second- order construct formatively formed by the three first-order constructs of service encounter quality, physical quality aspect and authentic souvenir quality.

Model assessment using PLS-SEM
Partial least squares-structural equation modelling (PLS-SEM) was used to test the proposed conceptual model (Figure 1) and for hypothesis testing. To avoid any possible confounding effects, gender, age and the previous experience of customers were used in the analysis as control variables. The results of the structural model are presented in Figure 2. First, the effect of perceived product quality on perceived value was tested. As shown in Table 3, the path coefficient between perceived product quality and perceived value was .54 (t = 10.97, p < .001), indicating that perceived product quality significantly influenced perceived value. Therefore, the evidence supports H2. Similarly, the path coefficient between perceived product quality and customer satisfaction was .46 (t = 6.28, p < .001), indicating that perceived product quality significantly and directly affected customer satisfaction. Thus, H3 was supported. With regard to H4 and H5, perceived product quality was positively related to repurchase intention and word-of-mouth. The findings presented in Table 3 for these two hypotheses support the predicted relationships for both H4 (β = .31, t = 3.00, p < .01) and H5 (β = .17, t = 2.00, p < .05).
Second, the effect of perceived value on customer satisfaction was explored. As presented in Table 3, the path coefficient was .27 (t = 3.47, p < .001), indicating that perceived value had a significant direct effect on customer satisfaction. Therefore, the evidence supports H6. Further, the effects of perceived value on repurchase intention and word-of-mouth were examined. The findings presented in Table 3 show that the path coefficients were .14 (t = 1.50, p > .05) and .23 (t = 2.89, p < .01), indicating that perceived value had a significant direct effect on word-of-mouth but not on repurchase intention. Therefore, H7 was not supported, but H8 was supported. Third, the effects of customer satisfaction on repurchase intention and word-of-mouth were investigated. The findings presented in Table 3 show that the path coefficients were .17 (t = 1.71, p > .05) and .35 (t = 4.04, p < .001), indicating that customer satisfaction had a significant direct effect on word-of-mouth but not on repurchase intention. Therefore, H9 was not supported, but H10 was supported.
To test the mediating effects, we followed the steps in Chin (1998) and Zhao et al. (2010). The results of the multiple mediating effects of perceived value and customer satisfaction on the impact of perceived product quality on repurchase intention and word-of-mouth are displayed in Table 4.
Regarding the results of the perceived value intervening between perceived product quality and its outcomes (e.g., repurchase intention and word-of-mouth), all the measures of the indirect effect of perceived product quality on repurchase intention and word-of-mouth demonstrated a 95% confidence interval that did not include zero; that is to say, the impact was significant. This finding indicated that perceived value and customer satisfaction with an authentic souvenir product did mediate the influence of perceived product quality on repurchase intention and wordof-mouth components. Therefore, the evidence supports H11 (a-b) and H12 (a-b) . Further, the results also presented that there was a significant relationship between the perceived value and both repurchase intention, and word-of-mouth in the direct model. However, these paths were significant, but there were a reduce in the regression weight for the direct relationship between perceived value and its outcomes. Therefore, customer satisfaction mediated the impact of perceived value on both repurchase intention, and word-of-mouth. Therefore, H13 (a-b) was supported.
As such, the R 2 value of the endogenous variables was used to compute the model fit Chin (1998). The R 2 of perceived value, customer satisfaction, repurchase intention and word-of-mouth were .28, .39, .27, and .37, respectively, indicating that the R 2 values were acceptable (Chin, 1998). The predictive relevance of the structural model was also assessed via the Stone-Geisser Q 2 criterion (see Figure 2). The Q 2 values of perceived value, customer satisfaction, repurchase intention and word-of-mouth were .12, .17, .11, and .22, respectively, indicating that Q 2 was greater than zero, supporting the predictive relevance of the research model (Hair et al., 2017). The model-fitting test showed that the standardized root mean square residual (SRMR) was well below the .10 threshold, SRMR = .07 (Hu & Bentler, 1999). Therefore, the results indicated that the formation of perceived product quality by the three components had a positive direct effect on perceived value, customer satisfaction, repurchase intention and word-of-mouth. Similarly, perceived value was found to have significant positive effects on customer satisfaction and word-of-  Note 2: this study followed the steps in Chin (1998) and Zhao et al. (2010) to test mediation approach. * = meets or exceeds criterion of t > 1.96, p < 0.05 (2-tailed), SE: Standard Errors in brackets.

Conclusion
After testing the conceptual model, the researchers found that 11 out of 13 hypotheses were supported. Customers' perceptions of product quality at Naka market significantly influenced how they perceive value and their satisfaction with Nakha products. To identify the perceptions of product quality, customers considered the following three aspects: the authentic souvenir, service encounter, and physical quality. Not only does the quality of products lead to customer satisfaction, but also customers' perceived value of products influences their satisfaction. Customers satisfied with Nakha products are more likely to recommend and encourage other people to visit the Nakha market. Plus, the findings show that customers' perceptions of product quality and valued products positively affect their WOM intention. However, customers only considered product quality as a factor to make a decision whether they would like to revisit the market or not. Furthermore, the results of this study illustrate that customers' perceived value and customer satisfaction mediated the relationship between customers' perceived quality and loyalty (repurchasing souvenir products and recommending souvenir products to others).

Theoretical implications
This study expands the customer perceived product quality concept in the context of authentic souvenir retail in the Nakha market. By conceptualizing and testing the second-order component of perceived product quality, it advances the customer-perceived product quality literature (Babakus et al., 2004;Das, 2014;San-Martín et al., 2017). The study demonstrates that customer perceived product quality consists of three components: authentic souvenir quality, souvenir encounter quality and physical quality. These components represent customers' past experiences of visiting the souvenir retail shop at the Nakha market, which contributes to the overall customer perceived product quality measurement. In addition, examining customer perceived product quality as a reflective second-order construct helps advance a higher level of theoretical *p < 0.05 (t = 1.960); **p < 0.01 (t = 2.576); ***p < 0.001 (t = 3.291) (2-tailed).

Path Coefficient
Step 1 Step 2 Step 3 Step Yes expansion. The collective effects (i.e., first-order effects) of perceived product quality (i.e., second-order effects) were used to evaluate customer satisfaction, repurchase intention, and word-of-mouth intention. Thus, developing a perceived value index that contains the firstorder components should be used as a perceived product quality measurement for the souvenir retail context.
Additionally, by investigating perceived product quality and customer satisfaction as mediators, it provides a meaningful model. This method has been used for theory development, which is beyond the use of a simple regression technique that tests the effect of independent variables on dependent variables. This study adds to the literature about how perceived product quality affects customers' word-of-mouth by assessing perceived value and customer satisfaction as mediators. By doing this, the contribution of the model provides a precise picture and knowledge of the relationships among perceived product quality, customer satisfaction and word-ofmouth intention. The focus on post-experience intentions (repurchasing and word-of-mouth intentions) helps explain customer behaviour and its effects on retail shops, such as increased profit and customer loyalty (Das, 2014;Kang, 2014). The results of this study also supported previous studies, which showed that the customer perceived quality of products led to customer satisfaction (De Leon et al., 2020), perceived value (Sweeney et al., 1999), positive word-ofmouth (Sivadas & Baker-Prewit, 2000) and repurchase intention (Das, 2014). Also, using perceived value as a mediator strengthened the effect of perceived product quality on word-ofmouth more than the direct effect of perceived product quality on word-of-mouth (Sivadas & Baker-Prewit, 2000).

Practical implications
The findings show direct connections between the early stages of the shopping experience and the likelihood of relaying a positive experience through word-of-mouth. The pre-purchase experience included perceptions of authentic souvenir quality, service encounter quality and place or physical quality. A previous study showed that customers tend to consider the "quality of product" as the most important factor in souvenir shopping (Amaro et al., 2020). These perceptions contribute to an overall perceived product quality mediated by perceived value and satisfaction. With this knowledge, souvenir providers would benefit by emphasizing the authenticity of their products. This may be accomplished in several ways, such as playing a video in their stores showing hand looms in action or even having live demonstrations. Providing a show room is also a great strategy to promote locally produced souvenirs that customers can personally observe being produced (Sthapit & Björk, 2017). Not only would it create product value, but it would also provide product knowledge for visitors. Souvenir providers should also consider their quality of product, including not only the initiation of pleasant encounters with customers but also by exhibiting a thorough knowledge of the products. This is especially true for products that emphasize authenticity because customers are likely to have higher expectations of the sellers' degree of knowledge. Of course, the physical setting also contributes to the perceived product quality. This may include cleanliness and available amenities, such as parking and rest areas. By addressing these variables within the shopping experience, souvenir providers can enhance customers' perceptions of product quality and thus increase the likelihood of positive post-purchase behaviours.

Limitations and future research
As with any study, there will be some limitations. For the present study, the researchers were unable to acquire as many surveys as predicted due to the effects of COVD-19 on tourist numbers in Thailand. Another effect of COVID-19 on tourism in Thailand was the tight restrictions on international travel. This severely limited the diversity usually present in the Nakha market. Plus, contactless and social distancing considerably impacted the response rate of this study. Thus, the present sample was not as representative as it might have been under normal circumstances. Another limitation might be that by nature, quantitative studies are less equipped to elicit a deeper understanding of personal experiences from shoppers and sellers. This information would be useful in applying concepts to particular locations, such as the Nakha silk market. Furthermore, the results might have some bias due to only collecting the data from a single market. Therefore, future studies should extend the collection of data to other local authentic markets and compare the results of customers' attitudes and behavior. Additionally, in-depth interviews from both customers' and sellers' perspectives would help inform market organizers to improve customer service experience.