Antecedents of Customer Loyalty and Perceived Service Quality: A SEM Analysis of Thai Restaurant Brands

Abstract

With over 15,000 Thai restaurants worldwide and 5,342 in the USA alone, Thai cuisine has been repeatedly internationally acclaimed for its taste, flavors, smell, and nutritional values. The Thai Ministry of Commerce has also established a global objective to find and award the best Thai food restaurants using Thai Select Premium (TSP), Thai Select Unique (TSU), and Thai Select brand labels. Therefore, using systematic random sampling, 620 diners were selected from certified Thai Select restaurants (TSR) in either Los Angeles, New York, Miami, or Chicago to ask for their opinions concerning factors important to each restaurant's service loyalty (SL). The four constructs, 18 observed variables, and six hypotheses were analyzed using LISREL 9.1. Results revealed a significant positive effect (88% R²) on the causal factor interrelationships on TSRs' SL. Additionally, the three factors affecting SL the most were service tangibility, taste perceptions, and brand image, with total effect values of 0.94, 0.43, and 0.31, respectively. Finally, when each US diner was asked why they chose to eat Thai food, respondents overwhelmingly indicated that taste (49.50%) and healthiness (32.90%) were most important. The study also adds to the literature by highlighting the importance of Thai cuisine contributing to healthier lifestyles.

1. Introduction

Thai cuisine has been defined as an amalgamation of the senses, from taste to smell. Thai dishes are delicious, refined, beautiful, and meticulously prepared. In addition to each dishes’ sensual enjoyment comes an added factor of nutritional value hidden beneath a wide range of savory flavors and beautiful oriental presentations.

Healthy foods have been stated to energize and nourish one’s mind, body, and soul (Thaik, 2013). Some researchers have even pointed out that certain Thai dishes can even influence the anti-aging process (Khanthapok & Suchada, 2019) due to their antioxidant (Tangkanakul & Trakoontivakorn, 2014) and anti-inflammatory (Sirikanokvilai et al., 2014), and antimicrobial effects. This agrees with other research suggesting that changing consumption behaviors by consuming more wholesome diets helps individuals achieve longer and healthier lives (Chang, 2017; Waratornpaibul, 2014).

Given the above factors, it most probably comes as no surprise that Thai cuisine has become a national treasure and is constantly finding a place within the top 10 ranked dishes worldwide (Sornsaruht, 2020). This is supported by an international survey from 2017 in which 35,000 travelers voted four Thai dishes into the top 10 list worldwide (Cheung, 2017). This is not surprising when one knows that research now indicates that there are over 15,000 Thai restaurants globally, with 5, 342 restaurants having been identified as ‘Thai’ within the United States (Karp, 2018). Finally, as of June 2022, in the US, there were 5, 342 Thai restaurants, of which 456 had been certified by Thailand’s Ministry of Commerce (MOC) as ‘Thai Select’ to denote authenticity and quality (Thai Emerald Restaurant, 2022).

Unfortunately, Thai restaurants do not always serve ‘Thai food, ’ with the government of Thailand noting this and taking steps to preserve and confirm what they consider to be a national treasure. With this goal in mind, Thailand’s MOC laid out a four-year plan which commenced in 2020, to identify and preserve restaurants that serve authentic Thai cuisine both in Thailand and worldwide (Boyle, 2018; Sornsaruht, 2020).

This plan called for the selection and certification of three different categories of ‘Thai Select’ cuisine, including Thai Select Premium, Thai Select Unique (TSU), and Thai Select (Thai Select Brandings To Spread, 2018). Moreover, each level is evaluated using six categories: taste, ingredients, sanitation, menu, restaurant decoration/atmosphere, and services.

As we can see, the Thai Select cuisine brand certification entails a multiplicity of factors, including taste which is ranked highest in award points towards certification. Namkung and Jang (2007) confirmed the importance of taste when they noted the significance of taste and presentation in customer satisfaction (CS) and behavioral intentions.

Therefore, from this overview, the authors expanded this information to identify and describe which factors could potentially play a role in a TSR’s service loyalty in Section 2’s literature review. In Section 3, the methods and materials for the study are outlined. In Section 4, the results are presented, while Section 5’s discussion expands on the results. In conclusion, interesting aspects of the study are highlighted. We hope the readers enjoy the fruits of our labor, and it whets their appetite for some unique and healthy Thai cuisine!

2. Literature Review

2.1. Service Tangibility (ST)

According to Were et al. (2020), a hotel’s beverage and food are tangible assets, with each customer’s perceptions derived from the tangibility of the food service quality (SQ). Nguyen et al. (2018) also added that tangibles within the UK fast-food sector, including assurance and responsiveness, play a crucial part in customer satisfaction.

2.2. Taste Perceptions (TP)

Much has been written about the connection between human taste, likes, and personality. A straight forward example is that people who eat chocolate ice cream are different from those who like vanilla ice cream (Atherley, 2021). However, the reality is that researchers have developed connections between personality characteristics and an individual’s desire for basic tastes such as sweet and sour, bitter and salty, and the seldom familiar taste of umami (translated from Japanese as “savory deliciousness”) (Cecchini et al., 2019; Tracy, 2018). Even the English language links taste to personality with expressions such as “he is a bitter old man” or “she is such a sweet little girl” (Spence, 2022).

So what tastes do Thai foods have that make them so appealing to so many worldwide? First, as the TSU certification suggests, Thai cuisine is ‘unique’ as recipes have been handed down from generations of cooks. Some might even say Thai food is a culinary art that creates a unique national identity. Thailand’s ‘street food’ has also become world-renowned due to its wide array of exotic tastes and smells, making Thai food one of the most sought after international cuisines (Sornsaruht & Sawmong, 2017).

Numerous studies have also pointed out the importance of regional dishes as they represent a unique element in a people’s culture and heritage (Berno et al., 2019). In Thailand’s case, Thai cuisine has become a strong national brand identity and an integral part of the Kingdom’s tourism promotion (King, 2009; Sornsaruht, 2020). Similarly, Sukalakamala and Boyce (2007) have stated that authentic ethnic food experiences are an important driving force in diners choosing ethnic restaurants. Wong and Baldwin (2018) have also pointed out that taste and sight play positive roles in emotional responses and perceived authenticity to food choices. A stroll can quickly validate these ideas through Bangkok’s street food stalls, where visitors are confronted with a seemingly endless selection of flavors and fragrances (Sornsaruht & Sawmong, 2017).

2.3. Brand Image (BI)

Sanglimsuwan and Songwathana (2021) found that the perceived food quality and brand image positively affected consumers’ decision to visit Thai cuisine restaurants. In another Thai study on Thai food, it was determined that a restaurant’s guest loyalty and value recognition were made up of three factors: their products, the image, and the customer’s psychology (Sornsaruht & Sawmong, 2017). In Malaysia, Zahari et al. (2010) detailed the importance of restaurant brand image, indicating BI is critical to survival, growth, and maintaining and increasing patron loyalty and trust.

2.4. Service Loyalty (SL)

Skogland and Siguaw (2004) researched customer satisfaction (CS), service quality (SQ), and hospitality user loyalty in the Thai hospitality sector using a SERVQUAL scale (Parasuraman et al., 1985, 1988). The findings revealed that the SERVQUAL dimensions assurance, tangibles, and reliability, played the most significant role in SQ and were related to loyalty.

Furthermore, Andaleeb and Conway (2006) added that restaurant diner satisfaction primarily depends on the service staff, the menu pricing, and the food quality. This is consistent with Armstrong and Kotler (2013) adding ‘price’ to the ‘4 Ps’ marketing mix model. Al-Tit (2015) showed that food taste and SQ positively affected diner satisfaction in Jordan’s small restaurants. Similarly, in Korea, Ahn (2015) showed that small, fast food-food restaurant diners viewed food quality as most important in their satisfaction. Interestingly, Espinosa et al. (2018) determined that customer brand loyalty is more crucial than satisfaction for restaurant revisit intentions. Finally, customer loyalty is essential, as various reports have determined that loyalty is the desired outcome of marketing (Phadungjit et al., 2020).

Therefore, the researchers found significant confirmation for the potential contribution of three constructs in a TSR’s SL. These included ST, TP, and BI and their 18 related observed variables outlined in Table 3. Furthermore, from the theory and literature, the authors propose the following six hypotheses for further evaluation and analysis:

H1: Service Tangibility (ST) directly influences Brand Image (BI).

H2: Service Tangibility (ST) directly influences Service Loyalty (SL).

H3: Service Tangibility (ST) directly influences Taste Perceptions (TP).

H4: Taste Perceptions (TP) directly influence Brand Image (BI).

H5: Taste Perceptions (TP) directly influence Service Loyalty (SL).

H6: Brand Image (BI) directly influences Service Loyalty (SL).

2.5. Research Objectives

1. To investigate the SEM’s interrelationships and their influences on a Thai Select customer’s service loyalty (SL).

2. To conduct goodness of fit (GOF) and confirmatory factor analysis (CFA) to confirm the model’s fit before the SEM.

3. The research results hope the authors can widen Thai cuisine’s knowledge, appeal, and nutritional aspects worldwide.

3. Methods

The authors’ University Human Ethics Committee was consulted and approved the research design before the study’s undertaking. After the questionnaire design, the experts who assisted with item validation were informed about their input and personal information privacy. The pretest sample group was also given an informed consent form acknowledging their privacy. Finally, all respondents in the final sample group were also informed of their privacy before their questionnaire item response, with each participant’s anonymity considered and ensured.

3.1. Population and Sample Size

The authors identified TSR diners in four major US cities as the population for the study. These cities included Chicago, Los Angeles, Miami, and New York City, which at the time of the survey included 812 restaurants. Sample size determination used the commonly accepted ratio method of observed variables to the respondent, as Schumacker and Lomax (2016) suggested. This is consistent with Hair et al. (2018), who added that sample size depends on model complexity and variables, but 200–400 respondents are usually sufficient.

3.2. Research Instruments

The instrument used for information collection was a questionnaire that contained five parts. Part 1 contained items concerning each individual’s personal and dining characteristics. Part 2 through Part 5 involved each diner’s opinions concerning each aspect contributing to service loyalty. Measurement and analysis used a five-level diner opinion scale in which 1.00–1.49 represented a minor agreement, 1.50–2.49 represented a small agreement, 2.50–3.49 represented moderate agreement, and 3.50–4.49 strong finally, 4.50–5.00 represented total agreement.

Part 2 contained nine items related to service tangibility (ST), whose reliability outcome was assessed as 0.94. Part 3 contained three items related to taste perceptions (TP), with a reliability outcome of 0.92. Part 4 contained three items related to the brand image (BI), with a reliability outcome of 0.86. Part 5 contained three items related to service loyalty (SL), with a reliability outcome of 0.83.

3.3. Research Survey Quality Evaluation

After the questionnaire’s design, a content validity (CV) evaluation was undertaken. Kelly (1999) has indicated that this is necessary to ascertain the study’s design strength and the accuracy of the selected variables and their measurement. Also, the CV questionnaire assessment process was conducted with the cooperation of five experts. Each item was then evaluated using the commonly accepted criteria from the indexes of item-objective congruence (IOC) rating (Turner & Carlson, 2003), from which IOC values were determined to range from 0.67 to 1.00. According to the experts ‘suggestions, an IOC value of 0.67 was advised. Therefore, values less than 0.67 were removed or re-written (Pimdee, 2020).

After that, questionnaire reliability was assessed using a pre-test group of 35 Thai Select diners in the US, with reliability assessment made using Cronbach’s alpha (α) (Nunnally, 1978), with α ≥ 0.80 being used as a strong indicator of the questionnaire’s reliability (Lance et al., 2006).

3.4. Data Collection

Table 1 details the TSR diner data collection process in the US from the four major metropolitan areas of Los Angeles, New York, Miami, and Chicago. The researchers collected data using an online questionnaire (Google form) from 81 TSRs and their 620 patrons randomly selected from every fifth customer between June 2019 and July 2019. This was made possible with the cooperation of selected restaurant staff members who were provided a laptop with a connection to the restaurant’s Wi-Fi Internet connection. Every fifth customer was asked if they would mind participating in an academic survey concerning the restaurant and Thai food. The reward for doing so would be a complimentary dessert of their choice. Participation using this method was nearly 100%.

Table 1: Thai Select Restaurant (TSR) Diner Sampling Process

3.5. Data Analysis

LISREL 9.10 analysis was done on the study’s data, four latent variables, and 18 observed variables.

4. Results

4.1. Participants’ Characteristics

Table 2 shows that questionnaire respondents were 57.10% women, with 46.90% indicating they were 21–30 years old. Another 59.70% of both genders indicated they were single. It also appears that TSR patrons are well educated, as 65.10% had obtained a university degree. Thai Select cuisine also attracts entrepreneurs as 44.20% indicated their private sector employment, with 49% marking income levels of $2, 201 monthly incomes or higher per month. Most importantly, when each TSR diner was asked what they liked most about their menu choices, they indicated they preferred taste number 1 (49.50%), followed by their perception of Thai food’s healthiness (32.90%).

Table 2: Respondents’ Characteristics and Preferences for Thai Food (n = 620)

Table 2: (Continued)

4.2. CFA Results

The results of the study’s goodness-of-fit appraisal were based on the recommended values from various studies, from which the analysis met or exceeded all recommended criteria. Therefore, the GoF used items suggested by Hooper et al. (2008), in which convergent validity (CV) testing is recommended to use the GFI (≥ 0.90), CFI (≥ 0.90), RMSEA (≤ 0.05), and the chi-square/df statistic (≤ 2.00). Schumacker and Lomax (2016) also suggested that GFI, AGFI, NFI, and CFI values ≥ 0.90. Also, Tabachnick and Fidell (2007) have also recommended that the values for Chi-square (χ²) are p ≥ 0.05 and for relative Chi-square (χ²/df) ≤ 2.00. Hu and Bentler (1999) have also suggested that values for RMSEA, RMR, and SRMR ≤ 0.05. Therefore, from these recommendations the model’s fit was shown to be excellent as Chi-square = 0.71, RMSEA = 0.00, NFI = 0.99, CFI = 1.00, RMR = 0.01, SRMR = 0.01, GFI = 0.99, AGFI = 0.97, and Cronbach’s Alpha = 0.83–0.94.

4.3. Construct Reliability and Validity

Hooper et al. (2008) have reported that R2 values should not be ≤ 0.20, while factor loadings should be ≥ 0.5, and construct reliability (CR) should also be ≥ 0.80 (Table 3) (Netemeyer et al., 2003). Analysis CR values from the analysis were 0.84– 0.94, exceeding this high criterion. After that, it is suggested that the model’s fit validity should also be tested using an AVE ≥ 0.5, which was also met as AVE = 0.64 to 0.79.

Table 3: CFA Construct Reliability and Validity

4.4. Mediation Effects

The model’s causal variables positively affected SL, which has an R² of 88% on SL when combined. Additionally, the three factors affecting SL hypothesized to affect SL the most were service tangibility (ST), taste perceptions (TP), and brand image (BI), with total effect values of 0.94, 0.43, and 0.31, respectively (Table 4).

Table 4: Standard Coefficients of Influence in the Causal Relationship Model of Service Loyalty of TSRs in the United States

* p ≤ 0.05, ** p ≤ 0.01.

4.5. Correlation Testing Results

The analysis results of r for each variable pair showed significant strength between all pair relationships, with TP to ST and BI to TP being the strongest (0.83). Moreover, Kim (2015) has suggested that data normality can be tested by considering the p-value skewness and kurtosis. The study’s results from the descriptive analysis of each latent variable using the kurtosis, skewness, mean, and standard deviation (SD) revealed that BI was strongest (mean = 3.85, SD = 0.98), followed by TP (mean = 3.84, SD = 0.94), then SL (mean = 3.75, SD = 0.91), and finally, ST (mean = 3.72, SD = 0.91). Skewness values for ST = –0.97, TP = –1.01, BI = –0.93, and SL = –0.83. Kurtosis values for ST = 0.35, TP = 0.69, BI = 0.22, and SL = 0.27. Significance for all factors was p ≤ 0.01. Curran et al. (1996) has also indicated that as values reach 2.0 for skewness or 7.0 for kurtosis, results become suspect.

4.6. Hypotheses Testing Results

The hypotheses testing results are shown in Table 5 and Figure 2, with all six conceptualized hypotheses determined to be consistent and supported. Moreover, based on recommendations from Pearson’s r of 0.10 to 0.29 as weak, 0.30 to 0.49 as moderate, and values from 0.50 to 1 as strong. Ranked from strongest to weakest, they were H₃ (r = 0.93), H₁ (r = 0.54), H₄ (r = 0.41), H₂ (r = 0.37) and finally, H₅ (r = 0.31) and H₆ (r = 0.31). Hair et al. (2018) added that CV could be judged appropriately when t-values ≥ 1.96. Sharma (1996) added that further validity could be ascertained when standardized factor loading is ≥ 0.60. From Figure 1 and Table 5, it is found that the causal relationship model development of service loyalty of Thai restaurants in the United States was accurate, with all six hypotheses being supported by the SEM analysis. Specifically, the following was determined:

H1: Service Tangibility (ST) directly influenced Brand Image (BI) with r = 0.54, t-test = 6.18, and p ≤ 0.01.

H2: Service Tangibility (ST) directly influenced Service Loyalty (SL) with r = 0.37, t-test = 3.17, and p ≤ 0.01.

H3: Service Tangibility (ST) directly influenced Taste Perceptions (TP) with r = 0.93, t-test = 23.17, and p ≤ 0.01. 

H4: Taste Perceptions (TP) directly influenced Brand Image (BI) with r = 0.41, t-test = 4.72, and p ≤ 0.01.

H5: Taste Perceptions (TP) directly influenced Service Loyalty (SL) with r = 0.31, t-test = 2.74, and p ≤ 0.01.

H6: Brand Image (BI) directly influenced Service Loyalty (SL) with r = 0.31, t-test = 3.05, and p ≤ 0.01.

Table 5: DSA for Service Tangibility (ST)

Note. 3.50–4.49, SA: strong agreement, SD: standard deviation

Figure 1: Final SEM Results for TSR SL

5. Discussion

Results revealed a positive effect (88% R²) between all the causal factors on Thai Select restaurant SL in the US. Additionally, the three factors affecting SL hypothesized to affect SL the most were service tangibility (ST), taste perceptions (TP), and brand image (BI), with TE values of 0.94, 0.43, and 0.31, respectively. Finally, when each diner was asked why they chose to eat Thai food, overwhelmingly taste (49.50%), and healthiness (32.90%) were noted as the most important reasons. Finally, from the use of the standard Pearson’s correlation coefficients strength interpretations and further analysis of other studies, the researchers interpreted the results of the study as follows:

5.1. Service Tangibility (ST) Results

The three hypotheses testing results for ST were positive and direct, with H₁’s relationship from ST to BI strong (r = 0. 0.54, t-test = 6.18, p ≤ 0.01). However, the interrelationship between TS to SL in H₂ was moderate (r = 0.37, t-test = 3.17, p ≤ 0.01). Finally, H₃’s interrelationship from TS to TP was judged to be the strongest of the six proposed hypotheses (r = 0.93, t-test = 23.17, p ≤ 0.01).

Additional support for the importance of service tangibility (ST) in Thai Select restaurant service loyalty (SL) can be found in the nine-item descriptive statistics analysis (DSA) details in which the Thai Select restaurant’s interior design was judged as appropriate for its image and price (x5) (mean = 3.93, SD = 1.09) (Table 5). Interestingly, the restaurant patrons judged the exterior design as most important next (x8) (mean = 3.93, SD = 1.09).

These results find support from Liu and Jang (2009), whose research determined that in highly competitive restaurant markets, innovative designs that stand out along with a unique dining environment contribute significantly to overall customer satisfaction. In Indonesia, Canny (2013) and Ryu et al. (2012) also determined the importance of a restaurant’s physical environment on a diner’s perception of their overall satisfaction.

5.2. Taste Perceptions (TP) Results

The two hypotheses testing results for TP were all positive and direct, with H₄’s relationship from TP to BI moderate (r = 0.41, t-test = 4.72, p ≤ 0.01). However, the interrelationship between TP to SL in H₅ was weak (r = 0.31, t-test = 2.74, p ≤ 0.01).

Additional support for the importance of taste perceptions (TP) in Thai Select restaurant service loyalty (SL) can be found in the three-item descriptive analysis details in which the Thai Select restaurant’s dishes, spices, and herbs were judged most important (y2) (mean = 3.93, SD = 1.09). This was closely followed by the patrons’ high opinions of the food’s overall delicious taste (y3) and blending of various ingredients (y1).

These results find support in a similar study concerning Chinese restaurants in the US by Liu and Jang (2009), in which the authors determined that taste, food presentation, and food temperature were essential to CS.

5.3. Brand Image (BI) Results

The single hypothesis testing results for BI were positive and direct, with H₆’s relationship from BI to SL weak (r = 0.31, t-test = 3.05, p ≤ 0.01). Additional support for the importance of brand image (BI) in Thai Select restaurant service loyalty (SL) can be found in the three-item descriptive analysis details in which the Thai Select restaurant’s reasonable prices were judged most important (y5) (mean = 3.89, SD = 1.12). This was closely followed by the patrons’ high opinions for the food’s interesting cuisine (y6) and being at the top of lists for the most delicious food in the world (y4).

These findings are interesting from a historical perspective when one knows about the considerable effort and costs the Thai government has made over the years in culinary diplomacy. Conceptualized as early as 2001 to increase Thai export and tourism, the Thai government created the Global Thai Restaurant Company, Ltd., to create 3, 000 Thai restaurants globally (Karp, 2018). Building on a previous Thai chef export program, the Thai MOC Export Promotion drew up prototypes for three varieties of ‘master restaurants’ that entrepreneurs could select as a prefabricated restaurant plan, from aesthetic to menu offerings to prices. Additionally, objets d’art were made available to enhance each restaurant’s interior and exterior design (see x5 and x8 of this study).

5.4. Service Loyalty (SL) Results

Support for the importance of Thai Select restaurant service loyalty (SL) can be found in the three-item descriptive analysis details in which the Thai Select restaurant’s staff’s focus on making their diners feel special was viewed as the most significant (y9) (mean = 3.81, SD = 1.05). Somewhat less important was the customer’s hesitancy to pay higher prices (y8) and the importance of how important it was to frequent the restaurant (y7).

The researchers interpreted this data as Thai Select restaurants are selected for their unique atmosphere for special occasions for important events such as anniversaries or birthdays. Price is an issue, so entrepreneurs must be aware that they must use a price window to remain competitive in the highly competitive markets in which the survey was undertaken. This is consistent with other restaurant studies that indicate the price is essential to revisit intention and customer satisfaction (Armstrong & Kotler, 2012). Furthermore, in a restaurant study from Vietnam, the authors also identified SQ’s significant importance to CL and that SQ was a precursor to customer loyalty (Dam & Dam, 2021).

Furthermore, numerous studies have reported the significance of developing and successful long-term maintenance of customer relationships (Baalbaki, 2012; Chow & Holden, 1997; Phadungjit et al., 2020; Seyanont, 2009).

Finally, the long-term support that various Thai government agencies have given to the development of Thai cuisine both domestically and internationally should be noted. Thai cuisine is more than just ‘food.’ Instead, it is a national treasure, with each Thai restaurant reflected in ‘Thainess’ and all the fantastic things the term implies. The Thai Select brand has become recognized for its quality, uniqueness, consistency, affordability, and healthiness. Therefore, Thai Select reflects Thai culture wherever restaurant doors open to its menu and exposes customers to the uniqueness of the brand’s image (Seyanont, 2009; Sukalakamala & Boyce, 2007).

6. Conclusion and Limitations

The study set out to determine which factors played a role in a Thai Select restaurant service loyalty in the United States. As the research progressed, it was quickly determined that besides the world-renowned awareness of Thai cuisine’s extraordinary taste and blending of unique flavors, it was also established that Thai food has exceptional nutritional and health benefits. As developed nations age, many are seeking ways to stay healthy. Thai food presents an option that is highly enjoyable and very affordable. With over 15, 000 restaurants worldwide presenting some version of ‘Thai cuisine, ’ it has become necessary for the Thai government to develop a certification and authentication process as Thai cuisine is viewed as a national treasure and part of Thailand’s export and tourism promotion. It was also determined that service tangibility played a significant role in service loyalty, with taste perceptions and brand image contributing factors. Finally, when each diner was asked why they chose to eat Thai food, most viewed taste (49.50%) and healthiness (32.90%) as the most important reasons.

The study is limited because the target sample was only four cities in the United States. With over 15, 000 Thai restaurants globally, the researchers could only tap a small number of the potential Thai food customer opinions. A similar survey in another region, such as Europe, might establish different opinions concerning what is vital in service loyalty and what makes an excellent Thai Select restaurant.