Modelling the Effect of Website Quality and Perceived Information Value on Tourist Satisfaction: A Survey of Hotels in Phuket, Thailand

This paper presents the investigation of the effect of website quality on tourist satisfaction. It focuses on an issue of the hotel industry in Phuket Province, Thailand: Does the website quality of hotels affect the relationship between the perceived value and tourist satisfaction? The results indicate that only the service quality aspect of website quality was affect positively associated with tourist satisfaction, whereas the service quality and information quality of hotel websites were not affect associated with the perceived information value. The perceived information value was affect strongly positively associated with tourist satisfaction.


Introduction
The evolution of internet technology and Web 2.0 has affected and altered the forms of communication between customers and businesses. Business transactions with online technology support effective interactions between customers and businesses. Furthermore, technology helps increase the opportunity for businesses to reach customers and is able to improve good service via the website.
The tourism industry in Thailand had been using digital technology to support the operation and expansion of the tourism industry. The National strategy Thailand 4.0 officially of the government aims to increase the business shift to the online world. Furthermore, the infrastructure development and facilities are in accordance with the lifestyle of the new generation of tourists that use more information to make decisions regarding product purchases and to verify information from the recommendations of others on social media.
Online travel agency (OTA) websites, such as Traveloka, Booking, Expedia, and Agoda, are popular for online reservations. Customers are able to make reservations via the internet, and it is easy for customers to change or cancel online reservations on the website. The OTA websites have an advantage over hotels websites in terms of data collection and integration for tourism. Moreover, OTA websites also allow customers to compare prices, check hotel ratings, and read comments written by previous customers. These information-seeking behaviours have helped reduce the uncertainty in online purchasing (Xu, 2017). However, hotel websites are still important for authentication and the reliability of hotels.
Regarding the hotel industry in Thailand, we could not find any study that examined website quality in hotels and how website quality affects the relationship between the perceived information value and tourist satisfaction. Thus, with this paper, we hope to contribute to the literature on website quality, perceived information value, and satisfaction.

Website Quality
Website quality is defined as the perception of the measurement and user evaluation that meets customer needs (Aladwani and Palvia, 2002). It includes the technical opinions on the features of the website (Hasan and Abuelrub, 2011). A recent study reported that the majority of the studies have focused on functional quality (service-delivery process), not technical quality (outcome of the service process; (Ladhari, 2010). Some previous studies measured website quality by focusing on the usability attribute (Teo et al., 2003), whereas others investigated website quality from an outcome perspective.
Furthermore, most studies have examined the dimensionality of website quality from the information systems (IS) perspective, which includes usability, accessibility, navigability, and information quality, whereas the marketing perspective examines dimensions such as advertising, promotion, online transactions, customer service, and order confirmation to assess the quality of the website. Loiacono et al. (2007), suggested that approaching the combined perspective of IS and marketing is important in evaluating website quality because technology continues to change, and new forms of applications are created for development of online channels. A number of researchers have also suggested ways to improve website design. It was found that, in studies that evaluated websites from the IS perspective, consumer satisfaction was driven by the information and system quality. However, in studies that were driven from the marketing perspective, consumer satisfaction was influenced by the retail experience (Chiou et al., 2010). For this study, we assess the website quality using the IS success model by DeLone and McLean (2003), that includes three dimensions: system quality, information quality, and service quality. DeLone and McLean (2003), described system quality as 'the desired characteristics of an e-commerce system', including the availability of the website, the ability to move through the website, and the download speed. In addition to the functionality of the website, the style and appearance of the website are included in system quality (Chang et al., 2005).

Information Quality
Information quality includes the content presented on a website and can be assessed regarding the informativeness and security dimensions (DeLone and McLean, 2003). Businesses should focus on aspects of information quality in the website design (Kim and Niehm, 2009) because the website design is a strategy to create value for customer satisfaction and to contribute to the goal of selling products. Therefore, the information presented on the website should be useful information for the customer. Dholakia and Zhao (2010), found that clear product information and the ease of finding information are two key factors of information quality. Hence, a good website design will help build credibility and convince customers to contact the business. Parasuraman et al. (1988), defined service quality as phases of customer interaction with a website that encourage effective online purchasing and delivery. DeLone and McLean (2003), suggested that system quality includes all of the support that is delivered to the customer during the purchase and post-purchase process.

Service Quality
The SERVQUAL model by Parasuraman et al. (2005), which is used in the assessment for this study, includes dimensions specific to the internet. The E-S-QUAL scale is used to assess the website quality for both services and products. Many studies have employed the five dimensions identified in SERVQUAL (e.g. tangibility, reliability, responsiveness, assurance, and empathy), but not all five dimensions are applicable. Based on the limitations of products and situations, it is necessary to consider post-purchase activities. In this study, service quality consists of two dimensions: reliability and responsiveness in the hotel website context.

Perceived Information Value
Perceived value is the in-depth study of the behaviour of the consumer to understand the real needs and user experience of using products and services in different situations. Woodruff (1997), said that the perceived value is a critical marketing tool for ensuring customer loyalty and is an important factor that provides a competitive advantage. Stahl et al. (2003), confirmed that the perceived value as an important driver of development and customer loyalty. The theory of perceived value is closely related to the theory of satisfaction and customer loyalty (Grönroos, 1997) to describe customer behaviour under conditions of risk and uncertainly (Gupta and Kim, 2010).  Kotler and Keller (2009), defined the meaning of perceived value as the difference between the value that customers receive from using goods or services and the total cost. Customers will buy from organisations that they perceive as offering the most value. Research studies about perceived value in the online context have found the estimated perceived value in various dimensions: information value, social value, and hedonic value (Alshibly, 2015;Yen, 2013). This study focuses on the perceived information value from hotel websites.

Satisfaction
Customer satisfaction is the key to building a competitive advantage and influencing customer repurchase intentions (Lin and Lekhawipat, 2014;Marinkovic et al., 2014). Most studies examining customer satisfaction creation have often identified two variables: perceived value and product/service quality, which are antecedents of satisfaction. Furthermore, the American Customer Satisfaction Index (ACSI) is a cause-and-effect model with indices for drivers of satisfaction (customer expectations, perceived quality, perceived value, satisfaction, customer complaints, and customer loyalty) that are often also used to study customer satisfaction. The ACSI consists of six components, and the indices are shown in Figure 2.

Figure-3. Proposed framework
Therefore, the following hypotheses were tested: Hypothesis 1: The effect of the website quality on the perceived information value. Hypothesis 1a: The effect of the system quality on the perceived information value. Hypothesis 1b: The effect of the information quality on the perceived information value. Hypothesis 1c: The effect of the service quality on the perceived information value. Hypothesis 2: The effect of the perceived information value on tourist satisfaction.

Methodology
The researcher used the descriptive research method to assess the construct validity of the website quality of hotels in Phuket Province, Thailand, regarding the perceived information value and tourist satisfaction. The research tools included questionnaires that consisted of 38 items from the literature review. All variables were measured using a 5-point Likert-type scale by testing the tools that consist of the content validity by assessing the Item-Objective Congruence Index: IOC value, considering 3 expert items and the reliability (coefficient value of Cronbach). Cronbach's alpha is equal to .970, which passes the criteria because its value should be higher than 0.6 (Hair et al., 2010).

Sampling
The populations in this research included 1,184,049 Thai and 3,041,042 foreign tourists who travelled and stayed at hotels in Phuket province in 2017, which totalled 4,225,073 persons. The sampling frame identified tourists who travelled and stayed at the hotels in Phuket Province, Thailand and who used the website of the hotels directly to reserve rooms or search for information for travel plans. Regarding the sample group appropriateness, the researcher chose the structural equation modelling (SEM) technique and used the method of determining the sample size according to Hair et al. (2010), who recommended that the sample sizes used in this research should be between 10 and 20 times of the observed variables in the research. For this study, the researcher used 38 observed variables. Hence, the sample sizes that are appropriate and sufficient for the research should be at least 380-760 at minimum. According to such calculations, for analysing the structural equation, there were 400 samples regarding the examination of validity in this study.  H1c H2 Concerning the sampling, quota sampling was used by considering the statistical data of Thai and foreign tourists who travelled to Phuket Province between 2016 and 2018 to construct the categorised tendency of the samples. The samples were divided into groups according to the quota proportions of both groups.

Data Collection
The data were collected by questionnaires from 400 tourists who travelled and stayed in hotels in Phuket Province, resulting in a response rate of 100%. After collecting the data according to the required targets, the researcher entered the data into the SPSS package software to assess the statistical results.

Questionnaire Development
The questionnaire included the website quality, perceived information value, and tourist satisfaction scales. Moreover, it included other background questions about tourist characteristics.

Website Quality
We developed a 31-item scale for website quality by adopting the scales from Aladwani and Palvia (2002), Kim and Stoel (2004); Parasuraman et al. (2005); and Egeln (2015). All items were separated to three groups: system quality (14 items), information quality (9 items), and service quality (8 items). A 5-point Likert scale was used to obtain responses for the 31 statements.

Perceived Information Value
The perceived information value was measured using three items. We adopted the scale by Yen (2013). These three items were rated by tourists on a 5-point Likert scale.

Satisfaction
We developed a four-item scale for tourist satisfaction by adopting all scales from Alshibly (2015), about information services on websites. These four items were rated by tourists on a 5-point Likert scale.

Measurement Variables
This study included several variables in the analysis, which are summarised in Table 4. 32. One can perceive the information of products and services easily from the website of the hotel.
Modified from Yen (2013) PIV2 33. The information on the products and services received is from the website of the hotel. PIV3 34. One can perceive the information of the products and services easily from the hotel website. Satisfaction: SAT SAT1 35. The hotel website provides services in accordance with expectations.
Modified from Alshibly (2015) SAT2 36. The hotel website offers information that one requires. SAT3 37. The user has a good experience using the hotel website. SAT4 38. Overall, the user feels satisfied with the services of the hotel via the hotel website and social media.

Measurement Models: Validity and Reliability of the Scales
The measurement validity and reliability were assessed with a confirmatory factor analysis: CFA carried out on 38 items. In the process of assessing convergent validity, items with factor loadings less than 0.40 (Nunnally and Bernstein, 1994), were deleted. In addition, items contributing to low reliability were deleted. None of the website quality, perceived information value, and tourist's satisfaction items were dropped. The overall fit for the measurement model without any modifications was good based on the fit indices, such as the chi-square χ 2 = 1555.67, root mean square error of approximation (RMSEA) = 0.058, normed fit index (NFI) = 0.97, goodness-of-fit index (GFI) = 0.83, comparative fit index (CFI) =0.99, incremental fit index (IFI) = 0.99, and standardised root mean square residual (SRMR) = 0.040. The convergent validity of the measures is evidenced by the large and significant factor loadings shown in Table 6, and the discrimination validity is indicated by the relatively high correlations between the dimensions of the same construct (shown in Table 7).

Results
The hypotheses were tested by estimating the SEM using LISREL 8.7.2, with system quality, information quality, and service quality as the three endogenous variables, and satisfaction as the exogenous variable. After assessing the overall fit of the model, the fit statistics were χ 2 = 1555.67, p = 0.00, with df = 658, GFI = 0.83, CFI = 0.99, NNFI =0.98, RMSEA = 0.058, IFI = 0.99, and SRMR = 0.040. Thus, the model was appropriate for the empirical data and indicates that the model had a concordance at a very good fit level. All items in the measurement model were statistically significant for each factor, and all standardised factor loadings are greater than the 0.50 cutoff. It mostly passed the consideration criteria, which should be greater than |0.5| (Hair et al., 2010); see Table 6.
As shown in Table 8, the results indicate that the effect of service quality is statistically significant (p<.01). Therefore, Hypothesis 1c is supported. However, the effects of the system quality and information quality are not statistically significant. Thus, Hypotheses 1a and 1b are not supported. For Hypothesis 2, we found that the effect of the perceived information value is statistically significant (p<.01). To summarise, our results indicate that the effect of website quality is only supported for service quality, whereas system quality and information quality are not supported. Moreover, the effect of the perceived information value is supported.

Conclusion
This study contributes to research on website quality, perceived value, and satisfaction by examining (1) the effect of website quality on the perceived information value and (2) the effect of the perceived information value on satisfaction. Some results in this study that should be addressed are as follows: 1) We find that the website quality of hotels located in Phuket, Thailand include system quality, information quality, and service quality, but only service quality is positively associated (.46) with the perceived information value. 2) The perceived information value is strongly positively associated (.92) with tourist satisfaction. The limitations of this study are related to the samples. Our target respondents were tourists who visited a 1-3 stars hotel, which does not cover all hotel standard levels, and this study is limited to one setting: Phuket Province.
Replications of this study done should be more study with samples from other provinces in the south of Thailand, such as Krabi, Trang, Surat Thani, and so on. Thus, we identify the limitations of this study to develop suggestions for future research. Moreover, our study can be expanded by including additional constructs, such as 'trust in the website' and 'loyalty to the website' or the need for purchase intention. We believe that our results can facilitate a better understanding of the effect of website quality of hotels in Phuket, Thailand regarding satisfaction and can be used to improve the website design in terms service quality to have more positive effects on satisfaction. For future research, we could compare the effect of website quality for hotels with 4-5 stars in order to apply more accuracy, leading to considering various result dimensions.