Destination attributes, improvement priority, and cultural influence on tourists' behavior evaluation using user-generated content in Langkawi Island, Malaysia

The attractiveness of a tourism destination determines its tourist arrivals and local tourism revenue. Currently, studies exploring destination attributes using user-generated content (UGC) are scarce, such as using machine learning to identify destination attributes, using dynamic methods to determine attribute improvement priorities, and the influence of country culture on tourist preferences and evaluations. Hence, this study focused on exploring the destination attributes that promote the attractiveness of Langkawi Island using UGC. This island was chosen as the case study of this research because, despite Langkawi Island's popularity, it is at a disadvantage in competition with the islands of neighboring countries (e.g., Bali and Phuket) and is considered dull and unattractive by tourists. This study sought to achieve three objectives, as follows: 1) to identify the destination attributes of Langkawi Island; 2) to track the dynamic trends of destination attributes of Langkawi Island; and 3) to establish the influence of culture on tourists’ preference and evaluation of destination attributes. Unlike traditional research, the data set used in this study was user-generated content (UGC), known as big data. The UGC was collected from Tripadvisor and contained both structured rating scores and unstructured text reviews of tourists’ travel experiences on Langkawi Island. Three main analyses were conducted to answer the research questions. First, this study extracted destination attributes from online reviews and compared them to pre- established attributes. The findings demonstrated that machine learning methods can extract more specific attributes from online reviews and are thus capable of identifying new attributes to supplement existing research. The specific island attributes extracted were ‘ticket purchasing’, ‘animal feeding’, ‘animal show’, ‘water fall, and ‘food market’. Second, the time dimension was introduced to importance-performance analysis (IPA) and asymmetric impact performance analysis (AIPA) to conduct dynamic importance- performance analysis (DIPA) and dynamic asymmetric impact performance analysis (DAIPA) that were used to determine destination attribute improvement priorities. According to the findings of the IPA and DIPA, the attributes that are in desperate need of improvement are those with increasing/high importance scores and decreasing/low performance scores, namely ’snorkeling’, ‘amenity’, ‘sky bridge walk’, ‘food market’, and ‘duty-free mall’. In addition, considering the asymmetric effect result from DAIPA, the attributes whose impact asymmetry and performance are both decreasing/low also need to be improved, which are ‘service’, ‘island view’, and ‘waterfall’. Finally, the relationships between Hofstede's cultural dimensions and tourist preferences and evaluations were investigated to explore the influence of national culture on tourist behavior. The results showed that national culture affects tourist preferences only for activities. That is, power distance is positively related to water activity, individualism is negatively related to animal feeding but positively related to water activity, uncertainty avoidance is negatively related to the sky bridge walk, long-term orientation is positively related to snorkeling, indulgence is positively related to animal feeding but negatively related to water activity, and masculinity is negatively related to water activity but positively related to the waterfall. National culture also demonstrated a stronger influence on rating evaluations than sentiment evaluations. In addition, power distance, individualism, and masculinity were found to have positive relationships with tourist evaluations. Conversely, uncertainty avoidance, long-term orientation, and indulgence exhibited negative relationships with tourist evaluations. The findings on the influence of national culture on tourists' preferences and evaluations provide useful guidelines on how managers should design and promote tourism offerings that appeal to the cultural background of tourists. For example, tourists from high power distance cultures such as Saudi Arabia, the Philippines, Qatar, and the United Arab Emirates can be recommended to visit high-end attractions or activities (e.g., parasailing and snorkeling). Additionally, managers should focus on improving the satisfaction of tourists from low individualism (such as Indonesia and China) and high indulgence (such as Mexico and Australia) cultures by providing them with interesting tourism activities, such as animal performance shows, museum and art gallery tours, excellent hospitality and services, as well as adventurous snorkeling activities. The important theory contributions are as follows: First, this study expanded the method of destination attribute identification using UGC. Second, Kano's three-factor theory was extended to a dynamic analysis of the impact of attribute performance on tourist satisfaction. Third, the role of Hofstede's cultural dimensions in tourists' preferences and evaluations in the online environment was explored. Finally, a method for quantifying tourists' preferences in text review was proposed.

Text SPE 2023 1 IR.pdf Download (3MB)

Actions (login required)

View Item