No abstract

Taiwanese tourism policy underwent a major change in 2008 when restrictions were gradually relaxed on Chinese tourists visiting Taiwan. According to the Tourism Bureau of Taiwan's statistics, the overall number of mainland tourists increased from 329,204 in 2008 to 4,184,102 in 2015; however, there was a 16.1% reduction (670,000) occurring in 2016. This significant event will cause more harm than good to Taiwan's all-important tourism industry. In response to such contractions, this study applied cluster analysis combined with entropy to derive suitable clusters useful towards identifying the best market performers among hoteliers through a measurement of large-scale tourist hotels' operational performance. This may signal a benchmark for the improvement of poor performance hotels. Entropy is used as an objective weight method to calculate the relative importance of all salient attributes by comparing the entropy values of each given attribute. Large-scale international tourist hotels have become the market mainstream in Taiwan; therefore, 17 tourist hotels with more than 5000 employees yearly were selected to become part of this study. Operational performance was measured by attributive means of occupancy rate, average room rate, average production-value-per-employee, total number of domestic tourists, and total number of foreign tourists (including overseas Chinese). A significant F value of the ANOVA analysis indicates that there is at least one significant difference found between the two clusters. Further post-hoc analysis uses the Scheffé method to identify any difference found between clusters and to determine the best performance cluster useful as a benchmark. The methods of this study are different from those of previous studies because of the use of a Data Envelopment Analysis (DEA) technique or mix, while there is also applied cluster analysis combined with entropy. Clear indicators are deemed useful for exacting improvement standards among under-performing tourist hotel properties.