Rasch model in data envelopment analysis: application in the international tourist hotel industry

The Rasch model can be used to transform ordinal data into interval data. Rasch has seldom been incorporated in data envelopment analysis (DEA) models. Therefore, this study attempts to apply the Rasch model in DEA to analyse the performance of seven districts including Taipei; Kaoshong; Taichung; Hwalian; Scenery; Taoyuan, Hsinchu, Miaoli; Others in Taiwan containing 56 international tourist hotels during 2009 using five criteria, including average occupancy rate, average room rate, average production value per employee in the accommodation department, average production value per employee in the catering department and operating income/expenses. Furthermore, this study also compared the results with and without the Rasch-bounded DEA model. As expected of the analytical results that applying the DEA model using the bounded formulas achieves better distinguishing power than applying the model without using the bounded formulas. Furthermore, the effectiveness value of each district of Taiwan (Taipei (1.000), Kaoshong (0.3661), Taichung (0.3494), Hwalian (0.2658), Scenery (0.6476), Tyoyuan, Hsinchu, Miaoli (0.5623), others (0.3127)) during 2009 is obtained.