Assessing preference elicitation methods in choice experiments : a case study of tourism facilities at Kenyir Lake, Malaysia

• Main Author: W Mohamad, W. N. B.
• Published: University of Newcastle upon Tyne 2018
• Online Access: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.765342

The aim of this thesis is to explore methodological issues in a choice experiment (CE); to ascertain how they might be used to improve the reliability of valuation estimates. Three methodological issues are explored; whether the status quo (SQ) is relevant as one of the alternatives in the CE choice sets; whether respondents ignore any of the attributes presented in the choice set, and the implications this has for estimating willingness to pay; and the effect of different distributional assumptions of random parameters in the Mixed Logit Model: does it matter what distributional assumption is employed? These issues were explored through a study of recreational visitors to Kenyir Lake in Malaysia. Currently, no entrance fee is charged to visitors using Kenyir Lake. But there are government plans to develop Kenyir Lake which involve public investment. Public authorities need to understand visitors' preferences towards facilities, and whether the benefits of improving these facilities justify the cost. The main findings are: 1) including the SQ on choice card does not affect the results substantially, 2) it is important to account for attribute non-attendance, and 3) except for the lognormal distribution, different specifications of the mixing distribution do not make that much difference in WTP values. This study delivers two fundamental contributions. Firstly, it demonstrates the importance of taking into account methodological issues in a CE, and in the analysis of the CE models. The study also provides methodological recommendations for future CE studies. Secondly, it investigates visitors' preferences for tourist facilities and offers policy recommendations regarding the improvement of these facilities. Accounting for methodological issues in a choice experiment is shown to help and provide a deeper understanding regarding the challenges of applying this method; and this thesis offers recommendations on how to apply CE in the future.