An application of a comparative analysis of revealed preference and stated preference methods to the air transport choice problem

This paper examines the travel behaviour of air transport passengers originating in Seoul and destined for six American and European cities: Los Angeles, New York, San Francisco, Frankfurt, Paris and London. The research surveyed several hundred travellers and modelled their behaviour with respect to travel mode choice using revealed preference and stated preference techniques. Revealed preference models were constructed in which four variables - journey time, fare paid, flight frequency and airline nationality - were utilised. Using separately executed surveys of passengers, slated preference techniques were used to build models of passengers mode choice using the same variables. Journey time, fare paid and flight frequency were found to be significant at the 99 percent level and airline nationality only at the 90 percent level. The two sets of models were compared and found to be reasonably comparable. It is recommended that slated preference techniques are used where possible to avoid the problems of multi-collinearity of variables which can occur with revealed preference techniques.