Treatment of reference alternatives in stated choice surveys for air travel choice behaviour

With the need for accurate forecasts of passenger demand, the airline sector is increasingly making use of behavioural models calibrated on data from stated choice surveys that allow for the analysis of hypothetical travel situations. To allow analysts to better frame the scenarios presented to respondents, the choice situations in such stated choice surveys often include a current trip as one of the travel options. Classically, these reference alternatives have been treated in the same way as the hypothetical alternatives. The applications presented in this paper show that this potentially leads to biased results, and that it is important to recognise the differences in the nature of the two types of alternatives.