In the Netherlands a major empirical study is underway to measure the value to society of travel time benefits and travel time reliability benefits in passenger and freight transport. The study is commissioned by the Dutch Ministry of Transport to a consortium consisting of RAND Europe (now succeeded by Significance), the VU University in Amsterdam, and John Bates. The resulting values of time (VOTs) and values of reliability (VORs) from the study will be used in cost benefit analyses (CBAs), in accordance with government directives. The VOTs will be updates of existing values from previous studies, the VORs will be the first of their kind for the Netherlands. The VORs will be estimated using Stated Preference (SP) experiments. In the past years several researchers have designed formats to present unreliability of travel times to respondents in SP experiments. These formats use concepts from statistics like average travel time, travel time variance, probability of arriving late, histograms, etc. The big question for us was whether these fairly advanced concepts are understandable for laymen travellers without a degree in statistics or even higher education. To the best of our knowledge, this issue has never been empirically put to test. It is, however, crucial to gain insight in the quality of the empirical results for VORs collected from interviews. In the analyses eight formats were tested. Respondents were stratified according to their education level. There were three groups of questions. First, questions about how respondents conceptualise unreliability themselves. Do they think in terms of average, minimum, maximum travel time or probability? And how complicated do they find these concepts? Secondly, respondents were prompted with questions to test whether they gave the right answer for the different presentation formats. Thirdly, there were questions about the respondents assessments of the eight presentation formats regarding clarity, ease of handling, and visual attractiveness, and which ones were preferred. The interviews supplied a clear winner among the eight formats. This format is not only preferred by a majority of respondents, but also equally by people with low and high levels of education. It contradicts the conventional wisdom that a picture says more than a thousand words. This format will, therefore, be used in the forthcoming main study. For the covering abstract see ITRD E137145.
Samenvatting