A smart Travel Survey. Results of a push-to-smart field experiment in the Netherlands

Travel Surveys are considered promising candidates to go ‘smart’. Respondents need to be both motivated and competent to correctly report all details of their travels for a specified time period. Location tracking offers options to remove burden and to improve quality of measurement. Adding contextual information, the collected location data may also be input to predictions of travel mode and travel purpose. However, location tracking may be perceived as privacy-sensitive by respondents. Furthermore, location data are subject to various types of error that (in part) can only be adjusted for with the help of respondents. It is, therefore, not evident that the promise of smart features will hold in practice. For this reason, Statistics Netherlands conducted a first field test in 2018 using a proof-of-concept travel app. Response rates clearly showed variation across relevant subgroups in the population, but were sufficiently high to justify further development and experimentation.

In 2022, Statistics Netherlands again fielded a travel-app assisted experiment, but this time including the regular online Travel Survey as a concurrent option. A population sample was offered the online option at different time points. These time points were randomized across the sample. Simultaneously, also the requested tracking period, one full day or one full week, was randomized.

In this paper, we discuss the design and outcomes of the field experiment. We focus on response and representation. Measurement data quality and the in-app behaviour of respondents are studied and reported in separate papers. Our main conclusion is that the concurrent option had a backfiring impact on response and needs to be introduced differently.

Downloads