Identifying poor response behaviour in (longitudinal) travel behaviour research

In travel behaviour research, longitudinal panels, such as the Netherlands Mobility Panel (MPN), have been set up to investigate travel behaviour over time. The data from these panels are ideally suited to model and understand the (causal) mechanisms underlying travel behaviour, and changes in it over time, at an individual level. It is important to ensure the quality of the data. However, there are several factors that can lead to low data quality. One of those factors concerns 'soft-refusal', which describes the tendency of some respondents to refuse participation in a 'soft' way, for example by reporting that they did not leave their house although they actually did, or by answering a questionnaire so quickly that it is not likely that one has read and understood all the questions and answers properly.

The Netherlands Institute for Transport Policy Analysis (KiM) studied several methods to identify possible soft-refusal in longitudinal travel research, based on: 1) predicting out-of-home activity 2) straightlining (giving the same answer to each item of a grid question), and 3) the speed with which the questionnaire is completed. All methods seem able to identify respondents with poor response behaviour. The study also provides insight into the profiles of respondents with poor response behaviour.

You can download the study from the TU Delft repository: Didn’t travel or just being lazy? An empirical study of soft-refusal in mobility diaries | TU Delft Repositories