A method to join data from a National Travel Survey of individuals into travel behaviour of families – with the driving pattern of the household cars as an example

In Denmark, as well as in the rest of the Nordic countries and most other European countries, the National Travel Surveys (NTS’s) are based on interviews with individuals. This is in contrast to for instance the British and American surveys, which are based on interviews with households. The advantage of a household survey is that it makes it possible to analyse the interaction between the household members which is for instance used in activity-based modelling.

However, in most cases the resulting data are used to analyse individual travel activity for which the diary for each household member is used separate from the rest of the household. The backside of household based surveys is that it is difficult to get contact to all the household members in the bigger households to interview them about their travel activities on the same day.

On the one hand, this results in incomplete households which cannot be used for household analyses and on the other hand in biased data with a lower response rate for big households. The advantage of individual surveys is that the resulting dataset is more representative for the population and the response rate can be increased because the contact period of the survey can be expanded.

When the main purpose of the National Travel Survey is to develop transport models and to forecast the effect on traffic volumes from infrastructure development and policy changes the focus is on representativeness. Therefore, the Nordic countries are not so interested in changing to a household survey methodology.

The backside is however that it is not possible to analyse how activities in the families are interrelated whereby some of the advantages of activity-based modelling disappear. A special problem with an individual-based survey is to get information about the driving pattern of the household car(s). In some NTS a special interview is conducted about the driving pattern of the car.

This is however not the case of the Danish NTS. The driving pattern has to be derived from the travel behaviour of the respondents, which is a problem if more than one household member drive the car.