The National Travel Survey (NTS), which is continuous, is carried out by Statistics Netherlands, the Dutch Central Bureau of Statistics (CBS). A sample of the population of the Netherlands is surveyed about their travel behaviour: how many kilometres, which modes of transport, during which period. The sample results are then extrapolated to represent `the whole Dutch population'. This report uses a method published by the CBS to estimate the uncertainties of the NTS data. Only those uncertainties resulting from the fact that NTS uses a sample, and not the whole population, are dealt with. For example, the method makes no estimation of the uncertainty which is a result of respondents estimating the distances, and their not actually being measured. The estimated uncertainty can, therefore, be best regarded as a minimum. The practical working out resulted in a relatively flexible SAS program. Various multidimensional tables were established quite easily. A practical example is given to show the effect of the uncertainty of further disaggregation of NTS data. (A)
Samenvatting