Analysis bias due to survey non response in the French National Travel Survey 2007-08

While nonresponse results in a reduced sample size, a more important concern of researchers is the possible impact of nonresponse bias. Bias is introduced when those that do not respond to the inquiry are systematically different from those that do respond on key estimates. In this circumstance, as the above section describes, the response mechanism is "confounded" with characteristics of the sample units. For example, if nonrespondents tended to take more trips, or longer trips than respondents, the estimates of travel produced from only survey respondents will be too low and not representative of the whole population (only those that responded). Getting the advantage that for the last French National Travel Survey (FNTS) 2007-2008 the sample was drawn directly from the census and the list of new residences built since the census, and therefore we have lots of information about respondent and non-respondent to the different survey instruments in the FNTS. We will quantify these biases by using auxiliary information in different calibration that we will produce.