Transportation Research Record Journal of the Transportation Research Board · 1982 · 25 citations · 0 references
Social SciencesBehavioral SciencesStatisticsSurveyingTransportData AnalysisMeasurement ErrorResponse BiasSurvey ResearchTravel BehaviorTravel BehaviourMissing Data In SurveysMobility SurveysNonresponse Error AnalysisSurvey ReliabilityTravel Activity PatternsTravel DiariesTravel FrequencyTravel Survey Analysis
The problem of nonreported has become the subject of substantive basic research activities. This has occurred in line with increasing concern about potential artifacts in the development of travel-behavior data attributable to the specific survey method used. In that context, three types of have to be distinguished, namely, (a) that were not reported by the respondents due to increasing lack of care in case of survey periods of several days' length, (b) that were not reported by the respondents because they forgot or considered them redundant, and (c) that the respondents did not want to report on the basis of their own deliberate decisions. It is relatively simple to check the loss of accuracy caused by type-a trips. A methodological experiment was conducted to disclose errors of type b and to gain at least a certain idea about errors of type c. The results confirm previous findings from the analysis of reporting losses for multiple-day surveys. Accordingly, in the analysis of trip volumes, a distinction has to be made first between the nonhome share and the number of for mobile persons. Use of the (artificial) measure trips per person would cover up important relationships. In addition, the level of underreporting of measured by this means can be set between 5 and 15 percent. This underreporting is not equally distributed across all transportation modes and trip purposes, but it occurs to a disproportionately large degree for discretionary and recreation travel, especially by nonmotorized modes.