The effects in survey sampling of nonadherence to the assumption that all elements of a selected sample provide the desired information are investigated. On the basis of a thorough survey sample that had four follow-up reminders (77 percent return rate) and a subsequent survey of nonrespondents, it is shown that substantial misrepresentations of mobile households, trip frequencies, mode-choice distributions, and certain trip purposes become evident. A number of precautions and remedies are suggested to deal with this problem in order to improve the quality of the information input used for the analysis of travel behavior. Not only is the nonresponse bias for low response rates substantially greater, but it also affects the trip structure (frequency, choice, purpose, and destination) more than is the case in a more-exhaustive survey sample. It is demonstrated that a systematic bias arises due to the underrepresentation of nonmobile persons. It is therefore essential to take steps that will increase the willingness of the nonmobile persons to respond to such surveys and that will generate more cost-effective methods to accomplish this objective. It is still necessary to aim for as large a response rate as possible, since the systematic nonresponse bias cannot be compensated for by sociodemographic weighting. A reduction in the follow-up reminders cannot be recommended.

Media Info

  • Media Type: Print
  • Features: References; Tables;
  • Pagination: pp 34-38
  • Monograph Title: Travel Demand Models: Application, Limitations, and Quantitative Methods
  • Serial:

Subject/Index Terms

Filing Info

  • Accession Number: 00334193
  • Record Type: Publication
  • ISBN: 0309031192
  • Files: TRIS, TRB
  • Created Date: Oct 28 1981 12:00AM