PANEL SAMPLING DESIGNS ---ANALYSES OF PANEL DATA, PROCEEDINGS OF THE ROUND TABLE CONFERENCE ON THE LONGITUDINAL TRAVEL STUDY, THE HAGUE, MAY 14-15, 1987

The panel sampling design has a great impact on three panel problems. First the design determines to which extent the problem of a changing population is overcome, i.e. To which extent the panel sample is updated. Next the panel design may contribute to combatting the other two panel problems, selective attrition and panel conditioning. These problems are especially severe if the panel also provides descriptive statistics. Rotation is a way of updating and maintaining a panel sample by refreshing the sample continuously. However rotation can not be applied if one wants to do long-term analysis. Then updating and maintaining will have to be done in another way, by recruitment from the families of panel members, by (another method of) random recruitment, by selective recruitment in underrepresented strata (quota sampling) or by a combination of these. After this some general considerations are addressed. The impact of the interviewing frequency is briefly treated. Furthermore it is argued that individual persons should be sampled, not aggregates like households or firms. Some obstacles are mentioned that one may encounter when drawing, updating and maintaining the sample. Reasons for not updating and maintaining the sample are dealt with and finally the issue of choosing a design is addressed. For the covering abstract of the conference see IRRD 818179.

  • Corporate Authors:

    Projectbureau Integrale Verkeers en Vervoerstudies

    Koninginnergracht 16-21
    The Hague,   Netherlands 
  • Authors:
    • Van DE POL, F
  • Publication Date: 1988

Language

  • English

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 00487360
  • Record Type: Publication
  • Source Agency: Transport and Road Research Laboratory (TRRL)
  • Files: ITRD, TRIS
  • Created Date: Sep 30 1989 12:00AM