An Introduction to Panel Surveys in Transportation Studies
1. EXECUTIVE SUMMARY
This report is a general introduction to the use of panel designs in surveys of travel behavior. It has four main objectives:
- to highlight the differences between cross-sectional and panel approaches to the study of travel behavior,
- to discuss the limitations of cross-sectional and panel data,
- to identify situations where panel data are preferable, and
- to provide guidelines for designing and maintaining a panel survey.
The report contains a number of recommendations concerning the conduct and use of panel designs in travel surveys. They are summarized below.
GUIDELINES FOR CONSIDERING A PANEL DESIGN
Consider using a panel design whenever the purpose of the travel survey is:
- to develop travel demand models and forecast future demand,
- to measure and understand trends in population behavior,
- to assess the impact of a change in transport policy or services, or
- to collect timely information on emerging travel issues.
GUIDELINES FOR DESIGNING AND CONDUCTING A PANEL SURVEY
- Use the household as the sampling unit and follow initial respondents as they move to new households.
- Collect data from respondents once a year unless more frequent data are required to meet the objectives of the survey.
- Add a supplemental sample of households to improve the representativeness of the panel if the composition of the population in the study area undergoes substantial changes during the survey period, or if the survey continues for five or more years.
- To reduce panel attrition, maintain contact with respondents between waves, develop a locating protocol for tracing respondents who move, give respondents small cash incentives in advance of their participation, and drop only hardcore refusals from the panel.
- Add new modules to the survey instruments as new issues arise, but change the core instruments only when absolutely necessary.
- Weight the data to produce unbiased estimates of population behavior.