On the Analysis of "intercept & Follow" Surveys

Studies of participation levels and destinations for economic activities such as recreation have traditionally been analyzed using random samples of households, with either cross-section observations or panel data on repeat choices obtained from diaries. In travel demand analysis, an alternative sampling design is to conduct intercept surveys at sites. This can result in substantial reductions in survey costs, and guarantee adequate sample sizes for sites of interest, but the statistical analysis must take into account the "choice based" sample frame. There is a well-developed theory for this analysis in the case of cross-section observations, where data is collected only on the intercept trip. However, recently sampling schemes have emerged in the literature on recreational site choice that combine interception at sites with diaries that provide panel data on intercept respondents on subsequent choice occasions. This paper provides a statistical theory for these "Intercept and Follow" surveys, and indicates where analyses based on random sampling or simple choice-based sampling require correction. The problem is closely related to length-biased sampling in multiple-spell duration modeling, where similar issues and solutions emerge. Acknowledgements: The statistical questions addressed in this paper arose from my analysis for the Atlantic Richfield Company of litigation-related submissions in the case of State of Montana v. Atlantic Richfield Company. This research was supported by the E. Morris Cox endowment at the University of California, Berkeley, and by the Atlantic Richfield Company. The sponsors are not responsible for the conclusions I draw in this paper. I am indebted to Charles Manski and James Walker for useful comments.