Augmenting travel cost models with contingent behavior data Poisson Regression Analyses with Individual Panel Data

Jeffrey Englin, Trudy Ann Cameron

Research output: Contribution to journalArticlepeer-review

154 Scopus citations


This paper proposes contingent behavior survey questions as a valuable supplement to observed data in travel cost models of non-market demand for recreational resources. A set of observed and contingent behavior results for each survey respondent allows the researcher to control for individual heterogeneity by taking advantage of panel data methods when exploring the nature of respondent demands. The contingent scenarios also provide opportunities to (a) test for differences between observed and contingent preferences and/or (b) assess likely demands under conditions beyond the domain of observed variation in costs or resource attributes. Most importantly, contingent scenarios allow the researcher to impose exogenously varying travel costs. Exogenous imposition of travel costs together with panel methods reduces the omitted variables bias that plagues observed-data travel cost models of recreational demand. Using a convenience sample of data for illustrative purposes, we show how to estimate the demand for recreational angling by combining observed and contingent behavior data. We begin with simple naive pooled Poisson models and progress to more theoretically appropriate fixed effects panel Poisson specifications.

Original languageEnglish (US)
Pages (from-to)133-147
Number of pages15
JournalEnvironmental and Resource Economics
Issue number2
StatePublished - 1996
Externally publishedYes


  • Contingent behavior surveys
  • Fixed effects models
  • Nonmarket valuation
  • Panel data methods
  • Poisson regression
  • Recreational demand modelling
  • Travel cost method

ASJC Scopus subject areas

  • Economics and Econometrics
  • Management, Monitoring, Policy and Law


Dive into the research topics of 'Augmenting travel cost models with contingent behavior data Poisson Regression Analyses with Individual Panel Data'. Together they form a unique fingerprint.

Cite this