Dichotomous-Choice Contingent Valuation

Description

Each respondent answers yes or no to paying one randomly assigned amount, optionally followed by a higher or lower amount, and the population WTP distribution is estimated by maximum likelihood.

Respondents are asked whether they would pay a stated amount, with the amount randomised across respondents. In the double-bounded version a yes is followed by a higher amount and a no by a lower one. True willingness to pay is assumed to follow a distribution, such as normal, lognormal if strictly positive, or a spike at zero plus a lognormal for non-payers. The probability of each answer pattern is the share of that distribution falling in the interval it implies, and the mean and spread are estimated by maximum likelihood, which is ordered-probit logic with cut-offs set by the questionnaire. Because each person answers one simple question, it mimics a real purchase or referendum decision better than open-ended price questions. The method is contested (Hausman, 1993), and double-bounded answers of yes-yes or no-no remain only one-sided.

Attributes

Loss type
Unmigrated
Effort
Weeks

Steps

  1. Choose 4-8 bid amounts spanning the plausible WTP range and randomise them across respondents.
  2. Ask whether the respondent would pay the assigned bid for a clearly described product or change (single-bounded).
  3. Optionally follow a yes with a higher bid and a no with a lower bid (double-bounded).
  4. Assume a WTP distribution: normal, lognormal, or a spike at zero plus lognormal.
  5. Write each answer pattern's probability as the WTP probability mass in the interval it implies.
  6. Maximise the log likelihood to estimate mean and spread of WTP and report confidence intervals.

Sources

  • Discrete Choice Methods with Simulation (Train, 2009)
  • Quantitative Social Research Methods (Singh, 2007)

Also known as

  • Dichotomous-choice willingness-to-pay elicitation
  • Contingent valuation
  • Referendum-format WTP question
  • Single-bounded dichotomous choice
  • Double-bounded dichotomous choice

Categories

  • Analytics & Behaviour
  • Behavioral Science & Economics