Objective: To compare two methods of measuring willingness to pay
(WTP): closed-ended questions with and without follow-up.
Methods: A measurement experiment based on dichotomous choice
contingent valuation survey data is reported. Marginal WTP estimates for
alleviation of rheumatoid arthritis symptoms resulting from treatment with a
novel anti-rheumatic agent, cA2 (TNF-α blockade), were calculated. Monte
Carlo simulations were undertaken to evaluate the methods with respect to
their statistical power.
Results: The estimated marginal WTP values using closed-ended
questions with and without follow-up were DKK 637 (US $91) and DKK 1,268 (US
$181), respectively. A Wilcoxon's signed-rank test showed that the
difference of DKK 631 was significant. Moreover, including a follow-up
question increases the precision of the result. Monte Carlo simulations showed
that trade-offs between power (i.e., the probability of a correct rejection of
a false null hypothesis), efficiency, and size may exist in the two models.
Conclusions: There was a significant difference between the WTP
estimates when using closed-ended questions with and without follow-up. When
choosing between the models, however, power, efficiency, and size could be
used as selection criteria.