The final objective was to be conservative in estimating WTP. When faced with
a decision between two wording, design, or analysis options, neither of which was
clearly preferred on the basis of theory or solid methodological grounds, we chose
the one that would, if it had any effect, lower the aggregate WTP amount. On
this basis, for example, pictures of oiled birds were not shown to respondents, a
one-time payment was used rather than installment payments, and "don't know"
responses were treated as "no" votes instead of dropping them from the sample.