The contingent valuation method is a stated preference technique that can be used to estimate non-market values in the economic evaluation of alternative public sector policies. The method is often employed to generate estimates of willingness to pay (WTP) values for a range of alternative policy options. Dichotomous choice contingent valuation surveys have sought to estimate the value of multiple options using a single sample of the population. Such an approach may be open to framing effects and the WTP estimates for the different options may be cross-correlated. In order to derive robust and independent estimates of the community's willingness to pay for different options, separate sampling of the population for each option being considered is desirable. However, this approach can rapidly escalate surveying costs. A compromise, which appears to have some merit, is to undertake full sampling for one of the options being evaluated and then sample at only one bid level for the other options. This permits a willingness to pay bid function to be estimated for the fully sampled option and shifts in the function to be inferred for the other options. An application of this approach to sewage treatment and disposal options for the Vaucluse, Diamond Bay, Dover Heights and Watsons Bay area in Sydney is reported in this paper. One potential problem with this approach, the case of nonparallel shifts in the bid curve, is briefly examined.