Willingness to pay (WTP) is increasingly being used as a measure of valuation in health technology assessment. A variety of formats for eliciting values are available, although the relative virtues of each remain the subject of methodological controversy. This article compares valuation results obtained using a WTP survey instrument in a closed-ended format with those obtained from instruments using open-ended and payment scale formats. Samples of subjects were drawn from a general population, and all were asked to value the same intervention--alternative methods of screening for colorectal cancer. It was discovered that, whereas the open-ended and payment scale formats produced broadly similar valuations, the closed-ended format produced significantly higher WTP valuations and different justifications for those valuations. It is hypothesized that anchoring and yea-saying effects explain these differences and that the closed-ended format triggers a different response mode in subjects.