How common is the 'prominence effect'? Additional evidence to Whynes et al.

In a recent issue of Health Economics Letters, Whynes et al. report an observation not previously reported in the willingness-to-pay (WTP) literature; that when people are asked to provide an estimate using payment scales they tend to disproportionately select prominent values (that is, 1, 2, 5, 10, 20, 50, 100, etc.). However, it remains an open question just how common this prominence effect actually is. Here we present data from several additional contingent valuation (CV) studies, which suggest that although prominence occurs, it does not reach statistical significance, as found by Whynes et al. A number of reasons are outlined to explain this.