This analysis uses the dichotomous choice contingent valuation methodology (CVM).
This technique was chosen over the multiattribute techniques, such as conjoint analysis/
choice modelling, as we were only considering one specific attribute; whether the
beef had been tested. A single-bounded logit model is used to explore factors affecting
WTP for BSE-tested beef.We also estimate the meanWTP for BSE-tested beef. There
are many other possible approaches to estimating non-market quality attributes, such
as experimental auctions or hedonic price analysis. Stated-preference techniques, such
as contingent valuation, are sometimes criticised because of the hypothetical nature
of the questions and the fact that actual behaviour is not observed (Mitchell and
Carson 1989). However, Adamowicz et al. (1994) criticise revealed preferencemethods
on the basis that the models of behaviour developed constitute amaintained hypothesis
about the structure of preferences that may not be testable. They also point out that
revealed preference methods can suffer from collinearity among attributes, precluding
identification of the marginal impact of factors that affect choice. A number of studies
test consistency or comparability between revealed and stated preferences with no
consensus of findings.4