Environmental Engineering Reference
In-Depth Information
believe that extensive methodological research and quality improvements have
already increased the reliability and feasibility of the contingent valuation approach.
An important stimulus to the use of CVM was the recommendations of the National
Oceanic and Atmospheric Administration's, or NOAA, panel on contingent valuation
(see NOAA
1993)
. This panel of eminent social scientists, co-chaired by two Nobel
Prize-winning economists, Kenneth Arrow and Robert Solow, specified an extensive
set of guidelines for CVM survey development, administration, and analysis. It also
found that CVM, if appropriately conducted, could convey useful information.
Choice Modelling
In the field of monetary valuation, choice modelling (CM) is being increasingly
applied as an alternative to CVM (Adamowicz et al.
1994
;
1998)
. Like CVM, CM
is a stated preference method and is capable of measuring consumer acceptance of
multi-attribute commodities. CM has benefited from the work carried out in the
field of psychology, economics, and consumer market research, where marketing
research scholars employed the technique to evaluate potential new products and
new markets for existing products (Garrod and Willis
1999a
; for thorough and critical
descriptions of CM, see Roe et al.
1996
; Farber and Griner
2000
, and Louviere
et al.
2000)
. Hanley et al.
(2001
: 436) define CM as '… a family of survey-based
methodologies for modelling preferences for goods, where goods are described in
terms of their attributes and of the levels that these take.' As such, the term choice
modelling is somewhat broader in coverage than the method of choice experiments;
or, to put it more precisely, choice experiments are a derivative of CM.
3
Generally, in a CM, respondents are confronted with a number of commodity
descriptions, or situations, that differ according to the attributes described (as in hedonic
pricing). Respondents are then asked to rank or rate the bundles of attributes, or select
the most-preferred one from the set. The basic premise underlying this method is
that commodities have a value because of their attributes. In order to decide which
commodity they want, people make trade-offs between the attributes (de Blaeij
2003)
.
CM is designed to determine the structure of preferences that underlie the judgement
of multi-attribute commodities and to that end, it measures the rates at which people
are willing to make such trade-offs (Shechter
2000)
. The inclusion of at least one
monetary attribute, such as the cost of provision or the WTP for conserving the coun-
tryside, allows for the derivation of implicit prices for each of the other attributes.
Unlike CVM, CM does not directly ask for a WTP, but offers the opportunity to
rank all of the alternatives from highest (best) to lowest (worst). Defenders of CM
therefore argue that the technique outperforms CVM on the point of strategic
behaviour. The indirect way of questioning in the CM allows respondents to
explicitly determine trade-offs between different attributes of the commodity under
3
The term choice modelling is often used interchangeably with the term conjoint analysis
(e.g. Hanley et al.
2001)
. Of these two terms, choice modelling seems to be the most commonly
used nowadays.
