Impact prediction, evaluation and mitigation - Introduction to Environmental Impact Assessment - page 152

(Adapted from DoE 1991, Winpenny 1991, Pearce & Markandya 1990, Barde & Pearce 1991.)

(preferences people have for using an environmental asset, such as a river for fishing) but
also non-user values (where people value an asset but do not use it, although some may
wish to do so some day). Of course, such techniques have their problems, for example the
potential bias in people's replies in the contingent valuation method (CVM) approach (for
a fascinating example of this, see Willis & Powe 1998). However, simply through the act
of seeking a value for various environmental features, such techniques help to reinforce
the understanding that such features are not “free” goods and should not be treated as
such.

5.3.3 Scoring and weighting and multi-criteria methods

Multi-criteria and multi-attribute methods seek to overcome some of the deficiencies of
CBA; in particular they seek to allow for a pluralist view of society, composed of diverse
“stakeholders” with diverse goals and with differing values concerning environmental
changes. Most of the methods use—and sometimes misuse—some kind of simple scoring
and weighting system; such systems generate considerable debate. Here we discuss some
key elements of good practice, and then offer a brief overview of the range of multi-
criteria/multi-attribute methods available to the analyst.
Scoring may use quantitative or qualitative scales, according to the availability of
information on the impact under consideration. Lee (1987) provides an example (Table
5.8) of how different levels of impact (in this example noise, whose measurement is in
units of L 10 dB A ) can be scored in different systems. These systems seek to standardize the
impact scores for purposes of comparison. Where quantitative data are not available,
ranking of alternatives may use other approaches, for example using letters (A, B, C, etc.)
or words (not significant, significant, very significant).
Weighting seeks to identify the relative importance of the various impact types for
which scores of some sort may be available (for example, the relative importance of a
water pollution impact, the impact on a rare flower). Different impacts may be allocated
weights (normally numbers) out of a total budget (e.g. 10 points to be allocated between
3 impacts). But by whom?
Multi-criteria/multi-attribute methods seek to recognize the plurality of views and
weights in their methods; the Delphi approach also uses individuals' weights, from which
group weights are then derived. In many studies, however, the weights are those
produced by the technical team. Indeed the decision-makers may be unwilling to reveal
all their personal preferences, for fear of undermining their negotiating
Table 5.8 A comparison of different scoring
systems

Method Alternatives Basis of score

A (no action) B C D

Ratio 65 62 71 75 Absolute L 10 dB A measure

Interval 0 −3 +6 +10 Difference in L 10 dB A using alternative A as base