There are three basic levels of criteria measurement in MCA represented by ‘ordinal’, ‘interval’ and ‘ratio’ scales of measurement. The scores of the options on the different criteria are often likely to be at not more than an interval scale or measurement, and some will only be at an ordinal level of measurement. This means that multiplication will produce mathematically meaningless results (Stevens 1946). In the case of ordinal level of measurement, addition or subtraction also result in mathematically meaningless results. So the tempting practice of multiplying the score of the option against a criterion by the weight given to the criterion, and summing the results, will often produce meaningless results.
In practice, it is likely better to avoid seeking to achieve more than an ordinal scale, one in which the numbers simply show whether one impact is larger or small than others, as to do more and employ interval scales requires defining a zero point. Seeking to define a zero point creates many different problems and there may little gain from seeking to address these adequately. It is unlikely to be possible to achieve ratio scales which would permit multiplication.
In considering measurement, it is necessary to point out the difference between economics and the economy. Economists do not define economics as the study of the economy , rather the study of human behaviour, but some of the stakeholders may expect that economists will be able to predict the consequences of the different options for the economy in terms of growth, employment, real incomes and so forth. In some other areas, economics does provide a means of attaching money values to some consequences but it is up to the stakeholders to determine in which cases this is helpful to them. It will only be useful if there is a basic consensus here, often regarding detail rather than just principles. For example, if the option being considered involves the loss of an ecologically important dryland ecosystem – perhaps by creating a flood storage area – but the gain of a potentially valuable wetland, it is probably not helpful to start by trying to attach money values to each. The general rule is: do not force a monetary valuation on to an asset or loss if does not help in understanding and sharing that understanding.