Design of experiments
Margaret Morrison and Mary Morgan (1999) have argued that models function as ‘instruments of investigation’. We can learn about the world from them and about theories because they involve some form of representation. They either represent an aspect of the world or an aspect of theories about the world, or both. As indicated in Chapter 1, we confine ourselves to models that provide mathematical representations of aspects of the empirical world. The relevant question about instruments is not ‘How true are they?’ but ‘How accurate are these instruments?’ In general, to assess an instrument’s accuracy we test it: that is, we investigate the correspondence between the representation itself and the aspect of the world that is to be represented. The accuracy of this correspondence depends on the complexity of the system under investigation and our ability to construct representations of it. This is, in principle, a technical problem, labelled by Kevin Hoover (1988: 21820) as the ‘Cournot problem’ after the man who formulated it explicitly, Antoine Augustin Cournot (1801-77):
The economic system is a whole of which all the parts are connected and react on each other. … It seems, therefore, as if, for a complete and rigorous solution of the problems relative to some parts of the economic system, it were indispensable to take the entire system into consideration. But this would surpass the powers of mathematical analysis and of our practical methods of calculations, even if the values of all the constants could be assigned to them numerically.