5.2 Measures of Model Specificity

5.2.2 Scope

We take the scope of a model to be the Cartesian product of all possible values of its dependent and independent variables. The scope of a model is increased by adding to (and is reduced by subtracting from) the set of variables over which it is defined. Increasing the scope of a model can vastly increase the specific forms that the model takes. In general, increasing model scope is to be avoided if at all possible because it necessarily increases the computational and information-processing capacities that the model will absorb. One of the advantages of generalizing models is that the generalized model may have fewer independent variables than the models it replaces and, so, a smaller scope.

Modelling Learning as Modelling - 23 FEB 98

[Next] [Previous] [Up] [Top] [Contents]

Generated with CERN WebMaker