Theory evaluation is related to the precision of predictions. It also involves assessing the range of phenomena a theory can predict; that is, its consilience. Other important features to evaluate are: the number of axioms and presuppositions built into the theory, the ease with which postulated entities can be measured, and the certainty with which phenomena can be predicted.
Within the empiricist doctrine, philosophers have tried to develop a framework for `correct' explanations. The idea is to evaluate the syntactic structure of an explanation, `a priori', separately from the empirical validity of the explanation. The only empirical issue that remains is the validation of propositions applied in the explanation. If the explanation conforms to this framework, and the applied propositions do hold empirically, then the theory is supposed to be able to explain the phenomena correctly. In this section I introduce some of these a priori criteria and the following empirical criteria; ambiguity of the encoding relations, predictive strength, satisfiability of assumptions, consilience, and simplicity.