A methodology is developed to assess distortions generated by scaling laws. This requires distinction between distortions inherent in a given scaling scheme [scaling law distortions (SLDs)] and the actual distortions (ADs) existing between prototypical behavior and the transposed behavior of a model. To develop the methodology, additional scaling concepts including “reference” and “resultant” similarity parameters and “required” and “assumed”’ conditions are defined. These parameters distinguish between conditions that are directly controllable and thus can be unequivocally determined by a scaling procedure and those that must rely to varying degrees on implied assumptions. In an illustrative example, it is shown that assessments of alternate scaling schemes can produce different conclusions when based on the results of an AD analysis as compared to an analysis of SLDs alone. The RELAP5 code is used to evaluate both prototypical and model behavior.