This study reviews a number of model evaluation studies. The statistical measures of model accuracy and performance presented in these studies have been grouped into sets reflecting the recommendations of the American Meteorological Society's (AMS) Woods Hole Workshop on judging air quality model performance. These statistics have been further aggregated into a generalized set of statements that describe model performance and accuracy and are presented in tables organized according to model type and use. Results indicate only four or five models that have a significant evaluation history; in addition, they illustrate several recurring tendencies of past model evaluation studies and ways of avoiding certain deficiencies in model evaluation.