Surveys to evaluate instructor effectiveness are commonly used in higher education. Yet the survey items included are often drawn from other surveys without reference to a theory of adult learning. The authors present the results from a validation study of such a theory-based survey. They evidence that an evaluation survey based on a theory that describes an effective instructor increases the possibility for validation, and that aligning survey items with a theoretically based description of effective teaching reduces the number of total items needed. As a result, there is less of a risk that halo rating errors will reduce subscore validity. (Contains 2 tables and 6 figures.)