This study discusses the National Assessment of Educational Progress's national survey of educational achievement in art and demonstrates that it is feasible to evaluate children's drawings obtained with free response assessment techniques. Four art production exercises, each calling for pencil drawing responses, were administered during 1974-1975 to a sampling of 9-, 13-, and 17-year-olds. Four scoring systems were developed. To be considered valid each dimension within a scoring system had to be made mutually exclusive and constant across age groups, cover the variety of responses generated by different ages, and be sufficiently replicable to be useful for assessing changes in performance. Persons of high academic ability who had previous experience in judging and codifying written test responses were selected as scorers. None of them had any art training. Two additional persons with advance training in art were selected to train the scorers. Results show that the four scoring systems met the conditions necessary for their adequacy. Careful attention to scoring system exercise development and scorer selection and training makes it possible to test art achievement. The study includes one sample exercise and a scoring guide. Appendices include scored sample responses. (Author/RM) Primary type of information provided by report: Procedures (Scoring); Results (Respondent Examples).