procedures were used with a less than 20% frequency (Clark & Zimmerman 1984). I have already discussed some disadvantages of portfolio review and interview as identification procedures. Informal art tests (tests designed and administered by local art teachers) and classroom teacher nominations also have disadvantages because such tests often are idiosyncratic and graded by subjective criteria unknown to the applicants. Similarly, unstructured classroom teacher nominations often are idiosyncratic and fail to identify giftedness in as many as 80% of cases (Pegnato & Birch 1959; Boston & Orloff 1980).
It was the purpose of the research reported here to develop and test a series of drawing tasks and a set of scoring criteria that could be used as a screening and identification instrument and as a research instrument in future inquiry. Drawing tasks were chosen because they are easy to administer, could be grounded in previous research by others, and are considered fundamental to other art abilities by many visual arts teachers. Two other art educators and I recently completed several years of inquiry about the history of research and testing in study of the development of children's abilities in the visual arts (Clark, Zimmerman & Zurmuehlen 1987). It was apparent from this inquiry that children's drawings were the most frequently cited source of evidence in a vast majority of studies carried out in both Europe and the United States from the early 1800s to the present. Drawing with a pencil or crayon on paper is the most frequently exercised art activity of most children and, therefore, the least intimidating art exercise for a testing situation.
For over seven years, from 1980 to 1987, various drawing tasks in various combinations were field tested in local classrooms at different levels, from the fifth grade to college graduate classes. Subjects included students in local