Interrater Reliability of the Handler Draw-a-Person Scoring

Abstract

Draw-a-Person test scoring systems developed prior to 1967 had low levels of predictive validity and reliability estimates for these scales are rarely found in the literature. Then, in 1967, Handler developed a set of scales for scoring anxiety indices in the Draw-a-Person test. The Handler scoring manual may offer more promise for future studies designed to validate the clinical and research use of human figure drawings. Reliability estimates reported in the present investigation provide independent support that Handler's scoring procedures have high interrater reliabilities. The Draw-a-Person indices had a median interrater reliability of .90, this being congruent with the median coefficient of .87 reported by Handler.