A new research paradigm to assess clinicians' ability to interpret draw-a-person test protocols
The diagnostic use of figure drawings has had a short history, beginning with Machover's l949 monograph. However, the specificity of her "sign and trait" interpretive equations supported a boom of validity studies which used these equations as research hypotheses. Yet, despite 30 years of "sign and trait" research, there is a paucity of empirical evidence supporting the Draw-A-Person Test. This was judged due to numerous, fatal methodological problems of the "sign and trait" research strategy, including an inherent rigidity which compromised its productivity and, on close inspection, its failure to actually measure whether clinicians can interpret the Draw-A-Person Test. Reliability studies have found global characteristics of human figure drawings stable, but the small details frequently used as "signs" were highly unreliable. As the Draw-A-Person Test remains one of the most widely used projective assessment devices, the critical need for empirical support of clinicians' ability to accurately use the test is apparent.
This research attempted to use a novel methodological approach which provided a direct measure of clinicians' abilities to interpret Draw-A-Person Test protocols and which was not subject to the disabling methodological flaws of "sign-trait" validity studies.
Five Ph.D. clinical psychologists experienced in interpreting the test were asked to respond to 45-item inventories on each of four patients on the basis of the patients' protocols. These patients had been carefully studied in advance by the experimenter, and the clinicians' responses were compared with this foreknowledge. A control group of five experienced clinicians responded to the inventories without the aid of the patients' protocols. The inventories were divided into five "areas" of personality description, Defense, Affect, Conflict, Self- Image, and Environment and into three "levels" of interpretive generalization, Global, Major Details, and Specific Details. While this inventory structure assured broad, relevant, and comparable descriptions of each individual patient, it allowed an investigation of this "topography" of clinicians' interpretive abilities, in order to outline specific areas of strength or weakness.
Both the experimental results and the success of the research methodology were evaluated. There were no significant differences between control and experimental group performances in respect to the entire battery, individual patients, "areas," or "levels," indicating the clinicians could not interpret Draw-A-Person Test protocols. The experimental methodology displayed potential sources of error, but these were felt amenable to experimental influence and control which were largely accomplished in this experiment. However, as only 10 subjects were used and only the inventories of neurotic patients yielded unbiased results, the consistent outcome of this pilot experiment can only be considered provocative and not definitive.
Thesis78.M592.pdf
4.12 MB
Adobe PDF
822db18bdb74cb12b4b939bf8e913c64