Lauren Hood

Olivia Flach

Lauren Champagne

Vanessa Flores


Research Paper on a Study of a Cautionary Study: Unwarranted

Interpretations of the Draw-A-Person Test.


Often, clinicians use tests that they have no training in or that have uncertain validity.  The APA guidelines list that clinicians only give services in areas that they have an adequate level of experience and training in.  A second obligation that clinicians have is that they interpret and place weights on assessments based on the testís validity.  This study was concerned with whether clinicians were following the APA guidelines or were interpreting assessments without adequate training or validity of the assessment. 

            The researchers hypothesized that clinicians, inexperienced and experienced, would form interpretations of a draw-a-person test that was included in a case study they were asked to review.  The researchers also hypothesized that clinicians would not express concerns with the lack of validity of the draw-a-person test.


            The study looked at 18 clinical psychologists and 18 counseling psychologists residing in Canada.  Half of the groupís members were psychologists with at least 5 years experience as a practicing clinician and the other half were currently enrolled in a doctorial or masters program in counseling psychology.  None of the students had any training in projective tests and only half of the experienced group had training in projective testing.  All participants were recruited for the study through fliers and phone calls.  This study was a descriptive correlational design because there was no random assignment used and the independent variable was not manipulated.

            Participants were given a case file on a middle-aged man who had been diagnosed with passive-aggressive personality disorder and elements of borderline personality disorder.  Statements that alluded to these diagnoses were placed throughout the file, at a rate of four to a page.  At the end of the file, a draw-a-person test was placed, which was drawn by someone unrelated to the study.  An experimenter audio taped the responses the clinicians gave on the case file, and the clinicians rated on whether they used the draw-a-person test to draw inferences and if those inferences were of a grand or specific nature.  The audio tapes from the sessions were transcribed, but the authors did not make clear what type of scales or rating systems they used, if they used any at all.


            The researchers found that of the 21 participants who had training in projective testing, 17 made inferences off of the draw-a-person test.  Of the 15 participants who had no training, only 5 made inferences of the draw-a-person test.  Only 1 participant out of the 36 mentioned that there was a lack of validity for the draw-a-person test.  Of the 22 participants who used the draw-a-person test as part of their assessment, 7 used global measures; whereas the other 20 used feature specific inferences to either substantiate their diagnosis, or to make new diagnoses from the draw-a-person test.  It was noted that of the 7 participants that used global measures, 5 also used feature specific measures.  Participants tended to look for evidence in the draw-a-person test to substantiate their hypotheses.


            The results concurred with previous research by fine (1992), which found that psychologists with little experience made gross generalizations on draw-a-person tests, even though they had not been trained on these assessments, and the tests lacked validity.  A study by Nisbett and Ross (1980; as cited in Smith and Dumont, 1995), which found that participants would scan draw-a-person test for anything that could concur with their previous findings, also was consistent with the results found in the study.  The picture used in this study was drawn by someone unrelated to the case file, which should have eliminated any possible connection to the case file. 

            The participants should have been randomly selected from the population, instead of calling possible participants, or placing fliers in areas frequented by psychologists.  Also, the participants should have been allowed to complete their evaluation away from the experimenter recording them.  This could have possibly led to the participants feeling pressured to evaluate all of the information provided in the file.  The authors also list that the participants could have been influenced by demand characteristics to respond to the draw-a-person test.  Another problem with this study is that these results are not necessarily correlated with whether the participants would actually use projective testing in their own practices, as was pointed out by the authors.           















Smith, D., & Dumont F. (1995).  A cautionary study: Unwarranted interpretations of the

draw-a-person test.  Professional Psychology: Research and Practice, 23, 298-303.