Psychologists use a number of methods to assess psychopathology and personality, including structured and unstructured interviews, brief self-rated and clinician-rated measures (such as the Beck Depression Inventory), projective techniques (e.g., the Rorschach Inkblot Technique), self-report personality inventories (e.g., the Minnesota Multiphasic Personality Inventory-2; MMPI-2), behavioral assessment methods (e.g., observational techniques and diary measures), outcome and treatment monitoring measures (e.g., the Outcome Questionnaire-45), and measures completed by peers or significant others (e.g., the Peer Inventory of Personality Disorders). This article describes research findings on the most scientifically controversial of these instruments, namely, projective techniques.
In comparison with other assessment methods, a clinician using a projective technique typically presents a client with an ambiguous stimulus (e.g., an inkblot), or asks the client to generate a response following open-ended instructions (e.g., “Draw a person”). Thus, for projective techniques, stimuli tend to be ambiguous and the nature of response options tends to be varied. The primary logic underlying these tests is the projective hypothesis—in the process of making sense of an ambiguous stimulus, the respondent presumably “projects” important aspects of his or her personality onto that stimulus. The test interpreter then works “in reverse” to infer the respondent’s personality traits.
Projective techniques can be placed into five broad categories: (1) association techniques including inkblot tests, (2) construction techniques including human figure drawing tests and story creation tests such as the widely used Thematic Apperception Test (TAT), (3) completion techniques including sentence completion tests, (4) arrangement or selection tests including the Szondi Test and the Luscher Color Test, and (5) expression techniques such as handwriting analysis, projective doll play, and puppetry.
A common argument made for using projective techniques is that they can circumvent a client’s conscious attempts to create a specific impression on these tests, as well as the client’s unconscious defenses. Clients may not know how to answer in a healthy or sick way when presented with an ambiguous stimulus such as an inkblot, so they will not be able to purposely overreport or underreport psychopathology. Similarly, when given an open-ended task such as drawing a human figure, clients may not be able to intentionally draw a figure that suggests that they have more or less psychopathology than they really have.
If clients’ responses are shaped by their personality traits and psychopathology, then projective techniques may be able to yield valuable information about the clients. In fact, a goal of using projectives is to learn things about clients that the clients themselves do not know. Proponents of projective techniques typically claim that the test results yield important insights into their clients’ unconscious processes.
Despite the claims made for projective techniques, the percentage of clinical psychologists using projective tests declined from 72% in 1986 to 39% in 2003. The decline in the popularity of projective techniques may be partly due to criticisms that have been leveled at them, which are described below. It can also be traced to the advent of managed care, which has made psychological testing, especially testing with questionable scientific support, less financially remunerative.
The scientific research on the three most popular projective instruments: the Rorschach, the TAT, and human figure drawings are described below. This entry concludes by addressing whether projective techniques can be used to circumvent a client’s defenses and evaluate unconscious motivations and conflicts.
Rorschach
Developed by the Swiss psychiatrist Hermann Rorschach in the 1920s, the Rorschach consists of 10 inkblots that are each printed on a separate card. During the first phase of the test (the response phase), the client is handed the cards one at a time and instructed to say what each blot resembles. In general, clients are allowed to give as many or as few responses as they wish. On average, clients make about 21 responses for the 10 cards.
During the second phase (the inquiry phase), each response is reviewed as the psychologist asks questions to clarify the nature of each response. For example, the psychologist tries to determine the exact location of the perception, and whether it was affected by the color or shading of the card or by other factors. Currently, the most widely used system for administering, scoring, and interpreting the Rorschach is Exner’s Comprehensive System (CS).
The controversy surrounding the Rorschach is complex and touches on a number of topics. Two of the most contentious topics are the adequacy of the normative data and evidence of the instrument’s validity.
Clinicians interpret test results for individual clients by comparing their test responses with normative data, that is, with those obtained for some meaningful comparison group (e.g., the general public or a group having a specific psychiatric diagnosis). Normative data were obtained for the Rorschach by administering it to individuals in the general community. Recent evidence indicates that the Comprehensive System norms are in error.
When results for relatively normal individuals were compared with results for the CS normative samples, the relatively normal individuals deviated markedly in a direction that supposedly indicates that they have serious psychopathology. Thus, interpreting the Rorschach using the CS norms tends to make many normal individuals look emotionally disturbed. This is likely to have harmful consequences. For example, clients in psychotherapy can be persuaded that they have problems they do not really have. Similarly, in forensic settings, clients may be given the Rorschach as part of a psychological evaluation. If they are described as having mental problems they do not really have, this could have deleterious consequences such as loss of custody of their children or denial of parole.
Validity describes whether tests measure what they are claimed to measure. There is general agreement among proponents and critics that at least some Rorschach scores are valid for their intended purposes. In particular, Rorschach scores are correlated with measurements of thought disorder, interpersonal dependency, and treatment outcome. They can also be correlated with diagnoses of mental disorders that are characterized by thought disorder such as schizophrenia. In addition, there is evidence that Rorschach scores are correlated with diagnoses of organic brain damage and measurements of hostility and anxiety.
At the same time, Rorschach advocates and Rorschach critics agree that many CS scores that clinicians interpret have not been studied adequately. These include the Coping Deficit Index, Obsessive Style Index, Hypervigilance Index, active-to-passive movement ratio, D-score, food content, anatomy and X-ray content, Intellectualization Index, and Isolation Index. In general, however, they disagree on the adequacy of the validity evidence. Critics judge a Rorschach score to have been adequately validated only if positive findings have been independently replicated, studies have been appropriately designed, and results for a test score have been consistently positive. Rorschach advocates are less explicit about the criteria they use, and at times they have defended the Rorschach on the basis of their clinical experiences.
Thematic Apperception Test
The Thematic Apperception Test (TAT), developed by Henry Murray and his student Christiana Morgan, consists of 31 cards, with a picture of an ambiguous situation on each card. For example, one picture shows a young woman grabbing the shoulders of a young man who seems to be pulling away from her. Respondents are instructed to look at each card and construct a story describing (a) the events occurring on the card, (b) what happened before, (c) what will happen in the future, and (d) what the characters are thinking and feeling. When testing a client, clinicians typically administer between 5 and 12 of the cards. The specific cards selected for administration varies greatly across examiners.
Rules have been developed for scoring TAT protocols, but clinicians rarely use these scoring systems. In general, validity is unimpressive when clinicians make judgments using the TAT, but better when judgments are made using formal scoring systems.
The best known scoring system for the TAT was developed by David McClelland, John Atkinson, Russell Clarke, and Edgar Lowell in 1953 to assess Henry Murray’s need for achievement. The scores generated by this system correlate modestly with real-world measures of achievement. The Social Cognition and Object Relations Scale, which can be used to assess object relations (i.e., mental representations of other people), also has received empirical support. Results have been mixed for other scoring systems, including the Defense Mechanisms Manual (a TAT-based index of the defense mechanisms of denial, projection, and identification).
Although a few TAT scoring systems appear to be promising, none of the scoring systems are appropriate for clinical use because adequate norms are not available. Norms are needed so that psychologists will not (a) diagnose psychopathology when it is not really present or (b) overlook psychopathology when it is present. Without adequate TAT norms the meaning of the TAT scores is not clear.
Human Figure Drawing Methods
Human figure drawing techniques include the Draw-A-Person, House-Tree-Person, Draw-A-Family, and Kinetic Family Drawing tests. The most frequently used version of the above drawing techniques is the Draw-A-Person test. Clients are simply instructed to draw a picture of a person, and afterward they are given a new sheet of paper and are instructed to draw a person of the opposite sex. For the House-Tree-Person test, a client may be instructed to draw a house, tree, and person all on one sheet of paper, or the client may be instructed to draw the house, tree, and person separately on three sheets of paper. For the Draw-A-Family, clients draw a picture of their whole family; for the Kinetic Family Drawing, clients are requested to draw a picture of their whole family “doing something.” In each instance the drawings are interpreted for signs of psychopathology.
There are two major approaches to scoring and interpreting human figure drawings. The sign approach draws inferences from isolated drawing features. For example, if a human figure drawing has large eyes, a clinician might infer that the client is suspicious or paranoid. Results for the sign approach have been largely negative; the overwhelming majority of validity coefficients have been negligible or zero. The global approach scores a number of features of a drawing and sums them to obtain a total score. Validity has been modest for the global approach. In particular, positive results were obtained when the Draw-A-Person: Screening Procedure for Emotional Disturbance was used to differentiate children and adolescents with conduct and oppositional disorders from normal children and adolescents.
Use of Projective Techniques
There is ample reason for skepticism concerning most widely used projective techniques. The lack of evidence of validity and of adequate normative data are stumbling blocks for many of the routine clinical uses of the Rorschach, TAT, and human figure drawing techniques. For example, in one study, Rorschach protocols were distributed to 90 psychologists who had completed formal Rorschach training. They were instructed to assign protocols to several diagnostic categories, including “normal.” Most protocols came from psychiatric patients, but some came from nonpatient adults. More than 75% of the normal adults were diagnosed as having mental disorders. For example, 12% were diagnosed as having major depression and 43% were diagnosed as having a personality disorder.
However, the results are not uniformly negative. Some Rorschach and human figure drawing scores are valid, as are some TAT scoring systems. If used cautiously, these scores can help clinicians describe some important psychological characteristics. For example, a client with a thought disorder may not be able to hide this condition on the Rorschach. Furthermore, the Washington University Sentence Completion Test (WUSCT), a measure of ego development (a construct that captures the complexity with which individuals view the world), has demonstrated impressive construct validity in a series of studies. Ironically, the WUSCT is almost never used in clinical practice, although it seems to have adequate norms and it is the most extensively validated projective technique.
There is a breach between clinical practice and scientific findings. Clinicians should not use a projective technique because it seems to work in their clinical practice. It can be difficult for clinicians to learn from clinical experience because accurate feedback is often unavailable, and also because their memories and cognitive processes are fallible. For example, clinicians who make a diagnosis typically do not receive objective feedback on the accuracy of the diagnosis. Because it can be difficult to learn from experience, clinicians should be guided by the scientific literature when deciding how to select and interpret projective techniques.
References:
- Exner, J. E. (2002). The Rorschach: Basic foundations and principles of interpretation (4th ed.). New York: Wiley.
- Garb, H. N. (2005). Clinical judgment and decision making. Annual Review of Clinical Psychology, 1, 67-89.
- Hamel, M., Shaffer, T. W., & Erdberg, P. (2000). A study of nonpatient preadolescent Rorschach protocols. Journal of Personality Assessment, 75, 280-294.
- Lilienfeld, S. O., Wood, J. M., & Garb, H. N. (2000). The scientific status of projective techniques. Psychological Science in the Public Interest, 1, 27-66.
- McClelland, D. C., Atkinson, J. W., Clark, R. A., & Lowell, E. L. (1953). The achievement motive. New York: Appleton-Century-Crofts.
- Meyer, G. J., & Archer, R. P. (2001). The hard science of Rorschach research: What do we know and where do we go? Psychological Assessment, 13, 486-502.
- Naglieri, J. A. (1988). Draw A Person: A quantitative scoring system. San Antonio, TX: Psychological Corporation.
- Shaffer, T. W., Erdberg, P., & Haroian, J. (1999). Current nonpatient data for the Rorschach, WAIS-R, and MMPI-2. Journal of Personality Assessment, 73, 305-316.
- Wood, J. M., Nezworski, M. T., Lilienfeld, S. O., & Garb, H. N. (2003). What’s wrong with the Rorschach? Science confronts the controversial inkblot test. San Francisco: Jossey-Bass.
See also: