Teaching Students How To Construct An Assessment Battery

Important sources of information will of course come from an interview with the patient and possibly with members of his or her family. Important history data and observations from these contacts form a significant core of data, enriched, perhaps, by information derived from other case records and from referral sources. In our clinical setting patients take the SCL-90-R before the intake interview. This self-report instrument allows the interviewer to note those physical and emotional symptoms or problems the patients endorse as particularly difficult problems for them. This information is typically quite useful in structuring at least part of the interview. The construction of a comprehensive assessment battery is typically the next step.

What constitutes a comprehensive assessment battery differs from setting to setting. Certainly, adherents of the five-factor model would constitute an assessment battery differently than someone whose theoretical focus is object relations. However, there are issues involved in assessment approaches that are far more important than one's theoretical orientation. No test is necessarily tied to any one theory. Rather, it is the clinician who interprets the test who may imbue it with a particular theory.

It is difficult to describe a single test battery that would be appropriate for everyone, because referral questions vary, as do assessment settings and their requirements; physical and emotional needs, educational and intellectual levels, and cultural issues might require the use of somewhat different instruments. Nevertheless, there are a number of guiding principles used to help students construct a comprehensive assessment battery, which can and should be varied given the issues described above.

Beutler and Berren (1995) compare test selection and administration in assessment to doing research. They view each test as an "analogue environment" to be presented to the patient. In this process the clinician should ask which types of environments should be selected in each case. The instructions of each test or subtest are the clinician's way of manipulating these analogue environments and presenting them to the patient. Responding to analogue environments is made easier or more difficult as the degree of structure changes from highly structured to ambiguous or vague. Some people do much better in a highly structured environment, and some do worse.

Assessment is typically a stressful experience because the examiner constantly asks the patient to respond in a certain manner or in a certain format, as per the test instructions. When the format is unstructured there is sometimes less stress because the patient has many options in the way in which he or she can respond. However, there are marked differences in the ways that people experience this openness. For some people a vague or open format is gratifying, and for others it is terrifying. For this reason it is helpful to inquire about the patient's experience with each format, to determine its effect.

Beutler and Berren make another important point in reference to test selection: Some tests are measures of enduring internal qualities (traits), whereas others tap more transitory aspects of functioning (states), which differ for an individual from one situation to another. The clinician's job is to determine which test results are measuring states and which reflect traits. When a specific test in some way resembles some aspects of the patient's actual living environment, we can assume that his or her response will be similar to the person's response in the real-world setting (Beutler & Berren, 1995). The assessor can often observe these responses, which we call stylistic aspects of a person's personality.

One question to be answered is whether this approach is typical of the patient's performance in certain settings in the environment, whether it is due to the way in which the person views this particular task (or the entire assessment), or whether it is due to one or more underlying personality problems, elicited by the test situation itself. It is in part for this reason that students are taught to carefully record verbatim exactly what the patient answers, the extratest responses (e.g., side comments, emotional expressions, etc.), and details of how each task was approached.

Important aspects of test choice are the research that supports the instrument, the ease of administration for the patient, and the ability of the test to tap specific aspects of personality functioning that other instruments do not tap. We will discuss choosing a comprehensive assessment battery next.

First, an intellectual measure should be included, even if the person's intelligence level appears obvious, because it allows the assessor to estimate whether there is emotional interference in cognitive functioning. For this we recommend the WAIS-III or the WISC-III, although the use of various short forms is acceptable if time is an important factor. For people with language problems of one type or another, or for people whose learning opportunities have been atypical for any number of reasons (e.g., poverty, dyslexia, etc.), a nonverbal intelligence test might be substituted if an IQ measure is necessary. The Wechsler tests also offer many clues concerning personality functioning, from the pattern of interaction with the examiner, the approach to the test, the patient's attitude while taking it, response content, as well as from the style and approach to the subtest items, and the response to success or failure. If these issues are not relevant for the particular referral questions, the examiner could certainly omit this test completely.

Additionally, one or more self-report inventories should be included, two if time permits. The MMPI-2 is an extremely well-researched instrument that can provide a great deal more information than the patient's self-perception. Students are discouraged from using the descriptive printout and instead are asked to interpret the test using a more labor-intensive approach, examining the scores on the many supplementary scales and integrating them with other MMPI-2 data. The PAI is recommended because it yields estimates of adaptability and emotional health that are not defined merely as the absence of pathology, because it has several scales concerning treatment issues, and because it is psychometrically an extremely well-constructed scale. Other possible inventories include the Millon Clinical Multiaxial Inventory-III (MCMI-III), because it focuses on Axis II disorders, and the SCL-90-R or its abbreviated form, because it yields a comprehensive picture concerning present physical and emotional symptoms the patient endorses. There are a host of other possible self-report measures that can be used, depending on the referral issues (e.g., the Beck Depression Inventory and the Beck Anxiety Inventory).

Several projective tests are suggested, again depending upon the referral questions and the presenting problems. It is helpful to use an array of projective tests that vary on a number of dimensions, to determine whether there are different patterns of functioning with different types of stimuli. We recommend a possible array of stimuli that range from those that are very simple and specific (e.g., the Bender Gestalt Test) to the opposite extreme, the DAP Test, because it is the only test in the battery in which there is no external guiding stimulus. Between these two extremes are the TAT, in which the stimuli are relatively clear-cut, and the Rorschach, in which the stimuli are vague and unstructured.

Although the research concerning the symbolic content in the interpretation of the Bender Gestalt Test (BG) is rather negative, the test nevertheless allows the assessor a view of the person's stylistic approach to the rather simple task of copying the stimuli. The Rorschach is a multifaceted measure that may be used in an atheoretical manner, using the Comprehensive System (Exner, 1993), or it may be used in association with a number of theoretical approaches, including self psychology, object relations, ego psychology, and even Jungian psychology. In addition, many of the variables scored in the Exner system could very well be of interest to psychologists with a cognitive-behavioral approach. The Rorschach is a good choice as a projective instrument because it is multidimensional, tapping many areas of functioning, and because there has been a great deal of recent research that supports its validity (Baity & Hilsenroth, 1999; Ganellen, 1999; Kubeszyn et al., 2000; Meyer, 2000; Meyer, Riethmiller, Brooks, Benoit, & Handler, 2000; Meyer & Archer, 2001; Meyer & Handler, 1997; Viglione, 1999; Viglione & Hilsenroth, 2001; Weiner, 2001). There are also several well-validated Rorschach content scoring systems that were generated from research and have found application in clinical assessment as well (e.g., the Mutuality of Autonomy Scale, Urist, 1977; the Holt Primary Process Scale, Holt, 1977; the Rorschach Oral Dependency Scale, or ROD, Masling, Rabie, & Blondheim, 1967; and the Lerner Defense Scale, Lerner & Lerner, 1980).

The TAT is another instrument frequently used by psychologists that can be used with a variety of theoretical approaches. The TAT can be interpreted using content, style, and coherence variables. There are several interpretive systems for the TAT, but the systematic work of Cramer (1996) and Westen (1991a, 1991b; Westen, Lohr, Silk, Gold, & Kerber, 1990) seems most promising.

One assessment technique that might be new to some psychologists is the early memories technique, in which the assessor asks the patient for a specific early memory of mother, father, first day of school, eating or being fed, of a transitional object, and of feeling snug and warm (Fowler et al., 1995, 1996). This approach, which can also be used as part of an interview, has demonstrated utility for predicting details of the therapeutic relationship, and it correlates with a variety of other measures of object relations. The approach can be used with a wide variety of theoretical approaches, including various cognitive approaches (Bruhn, 1990, 1992).

Additional possible tests include various drawing tests (e.g., the DAP test and the Kinetic Family Drawing Test, or K-F-D). The research findings for these tests are not consistently supportive (Handler, 1996; Handler & Habenicht, 1994). However, many of the studies are not well conceived or well controlled (Handler & Habenicht, 1994; Riethmiller & Handler, 1997a, 1997b). The DAP and/or the K-F-D are nevertheless recommended for possible use for the following reasons:

1. They are the only tests in which there is no standard stimulus to be placed before the patient. This lack of structure is an asset because it allows the examiner to observe organizing behavior in situations with no real external structure. Therefore, the DAP taps issues concerning the quality of internal structuring. Poor results are often obtained if the person tested has problems with identity or with the ability to organize self-related issues.

2. Drawing tests are helpful if the person being assessed is not very verbal or communicative, because a minimum of talking is required in the administration.

3. Drawing tests are quick and easy to administer.

4. Drawings have been demonstrated to be excellent instruments to reflect changes in psychotherapy (Handler, 1996; Hartman & Fithian, 1972; Lewinsohn, 1965; Maloney & Glasser, 1982; Robins, Blatt, & Ford, 1991; Sarel, Sarel, & Berman, 1981; Yama, 1990).

Much of the research on drawing approaches is poorly conceived, focusing on single variables, taken out of context, and interpreted with a sign approach (Riethmiller & Handler, 1997a, 1997b). There is also confusion between the interpretation of distortions in the drawings that reflect pathology and those that reflect poor artistic ability. There are two ways to deal with these problems. The first is to use a control figure of equal task difficulty to identify problems due primarily to artistic ability. Handler and Reyher (1964, 1966) have developed such a control figure, the drawing of an automobile. In addition, sensitizing students to the distortions produced by people with pathology and comparing these with distortions produced by those with poor artistic ability helps students differentiate between those two situations (Handler & Riethmiller, 1998).

A sentence completion test (there are many different types) is a combination of a self-report measure and a projective test. The recommended version is the Miale-Holsopple Sentence Completion Test (Holsopple & Miale, 1954) because of the type of items employed. Patients are asked to complete a series of sentence stems in any way they wish. Most of the items are indirect, such as "Closer and closer there comes ...," "A wild animal...," and "When fire starts " Sentence completion tests also provide information to be followed up in an interview.

Was this article helpful?

0 0
Anxiety Away

Anxiety Away

The strategies revealed within Anxiety Away are fast acting, simple and guaranteed to work even if you have suffered from anxiety for a long time!

Get My Free Ebook


Post a comment