Practical and Philosophical Jibberish about Testing

There is no consensus of opinion on the "correct" way to write a psychological report. The report will vary considerably depending upon the training and experience of the psychologist, the theoretical approach (or lack thereof) to assessment, the problems presented by the client, the training/ orientation of the referral source, the intended use for the results, etc. 

The purpose of the current meanderings is to discuss some of the possible approaches to assessment and report writing. The goal is to offer an improved sense of purpose and direction to the process of producing a useful psychological evaluation. 

This discussion will begin with a presentation of several "models" of psychological reports. The strengths and weaknesses of each model will be noted. An argument will be made for using one particular model of report (i.e. the Hypothesis Testing model) over others for most evaluations conducted at large inpatient psychiatric hospitals. The discussion will then turn to the different "levels" of reports. I have chosen the word "levels" to differentiate these reports in order to communicate a personal bias toward the third, or highest, level, which I believe is a goal worth pursuing in most reports. Finally, I have included a link to a page describing a recommended report format which has evolved with input from a number of peers. 
 

The three models for psychological reports to be discussed are the Test Oriented Model, the Domain Oriented Model, and the Hypothesis Oriented Model. 

In the Test Oriented Model, results are discussed on a test-by-test basis. Each test is listed by name and significant results for that test are presented. Each test is generally discussed in a separate paragraph. Little or no effort is made to compare and contrast data between the various tests (at least not in the "Results of Assessment" section). The strength of this approach is that it makes clear the source of each piece of data. This could be important in certain settings, such as forensic reports. The weakness of this model is that the reader's attention becomes focussed on the tests, rather than on the client's adaptive functioning. It also communicates to the reader that psychological assessment is a low-level, technical skill which involves little more than giving the test and copying some interpretive statements out of a manual. It ignores the role of the psychologist as the integrator of the test data; a professional who brings to bear his knowledge of how the test was constructed, how it was normed, limits to generalizability of test data, and how to use the data in a theoretical/conceptual manner to better understand the client. The Test Oriented Model was used extensively in past, but has become increasingly unpopular in recent years. 

In the Domain Oriented Model, results are grouped according to abilities or "functional domains". Separate paragraphs are usually devoted to such topics as intellectual ability, interpersonal skills, psychosocial stressors, coping techniques, intrapersonal needs, motivational factors, depression, psychotic features, etc. This model is useful when there is no specific referral question and you're not certain what use will be made of your data. For example, little background information may be available on a newly admitted patient. You're not sure why he was admitted or what factors precipitated the admission. Therefore, its hard to know which portions of your data will be useful to the treatment team. The Domain Oriented Model is also common in neuropsychological reports, where a variety of providers may eventually become involved in the case. Each provider will focus on separate parts of the report to assist in a specific aspect of intervention. This approach is also helpful when assessment is being used to monitor treatment progress. It allows you to monitor changes in the client's functioning across a wide variety of areas. The weakness of the Domain Oriented approach is that the reader may be presented with a lot of information that has little relevance to his intended intervention. He may become so distracted by parts of the report he doesn't understand, that he fails to focus on information which could be helpful to him. This model is sometimes pejoratively referred to as a "shotgun" approach, referring to its apparent effort to hit all the possible target issues. 

In the Hypothesis Testing Model, results are focussed on possible answers to the referral question(s). The idea is to present a hypothesis in the "Purpose for Evaluation" section, then present data systematically to support or refute the hypothesis. Separate paragraphs in the "Results of Evaluation" section address theoretical/conceptual issues by integrating data from the history, mental status exam and behavioral observations with data from all the tests. Tests are rarely mentioned by name. For example, information from scale 2 on the MMPI-2 may be combined with interpretive data from the MCMI dysthymia scale. If the integration of this information is consistent with the history and the mental status exam, it is included in a paragraph dealing with depression. The strength of this model lies in its efficiency and concise focus on the referral problem. The reader isn't distracted by unrelated details. The primary weakness of the model is that you don't report some of the information which is unrelated to the "purpose of the evaluation" but which could potentially be useful to other disciplines.  
  

Having covered the issue of report Models, this discussion will now turn to "levels" of reports. Three levels of reports will be covered. 

A "Level One" report is the copied-out-of-the-manual level. The interpretive data come directly from the manual (or computer print out) and usually follow the format...."People with similar profiles tend to....". There is often little or no effort to personalize the evaluation or to differentiate which of the interpretive hypotheses are (or are not) consistent with the history and mental status exam. This makes for a conceptually weak report and may actually do more harm than good for the client. Keep in mind that many of the referral agents will have little understanding of the limits to generalizability and external validity of "raw" test data. This level of report is only appropriate when there are extenuating circumstances which make it impossible to interview the patient or to obtain background information. In those cases the report should be clearly qualified with a statement to the effect that...."These results represent a blind interpretation of test data and should be considered tentative until confirmed by subsequent clinical data or background information". 

A "Level Two" report represents the minimum level of conceptual input which should be used for most purposes. Of all the possible interpretive hypotheses generated by the test, the only ones included in the "Results of Evaluation" are those that have been confirmed (either by the history or in the clinical interview). It is best to avoid comments like "persons who obtain similar results tend to....". Rather, personalize the interpretative statement with something like.... "Mr. Jones tends to....". Then follow your interpretive comment with an example which is specific to the client. For example: 
 

• describe the nature of the problem and how it developed over time 
• describe factors which influence and reinforce the problem 
• describe any recent exacerbating factors which led to the referral 
• provide suggestions for intervention based on the client's strengths, weaknesses, and coping skills; 

Potential uses for the evaluation: 
     1. Answer the referral question(s). 
     2. Provide the referral source with information about the client which 
         can be used in his treatment. 
     3. To provide treatment recommendations and/or facilitate diagnosis. 
     4. Establish an objective record of the patient's current functional and 
         adaptive status, which can serve as a baseline for comparison after 
         treatment has been implemented. 
   

1. Specify the purpose of clinical assessment: 
          a. Diagnosis / Screening for psychopathology 
          b. Problem description / Case conceptualization 
          c. Selection of treatment procedures 
          d. Evaluate treatment outcome and progress 

2. Specify all sources of clinical information: 
           a. Clinical Interview 
           b. Observation of the client 
           c. Interview with significant others 
           d. Psychological testing 
           e. Self-report / Other-report forms 
           f. Medical records 
           g. Physiological measurement 

3. A "conceptualization" or "dynamic formulation" is a general impression of what the client is like as a person and how he got to be that way. This conceptualization is theory based and may emphasize behavioral, social learning, dynamic, family systems, or other theories. A relatively behavioral conceptualization will focus on explicit problem description, factors which elicit the problem behavior, and factors which maintain or reinforce the behavior. A relatively dynamic conceptualization will focus on the nature of the conflict the client is experiencing, the defenses he uses to cope with anxiety, his style of coping with stress, and his attitudes towards himself and others. 

4. The goal of a thorough assessment is to identify the type of conceptualization which will be most beneficial to a given client. This decision will be influenced by the client's intelligence, personality/coping style, capacity for insight, and motivation for intervention. However, it must also take into consideration the skill level and theoretical orientation of the intended therapist or counselor. Ideally, the case needs to be conceptualized in a way that both the client and his counselor can easily understand and which will direct the course of treatment.