IN THE BEGINNING 166 Who Is the Client? 166 Informed Consent 167 Test User Competence 168 Planning the Evaluation 170 Selection of Instruments 170 Adequacy of Instruments 170
Appropriate Assessment in a Multicultural Society 170 Getting Around Language Barriers 171 What's in a Norm? 171 Bases for Assessment 172 CONDUCTING THE EVALUATION 172 Conduct of the Assessment 172 Data Collection and Report Preparation 173
Automated Test Scoring and Interpretation 173 AFTER THE EVALUATION 173 Feedback Requests 174 Requests for Modification of Reports 174 Release of Data 175 Test Security 175 SPECIAL ISSUES 176
The Testing Industry 176 Automated Testing Services 177 High-Stakes Testing 177 Teaching Psychological Testing 178 RECOMMENDATIONS 179 REFERENCES 179
Psychological assessment is unique among the services provided by professional psychologists. Unlike psychotherapy, in which clients may come seeking help for themselves, psychological evaluation services are seldom performed solely at the request of a single individual. In the most common circumstances, people are referred to psychologists for assessment by third parties with questions about school performance, suitability for potential employment, disability status, competence to stand trial, or differential clinical diagnosis. The referring parties are invariably seeking answers to questions with varying degrees of specificity, and these answers may or may not be scientifically addressable via the analysis of psychometric data. In addition, the people being tested may benefit (e.g., obtain remedial help, collect damages, or gain a job offer) or suffer (e.g., lose disability benefits, lose custody of a child, or face imprisonment) as a consequence of the assessment, no matter how competently it is carried out.
Psychological assessment is founded on a scientific base and has the capacity to translate human behavior, characteristics, and abilities into numbers or other forms that lend themselves to description and comparison across individuals and groups of people. Many of the behaviors studied in the course of an evaluation appear to be easily comprehensible to the layperson unfamiliar with test development and psychometrics
(e.g., trace a path through a maze, perform mental arithmetic, repeat digits, or copy geometric shapes)—thereby implying that the observed responses must have some inherent validity for some purpose. Even common psychological assessment tasks that may be novel to most people (e.g., Put this unusual puzzle together quickly. What does this inkblot look like to you?) are imbued by the general public with some implied valuable meaning. After all, some authority suggested that the evaluation be done, and the person conducting the evaluation is a licensed expert. Unfortunately, the statistical and scientific underpinnings of the best psychological assessments are far more sophisticated than most laypersons and more than a few psychologists understand them to be. When confronted with an array of numbers or a computer-generated test profile, some people are willing to uncritically accept these data as simple answers to incredibly complex questions. This is the heart of the ethical challenge in psychological assessment: the appropriate use of psychological science to make decisions with full recognition of its limitations and the legal and human rights of the people whose lives are influenced.
In attempting to address the myriad issues that challenge psychologists undertaking to conduct assessments in the most ethically appropriate manner, it is helpful to think in terms of the prepositions before, during, and after. There are ethical considerations best addressed before the assessment is begun, others come into play as the data are collected and analyzed, and still other ethical issues crop up after the actual assessment is completed. This chapter is organized to explore the ethical problems inherent in psychological assessment using that same sequence.
In the beginning—prior to meeting the client and initiating data collection—it is important to consider several questions. What is being asked for and by whom (i.e., who is the client and what use does that person hope to make of the data)? Is the proposed evaluator qualified to conduct the evaluation and interpret the data obtained? What planning is necessary to assure the adequacy of the assessment? What instruments are available and most appropriate for use in this situation?
As one proceeds to the actual evaluation and prepares to undertake data collection, other ethical issues come to the fore. Has the client (or his or her representative) been given adequate informed consent about both the nature and intended uses of the evaluation? Is it clear who will pay for the evaluation, what is included, and who will have access to the raw data and report? What are the obligations of the psychologist with respect to optimizing the participants' performance and assuring validity and thoroughness in documentation? When the data collection is over, what are the examiner's obligations with respect to scoring, interpreting, reporting, and explaining the data collected?
Finally, after the data are collected and interpreted, what are the ongoing ethical responsibilities of the psychologist with respect to maintenance of records, allowing access to the data, and providing feedback or follow-up services? What is appropriate conduct when one psychologist is asked to review and critique another psychologist's work? How should one respond upon discovery of apparent errors or incompetence of a colleague in the conduct of a now-completed evaluation?
This chapter was completed during a period of flux in the evolution of the American Psychological Association's (APA) Ethical Standards of Psychologists and Code of Conduct. The current, in-force version of this document was adopted in 1992 (APA, 1992), but work on a revision is nearing completion (Ethics Code Task Force, 2001); a vote on adoption of the newest revision should take place in 2002. Many of the proposed revisions deal with psychological assessment. The Ethics Code Task Force (ECTF) charged with revising the code set out to avoid fixing that which is not broken—that is, the proposed changes focused on problems raised in a critical incident survey of psychologists, intended to ascertain where clarification and change were needed to improve professional and scientific practices. We have chosen to focus this chapter on key ethical foundations, but we also identify areas of controversy throughout. Whenever possible, we discuss trends and likely policy decisions based on the work of the ECTF; however, readers are encouraged to visit the APA Web site (http://www.apa.org/ethics) to view the most current version of the code and standards, which will continue to evolve long after this chapter appears in print.
IN THE BEGINNING Who Is the Client?
The first step in undertaking the evaluator role is often seductively automatic. The simple act of accepting the referral and setting up the appointment may occur almost automatically; not much thought may be devoted to the question of what specific duties or professional obligations are owed to which parties. Is the client simply the person to be evaluated, or are there layers of individuals and institutions to whom the psychologist owes some degree of professional obligation? For example, is the person to be evaluated a legally competent adult? If not—as in the case of children or dependent adults— the party seeking the evaluation may be a parent, guardian, government agency, institution, or other legally responsible authority. The evaluator must pause to consider what rights each layer of authority has in terms of such factors as the ability to compel cooperation of the person to be assessed, the right of access to test data and results, and the right to dictate components of the evaluation or the manner in which it is conducted. Sometimes there is uniform agreement by all parties, and no conflicts of interest take place. In other circumstances, the person being evaluated may have had little choice in the matter or may wish to reserve the right to limit access to results of the evaluation. In still other instances, there may be direct conflicts between what one party is seeking and the objectives of another party with some degree of client status.
Evaluations conducted in the context of the educational system provide a good example of the complex layers of client status that can be involved. Suppose that a schoolchild is failing and an assessment is requested by the child's family for use in preparation of an individualized educational plan (IEP) as specified under state or federal special education laws. If a psychologist employed by the public school system undertakes the task, that evaluator certainly owes a set of professional duties (e.g., competence, fairness, etc.) to the child, to the adults acting on behalf of the child (i.e., parents or guardians), to the employing school system, and—by extension—to the citizens of the community who pay school taxes. In the best of circumstances, there may be no problem—that is to say, the evaluation will identify the child's needs, parents and school will agree, and an appropriate effective remediation or treatment component for the IEP will be put in place.
The greater ethical challenge occurs when parents and school authorities disagree and apply pressure to the evalua-tor to interpret the data in ways that support their conflicting demands. One set of pressures may apply when the psychologist is employed by the school, and another may apply when the evaluation is funded by the parents or another third party. From an ethical perspective, there should be no difference in the psychologist's behavior. The psychologist should offer the most scientifically and clinically sound recommendations drawn from the best data while relying on competence and personal integrity without undue influence from external forces.
Similar conflicts in competing interests occur frequently within the legal system and the business world—a person may agree to psychological assessment with a set of hopes or goals that may be at variance or in direct conflict with the data or the outcomes desired by another party in the chain of people or institutions to whom the evaluator may owe a professional duty. Consider defendants whose counsel hope that testing will support insanity defenses, plaintiffs who hope that claims for psychological damages or disability will be supported, or job applicants who hope that test scores will prove that they are the best qualified. In all such instances, it is critical that the psychologist conducting the assessment strive for the highest level of personal integrity while clarifying the assessment role and its implications to all parties to whom a professional duty is owed.
Other third parties (e.g., potential employers, the courts, and health insurers) are involved in many psychological assessment contexts. In some cases, when the psychologist is an independent practitioner, the third party's interest is in making use of the assessment in some sort of decision (e.g., hiring or school placement); in other cases, the interest may simply be contract fulfillment (e.g., an insurance company may require that a written report be prepared as a condition of the assessment procedure). In still other situations, the psychologist conducting the evaluation may be a full-time employee of a company or agency with a financial interest in the outcome of the evaluation (e.g., an employer hoping to avoid a disability claim or a school system that wishes to avoid expensive special education placements or services). For all these reasons, it is critical that psychologists clearly conceptualize and accurately represent their obligations to all parties.
The current revision of the ECTF (2001) has added a proposed standard referring specifically to obtaining informed consent for psychological assessment. The issue of consent is also discussed extensively in the professional literature in areas of consent to treatment (Grisso & Appelbaum, 1998) and consent for research participation, but references to consent in the area of psychological assessment had been quite limited. Johnson-Greene, Hardy-Morais, Adams, Hardy, and Bergloff (1997) review this issue and propose a set of recommendations for providing informed consent to clients. These authors also propose that written documentation of informed consent be obtained; the APA ethical principles allow but do not require this step. We believe that psychologists would be wise to establish consistent routines and document all discussions with clients related to obtaining consent, explaining procedures, and describing confidentiality and privacy issues. It is particularly wise to obtain written informed consent in situations that may have forensic implications, such as personal injury lawsuits and child custody litigation.
Psychologists are expected to explain the nature of the evaluation, clarify the referral questions, and discuss the goals of the assessment, in language the client can readily understand. It is also important to be aware of the limitations of the assessment procedures and discuss these procedures with the client. To the extent possible, the psychologist also should be mindful of the goals of the client, clarify misunderstandings, and correct unrealistic expectations. For example, parents may seek a psychological evaluation with the expectation that the results will ensure that their child will be eligible for a gifted and talented program, accident victims may anticipate that the evaluation will document their entitlement to damages, and job candidates may hope to become employed or qualify for advancement. These hopes and expectations may come to pass, but one cannot reasonably comment on the outcome before valid data are in hand.
Whether the assessment takes place in a clinical, employment, school, or forensic settings, some universal principles apply. The nature of assessment must be described to all parties involved before the evaluation is performed. This includes explaining the purpose of the evaluation, who will have access to the data or findings, who is responsible for payment, and what any relevant limitations are on the psychologist's duties to the parties. In employment contexts, for example, the psychologist is usually hired by a company and may not be authorized to provide feedback to the employee or candidate being assessed. Similarly, in some forensic contexts, the results of the evaluation may ultimately be open to the court or other litigants over the objections of the person being assessed. In each case, it is the psychologist's responsibility to recognize the various levels of people and organizations to whom a professional duty may be owed and to clarify the relationship with all parties at the outset of the assessment activity.
The key elements of consent are information, understanding, and voluntariness. First, do the people to be evaluated have all the information that might reasonably influence their willingness to participate? Such information includes the purpose of the evaluation, who will have access to the results, and any costs or charges to them. Second, is the information presented in a manner that is comprehensible to the client? This includes use of appropriate language, vocabulary, and explanation of any terms that are confusing to the client. Finally, is the client willing to be evaluated? There are often circumstances in which an element of coercion may be present. For example, the potential employee, admissions candidate, criminal defendant, or person seeking disability insurance coverage might prefer to avoid mandated testing. Such prospective assessment clients might reluctantly agree to testing in such a context because they have no other choice if they wish to be hired, gain admission, be found not criminally responsible, or adjudicated disabled, respectively. Conducting such externally mandated evaluations do not pose ethical problems as long as the nature of the evaluation and obligations of the psychologist are carefully delineated at the outset. It is not necessary that the person being evaluated be happy about the prospect—he or she must simply understand and agree to the assessment and associated risks, much like people referred for a colonoscopy or exploratory surgery.
Additional issues of consent and diminished competence come into play when a psychologist is called upon to evaluate a minor child, an individual with dementia, or other persons with reduced mental capacity as a result of significant physical or mental disorder. When such an evaluation is undertaken primarily for service to the client (e.g., as part of treatment planning), the risks to the client are usually minimal. However, if the data might be used in legal proceedings (e.g., a competency hearing) or in any way that might have significant potentially adverse future consequences that the client is unable to competently evaluate, a surrogate consent process should be used—that is to say, a parent or legal guardian ought to be involved in granting permission for the evaluation. Obtaining such permission helps to address and respect the vulnerabilities and attendant obligations owed to persons with reduced personal decision-making capacities.
Before agreeing to undertake a particular evaluation, the clinician should be competent to provide the particular service to the client, but evaluating such competence varies with the eye of the beholder. Psychologists are ethically bound not to promote the use of psychological assessment techniques by unqualified persons, except when such use is conducted for training purposes with appropriate supervision. Ascertaining what constitutes competence or qualifications in the area of psychological assessment has been a difficult endeavor due to the complex nature of assessment, the diverse settings and contexts in which psychological assessments are performed, and the differences in background and training of individuals providing psychological assessment services. Is a doctoral degree in clinical, counseling, or school psychology required? How about testing conducted by licensed counselors or by individuals with master's degrees in psychology? Are some physicians competent to use psychological tests? After all, Hermann Rorschach was a psychiatrist, and so was J. Charnley McKinley, one of the two originators of the Minnesota Multiphasic Personality Inventory (MMPI). Henry Murray, a nonpsychiatric physician, coinvented the thematic apperception test (TAT) with Christiana Morgan, who had no formal training in psychology.
Historically, the APA addressed this issue only in a very general manner in the Ethical Principles and Code of Conduct for Psychologists. In earliest versions of the Ethical Standards of Psychologists (APA, 1953), the ethical distribution and sale of psychological tests was to be limited to unspecified "qualified persons." A system of categorization of tests and concordant qualifications that entailed three levels of tests and expertise was subsequently developed. Vocational guidance tools, for example, were at the low end in terms of presumed required expertise. At the high end of required competence were tests designed for clinical assessment and diagnoses such as intelligence and personality assessment instruments. The rationale involved in this scheme was based on the need to understand statistics, psychopathology, and psychometrics in order to accurately draw clinical inference or make actuarial predictions based on the test data. Although the three-tier system is no longer discussed in APA ethical standards, it was adopted by many test publishers, and variations continue in use. When attempting to place orders for psychological test materials, would-be purchasers are often asked to list their credentials, cite a professional license number, or give some other indication of presumed competence. Decisions about the actual sale are generally made by the person processing the order—often a clerk who has little understanding of the issues and considerable incentive to help the test publisher make the sale. Weaknesses and inconsistencies in the implementation of these criteria are discussed in a recent Canadian study (Simner, 1994).
Further efforts to delineate qualifications of test users included the lengthy efforts of the Test User Qualifications Working Group (TUQWG) sponsored by an interdisciplinary group, the Joint Committee on Testing Practices, which was convened and funded by the APA. To study competence problems, this group of professionals and academics attempted to quantify and describe factors associated with appropriate test use by using a data-gathering (as opposed to a specification of qualifications) approach (Eyde, Moreland, Robertson, Primoff, & Most, 1988; Eyde et al., 1993; Moreland, Eyde, Robertson, Primoff, & Most, 1995).
International concerns regarding competent test use— perhaps spurred by expansion of the European Economic Community and the globalization of industry—prompted the British Psychological Society (BPS) to establish a certification system that delineates specific competencies for the use of tests (BPS, 1995, 1996) in employment settings. Under this system, the user who demonstrates competence in test use is certified and listed in an official register. Professionals aspiring to provide psychological assessment services must demonstrate competence with specific tests to be listed. The International Test Commission (2000) recently adopted test-use guidelines describing the knowledge, competence, and skills.
Within the United States, identifying test user qualifications and establishing competence in test use have been complicated by political and practical issues, including the potential for nonprofit professional groups to be accused of violating antitrust laws, the complexity of addressing the myriad settings and contexts in which tests are used, and the diversity of experience and training in assessment among the professionals who administer and interpret psychological tests. Further complicating matters is the trend in recent years for many graduate programs in psychology to de-emphasize psychological assessment theory and practice in required course work, producing many licensed doctoral psychologists who are unfamiliar with contemporary measurement theory (Aiken, West, Sechrest, & Reno, 1990).
In October of 1996, the APA formed a task force to develop more specific guidelines in the area of test user qualifications and competence (Task Force on Test User Qualifications). Members of the task force were selected to represent the various settings and areas of expertise in psychological assessment (clinical, industrial/organizational, school, counseling, educational, forensic, and neuropsychological). Instead of focusing on qualifications in terms of professional degrees or licenses, the task force elected to delineate a core set of competencies in psychological assessment and then describe more specifically the knowledge and skills expected of test users in specific contexts. The core competencies included not only knowledge of psychometric and measurement principles and appropriate test administration procedures but also appreciation of the factors affecting tests' selection and interpretation in different contexts and across diverse individuals.
Other essential competencies listed by the task force included familiarity with relevant legal standards and regulations relevant to test use, including civil rights laws, the Americans with Disabilities Act (ADA), and the Individuals with Disabil ities Education Act (IDEA). Public commentary in response to the Task Force's preliminary report emphasized the variety of settings and purposes involved and generally argued against a focus on degrees and licenses as providing adequate assurance of competence. The Task Force delineated the purposes for which tests are typically used (e.g., classification, description, prediction, intervention planning, and tracking) and described the competencies and skills required in specific settings (e.g., employment, education, career-vocational counseling, health care, and forensic). The task force's recommendations were drafted as aspirational guidelines describing the range of knowledge and skills for optimal test use in various contexts. The task force report also expressed the hope that the guidelines would serve to bolster the training of future psychologists in the area of assessment. The final report of the task force was approved by the APA Council of Representatives in August 2000.
The chief ethical problems in this area involve matters of how to objectively determine one's own competence and how to deal with the perceived lack of competence in others whose work is encountered in the course of professional practice. The key to the answer lies in peer consultation. Discussion with colleagues, teachers, and clinical supervisors is the best way to assess one's emerging competence in assessment and focus on additional training needs. Following graduation and licensing, continuing professional education and peer consultation are the most effective strategies for assessing and maintaining one's own competence. When in doubt, presenting samples of one's work to a respected senior colleague for review and critique is a useful strategy. If one's competence is ever challenged before a court, ethics committee, or licensing board, the expert testimony of senior colleagues will be used in an effort to prove incompetence or negligence. By consulting with such colleagues regularly, one can be continuously updated on any perceived problems with one's own work and minimize the risk of criticism in this regard.
Dealing with the less-than-adequate work of others poses a different set of ethical concerns. At times psychologists may become aware of inadequate assessment work or misuse of psychological tests by colleagues or individuals from other professions (e.g., physicians or nonpsychologist counselors). Such individuals may be unaware of appropriate testing standards or may claim that they disagree with or are not bound by them. Similarly, some psychologists may attempt to use assessment tools they are not qualified to administer or interpret. The context in which such problems come to light is critical in determining the most appropriate course of action. The ideal circumstance is one in which the presumed malefactor is amenable to correcting the problem as a result of an informal conversation, assuming that you have the consent of the party who has consulted you to initiate such a dialogue. Ideally, a professional who is the recipient of an informal contact expressing concern about matters of assessment or test interpretation will be receptive to and interested in remedying the situation. When this is not the case, the client who has been improperly evaluated should be advised about potential remedies available through legal and regulatory channels.
If one is asked to consult as a potential expert witness in matters involving alleged assessment errors or inadequacies, one has no obligation to attempt informal consultation with the professional who rendered the report in question. In most cases, such contact would be ethically inappropriate because the client of the consulting expert is not the person who conducted the assessment. In such circumstances, the people seeking an expert opinion will most likely be attorneys intent on challenging or discrediting a report deemed adverse to their clients. The especially complex issues raised when there are questions of challenging a clinician's competence in the area of neuropsychological assessment are effectively discussed by Grote, Lewin, Sweet, and van Gorp (2000).
As an essential part of accepting the referral, the psychologist should clarify the questions to be answered in an interactive process that refines the goals of the evaluation in the context of basic assessment science and the limitations of available techniques; this is especially important when the referral originates with nonpsychologists or others who may be unaware of the limitations of testing or may have unrealistic expectations regarding what they may learn from the test data.
In attempting to clarify the ethical responsibilities of psychologists conducting assessments, the APA's ECTF charged with revision of the code concluded that psychologists should base their assessments, recommendations, reports, opinions, and diagnostic or evaluative statements on information and techniques sufficient to substantiate their findings (ECTF, 2001). The psychologist should have a sound knowledge of the available instruments for assessing the particular construct related to the assessment questions. This knowledge should include an understanding of the psychometric properties of the instruments being employed (e.g., their validity, reliability, and normative base) as well as an appreciation of how the instrument can be applied in different contexts or with different individuals across age levels, cultures, languages, and other variables. It is also important for psychologists to differentiate between the instrument's strengths and weaknesses such that the most appropriate and valid measure for the intended purpose is selected. For example, so-called floor and ceiling constraints can have special implications for certain age groups. As an illustration, the Stanford-Binet, Fourth Edition, has limited ability to discriminate among children with significant intellectual impairments at the youngest age levels (Flanagan & Alfonso, 1995). In evaluating such children, the use of other instruments with lower floor capabilities would be more appropriate.
Consistent with both current and revised draft APA standards (APA, 1992; ECTF, 2001), psychologists are expected to develop, administer, score, interpret, or use assessment techniques, interviews, tests, or instruments only in a manner and for purposes that are appropriate in light of the research on or evidence of the usefulness and proper application of the techniques in question. Psychologists who develop and conduct research with tests and other assessment techniques are expected to use appropriate psychometric procedures and current scientific or professional knowledge in test design, standardization, validation, reduction or elimination of bias, and recommendations for use of the instruments. The ethical responsibility of justifying the appropriateness of the assessment is firmly on the psychologist who uses the particular instrument. Although a test publisher has some related obligations, the APA ethics code can only be enforced against individuals who are APA members, as opposed to corporations. The reputations of test publishers will invariably rise or fall based on the quality of the tools they develop and distribute. When preparing new assessment techniques for publication, the preparation of a test manual that includes the data necessary for psychologists to evaluate the appropriateness of the tool for their work is of critical importance. The psychologist in turn must have the clinical and scientific skill needed to evaluate the data provided by the publisher.
In countries with a citizenry as diverse as that of the United States, psychologists are invariably confronted with the challenge of people who by reason of race, culture, language, or other factors are not well represented in the normative base of frequently used assessment tools. The Reynolds and Ramsay chapter in this volume considers these issues in detail. Such circumstances demand consideration of a multiplicity of issues. When working with diverse populations, psychologists are expected to use assessment instruments whose validity and reliability have been established for that particular population. When such instruments are not available, the psychologist is expected to take care to interpret test results cautiously and with regard to the potential bias and potential misuses of such results. When appropriate tests for a particular population have not been developed, psychologists who use existing standardized tests may ethically adapt the administration and interpretation procedures only if the adaptations have a sound basis in the scientific and experiential foundation of the discipline. Psychologists have an ethical responsibility to document any such adaptations and clarify their probable impact on the findings. Psychologists are expected to use assessment methods in a manner appropriate to an individual's language preference, competence, and cultural background, unless the use of an alternative language is relevant to the purpose of the assessment.
Some psychologists incorrectly assume that the use of an interpreter will compensate for a lack of fluency in the language of the person being tested. Aside from the obvious nuances involved in vocabulary, the meaning of specific instructions can vary widely. For example, some interpreters may tend to simplify instructions or responses rather than give precise linguistic renditions. At other times, the relative rarity of the language may tempt an examiner to use family or community members when professional interpreters are not readily available. Such individuals may have personal motives that could lead to alterations in the meaning of what was actually said, or their presence may compromise the privacy of the person being assessed. Psychologists using the services of an interpreter must assure themselves of the adequacy of the interpreter's training, obtain the informed consent of the client to use that particular interpreter, and ensure that the interpreter will respect the confidentiality of test results and test security. In addition, any limitations on the data obtained via the use of an interpreter must be discussed in presenting the results of the evaluation.
Some psychologists mistakenly assume that they can compensate for language or educational barriers by using measures that do not require verbal instructions or responses. When assessing individuals of diverse cultural and linguistic backgrounds, it is not sufficient to rely solely on nonverbal procedures and assume that resulting interpretations will be valid. Many human behaviors, ranging from the nature of eye contact; speed, spontaneity, and elaborateness of response; and persistence on challenging tasks may be linked to social or cultural factors independent of language or semantics. It has been demonstrated, for example, that performance on nonverbal tests can be significantly affected both by culture (Ardila & Moreno, 2001) and by educational level (Ostrosky, Ardila, Rosselli, Lopez-Arango, & Uriel-Mendoza, 1998).
What's in a Norm?
Psychologists must have knowledge of the applicability of the instrument's normative basis to the client. Are the norms up-to-date and based on people who are compatible with the client? If the normative data do not apply to the client, the psychologist must be able to discuss the limitations in interpretation. In selecting tests for specific populations, it is important that the scores be corrected not only with respect to age but also with respect to educational level (Heaton, Grant, & Matthews, 1991; Vanderploeg, Axelrod, Sherer, Scott, & Adams, 1997). For example, the assessment of dementia in an individual with an eighth-grade education would demand very different considerations from those needed for a similar assessment in a person who has worked as a college professor.
Psychologists should select and interpret tests with an understanding of how specific tests and the procedures they entail interact with the specific individual undergoing evaluation. Several tests purporting to evaluate the same construct (e.g., general cognitive ability) put variable demands on the client and can place different levels of emphasis on specific abilities. For example, some tests used with young children have different expectations for the amount of language used in the instructions and required of the child in a response. A child with a specific language impairment may demonstrate widely discrepant scores on different tests of intelligence as a function of the language load of the instrument because some tests can place a premium on verbal skills (Kamphaus, Dresden, & Kaufman, 1993).
It is important to remember that psychologists must not base their assessment, intervention decisions, or recommendations on outdated data or test results. Similarly, psychologists do not base such decisions or recommendations on test instruments and measures that are obsolete. Test kits can be expensive, and more than a few psychologists rationalize that there is no need to invest in a newly revised instrument when they already own a perfectly serviceable set of materials of the previous edition. In some instances, a psychologist may reasonably use an older version of a standardized instrument, but he or she must have an appropriate and valid rationale to justify the practice. For example, a psychologist may wish to assess whether there has been deterioration in a client's condition and may elect to use the same measure as used in prior assessments such as the Wechsler Adult Intelligence
Scales-Revised (WAIS-R), even if a newer improved version such as the WAIS-III is now available. The chapter in this volume by Wasserman and Bracken discusses psychometric issues relevant to such comparisons.
The currentAPAethics code holds that psychologists typically provide opinions on the psychological characteristics of individuals only after conducting an examination of the individuals that is adequate to support the psychologists' statements or conclusions. This provision is confusing in some contexts. At times such an examination is not practical (e.g., when a psychologist serving as an expert witness is asked to offer hypothetical opinions regarding data sets collected by others). Another example would occur when a psychologist is retained to provide confidential assistance to an attorney. Such help might be sought to explore the accuracy of another's report or to help the attorney frame potential cross-examination questions to ask the other side's expert. In such situations, psychologists document the efforts they made to obtain their own data (if any) and clarify the potential impact of their limited information on the reliability and validity of their opinions.
The key to ethical conduct in such instances is to take great care to appropriately limit the nature and extent of any conclusions or recommendations. Related areas of the current APA code include Standards 2.01 (Boundaries of Competence) and 9.06 (Interpreting Assessment Results). When psychologists conduct record review and an individual examination is not warranted or necessary for their opinions, psychologists explain this situation and the bases upon which they arrived at this opinion in their conclusions and recommendations. This same issue is addressed in the pending revision (ECTF, 2001) as part of a new section—Section 9 (Assessment). Subsection c of 9.01 (Bases for Assessments) indicates that when despite reasonable efforts, examination of an individual is impractical, "psychologists document the efforts they made and the result of those efforts, clarify the probable impact of their limited information on the reliability and validity of their opinions, and appropriately limit the nature and extent of their conclusions or recommendations" (ECTF, 2001, p. 18).
Subsection d addresses the issue of record review. This practice is especially common in the forensic arena; attorneys often want their own expert to examine the complete data set and may not wish to disclose the identity of the expert to the other side. When psychologists conduct record reviews and when "an individual examination is not warranted or necessary for the opinion, psychologists explain this and the bases upon which they arrived at this opinion in their conclusions and recommendations" (ECTF, 2001, p. 18).
This circumstance of being able to ethically offer an opinion with limitations—absent a direct assessment of the client—is also germane with respect to requests for the release of raw test data, as discussed later in this chapter.
Was this article helpful?