Essentials of a Forensic Child Custody Evaluation
By Alan M. Jaffe and Diana Mandeleew
Before an expert can begin to conduct a good custody evaluation, he or she must have a full understanding of the various types of divorce, including the effects of divorce on children of different ages, both in the short- and long-term. The expert must also demonstrate a good legal knowledge of the types of custody and visitation arrangements that can be recommended. Although the evaluator should definitely have some experience in the treatment of children and adolescents, it is most important that the evaluator have advanced skills in the assessment of child and adolescent personality, mental illness, family dynamics, and parenting skills required to provide a healthy environment for growth and development of children. A competent evaluator also will be familiar with the legal aspects of custody procedure and understand the various legal definitions of custody as reflected in state law.
When conducting a custody evaluation, the procedure should be equitable and offer fair treatment to all parties by administering the same procedures with each party. Specifically, it is important to use interviewing, psychological testing, home visits, the utilization of collateral informants, observation of parents with children, and the amount of time children spend in a consistent way with all parties.
Who Should Be Included?
A thorough custody evaluation should include not only parents but also any other adults directly responsible for the daily care of children, such as stepparents, grandparents, and either parent’s significant other. Any other party living in the custodial or visiting home also should be seen, such as step- or half-siblings. It is generally a good practice for daycare providers as well as medical professionals, psychotherapists, and school personnel to be included. However, it is not always in the best interests of the children to include these collaterals for various reasons.
A good evaluation and a seasoned evaluator may elect not to include specific collaterals at certain times. If consulting with any of these individuals is not ultimately in the best interests of the children and/or would cause a negative result in the day-to-day life of the children, then a competent evaluation need not include them.
For example, (1) the daycare provider who is concerned that what is said may interfere with his or her ability to continue working with the children due to the inadvertent alienation of a parent or an inability to communicate with either parent going forward; (2) the psychotherapist who risks upsetting the course of treatment by taking a position that would either risk confidentiality with his patient or risk the unfavored parent’s discontinuing therapy; or (3) school personnel who are concerned that close scrutiny will result in a modification of the curriculum or approach to the student, resulting from the fear of repercussions from either parent. Assuming the use of collateral informants will not interfere with the best interests of the children, a good evaluation will include them.
A thorough evaluation should include a good life history. Although there has been a good deal of discussion among judges about the relevance of early historical information, it is important to include as much early history as possible. A good evaluation should demonstrate the expert’s deep understanding and working knowledge of each parent’s individual psychology and philosophy of child rearing. To truly understand who a parent is and how he/she “arrived” at where the parent is today, a competent evaluator should demonstrate that he or she understands the events that led up to the current crisis. (Those who were political science majors instead of psychology majors would agree. Understanding national and world politics today is impossible without the context of history.) Evaluations differ in the amount of history provided, but a good evaluation should provide a significant amount of relevant history.
The evaluation should include evidence of document review. It is not necessary for the evaluation to summarize all pleadings and the court-related matter, but it should include reference to relevant medical records, school records, encounters with the police, and other issues that affect the well-being and placement of children.
There is some debate about the pros and cons of psychological testing. But a good evaluation includes at least some psychological testing. Experience has shown that testing is not only appropriate and relevant when used correctly, but also essential to a comprehensive evaluation. The tests permit a comparison of each party’s performance with the performance of the general population.
Some psychological tests and scales specifically measure the test-taker’s approach to the test, that is, whether the tests or scales exaggerate or minimize the test-taker’s problems or symptoms. This is especially valuable in a custody situation where there is much at stake and the parties have an interest in appearing problem-free. The fact is that some people do better in interview situations than others. Even though the evaluator makes use of trained clinical interview skills, without the testing, the evaluator is relying entirely on what he or she is being told.
Evaluators are not mind readers. Even with the limitations inherent in psychological testing, it is more information for the courts, and it is based on scientific research. A good evaluation includes psychological testing widely used in custody situations and can demonstrate for the courts, if necessary, how the tests are relevant.
There is a multitude of psychometric measures from which a psychologist can choose. Typically, widely used tests with established validity and reliability measures, as well as those supported by a substantial research body, are better choices. When examining psychometric measures, it is important to look at the content as well as statistical parameters, including validity and reliability. Validity is the degree to which a test measures what it was designed to measure, and reliability is the degree to which the results of a test remain consistent over repeated administrations under identical conditions.
In child custody evaluations, most psychometric measures tend to fall within the following categories: cognitive functioning tests, objective personality tests, projective personality tests, and parenting assessment tests. A comprehensive evaluation will contain a battery of tests from numerous categories. Some of the more common tests used in child custody evaluations follow.
In child custody evaluations, the purpose of cognitive functioning tests given to parents is to determine whether their intellectual skills are adequate to meet parenting demands. Since these tests often are time consuming and not high predictors of custody placement, they are often omitted in evaluations of high-functioning parents. When cognitive tests are administered, it is important to keep in mind that they measure only aptitude or achievement of an individual and might not fully correspond to the multifaceted intelligence of an individual.
The Wechsler Adult Intelligence Scale—Third Edition (WAIS-III) is a comprehensive measure of intelligence composed of verbal and nonverbal tasks (Wechsler, 1997). Examinee’s scores are compared with norms of his or her peer group and are calculated into a standard score with a mean of 100 and a standard deviation of 15. In addition to the Full Scale Intelligence Quotient (FSIQ) score, Verbal and Performance IQ scores are generated. Scores are further broken down into the Verbal Comprehension Index (VCI), the Working Memory Index (WMI), the Perceptual Organization Index (POI), and the Processing Speed Index (PSI). The WAIS-III can be administered to examinees over 16 years of age.
To assess a child’s cognitive abilities, the Wechsler Intelligence Scale for Children-IV (WISC-IV) can be administered (Wechsler, 2003a; Wechsler, 2003b). Similar to the WAIS-III, this test is composed of a number of verbal and nonverbal tasks. The Full Scale IQ score (FSIQ) can be broken down into four indices: Verbal Comprehension (VCI), Working Memory (WMI), Perceptual Reasoning (PRI), and Processing Speed (PSI). A child might be administered an intelligence measure when it is suspected that he or she has a much lower intelligence than average and, as such, requires additional parental support.
Assessing Academic Achievement
The Wide Range Achievement Test (WRAT-IV) is a measure of academic achievement and includes Reading, Comprehension, Spelling, and Mathematics subtests (Wilkinson & Robertson, 2006). This test can be administered to children, adolescents, and adults and has strong validity and reliability coefficients. The resulting scores compare the examinee with a normative sample of peers, and the results can be expressed either in grade level or age level.
For individuals who have a limited English-speaking ability or whose verbal or fine-motor skills might undermine their true cognitive functioning, a nonverbal measure of cognitive ability can be administered. An example of such a test is General Ability Measure for Adults (GAMA), which yields an IQ score (Naglieri & Bardos, 1997). It consists of 66 pictorial puzzles that require the examinee to indicate which of the six possible answers is correct. GAMA takes only 25 minutes to administer, compared with the much lengthier Wechsler Scales, which can take hours to complete. An obvious drawback of a nonverbal test is that it does not assess verbal expressive abilities.
Objective Personality Tests
Objective measures assess personality and socio-emotional functioning, including broadband comprehensive measures (such as MMPI-II, MCMI-III, and PAI) and narrowband measures (such as the Beck Depression Inventory-II). Typically these tests are designed to screen for clinical symptoms and personality disorders, consistent with the criteria in the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV-TR).
The Minnesota Multiphasic Personality Inventory (MMPI-II) is an objective inventory of adult personality designed to provide information on critical clinical variables (i.e., depression, social introversion, hypochondriasis, schizophrenia, etc.) (Hathaway & McKinley, 1989). It contains nine Validity Scales, five Superlative Self-Presentation Subscales, 10 Clinical Scales, 31 Clinical Subscales (Harris-Lingoes and Social Introversion Subscales), nine Restructured Clinical (RC) Scales, 15 Content Scales, 27 Content Component Scales, and 20 Supplementary Scales.
The MMPI-II is based on a large normative sample of thousands of individuals from various communities in the United States. This test incorporates recent trends in mental health diagnosis and includes many common mental health disorders. It is one of the most widely used psychometric measures and, although there are some concerns regarding its validity in testing nonpsychiatric individuals, it has well-established validity and reliability (Friedman, Lewak, Nichols, & Webb, 2001). The drawback to administering the MMPI-II is that it contains 567 true or false items, which can be lengthy to administer.
The Millon Clinical Multiaxial Inventory-III (MCMI-III) is a personality measure for adults, which is composed of 175 true or false questions (Millon, Davis, & Millon, 1997). This instrument can be completed in approximately 30 minutes and can provide numerous subscales for interpretation. It is more sensitive to Axis 2 psychopathology.
The Personality Assessment Inventory (PAI) is an objective inventory of adult personality, which contains 344 items (Morey, 1991). It was designed to provide information on critical clinical variables (i.e., depression, anxiety, schizophrenia, antisocial tendencies, alcohol and drug problems). It contains four validity scales, 11 clinical scales, five treatment scales (including possible areas of interventions, such as suicide or anger), and two interpersonal scales (whether the examinee tends to be domineering or supportive in his or her interactions). It is based on a large database and includes many common mental-health disorders.
The Beck Depression Inventory–Second Edition (BDI-II) is a 21-item self-report instrument that assesses the existence and severity of depressive symptoms, including cognitive, affective, and physiological factors over the past two weeks (Beck, Steer, & Brown, 1996). The time period and the areas of functioning reflect the DSM-IV-TR criteria for depression. The measure’s construct validity has been established, and research indicates that this measure can be used to differentiate between depressed and nondepressed patients. However, this test has a high face validity, which means that its purpose easily can be determined from reading the items. As such, the examinee can respond so as to appear to be either more or less pathological than he or she truly is.
Projective Personality Tests
The Rorschach Inkblot Test is a projective measure of emotional functioning and personality characteristics (Rorschach, 1942). The test contains ten inkblots: some are achromatic, and some are multicolored. The individual is first asked what he or she sees in each of the cards, what makes it look like that, and where the image is located. Some evaluators look at the content and common themes of the Rorschach responses. Alternatively, the Exner Scoring System can be used for scoring and interpretation (Exner, 2002). Although some clinicians still incorporate this test, it generally has been abandoned because the concepts employed in the interpretation are too abstract for the courtroom.
The Thematic Apperception Test (TAT) is a projective measure that requires the examinee to tell stories about a series of pictures (Murray, 1971). For each picture, the individual is asked to tell a story with a beginning (what led to the event), a middle (what is happening now), and an end (what will be the outcome). The examinee is asked what the character(s) might be thinking or feeling. It generally is believed that characters in the stories represent projected aspects of the self. The evaluator looks for common themes among the stories.
The Sentence Completion Series–Adult Form (Brown & Unger, 1998) consists of sentence stems on a variety of topics, which the individual is asked to complete. It is designed to gauge areas of concern and distress. The responses can be analyzed based on themes; conflicts; conflict resolution styles, wishes, and fears; and the presented world view.
Projective drawings also are part of the projective personality tests. For example, in the House–Tree–Person Technique, the examinee is asked to draw a house, a tree, and a person on paper (Buck, 1970). In the Kinetic Family Drawing Technique, the examinee is asked to draw his or her family performing some activity (Burns & Kaufman, 1972). There are different ways to interpret projective drawings (i.e., Ogdon, 1998). For example, some evaluators view the drawing of a person (part of the House–Tree–Person Technique) to be indicative of how the individual views him- or herself, including ideas about gender roles. The evaluator looks at the details of the drawings, the placement of the drawing on the page, as well as the verbal description provided by the examinee.
Parenting Assessment Tests
The Bricklin Perceptual Scales (BPS) is a measure that was designed for child custody evaluations (Bricklin, 1984). A child over the age of six is asked 32 questions about both parents (64 questions in total). The four parenting areas gauged by this measure include Supportiveness, Competence, Follow-up Consistency, and Possession of Admirable Personality Traits. A limitation of this test is that it uses a child’s report, which can change over time and might be a function of the child’s current mood or parental influence. Limited research has made this an instrument beneficial for information gathering, rather than relying on the classifications.
The Ackerman–Schoendorf Scales for Parent Evaluation of Custody (ASPECT) also was designed specifically for child custody evaluations (Ackerman & Schoendorf, 1992). This measure includes a parental questionnaire and incorporates the results of a variety of other tests (i.e., MMPI-II, parents’ and child’s IQ scores, TAT, projective drawings, etc.). In addition to the global Parental Custody Index, Observational, Social, and Cognitive–Emotional Scales can be used to compare the parenting effectiveness of both parents.
Psychologists choose from a variety of psychometric tests for a child custody evaluation. General trends change over time. For example, in 1986, the three most common psychometric measures administered to adults in custody evaluations were MMPI-II, Rorschach, and TAT (Keilin & Bloom, 1986). A similar study in 2001, found 92 percent of evaluators had reported administering MMPI-II, and relied much less on objective personality and cognitive tests than did evaluators 15 years previously (Quinnell & Bow, 2001). Today, children are being tested less frequently than before, and when they are, evaluators tend to administer projective rather than objective measures (Quinnell & Bow, 2001).
Since both parents in a child custody evaluation often are motivated to present themselves in the best possible light, the results of the psychometric measures must be considered carefully and compared with other information obtained during the evaluation. Similarly, psychometric measures that contain validity scales, such as the MMPI-II, can be useful in determining the degree of consistency between the examinee’s report and their true functioning.
When assessing a bilingual client, it is important to ask which is his or her preferred language. The client may feel more comfortable conversing in a native tongue. When an examiner fluent in the examinee’s native tongue is unavailable, the services of a translator may be sought. Family members, and especially minors, should not be used for translations of “sensitive and confidential conversations” between the assessor and the examinee (Raso, 2006, p. 56).
Keep in mind that cultural factors may influence the examinee’s performance on psychometric measures, particularly those that assess verbal expression and culture-bound knowledge. In such circumstances, the psychological report must contain a disclaimer to explain this limitation.
When making recommendations about custody matters, each parent, guardian, stepparent or any adult who physically lives or could potentially live with the children should be clinically evaluated. These are the people who will have the most influence on the children. A report should show that clinical interviews have been given to anyone in a position of parental responsibility and that they have been carefully examined. Although there is no magic number of clinical interviews each parental figure should have, a good report demonstrates that an adequate number has been given.
A good evaluation should include observations of the children with their parents and other live-in significant others. Some evaluators conduct these observations at the parent’s home, while most are conducted in the evaluator’s office. The observation sessions allow the evaluator to see children relating to and interacting with their parents at a moment in time. It is at the discretion of the evaluator as to whether these observation sessions are open-ended, task structured, or a combination of both. These observation sessions are important, will bolster the credibility of the final report, and will demonstrate to the court that the evaluator has spent time in the same room with the parents and children who are the subject of recommendations for the future.
The custody evaluation should include individual clinical interviews with the children as long as such interviews do not create undue stress for a given child. The evaluation should include how the child spends time in general with each parent, what he or she likes and dislikes about each parent, the kinds of activities parent and child engage in together, and how discipline is administered. The evaluator should demonstrate competence at eliciting information from children without having to be too direct about controversial issues. It is not the responsibility of the evaluator to extract a statement of preference from a child unless it is clear that the child is old enough and free from all of the other psychological and emotional consequences that could occur.
Psychological testing of children is not necessary unless there are questions raised that require deeper exploration of the child’s mental health. By the time the custody evaluation is underway, this generally has been accomplished by a school or outside agency. However, a good evaluation provides enough information about the children’s behavior through clinical interviews with parents or by having parents fill out checklists or inventories regarding their children.
A home visit usually is at the discretion of the evaluator. Because the home visit is an additional expense to one or both parties, it should not be conducted if both parties stipulate that the other’s living situation is adequate. But where allegations have been made that a home environment is substandard or undesirable for any reason, a home visit should be included.
Sidebar: Summary and Recommendations
Ultimately a good report should provide a summary section and a list of conclusions and recommendations. The report should emphasize to the judge how conclusions and recommendations were reached, based on the facts gathered throughout the evaluation. The evaluator should be free to express his or her opinion as an expert, but also should expect to demonstrate the foundation for the opinions. In the best reports, little or no additional explanation is required, because the conclusions follow naturally from the foregoing information. However, the evaluator should explain how he or she processed and interpreted the information to reach the final recommendations. Every expert has a particular style of writing a report, and there is room for differences in style and written expression. A good evaluation should be written so that a layperson can readily understand information in it. When it is necessary for a report to contain theoretical information, it is important to explain in layman’s terms what the expert is attempting to communicate.
All things considered, a good report walks the fine line of taking into account that the court requires a demonstration that all relevant information has been obtained and a showing that appropriate and logical conclusions have been reached.
This should be accomplished without burdening the court with every word said during each clinical interview and every other encounter. The court appreciates thorough work, but expects the expert to distill the information into a manageable form.
In the end, after the hard work of testing, interviewing, and evaluating has been completed and objectives reached, the best experts keep in mind as they draft the final report that someone else will be reading it.
Alan M. Jaffe, Psy.D., is on the faculty of Northwestern University Feinberg School of Medicine in Chicago. He has authored book chapters and articles and has published internationally in the area of psychological testing, substance abuse, and compulsive disorders. He is a Fellow of the American College of Forensic Examiners. Diana Mandeleew, M.A., has extensive experience conducting psychological assessments, including forensic, personal injury, and custody evaluations. She is completing her doctorate at the Adler School of Professional Psychology.
This article previously appeared in Family Advocate, Vol. 30, No. 4 (Spring 2008). Copyright 2011© by the American Bar Association. Reprinted with permission. All rights reserved. This information or any or portion thereof may not be copied or disseminated in any form or by any means or stored in an electronic database or retrieval system without the express written consent of the American Bar Association.
© Copyright 2011, American Bar Association.