Examinations
Psychological tests in custody evaluations: what can the scores show, and which tests hold up?
Psychological tests are one source of data in a custody evaluation, and the 2022 APA guidelines say they are not used in isolation. California requires the evaluator to explain in advance what role test results will play. A 2007 review in Family Court Review rated only the MMPI-2 as meeting scientific and evidentiary standards among the tests it examined and rated the Rorschach and several custody-specific instruments as falling short.

What role do psychological tests play in a custody evaluation?
They are one source among several. The 2022 APA guidelines call for a multimethod, multitrait assessment and state that tests are typically not used in isolation but as part of a comprehensive assessment.
The American Psychological Association adopted its Guidelines for Child Custody Evaluations in Family Law Proceedings as policy in February 2022. There are 23 guidelines, and they lapse ten years after adoption. Under Guideline 10, psychologists strive to build an evidence-based, multimethod and multitrait assessment format. The explanation reads: "Multitrait and multimethod assessments help balance the limitations on reliability and validity of single measures."
Guideline 9 states that tests "are typically not used in isolation, but are part of a comprehensive assessment." Guideline 18 asks psychologists to include an observation of parent-child interactions and warns that such observations "are not in and of themselves 'attachment' evaluations," which require special training and settings.
The earlier version of the guidelines, adopted on February 21, 2009 and now archived on the APA site, listed psychological testing and clinical interviews among the direct methods of data gathering (Guideline 10). Its Guideline 3 focused the evaluation on "parenting attributes, the child's psychological needs and the resulting fit" and stated: "Comparatively little weight is afforded to evaluations that offer a general personality assessment without attempting to place results in the appropriate context." A personality test result not placed in the context of parenting carried little weight under that version, and the 2022 version keeps tests inside a comprehensive assessment.
What does California require the evaluator to say about tests?
Rule 5.220 requires the evaluator to explain the procedures and, if psychological tests will be used, the role of the results in confirming or questioning other information or previous conclusions.
Rule 5.220(e)(1)(B) of the California Rules of Court, last amended effective January 1, 2023, covers the "Procedures used and the time required to gather and assess information and, if psychological tests will be used, the role of the results in confirming or questioning other information or previous conclusions." Under (e)(2), data collection and analysis must let the evaluator consider each party in comparable ways, and the methods the rule lists as possible steps include review of documents, observation of parent-child interaction, interviews with the parties and children, and consultation with other experts when a matter is outside the evaluator's expertise.
Subdivision (j)(1) requires the evaluator to "Maintain objectivity, provide and gather balanced information for both parties, and control for bias." Under (j)(6), the evaluator works within his or her training and experience and discloses limitations or bias that would affect the evaluation.
These provisions give a party two documents to compare: what the evaluator announced about the role of testing before the evaluation, and what the report did with the scores. If both parents were tested with different instruments, the comparability requirement in (e)(2) is the starting point for a question.
Which test is used most often, and how does the MMPI-2 fare?
The MMPI-2 is the test given most often in family court evaluations; in a 1997 survey of 202 evaluators, 92 percent used it frequently. A 2007 review rated it the only examined test meeting scientific and evidentiary standards.
Erickson, Lilienfeld and Vitacco published the review as A critical examination of the suitability and limitations of psychological tests in family court (Family Court Review, volume 45, issue 2, 2007, pages 157-174). Its Table 2 rates each test against scientific and evidentiary standards; the MMPI-2 met them on the strength of its reliability and validity data.
The authors found that most custody examinees answer defensively, so the evaluator has to choose the validity scales that detect this most accurately; on the basis of a 2002 study by Baer and Miller they favored the Wsd and Pm scales. They wrote that "once validity thresholds are exceeded, the veracity of any subsequent interpretations are highly suspect," that ad hoc adjustment of suppressed clinical scales has little empirical support, and that claims of deception "should only be made when other data support such a conclusion."
The 2009 APA guidelines made the same point about custody litigants in general: "child custody litigants often display increased elevations on such scales" (Guideline 11). A report that treats a defensive profile as a finding about the parent's character, without other data, departs from both sources.
What about the MCMI-III and other personality inventories?
The 2007 review rated the MCMI-III equivocal. In one study only three of its scales showed adequate construct validity, and several scale elevations common in custody litigants can have explanations unrelated to psychopathology.
Erickson and colleagues cite Rogers and co-authors (1999), who found adequate construct validity for only the Avoidant, Schizotypal and Borderline scales of the MCMI-III. Elevations on the Compulsive, Histrionic and Narcissistic scales are common among custody examinees and, citing Halon (2001), can have explanations unrelated to psychopathology.
Table 2 of the review lists for the MCMI-III problems with construct validity, a known error rate and a good publication record, and rates it equivocal. For a party reading the report, the question is which scale elevations the evaluator relied on and whether the report says that elevations of this kind are frequent in custody litigation.
Do projective tests such as the Rorschach, drawings and the TAT meet the standard?
Not in the 2007 review. Its authors rated the Rorschach, human figure drawings and the Thematic Apperception Test as not meeting scientific and evidentiary standards and called continued Rorschach use in family court unwise.
For the Rorschach, the authors wrote that the Comprehensive System norms are inaccurate and tend to overestimate psychopathology, so healthy adults can appear maladjusted. Inter-rater reliability is poor for some scores, and validity is moderate for thought disorder and perceptual disturbance but negligible for most personality traits and major disorders. Their conclusion: "The continued pervasive use of the Rorschach test in family court evaluations is unwise at best and unethical at worst."
Human figure drawings were described as unreliable and invalid. The authors criticized evaluators who infer a child's estrangement from a parent on the basis of a family drawing, writing that there is "no credible scientific support" for that conclusion.
For the Thematic Apperception Test, the second most used personality measure with children in custody evaluations, they found inconsistent administration and "no published evidence validating the use of the TAT" for assessing a child's mental stability, parental preference or family cohesion in forensic settings. To the argument that a projective test is just one more piece of data they replied that "more is not necessarily better."
Are custody-specific instruments better?
Not those the 2007 review examined. The Bricklin Perceptual Scales, the Perception of Relationships Test, the Parent Awareness of Skills Survey and the Parent Perception of Child Profile were all rated as not meeting standards.
The Bricklin Perceptual Scales have no norms, measure a minority of relevant parental capacities and lack validity data; the authors wrote that "there is no scientific evidence" that a child's projective responses reveal unconscious preferences. For the Perception of Relationships Test they concluded that its "fundamental psychometric deficiencies preclude the usefulness of this test."
The Parent Awareness of Skills Survey lacks reliability and validity data, norms and defined scoring, and its author suggests that evaluators apply their own scoring standards, which the review says "probably violates established ethical canons." The Parent Perception of Child Profile has no published reliability or validity data. Anatomically detailed dolls, which the APA found do not meet most minimal criteria for a valid psychological test, "should be avoided by mental health experts in family court evaluations."
Earlier, Otto, Edens and Barcus (Family Court Review, 2000) reviewed the use of tests in custody evaluations, urged evaluators to examine each test carefully before using it and proposed a template for that assessment. A report that names a custody-specific instrument without its norms and validity data invites the same examination.
Why do norms and the custody context matter?
Because parents in a custody dispute are tested under stress and have a stake in the result. The 2022 guidelines ask psychologists to know normative data for divorced parents and to read results as possibly temporary and situational.
Guideline 17 asks psychologists to "strive to be aware of normative data for divorced parents" and to avoid assessment measures "that introduce, perpetuate, or otherwise contribute to bias." The same guideline lists factors relevant to admissibility, among them "history of peer review, cultural relevance, and known error rates."
Guideline 21 asks psychologists to integrate and analyze evaluation data "in a contextually informed fashion." It notes that results may stem from temporary, situationally determined states, and that relationship conflict and the evaluation process itself "can be exceptionally stressful." Guideline 14 adds that parental impression management during interviews may require confirmation from other sources.
A score compared only with general population norms says how a parent answered on a stressful day in litigation. A review checks whether the evaluator compared the score with data for parents in the same position and whether the report says so.
Should children be tested?
Not always. Guideline 13 of the 2022 APA guidelines says testing of a child may not be necessary or appropriate if it does not help elucidate the child's best interests.
Guideline 13 states that testing "may not be necessary or appropriate if such testing does not help elucidate the best interests" of the child. Guideline 18 lets the psychologist postpone or skip a parent-child observation to protect the child's safety.
Language and culture affect the choice of tests. Guideline 6 asks psychologists to practice in a culturally competent way and mentions procedures "such as the use of interpreters and test translations"; interpreters should be qualified and professionally trained. When a test is adapted, for example translated, Guideline 17 asks the psychologist to consider the effect on its reliability and validity.
In practice the question for a child's test is what it added. If the report cites a child's drawing or story test as the basis for a preference or an alienation finding, the 2007 review gives the reply quoted above: no published evidence supports that use.
Can an evaluator rely on a computer-generated interpretive report?
Only as a source of hypotheses. Guideline 17 says such reports may generate helpful hypotheses but are not meant to supplant the psychologist's clinical and forensic judgment.
Guideline 17 states that "computer-generated interpretive reports may generate helpful hypotheses," that they must still be evaluated, and that they are "not meant to supplant the psychologist's clinical and forensic judgment." A sentence copied from a scoring program into the evaluation, without data from interviews or observation behind it, is a hypothesis presented as a finding.
The 2009 guidelines required that "Test and interview data are documented with an eye toward their eventual review by other qualified professionals" (Guideline 14). A reviewing psychologist therefore asks for the score reports, the interpretive output and the evaluator's notes, and compares the wording of the evaluation with the program's wording.
How can a party challenge test-based opinions in court?
By asking whether each test has been tested, peer reviewed, has a known error rate and is generally accepted. The 2007 authors recommended motions in limine against testimony based on tests with serious psychometric defects, and vigorous cross-examination.
Erickson and colleagues applied the Daubert factors to each instrument: testability, peer review, known error rate and general acceptance. In jurisdictions that follow Frye, the question is whether the method is generally accepted, and the authors note that many such jurisdictions have adopted Daubert principles. They argue that the burden lies with the proponent of the testimony to show that the method is acceptable.
The California rule gives a second line of questions: what the evaluator said in advance about the role of the test results, whether both parties were assessed in comparable ways, and whether the evaluator disclosed limitations. The 2022 APA guidelines give a third: whether the evaluator combined several methods, used normative data for divorced parents and read the results in the context of the dispute.
What does IAE do when a custody evaluation relies on psychological tests?
A psychologist from the IAE registry writes an expert report on the client's questions or a review of the opposing psychologist's evaluation. Delivery takes up to 5 business days after the format and cost are confirmed.
The IAE page on forensic psychological evaluation covers child custody and parenting time, decision-making capacity in contract and will disputes, and emotional distress claims in civil cases. For custody it cites the 2022 APA guidelines on multimethod assessment, standardized measures and observation of parent-child interaction (Guidelines 10, 17 and 18). Questions are framed from the pleadings, the records and the opposing evaluator's report and are confirmed with the client before work starts.
The client sends the materials and describes the task. The secretariat confirms receipt the same day and, within one business day, confirms the format, cost, deadline and the specialist from the registry; the client approves the candidate and the CV before work starts. For tight deadlines there is a 48-hour rush.
The specialist is listed in the public IAE registry and can be verified by Member ID or QR code. IAE is not a law firm, and neither IAE nor the psychologist guarantees the court's decision.
Sources
- American Psychological Association, Guidelines for Child Custody Evaluations in Family Law Proceedings (adopted February 2022), PDF copy published by Infocop (Consejo General de la Psicología de España)
- American Psychological Association, Guidelines for Child Custody Evaluations in Family Proceedings (adopted February 21, 2009, archived page)
- California Rules of Court, rule 5.220, Court-ordered child custody evaluations (amended effective January 1, 2023)
- Erickson S.K., Lilienfeld S.O., Vitacco M.J., A critical examination of the suitability and limitations of psychological tests in family court, Family Court Review 45(2) (2007) 157-174
- Otto R.K., Edens J.F., Barcus E.H., The use of psychological testing in child custody evaluations, Family Court Review (2000), abstract (Digital Commons, University of South Florida)
Read also
All articles
Undue influence expert witness: what can a psychologist assess in a will, trust or contract dispute?
In California a will, or a part of one, is ineffective to the extent undue influence procured it, and that claim is separate from a claim that
Capacity evaluation for guardianship: who examines the person, and what must the report contain?
Before a court appoints a guardian for an adult, professionals examine the person and report on which rights he or she can still exercise. Florida
Supervised visitation: when does a court order it, and what does a psychologist assess?
A court orders supervised visitation when contact with a parent is in the child's interest but carries a risk that a neutral third person can manageExpert examinations
Ready to discuss your case?
Send the materials: the secretariat will confirm the format, cost, deadline and the specialist within one business day.

