The MMPI-2 is the most frequently used psychological test with parents undergoing child custody evaluations (CCE's). Studies of parents who completed the MMPI-2 in the CCE context have consistently found that, on average, parents portray a very favorable image of themselves as both virtuous and psychologically well-adjusted. Also, on average, Clinical Scale 6 (Paranoia) and its counterpart on the Restructured Clinical Scales, RC6, have been the most frequently elevated of the clinical scales on the MMPI-2 in samples of child custody litigants, albeit, the reported scores on these scales are well within the norm. This paper examines the significance of clinically elevated scores on Scale 6 (and RC6) in parents undergoing CCE's, with consideration of both contextual and psychometric features. Based on these considerations, this writer suggests that custody evaluators may encounter the paradox that some parents whose clinical profile suggests mistrust and suspiciousness in interpersonal relations may actually trust enough to report their perceptions in a candid and forthcoming manner. At least some of these parents are not mistrustful in general and are not inclined to model or influence their children toward the development of mistrust. In contrast, other parents may respond to the test in a more defensive and deceptive manner resulting in a "normal" clinical profile, which may be specific to the evaluation context, but may also reflect moral righteousness and a more insidious tactic. The custody evaluator is advised to use caution when interpreting clinically elevated scores on the MMPI-2.

The American Psychological Association Guidelines for Child Custody Evaluations in Family Law Proceedings,¹ and the Association of Family and Conciliation Courts Model Standards of Practice for Child Custody Evaluation,² recommend the use of a multi trait/multi method assessment procedure in child custody evaluations (CCE’s). These procedures generally include clinical interviews of each parent, observations or interviews of the children, parent-child observation or interview, collateral interview of relevant third parties, extensive review of relevant documentation, and, in some cases home visits; and, most psychologists who conduct CCE’s administer standardized, objective self-report personality assessment measures.^3‒5 The Minnesota Multiphasic Personality Inventory-2 (MMPI-2)⁶ has been identified in surveys of forensic psychologists as the most frequently used test in forensic assessment of adults,⁷ and in the assessment of parents in CCE’s,^3‒4 and as receiving the highest ratings in meeting Daubert criteria⁸ among tests, techniques, and inventories used in CCE’s.⁹ The use of personality testing of parents does not permit direct inferences about parenting behavior, and there is no MMPI-2 profile of a good vs. bad parent.¹⁰ However, as one component of a CCE, standardized, validated psychological tests such as the MMPI-2 may be very useful in the assessment of personality qualities and psychopathology that may be relevant to issues in the realm of parenting. These issues may include regulation of emotions, impulse control, the presence of severe depressive or anxiety symptoms, paranoid delusions, problems with misuse of alcohol or other substances, severe narcissistic qualities, and antisocial behavior, as well as the assessment of how stress might be impacting the parent’s relationship with the child.¹¹

Over the past two decades, several published studies reported descriptive data on the MMPI-2in parents involved in child custody litigation.^12‒18 Such studies indicate that, on average, parents who are evaluated in the context of custody litigation obtain modestly elevated scores, ranging from one half to 1 standard deviation above the mean score of 50 on the standard validity scales designed to assess under-reporting (Scales L, K, and S). At modestly elevated score levels, e.g., T 55-60, the K and S scales may suggest an accurate portrayal of positive emotional adjustment. However, at increasingly elevated levels, the K and S scales suggest defensiveness, denial of any problems, and unrealistic claims of superior adjustment, which may seriously undermine the veracity of the test profile. At increasingly elevated levels the L scale suggests naïve efforts to claim virtues that are uncommon in most people. Pope et al.,¹⁸ reported average T scores of 56.5, 56.5, and 58.8 on the L, K, and S scales, respectively, based on a total of 1,799 parents. Hynan¹⁶ reported average T scores of 60.88 and 62.14 on the L and K scales, respectively, based on a total of 557 parents. Simultaneously, on average, parents undergoing CCE’s typically obtain scores within the average range on the clinical scales assessing psychopathology. In essence, parents that are evaluated in the context of CCE’s tend to adopt a test response style in which they portray a favorable image of themselves as being well-adjusted, virtuous, and free of any mental health problems. In studies as cited above, Clinical Scales 3 (Hy: Hysteria), 4 (Pd: Psychopathic Deviate), and 6 (Pa: Paranoia), have been the only clinical scales that exceeded the standardization mean T score of 50, albeit mean scores on these scales were well within the average range.

Butcher,¹⁴ and Pope et al.,¹⁸ reported that Scale 6 as a high-point clinical scale (i.e., the highest clinical scale in the profile) occurred in 29.4 % of male, and 26.6 % of female custody litigants in their sample. Scale 6 also was the most frequent well-defined (i.e., higher than other clinical scales beyond the standard error of measurement) high-point clinical scale at or above a T score of 65, occurring in 8.5 % of men and 7.6 % of women in their sample of parents. Bagby et al.,¹² found Scale 6 to have the highest elevation among the clinical scales in samples of parents undergoing child custody evaluations (M = 56.14, SD = 12.59). Bagby et al. also found that of the parents in their sample who were identified as non-under reporters on the basis of two different criteria, 25 % to 33 % of such parents obtained elevations on Scale 6. Similarly, Archer et al.,¹⁹ using the Minnesota Multi phasic Personality Inventory-2 Restructured Form (MMPI-2-RF),²⁰ found that Restructured Clinical Scale 6 (RC6: Ideas of Persecution), was the most frequently elevated RC scale in a sample of custody litigants, with 18 % of female custody litigants and 15.1 % of male litigants scoring at or above a T score of 65.

Individuals who obtain modestly elevated scores on Scale 6 of the MMPI-2 are often characterized as overly sensitive and responsive to the opinions of others, especially to criticism, they are easily hurt, they tend to externalize blame and to see others as responsible for their problems, they may see themselves as unfairly treated and unjustly blamed for others’ problems, they tend to be moralistic and rigid in their opinions and attitudes, they may be suspicious and guarded and concerned that others may take advantage of them, they may be occasionally touchy and argumentative, they may have episodes of mistrust, hostility and anger, and their sensitivity may have a paranoid tintas when they are unsure who is for or against them.^{10‒11,21‒22} High scores, in the range of T 75 to T 85, increasingly suggest the presence of delusional beliefs.^11,21‒22 However, moderately elevated scores in non-clinical settings may be seen in persons described as interpersonally sensitive, emotional, rational, and clear thinking.²¹ Persons who obtain moderately elevated scores on Scale 6 and who are not feeling vulnerable to negative judgment, may also be viewed as fair-minded, rational and analytic, and they can be emotionally responsive and empathic.²³

As noted by various investigators such as Bathurst et al.,¹³ heightened interpersonal sensitivity, mistrust, suspiciousness, anger and/or resentment are not uncommon in the context of child custody evaluation. In a review of the extant literature on personality traits of parents involved in post-divorce conflict, Ellis concluded: “The most common elevation was on Scale 6, which is understandable in that parents in custody disputes often are the subjects of scorn by others and do justifiably feel that others are plotting against them”.²⁴ Gould²⁵ noted that elevated MMPI-2 scores can be “…the results of artifacts of the custodial dispute and/or other factors that may be unrelated to parenting and co-parenting”. As suggested by Caldwell,¹¹ the modest but frequently seen elevation on Scale 6 may reflect an exacerbation of certain pre-existing sensitivities to feeling wrongly treated in the context of highly stressful family and legal circumstances. Furthermore, in a review of the literature on the MMPI-2 in CCE’s involving battered women, Erickson²⁶ opined that the commonly found elevation on Scale 6 in women survivors of domestic violence may be expected in view of the fearfulness and suspicion associated with the abuse and betrayal of trust in an intimate relationship.

In conjunction with contextual factors to be considered when interpreting elevated scores on Scale 6 and/or RC6 in a CCE, there are certain psychometric considerations that should be regarded. First, the endorsement of three out of 17 items on RC6 results in a T score of 66, in the clinical range. While the RC scales are well-designed, psychometrically efficient and they have been well validated,²⁰ two or three items on RC6 (these same items are also contained within the Pa1 subscale of Clinical Scale 6) are relatively benign in their content and may be endorsed by some parents on a context-specific basis. It is noted, though, that a low percentage of the custody litigant comparison group responds in the scored direction even on such fairly benign test items on Scale RC6²⁰ and Clinical Scale 6.^14,18 With regard to Clinical Scale 6, the evaluator would be prudent to examine the three component Harris-Lingoes subscales. Pa1 (Ideas of External Influence) affirms ideas of persecution, Pa2 (Poignancy) conveys intense emotional sensitivity to criticism, and Pa3 (Naivete/Moral Virtue) affirms a trusting, virtuous outlook about people and denies cynicism and distrust (raw score points are counted on Pa3 by responding “False” to items which suggest that people generally are self-seeking). Thus, the item content of Pa1 and Pa3 suggest the endorsement of opposite perceptions (i.e., affirmation of mistrust and cynicism on Pa1, and disavowal of mistrust and cynicism on Pa3). Not surprisingly, Pa1 and Pa3 have a moderate, negative correlation of -.44.²² Nichols²² asserts thatPa3 may function to suppress false positives on Scale 6.

Various explanations have been expounded to account for the contribution of Pa3 items to a scale designed to assess a paranoid orientation. For example, high scores on Pa3 may reflect the portrayal of high moral standards and stern moral judgment that follows when a person who holds such high, rigid moral expectations judges that someone else has failed to meet such expectations.^22,27 Caldwell¹¹ opined that it is particularly the moral righteousness conveyed by high scores on Pa3 that feed and contribute to a parent’s sense that others are either “for or against” them, and might be associated with alienating tendencies. Notwithstanding those very reasonable considerations, given the non-clinical context of a CCE that pulls for defensiveness, it is also important to note that Pa3 is highly correlated with K scale (r = .64), whereas Pa1 and Pa2 are negatively correlated with K scale. Pa3 is also highly correlated with the Superlative (S) Scale (r = .66), and very highly correlated (negatively) with the Cynicism content scale (CYN) and RC3 (Cynicism) scales (r = -.83 and - .81, respectively).²¹ Thus, in the CCE context, parents who are attempting to portray a very well-adjusted, problem-free image of themselves (high scores on L, K or S scales) may also be more inclined to endorse the socially desirable view (whether or not such endorsement accurately reflects their view) that people are inherently honest, righteous and trustworthy, while also denying that they personally hold cynical, suspicious, blaming, paranoid ideas or that they are painfully affected by criticism from others (as may be seen in a configuration of lower scores on Pa1 andPa2, and higher scores on Pa3). A defensive parent may endorse no items, or one or two items on both Pa1 (composed of 17 items) and Pa2 (composed of 9 items), and may endorse eight or all nine items on Pa3 and obtain a T score in the average range on Scale 6.

Perhaps paradoxically, some parents in the CCE context do not attempt to portray an overly well-adjusted image of them (as suggested by average scores on L, K, and S scales). Yet, in the context of intense stress and emotional pain associated with the dissolved marital or intimate relationship and acrimonious custody litigation, some may openly endorse items pertaining to feeling that they are being targeted by someone who is against them, feeling vulnerable, sensitive, threatened, and easily hurt, while also endorsing a fair balance of skepticism about human nature, but absent the wholesale belief that others are egocentric. In such cases, if a parent endorses four to six items on Pa1, three or four items on Pa2, and five to seven items on Pa3, such a parent would obtain a T score in the clinical range on Scale 6, even with the absence of endorsement of patently paranoid or bizarre ideation. Furthermore, what should raise a greater flag of concern would be a parent who obtains an elevated score on Pa1 and a particularly low score on Pa3. Essentially the same concern should be raised by the finding of elevated scores on both RC6 and RC3 (which have correlations of .42 and .44, in the normative sample of men and women, respectively).¹⁹ The high Pa1 – low Pa3, or high RC6 and RC3 type of profile may well pose greater concern for the possibility that such a parent’s misanthropic, debased view of people may have a contaminating influence on the child.

The above psychometric considerations should alert evaluators to the potential “mismatch” in test-taking approach (this term was related to this writer by Y.S. Ben-Porath, personal communication, February 20, 2015) between pairs of custody litigants. In some cases, Parent A responds to the test inventory with a fairly open and non-defensive test approach and obtains a modestly elevated score on Scale 6 that falls within the range of clinical interpretation. At the same time, Parent Bmay respond to the test inventory with a defensive test-taking approach and his or her scores on all of the clinical or substantive scales are either suppressed or are within the average range. Here, we may see the possibility of a paradox in that Parent A trusts enough to candidly report his or her suspicious, distrusting thoughts of being wrongly treated or feeling threatened, but such parent’s distrustful perceptions may be contextually specific to the other parent, (and may also have a sound basis in reality), while not reflecting a general lack of trust in people. Such a parent may have no inclination or motivation to model or otherwise influence the child toward the development of mistrust, in general, or to convey alienating attitudes to the child about the other parent, specifically. Parent B, meanwhile, may be attempting to conceal or suppress hostile, mistrusting, grudging, acrimonious ideas and feelings viz. the other parent and may be savvy enough to deny any such thoughts or sentiments on the test. Such parent may be defensively denying any cynicism and/or proclaiming excessive moral virtue. Such parent may be of a type with an agenda who scaffolds either clearly negative, or distorted, or mixed messages to the child. Such messages can have the effect of eliciting confusion and doubt on the part of the child.

In conclusion, in the absence of very high scores on Scale 6 or RC6, where such high scores are elevated by the person’s endorsement of particular item content so that the probability of paranoid and possibly delusional ideation is increased, caution is particularly warranted in the interpretation of the elevated score. Even with the finding of a very high clinical score on Scale 6 or RC6 the evaluator needs to assess whether independent data may confirm or disconfirm the presence of paranoid or delusional ideation. This is not to say that one disregards a moderately clinically elevated score (e.g., T = 65) or that such score should be considered “normal” for parents involved in custody litigation. Rather, the evaluator should carefully assess whether inferences generated about a parent on the basis of his or her elevated test score on this particular scale are consistent with or discrepant from other, independent sources of data about this particular individual, and whether these inferences are more contextual and circumscribed or more pervasive. The same caution in interpretation applies of course to other test scales, and, more generally, caution has been urged in the interpretation of the MMPI-2 in CCE’s with special emphasis on contextual considerations.^28,29 And when an evaluator finds a mismatch between the MMPI-2 clinical profiles in the pair of parent litigants, it would be wise to particularly assess response styles and whether implications of the test profiles are supported by other data. Crandall et al.,³⁰ opined that the need to appear honest on self-report measures is one aspect of the need to appear socially acceptable. Crandall et al.,³⁰ proposed the existence of a “measurement paradox” in that the degree of a person’s “true honesty” in conjunction with his or her need to appear honest would detract from his or her measured social desirability scale score. Extending this concept to the CCE context, it is possible that some parents obtain average scores on the MMPI-2 validity scales and elevated scores on clinical scales such as Scale 6 (or RC6) based on both their level of “true honesty” and the desire to impress that they are being open and non-defensive. However, such a parent might be one in a pair of litigants in which the other parent’s level of “true honesty” (which is intertwined with his or her degree of self-awareness and self-deception) is overridden by his or her contextual need to appear socially acceptable; and, for such a parent, the demonstration of social acceptability, even to a degree that suggests a naively unrealistic portrayal, does not take much account of acting or at least in appearing to act in an honest or candid and non-defensive manner.

There is no scientific basis from which one can disentangle evaluation context, response style, and individual characteristics, with complete and unquestionable accuracy. Rather, the evaluator utilizes all sources of information obtained in CCE’s to generate and test hypotheses, and consider alternative hypotheses, in order to address the issues identified by the Court Order.

One possible area for future study concerns the statistical associations between scores on the MMPI-2 for litigating couples. Han et al.,³¹ studied dyadic agreement on MMPI-2 scales for 813 couples from the normative sample and 338 couples from a marital counseling sample. The highest correlations between couples, which were positive and were of medium magnitude, were on scales assessing attitudes about substance use, domestic satisfaction, and cynical mistrust. Higher dyadic agreement was found for the normative couples than for the counseling couples on most scales, suggesting greater “matches” on certain measures of personality for the normative couples in contrast to couples seeking counseling. Litigating couples probably represent a more extreme variation of marital/intimate relationship distress or dysfunction than couples who are seeking counseling. The direction and strength of MMPI-2 scale correlations (or degree of match/mismatch) for litigating couples remains to be studied.

Robert A Semel, Psy.D. is a licensed psychologist who conducts Family Court evaluations under the auspices of the New York State Office of Mental Health.

None.

Author declares there are no conflicts of interest.

None.

1. American Psychological Association. Guidelines for child custody evaluations in family law proceedings. American Psychologist. 2010;65:863‒867.
2. Association of Family and Conciliation Courts (AFCC). Model standards of practice for child custody evaluation. Wisconsin, USA. 2006.
3. Ackerman MJ, Ackerman MC. Custody evaluation practices: A survey of experienced professionals (revisited). Professional Psychology: Research and Practice. 1997;28:137‒145.
4. Ackerman MJ, Pritzl TB. Child custody evaluation practices: A 20-year follow-up. Family Court Review. 2011;49(3):618‒628.
5. Bow JN, Quinnell FA. A critical review of child custody evaluation reports. Family Court Review. 2002;40(2):164‒176.
6. Butcher JN, Graham JR, Ben-Porath YS, et al. MMPI-2 (Minnesota Multiphasic Personality Inventory 02): Manual for administration, scoring, and interpretation In: Ben-Porath YS (Ed.), University of Minnesota Press, Minneapolis, Minnesota, USA. 2001.  
7. Archer RP, Buffington-Vollum JK, Stredny RV, et al. A survey of psychological test use patterns among forensic psychologists. J Per Assess . 2006;87(1):84‒94.
8. Daubert v. Merrell Dow Pharmaceuticals Inc. 1993;509, U.S. 579, 113 S.Ct. 2786.
9. Bow JN, Gould JW, Flens JR, et al. Testing in child custody evaluations – Selection, usage, and Daubert admissibility: A survey of psychologists. Journal of Forensic Psychology Practice. 2006;6:17‒38.  
10. Graham JR. MMPI-2: Assessing personality and psychopathology (5th edn), Oxford University Press, New York, USA. 2012. p.688.
11. Caldwell A. How can the MMPI-2 help child custody examiners? Journal of Child Custody. 2005;2(1):83‒117.
12. Bagby RM, Nicholson RA, Buis T, et al. Defensive responding on the MMPI-2 in family custody and access evaluations. Psychological Assessment . 1999;11:24‒28.
13. Bathurst K, Gottfried AW, Gottfried AE. Normative data for the MMPI-2 in child custody litigation. Psychological Assessment . 1997;9(3): 205‒211.
14. Butcher J. Frequency of MMPI-2 scores in forensic evaluations. MMPI-2 News and Profiles. 1997;8(1):2‒4.
15. Ezzo FR, Pinsoneault TB, Evans TM. A comparison of MMPI-2 profiles between child maltreatment cases and two types of custody cases. Journal of Forensic Psychology Practice. 2007;7(2):29‒43.
16. Hynan DJ . Child custody evaluation: New theoretical applications and research. Springfield, Charles C. Thomas Publisher Ltd., Illinois, USA. 2014.
17. Posthuma AB, Harper JF.Comparison of MMPI-2 responses of child custody and personal injury litigants. Professional Psychology: Research and Practice. 1998;29:437‒443.
18. Pope KS, Butcher JN, Seelen J. The MMPI, MMP:I-2, & MMPI-A in court: A practical guide for expert witnesses and attorneys (3rd edn), American Psychological Association, Washington DC, USA. 2006.
19. Archer EM, Hagan LD, Mason J, et al. MMPI-2-RF characteristics of custody evaluation litigants. Assessment. 2012;19(1):14‒20.
20. Ben-Porath YS, Tellegen A. MMPI-2-RF (Minnesota Multiphasic Personality Inventory-2 Restructured Form) manual for administration, scoring, and interpretation. University of Minnesota Press, Minneapolis, Minnesota, USA. 2008.
21. Greene RL. The MMPI-2/MMPI-2-RF: An interpretive manual (3rd edn), Allyn & Bacon, Boston, USA. 2011.
22. Nichols DS . Essentials of MMPI-2 assessment (2nd edn), John Wiley & Sons, Hoboken, New Jersey, USA. 2011. p.396.
23. Levak RW, Siegel L, Nichols DS, et al. Therapeutic feedback with the MMPI-2: A positive psychology approach. Routledge, New York, USA. 2011.
24. Ellis E. Divorce wars: Interventions with families in conflict. American Psychological Association, Washington DC, USA. 2000.
25. Gould J . Use of psychological tests in child custody assessments. Journal of Child Custody. 2005;2:49‒69.
26. Erickson NS. Use of the MMPI-2 in child custody evaluations involving battered women: What does psychological research tell us? Family Law Quarterly. 2005;39:87‒108.
27. Friedman AF, Lewak R, Nichols DS, et al. Psychological Assessment with the MMPI-2. Lawrence Erlbaum Associates Inc., Mahwah, New Jersey, USA. 2001.
28. Ellis EM. Are MMPI-2 scale 4 elevations common among child custody litigants? Journal of Child Custody. 2012;9:179‒194.
29. Archer RP, Wygant DB. Child custody evaluations: Ethical, scientific, and practice considerations. Journal of Psychological Practice. 2012;17:1‒70.
30. Crandall VC, Crandall VJ, Katkovsky W. A children’s social desirability questionnaire. J Consult Psychol . 1965;29:27‒36.
31. Han K, Weed NC, Butcher JN. Dyadic agreement on the MMPI-2. Personality and Individual Differences. 2003;35:603‒615.