The effects of domestic violence on survivors, who are primarily women, may be severe. Battered women's advocates often note that, in custody cases, the batterer often "looks better" to the court than the victim does because he is confident and calm, whereas she is still suffering the effects of his abuse and therefore may appear hysterical, weepy, angry, or otherwise not "together." n1
n1 LUNDY BANCROFT & JAY SILVERMAN, THE BATTERER AS PARENT 122-23 (2002); Joan S. Meier, Domestic Violence, Child Custody, and Child Protection: Understanding Judicial Resistance and Imagining the Solutions, 11 AM. U.J. GENDER SOC. POL'Y & L. 657, 690 (2003); Joan S. Meier, Notes from the Underground: Integrating Psychological and Legal Perspectives on Domestic Violence in Theory and Practice, 21 HOFSTRA L. REV. 1295, 1312 (1993).
When a custody evaluation is conducted by a psychologist, the revised version of the Minnesota Multiphasic Personality Inventory (MMPI-2) is often used as part of the evaluation process. n2 The MMPI-2, like other traditional psychological tests, was not designed for use in custody evaluations [*88] and arguably should not be used for such purpose except "when specific problems or issues that these tests were designed to measure appear salient in the case." n3 However, if an evaluator chooses to use it, great care should be taken to make sure that it is not misinterpreted. A misinterpretation could result in placing custody of a child with a batterer, which could put the child at severe risk. Additionally, for many parents, especially those who have been primary caretakers, loss of custody of their children is the most frightening thing they can imagine, short of death. Loss of such an important liberty interest should not occur because of flawed information presented to the court by anyone, including one deemed to be an expert.
n2 Marc J. Ackerman & Melissa C. Ackerman, Child Custody Evaluation Practices: A 1996 Survey of Psychologists, 30 FAM. L.Q. 565, 573 (1996).
n3 GARY MELTON ET AL., PSYCHOLOGICAL EVALUATIONS FOR THE COURTS: A HANDBOOK FOR MENTAL HEALTH PROFESSIONALS AND LAWYERS 503 (2d ed. 1997).
Abusers typically disavow any wrongdoing and claim the mother is "crazy" or unfit in some other way. n4 The MMPI-2 cannot disprove a batterer's claim of innocence, because there is no known MMPI-2 abuser "profile." n5 In fact, many MMPI-2 profiles of batterers do not reveal any psychopathology. n6
n4 Judith L. Herman, Crime and Memory, in TRAUMA AND SELF 3, 11-12 (Charles B. Strozier & Michael Flynn eds. 1996); cf. Catherine Ayoub et al., Alleging Psychological Impairment of the Accuser to Defend Oneself Against a Child Abuse Allegation: A Manifestation of Wife Battering and False Accusation, in ASSESSING CHILD MALTREATMENT REPORTS: THE PROBLEM OF FALSE ALLEGATIONS 191, 201-03 (Michael Robin ed. 1991) (finding that in a study of mothers who alleged fathers abused their children, but where fathers alleged the mothers were "crazy," investigation revealed that all the fathers were abusive to the mothers).
n5 Randy K. Otto & Robert P. Collins, Use of the MMPI-2/MMPI-A in Child Custody Evaluations, in FORENSIC APPLICATIONS OF THE MMPI-2 (Yossef S. Ben-Porath et al., eds. 1995). Research using the other most commonly used personality assessment tool, the Millon Clinical Multiaxial Inventory (MCMI) likewise has not demonstrated any one batterer profile. Robert J. Craig, Use of the Millon Clinical Multiaxial Inventory in the Psychological Assessment of Domestic Violence: A Review, 8 AGGRESSION & VIOLENT BEHAVIOR 235, 240 (2001).
n6 Amy Holtzworth-Munroe & Gregory G. Stuart, Typologies of Male Batterers: Three Subtypes and the Differences Among Them, 116 PSYCHOLOGICAL BULL. 476 (1994) and sources cited therein; Gayla Margolin, Interpersonal and Intrapersonal Factors Associated with Marital Violence, in FAMILY ABUSE AND ITS CONSEQUENCES: NEW DIRECTIONS IN RESEARCH (Gerald T. Hotaling et al., 1988); Catherine Wall, Battered Women and their Batterers: Personality Variables and Attitudes Toward Violence (1993) (unpublished Ph.D. dissertation, California Institute of Integral Studies) (U.M.I. No. 9324523).
Battered women, however, based on the results of the MMPI-2, may appear to be suffering from various psychopathologies, including but not limited to borderline personality disorder, paranoia, histrionic personality disorder, or even schizophrenia. n7 The custody evaluator may conclude that the mother's apparent psychopathology is a personality disorder and [*89] therefore characterological (a "trait"). Personality disorders are viewed by many psychologists as highly treatment resistant and therefore curable, if at all, only with very long-term therapy and often psychotropic drugs. n8 The custody evaluator might even conclude that the mother's apparent "psychopathology" caused the physical conflict between the parents.
Clinicians inadvertently maintaining such assumptions may examine a battered woman's profile and conclude, "Oh, no wonder she gets beat up. She's crazy, schizophrenic, borderline, and unstable," and the clinician may fail to investigate alternative conceptualizations for the woman's psychological presentation. n9
n7 Lynne B. Rosewater, Battered or Schizophrenic? Psychological Tests Can't Tell, in FEMINIST PERSPECTIVES ON WIFE ABUSE (KERSTI YLLO & MICHELE BOGRAD EDS., 1988); LENORE E. WALKER, ABUSED WOMEN AND SURVIVOR THERAPY: A PRACTICAL GUIDE FOR THE PSYCHOTHERAPIST 75 (1994) (see also 111-13 and 379 for further discussion of problems of mis-diagnosis).
n8 THEODORE MILLON & ROGER DAVIS, PERSONALITY DISORDERS IN MODERN LIFE 87 (2000).
n9 John Morrell & Linda Rubin, The Minnesota Multiphasic Personality Inventory-2, Posttraumatic Stress Disorder, and Women Domestic Violence Survivors, 32 PROF. PSYCHOL: RES. & PRAC. 151 (2001).
Failure to investigate other possible causes could even lead the custody evaluator to doubt whether the woman was abused at all--perhaps someone so unstable has made false allegations or perhaps she has attacked her partner and he has simply acted in self-defense.
An "alternative conceptualization" is that the woman's psychological presentation is a reaction to the abuse she has suffered (a reactive "state"). If battered women's MMPI elevations are reactive, one would expect that their MMPIs prior to being battered would be relatively "normal," that their MMPIs during the battering relationship would be elevated, and that their MMPI elevations would decrease after the abuse ended. Additionally, it might be expected that the severity of the abuse suffered by the woman or the length of time she was abused might correlate with the MMPI elevations.
This article surveys the available research on battered women's MMPI/MMPI-2 profiles. That research tends to support the hypothesis that a battered woman's MMPI-2 profile often is a result of the abuse she has suffered (a reactive "state") and therefore should not be viewed by child custody evaluators as evidence that she has personality traits indicating that she would not be a fit parent. n10
n10 Mention should be made of the assumption by some mental health practitioners that domestic violence causes the quality of the victim's parenting capabilities to suffer. See Cris M. Sullivan et al., Beyond Searching for Deficits: Evidence that Physically and Emotionally Abused Women Are Nurturing Parents, 1 J. EMOTIONAL ABUSE 51, 52 (2000). If one accepted this assumption, one could draw the conclusion that custody to the battered mother might not be in the best interests of the child. Even putting aside the important and obvious, but often overlooked, question of whether custody to the abuser of the child's mother would be more likely to be in the best interests of the child (see generally BANCROFT & SILVERMAN, supra note 1), research does not support any assumption that the mother's parenting capabilities are substantially impaired by the domestic violence she suffered. Id. Individual battered women, like individual nonbattered women, may have parenting deficits (as well as strengths), but no assumptions should be made that all battered women are incapable of being good parents to their children, especially after the abusers have been removed from the home and the battered mothers have had time to recover from the abuse. Alytia Levendosky & Sandra Graham-Bermann, Behavioral Observations of Parenting in Battered Women, 14 J. FAM. PSYCHOL. 80 (2000). Nor should a battered mother be viewed as neglectful simply because domestic violence occurred in her home. In re Nicholson, 181 F. Supp. 2d 182 (E.D.N.Y. 2002) (granting a preliminary injunction against the city of New York, prohibiting the city from removing children from mothers who were victims of domestic violence because, as victims, they "engaged in domestic violence"). In Nicholson v. Williams, 203 F. Supp. 2d 153 (E.D.N.Y. 2002), Judge Weinstein elaborated on the grounds for the injunction. In Nicholson v. Scoppetta, 3 N.Y.3d 357, 820 N.E.2d 840 (N.Y. 2004), the New York Court of Appeals addressed three questions certified to it by the federal court, and held that exposing a child to domestic violence is not presumptively neglectful.
[*90] II. What Is the MMPI-2?
The original (1942) MMPI was a self-report questionnaire, containing 566 questions to be answered yes or no. n11 The MMPI-2 (1989) is a revised version, containing 567 questions. Based on the answers the testee gives to those questions, the testee's score is generated for each of the MMPI clinical scales. The names of the scales, along with approximate ordinary language descriptions of what the scales appear to measure, are as follows:
1. Hypochondriasis (Hs): concerns about one's body and health.
2. Depression (D): depression.
3. Hysteria (Hy): repression or tendency to sit on one's feelings.
4. Psychopathic Deviate (Pd): anger
5. Masculinity-Femininity (Mf): a high score may indicate lack of comfort with one's biological sex or cultural sex roles. n12
6. Paranoia (Pa): fearfulness, suspicion
7. Psychasthenia (Pt): anxiety, worry, or tension.
8. Schizophrenia (Sc): confusion in thought processes; feelings of being overwhelmed.
9. Hypomania (Ma): a high score may indicate high activity level
10. Social Introversion (Si): high indicates introverted; low indicates extroverted.
n11 Except as otherwise noted, the following discussion of the MMPI and the MMPI-2 is taken from ALAN F. FRIEDMAN ET AL., PSYCHOLOGICAL ASSESSMENT WITH THE MMPI-2 (2001).
n12 Scale 5 was added after the MMPI was originally published--as was scale 0--and is not truly a clinical scale in the same sense as scales 1-4 and 6-9. See FRIEDMAN ET AL., supra note 11, at 112-19 and 316-20. This article will not attempt to analyze battered women's scores on scale 5.
The names of scales 1-4 and 6-9 are labels that were standard diagnoses at the time the MMPI was first being developed, but improvements in diagnosis and treatment of mental illnesses since that time have made many of the names outmoded and misleading. Additionally, elevations on a particular scale do not necessarily indicate the testee can be diagnosed as having the mental disorder or condition matching the "name" of that scale. The categories of mental disorders currently commonly used by psychiatrists and psychologists are those found in the fourth edition of the Diagnostic [*91] and Statistical Manual of Mental Disorders (DSM-IV), published in 1994, to which text revisions were added in 2000 (DSM-IV-TR).
The MMPI was first published in 1942. The MMPI revisors did not attempt to track the symptoms of the mental disorders as categorized in any version of the DSM. Therefore, for example, a testee with clinical elevations on the MMPI-2 8 (Sc) scale may not meet the DSM-IV-TR diagnostic criteria for schizophrenia. For these reasons, numbers rather than names are now used to describe the MMPI-2 clinical scales. n13
n13 See FRIEDMAN ET AL., supra note 11, at 9; JOHN R. GRAHAM, MMPI-2: ASSESSING PERSONALITY AND PSYCHOPATHOLOGY 225-26 (1993); ROBERT M. KAPLAN & DENNIS P. SACCUZZO, PSYCHOLOGICAL TESTING: PRINCIPLES, APPLICATION, AND ISSUES 412 (4th ed. 1999).
A testee's raw scores on the scales are converted into uniform T scores so that scores on the different scales can be compared. Someone who achieves a uniform T score of fifty on a clinical scale has scored at about the fifty-fifth percentile, which means that about 45% of the population would score above fifty. It is usually recommended that a T score of sixty-five or above be considered in the "clinically significant" range. n14 Some prefer to use seventy as the cutoff. A uniform T score of sixty-five is at the ninety-second percentile, which means that only 8% of the population would score above sixty-five. Similarly, only 1% would score above eighty.
n14 FRIEDMAN ET AL., supra note 11, at 21-22, referencing J.N. BUTCHER & C.L. WILLIAMS, ESSENTIALS OF MMPI-2 AND MMPI--A INTERPRETATION (1992).
In addition to the clinical scales, the MMPI also has "validity" scales, which seek to measure the mindset of the testee toward taking the test. For example, most testees trying to convince a judge or jury of an insanity defense would try to "fake bad." Most parents involved in child custody evaluations would want to try to look as good as possible. n15 The issue of how batterers and battered women score on the validity scales is an important one, but is outside the scope of this article.
n15 Kay Bathurst, et al., Normative Data for the MMPI-2 in Child Custody Litigation, 9 PSYCHOL. ASSESSMENT 205 (1997); Allan Posthuma & James Harper, Comparison of MMPI-2 Responses of Child Custody and Personal Injury Litigants, 29 PROF. PSYCHOL.: RES. & PRAC. 437 (1998); Jeffrey Siegel, Traditional MMPI-2 Validity Indicators and Initial Presentation in Custody Evaluations, 14 AM. J. FORENSIC PSYCHOL. 55 (1996); Otto & Collins, supra note 5, at 245 (noting that the MMPI validity scales can be useful even when both parents respond defensively if one parent responds much more defensively than the other).