Biased assimilation belief perseverance group think and the

Biased assimilation, belief perseverance, group think and the DV paradigm Sacramento, CA. Feb 16, 2008

The human understanding when it has adopted an opinion draws all things else to support and agree with it. And though there may be a greater number and weight of instances to be found on the other side, yet these it either neglects and despises, or else by some distinction sets aside and rejects, in order that by this great and pernicious predetermination the authority of its former conclusion may remain inviolate. ----- Francis Bacon : The New Organon and Related Writings (1620).

The Woozle Effect l l l Langley & Levy 1977, (Wife Beating: the Silent Crisis) reported that half the women in the US were abused and cited a Gelles & Straus study as the basis for their inappropriately extrapolated statistic but the study had been conducted in a shelter. In 1980, Linda Macleod published a book called Wife Battering in Canada: The Vicious Circle in which she claimed (p. 21), every year 1 in ten Canadian women in a relationship are battered. The first Canadian domestic violence incidence survey was done in 2000. Mac. Leod’s figure was apparently based on the proportion of women that a shelter in Windsor, Ontario said they had to turn away.

Beyond Woozle l l Arias et al. (2002), quoting Stets & Straus (1992 a) as a source, claimed “women were seven to fourteen times more likely to report that intimate partners had beaten them up, choked them, threatened them with weapons, or attempted to drown them”. (p. 157). Of course, Stets & Straus say no such thing. There is no action by action analysis reported (such as choking or drowning) and they conclude that male and female violence rates are identical.

Beyond Woozle Neil Jacobson on Oprah: there are 2 kinds of male abusers: cobras and pitbulls l (to sell his book, When Men Batter Women, Jacobson and Gottman 1998) l

The Current Climate ABA Website – 85% of perpetrators are male l American Psychologist (Bornstein 2006) “studies indicate that more than 95% of abuse perpetrators are men” (p. 595). l

The Paradigm all IPV (intimate partner violence) is male perpetrated “ 95%” l Also “normative “ (Dobash & Dobash) to sustain male dominance l Female IPV is self defense or preemptive strikes -> still originates in male Hx of abuse l l

Feminist Sociology: Power & Control l As Dobash and Dobash (1979) put it, "Men who assault their wives are actually living up to cultural prescriptions that are cherished in Western society--aggressiveness, male dominance and female subordination-and they are using physical force as a means to enforce that dominance" (p. 24).

Main Beliefs l l l All male battering is for “power and control” and instrumental All female violence is self defensive, therefore, couples therapy is ruled out, too dangerous for women Male violence escalates if unchecked, hence, males need to be more “accountable”. Male violence is “normal violence” therefore, therapy is not warranted. Focuses on confrontation of sexist beliefs which “cause violence” and “male privilege” Source: Pence & Paymar (1993) Education groups for men who batter: The Duluth Model (Springer).

Main Beliefs 2 All “psychological causes” of IPV in males are excuses l All forms of couples interaction causes are “blaming the victim” l

Feminist Sociology: Power & Control (Yllo & Straus 1990) Focus on male violence (in general) towards women l Patriarchy: Male violence as “normal” control l Emphasis on gender relations and power l Bograd (1988) “All feminist researchers, clinicians, and activists address a primary question: “Why do men beat their wives? ”…why men in general use physical force against their partners and what functions this serves for a given society “ l

Impact on Court Mandated “Intervention” (not Treatment ) “Using slavery, a colonial relationship, or an oppressively structured workplace as an example, the facilitator can draw a picture of the consciousness of domination”. l – E. Pence and M. Paymar, Education Groups for Men who Batter: The Duluth Model, p. 49. ( see RDV chapter 14) l

Where did the “paradigm” originate? l l “Sexuality is to feminism what work is to Marxism…the molding, direction, and expression of sexuality organizes society into two sexes: women and men. This division underlies the totality of social relations. “ ---- Catherine Mac. Kinnon, Toward a feminist theory of the state (p. 3) Mac. Kinnon CA. Toward a feminist theory of the state. Harvard University Press. : Cambridge , Massachusetts, 1989.

Corvo and Johnson (2003) l the bedrock view of feminist thought “that battering (by males) is NEVER. . . provoked, hereditary, out of control, accidental, an isolated incident. It is not caused by disease, diminished intellect, alcoholism/addiction, mental illness or any external person or event. It is a means for men to systematically dominate, disempower, control and devalue women…. it is greater than an individual act, it supports the larger goal of oppression of women”

Normative? Acceptance of wife assault is not normative l Only 2% of North American men agree with “ its’ ok to hit your wife to keep her in line” (Simon et al 2001, n = 5238+) l Most marriages are not patriarchal – only 9. 4 % of North American marriages are “male dominant” ( Coleman and Straus 1986) l

More buts Violence rates in victim surveys using the CTS show equivalent levels of violence l This is true even when level of severity is assessed (Stets and Straus 1989/92) l In younger age samples, female violence rates are higher than male rates (Whittaker, Morse, Capaldi) l

The rejoinder CTS surveys take the violence “out of context” l Contextual patterns of domination are still exclusively male l However, when “context” ( control, motives for IPV) is measured too, the differences are not so black and white ( La. Roche 2005, Follingstad 2002) l

Lesbian Violence l l l Lesbians in currently aggressive relationships reported on past relationships with both men and women. Past relationships with women were more abusive Sample = 1099 in Phoenix Violence rates higher than in heterosexual relationships Suggest intimacy is a factor (and psychological problems that are triggered by intimacy)

Belief perseverance l l Lord, Lepper and Ross (1979) : Stanford study found that when presented with research contradicting our beliefs (in this case on the deterrence effect of capital punishment) , we are more likely to disparage the methodology than when it confirms our beliefs Ss judged research methods less acceptable when results contradicted their beliefs (regardless of whether they were pro or anti capital punishment)

Belief Perseverance 2 In IPV studies, this has taken the form of criticizing the CTS as failing to assess “context” l “context” is taken to mean instrumental use of violence, controlling violence, self defense l But the CTS is 16 X as sensitive as “crime victim” surveys and context has been assessed (see below) l

So how does the paradigm survive? 1) don’t ask, don’t tell l 2) ecological fallacy l 3) cherry pick the data l 4) drop female violence from analyses l 5) the “crime data” problem l 6) “evidence “ by citation l

How does the paradigm survive? 1 1) don’t ask, don’t tell l - surveys of women’s victimization, and of men’s perpetration l Initially only government survey (VAWS) was presented as a survey of criminal victimization of women l

Method 1 VAWS survey often cited as evidence for “male violence” but l Straus 1999 survey found VAWS only 1/16 as sensitive as CTS surveys l CTS surveys generate equivalent rates of violence by gender l “context” issue – self defense, instrumental violence equal by gender l

Incidence of IPV “victim surveys” (e. g. VAWS) generate low baseline reports of victimization but show a gender difference in victimization rate l E. g. the VAWS (US). 03% (for males) and 1. 1% (for females) l Advocates like to cite “males are 3 X as violent as females” – its. 03 vs 1. 1 % l Actually, these surveys have filters-> need to define IPV as crime and self as victim l

Hence Men had filters on answering that they were “crime victim” in a survey of crimes against women l Generated a low report rate from both genders but especially men l l

Incidence data Are typically seen and interpreted as unilateral victimization l E. g. if “repeat severe violence” is reported by 8% of US women are they all “battered women”? l

Battering Repeat, severe, male perpetrated violence is what we typically call battering l “wife battering” is this type of violence perpetrated by the male l What would “husband battering “ be? l How about the use of severe violence against a non-violent man l What about bilateral violence of equal severity? l

Reality check 1 Stets and Straus l Archer l Whittaker l La. Roche l

Stets and Straus 1989/92 Examined 1985 US National survey data, focusing on violence levels in married couples vs. couples from other survey who were cohabitating or dating l Females more likely to use unilateral violence than males n= 5005 married couples, 526 dating couples, 237 cohabiting couples l

Stets and Straus 1989/92 l l Data showed that when any violence was reported on a CTS survey, 10% of those married couples reporting (married) and 13% of cohabitating couples reported females using severe violence when male partner was non-violent (husband battering) Reverse pattern (wife battering) was 6% -7% So husband battering was about double wife battering

Stets and Straus 1992 Table 13. 2 with violence levels

Archer Psychological Bulletin 2000 Most comprehensive study of IPV by gender ever done l Meta-analysis, combined all previous studies into 1 combined analysis l Overall sample size was 64000+ l Calculated a measure (d’) that estimates the difference size in terms of a standard deviation (d’ =1 is 1 sd difference) l

Meta analytic study of sex differences in aggression (Archer 2000) l N men N women Effect Size l violence 30, 434 34, 053 -. 05* l Injury 7, 011 7, 531 . 15 l Medical treatment 4, 936 6, 323 . 08 l * women slightly more likely to be violent

Archer Key Findings Women are slightly more likely than men to use intimate violence d’ =. 05 l Women are slightly more likely than men to be injured d’ =. 15 l d’ =. 15 is about 1/6 of a standard deviation difference l

CDC (Centers for Disease Control) 2007 l l l Whittaker et al (2007) US national study n = 11, 370 age 18 -28 National Longitudinal Study on Adolescent Health (representative cohort sample) 24% reported some IPV using the CTS 2 Of those, half were reciprocal (49. 7%) Of those unilateral, 70% of perpetrators were female Most female injury resulted from reciprocal IPV

Whittaker 2007 l Regarding injury, men were more likely to inflict injury than were women and reciprocal intimate partner violence was associated with greater injury than was nonreciprocal intimate partner violence regardless of the gender of the perpetrator

Conclusion of incidence studies Most common form of IPV is mutual (Whittaker- 50%, Stets and Straus – 4852%) l Second most common form is female perpetrated (Whittaker 35%, S&S 28%) l Third most common form is male perpetrated (wife battering) (Whittaker – 15%, S&S 22%) l

Not just Reporting l In surveys, using representative community samples, the same results are obtained regarding relative frequency of male and female violence, regardless of whether the respondent is male or female (Stets & Straus 1992; Douglas & Straus 2003), hence, lack of agreement by gender is a non-issue.

so 1989 (n=5, 000+) 2007 (n=11, 000) l Stets and Straus Whittaker et al. l l Both 50% l Female only 28% 35% l Male only 22% 15%

Limitations of Feminist Position Female intimate violence more frequent than male (Archer 2001) 2. Female violence against non-violent males (Stets & Straus 1990) is not “self defense” 3. Lesbian abuse/violence greater than heterosexual male violence (Lie et al. 1991) 4. “Male dominance” in only 9. 4% of American Families (Coleman & Straus 1985) based on “final say” measures. 1.

Reaction to the CTS surveys criticized the reporting of conflict tactics out of context l i. e. that women were exposed to more instrumental forms of violence (patriarchal terrorism: M. Johnson, 1995) l Also argued that male violence was “qualitatively different” (e. g. instrumental vs. defensive) l

Canadian General Social Survey l l Canadian General Social Survey of 25, 876 respondents, equally split by gender. In this survey (Laroche, 2005) the “crime victim” filter was dropped and the focus was on “perceptions of crime”. 654, 000 (8% of all Canadian women) women and 546, 000 (7%) men reported being physically abused at least once in last 5 years

La. Roche l intimate terrorism described by Johnson (1995) : repeat, severe, instrumental violence by partner l was measured by questionnaire in this survey l Reported by 2. 6% of men and 4. 2% of women l In Canadian GSS Survey of 2004.

In other words l l l The intimate partners of 95. 8% of men and 97. 4% of women say they do not use instrumental intimate violence So, why then is gender given the weight that it is? There may be some stereotypical cases of IPV that are gender non-reversible But are they a small minority of cases? Are our policies “drift net” approaches to these cases?

Motives for IPV l Instrumental use of IPV by females contradicts the feminist view that all female IPV is for self-defense

Is female violence self defense? De. Keseredy & Schwartz (1998) in a survey of young adults found that l 62. 3% of women said their violence perpetration was never in self defence l 6. 9% said it was always in self defence. l The authors concluded that female violence was “self- defensive” !!!!! l

When “mistakes” occur l When “mistakes” occur repeatedly in the framing of research questions, the selection of which research findings are important or the actual interpretation of the data, it raises the question of whether these are just “mistakes”

Self Defense Follingstad, Wright, Lloyd, and Sebastian (1991) asked undergraduate subjects in South Carolina about assault experiences l of the total sample of 495 l 115 respondents reported they had been victimized by a partner using physical force l their perceptions of their assaulters’ and their own motivations were assessed. l

Follingstad, Wright, Lloyd, and Sebastian (1991) l l Women reported being victimized and perpetrating physical aggression twice as often as men. Furthermore, a greater percentage of women than men reported using aggression to feel more powerful (3. 4% vs. 0), to get control over the other person (22. 0% vs. 8. 3%), or to punish the person for wrong behavior (16. 9% vs. 12. 5%).

Douglas and Straus (2006) Was the Follingstad et al data atypical? l Douglas and Straus Examined partner violence in a college student sample in 19 countries (n = 9549) l On average female partner violence was 21% higher incidence than male partner violence l

NIJ Study of Coercive Control l Development and Validation of a Coercive Control Measure for Intimate Partner Violence: Final Technical Report l N= 757 - found equal coercive control by gender l Author(s): Mary Ann Dutton ; Lisa Goodman ; R. James Schmidt (2006)

Dutton (M. A. ), Goodman & Schmidt A valid measure of nonviolent coercive control was developed l The validity of the coercive control measure was assessed in a sample that included both males (n=302) and females (n=448) l Victimization was equal by gender l

Who struck first? l l Fiebert & Gonzalez(1997)- 29% of sample of 968 college age women in California reported initiating assaults on male partner. Bland Orn (1986) telephone survey conducted in Canada: Of the women who reported using violence against their husbands, 73. 4% said they used violence first. Stets & Straus (1992 a) reported that females said they struck first 52. 7% of the time Gondolf (2006) – 40% of female partners of men in court mandated therapy said they (the female) had struck first

Who fights back? l Stets and Straus if assaulted by their partner: l 25% of women l 15% of men hit back l

Method 2: Ecological Fallacy l l Draw Ss male S populations from courtmandated treatment groups of female populations from transition houses -> then generalize to the entire population without mentioning the selection factors that make these distinct populations Example Jaffe, Lemon & Poisson (2003) and Bancroft (2002) draw conclusions based on these populations for custody assessors

How does the paradigm survive? 2 citing self selected samples and then generalizing to society as a whole l - JCC papers: Jaffe , Bancroft l l

Domestic Violence and Child Custody l l l Books intended for custody assessors by Jaffe and Bancroft cite data drawn from shelter houses (Jaffe) and BIP’s (Bancroft) And then generalize to the entire population Both use “he” =“batterer” And “she” – victim What Kahneman and Tversky (1982) call a “representative heuristic” is developed; batterers have the attributes of maleness, and they alone pose a risk to the child.

Even worse Both suggest that if the woman is battered, the children will be too l Cite 40% overlap of wife/child abuse l No mention of the possibility of female spouse/child abuse is made in either book l Jaffe dismissed violence toward men in 1 paragraph as “qualitatively different” l

Even worse Warn the assessor to expect the batterer to be in denial and lie -> blueprint for a witchhunt (batterer will be male and will lie – guilt by gender) l This material is given to judges who must make these decisions: l State Justice Institute. Navigating Custody and Visitation Evaluations in Cases with Domestic Violence: A Judge’s Guide l

Judges Guide l The Family Violence Department of the National Council of Juvenile and Family Court Judges (NCJFCJ) l Cites Peter G. Jaffe et al. , Common Misconceptions in Addressing Domestic Violence in Child Custody Disputes, 54 JUV. & FAM. CT. J. 57 (2003). Promotes view that male is sole perpetrator, if wife abuse then child abuse, etc. l

Reality Check The actual overlap of physical wife/child abuse is 6% l actual rate is 6%: in “community samples” (Appel & Holden 1998: occurs mainly for least violent items (e. g. slapped)) l

Real Risks to Children l Statistically what are the real risk sources for children?

Mc. Donald, Jouriles et al (2006) Sampled 1615 dual parent households l Main focus was on exposure to violence by children but reported violence by severity and gender of perpetrator. l Used multi- stage probability sample l Response rate of 85% l

Mc. Donald et al (2006) l l l l l Table 1 Prevalence of Partner-Violent Acts Committed During the Past Year Within Couples and Separately for Men and Women Violent act Couple Male-to-female Female-to-male 1. Threw something 12. 9 4. 9 11. 6 2. Pushed, grabbed, or shoved 15. 6 11. 2 12. 2 3. Slapped 6. 8 2. 5 5. 6 4. Kicked, bit, or hit 4. 9 1. 8 4. 3 5. Hit or tried to hit with something 6. 4 2. 0 5. 6 6. Beat up 0. 8 0. 5 0. 3 7. Choked 0. 9 0. 7 0. 3 8. Burned or scalded 0. 5 0. 3 0. 1 9. Forced sex 1. 0 0. 7 0. 6 10. Threatened with a knife or gun 1. 1 0. 4 0. 9 11. Used a knife or gun 0. 4 0. 3 0. 1 Any violence 21. 45 13. 66 18. 20 Severe violence 8. 64 3. 63 7. 52 Note. Any violence one or more of items 1– 11 reported; severe violence one or more of items 4– 11 reported.

2 conclusions from study l l l 1 having children was a risk factor for IPV (presence of children upped the rate amongst couples reporting violence to 59% from 41%) 2 children were more exposed to violence from mothers 4. 3% exposed to severe male to female violence 11. 4% exposed to severe female to male violence

Health Canada Survey l study of 135, 573 child maltreatment investigations conducted by Health Canada and Published by the National Clearing House on Family Violence (Trocme and al. 2001) l The data tell a very different picture than that presented by the paradigmatic studies by Jaffe and Bancroft.

Data on Risk to Children: Health Canada 2001 This is a study of 135, 573 child maltreatment investigations conducted by Health Canada and Published by the National Clearing House on Family Violence (Trocme and al. 2001). l The study designates the abuse type as physical abuse, sexual abuse, neglect, emotional maltreatment and “multiple categories”. l

Data on Risk to Children: Health Canada 2001 Biological mothers (as compared to biological fathers) are the more likely substantiated perpetrator of l physical abuse (47 vs. 42%), l neglect (86% vs. 33%), l emotional maltreatment (61% vs. 55%) l and multiple categories (66% vs. 36%). l The biological father is the most likely perpetrator of sexual abuse (15% vs. 5%). l

US National Survey on Child Maltreatment : Health and Human Services Report 2004 l Children's Bureau Statistics and Research Report l http: //www. acf. hhs. gov/programs/cb/pubs/cm 04/index. html Sample size 718, 948 l For 2004, 57. 8 percent of the perpetrators were women and 42. 2 percent were men. l

Types of Maltreatment l During 2004, 62. 4 percent of victims experienced neglect, 17. 5 percent were physically abused, 9. 7 percent were sexually abused, 7. 0 percent were psychologically maltreated, and 2. 1 percent were medically neglected. 5 In addition, 14. 5 percent of victims experienced such "other" types of maltreatment as "abandonment, " "threats of harm to the child, " or "congenital drug addiction. " Source: http: //www. acf. hhs. gov/programs/cb/pubs/cm 04/index. html l Health and Human Services l l

Health and Humans Services Report

HHS Of all children under age 5 murdered from 1976 -2004 -- l 31% were killed by fathers l 30% were killed by mothers l 23% were killed by male acquaintances l 7% were killed by other relatives l 3% were killed by strangers l

Custody assessments are forensic assessments l l l the American Psychological Association Guidelines forensic evaluation summarised in Weissman and De. Bow (2003). The forensic evaluation must begin with a “cognitive set and evaluative attitude” of the assessor that is “neutral, objective and detached” (p. 39). The gender paradigm (Jaffe et al, Bancroft et al, etc) make adherence to this principle impossible.

Method 3 Cherry picked Data l Morse

US National Youth Survey Morse 1995 l l l Barbara Morse (1995) reported data from the U. S. National Youth Survey (NYS), a longitudinal study of self reported problem behavior involving a national probability sample of 1, 725 respondents. Violence was measured using the eight-item subscale from the CTS, injury was also assessed. For the years 1983, 1986, 1989, and 1992, female to male violence and severe violence was about double the rate of male to female violence and severe violence.

Morse l To demonstrate, in 1992 female to male severe violence was reported by 13. 8% of respondents, male to female by 5. 7% (Morse, 1995, Table 1, p. 255). l There was little or no change in the ratio of female to male vs. male to female violence over the years, with slightly over twice as much female initiated as male initiated violence.

Morse (1995) Frequency of “severe violence” by women is double that of men (table 3) l “used knife or gun” 4 X as frequent for women (4. 3% vs. . 1%) l

Morse 1995 l e Morse concludes “men were more likely than women to beat up their partner during the course of a year” Actual data: men: 2% did it 3. 3. times l women: . 9% did it 1. 3 times l Why pick 1 action? l Why generalize to “men” from 2% l

Method 4: Drop female violence from equation Wolfe and Reitzel-Jaffe, Gondolf, Jacobson et al (UW) – study male predictors of “dating violence” and “treatment outcome” without assessing whether female in couple is violent l BUT Douglas & Straus (2006) – females at UWO ( Jaffe’s sample) used 171% as much IPV as males l

Dropping Female Violence Gondolf- 40% of female partners of men in court mandated treatment said they (the women) initiated IPV l Jacobson et al (1994) 40% of the partners of violent men used severe violence tactics on the CTS but were not the focus of the research. l

Gondolf Performed a multi-site study of predictors of recidivism after BIP (psychoeducational) l Focused only on individual aspects of men l l (Gondolf, 1996, 1999, 2000 a, b; Gondolf & Jones, 2001; Jones & Gondolf, 2002)

Gondolf 2 l As is typical with the gender paradigm, his research questions focus on male violence, and whether or nor the men in these programs have violent female partners is de-emphasized. l However, even through this filter the following data emerged: l “ 66% of the women reported being physically aggressive toward their partners prior to the initial arrest” and 15% were also arrested when the man was, 25% were heavy drinkers, 40% had hit the man first (Gondolf, 1996, p. 39– 40). l www. mincava. umn. edu/documents/gondolf/batchar. html l

The missing other half Gender paradigm studies routinely ignore and discount female violence. l Was female violence included then in the Gondolf's structural model used to predict recidivism (Jones & Gondolf, 2002)? l No, although 40% of the men had wives who “hit them first” and the overall recidivism rates was 40%. l

Method 4: Jacobson et al reported “cobras and pit bulls” (male batterers types on Oprah Winfrey) l But “according to the wives themselves, almost half (28 / 57) would have qualified for the DV group if wife violence had been the criterion” (Jacobson et al. , 1994, p. 983). l

Jacobson et al l l No measures were taken of the wives' use of violence and all independent variables focused on male violence as though it were being produced unilaterally in all relationships (even though it clearly was not). In fact females in sample used more aversive acts in interaction than did males ( Cordova et al 1993) But study had been reported and publicized (Oprah)as example of “male violence” Book is called When Men Batter Women (Jacobson and Gottman)

Risk prediction: why include females? Longitudinal studies (Serbin, Capaldi) found “assortative mating” – aggressive women (assessed in Grade 1) and again in teens, more likely to l 1) aggress against male partner in late teens, early twenties l 2) choose violent male partner l 3) have children who make more emergency visits for injuries l

Why include females? The inclusion of females in risk prediction has been categorically ruled out as “victim blaming” l This overlooks the data that say female violence may contribute to overall risk of continuing violence in couple l Also “risk prediction” based solely on male characteristics is not very accurate l

Method 5: The “crime data “ problem

Crime Data l l l Police reports skew gender differences because males do not call as frequently for IPV males call only 1/3 to 1/10 as often as do females Stets and Straus (1992) females called 10 X as frequently La. Roche : for Canada Laroche found males 6%, females 16% but even when called , police are less likely to arrest a female ( Brown 2004)

Males Calling Police Males call less (Stets and Straus 1989 - 1%) because they are unlikely to be taken seriously by police (Brown, Buzawa) l Brown (2004) found that even injured males were unlikely to have their female perpetrator-partner arrested l

Brown 2004 l l l traced outcome of 2044 cases of domestic violence to Edmonton police When male injured, female charged 60% of time When female injured, male charged 90% of time When male calls but not injured, female charged 13. 2% of time When female calls but not injured, male charged 52. 5% of time.

Hence All statistics based on criminal justice system data re self-selected for male violence l They inflate ratios of male to female perpetrators l

Method 5 a: data inflation Certain items inflate rates of “violence” against women l Gilbert (2005) mary Koss included “attempted unwanted sex” in survey of campus rape l Rape rates project to 3000 per year at Berkeley l But 40 -80 per year seel guidance from rape counseling center, 2 report to police l

Method 5 a When Koss’ victims were asked directly, l 73% who had attempted unwanted sex did not believe they had been aped l 42% had sex later with the man who “raped “ them l The latest CDC report (Feb 2008) included this item as part of a measure of IPV l “threatened, attempted or completed physical violence or unwanted sex” l

Method 5 a Rates of IPV to women now exceeded those to men l in contrast to Whittaker’s study that did not use this measure l ABA website uses “reports to police” as true measure of incidence of DV, even in custody cases l

Dutton : RDV 47/47 studies made methodological errors in the direction of favoring the paradigm l 0/47 made errors in the opposite direction l

Violence Distributions According to the Paradigm

Violence Distributions According to the Data

Method 6 “Evidence” by Citation Dekeseredy cites Jaffe, Dobash, Gondolf who all mutually cite each other l Worse still the ABA Website cites the APA Website cites the National Judicial training Institute on the issue of abusive men seeking and getting sole custody l l The problem: there are no data/empirical studies on this issue

Main “spins” put on dv data l l l 1) sample questions biased (VAWS) 2) ecological fallacy: selected sample (transition house females, BIP males) generalized to entire population (e. g. Jaffe et al, Bancroft, Dobash &Dobash 2004, others) 3) data “cherry picked” (e. g. Morse) 4) data mis reported (e. g. Dekeseredy) 5) female violence not measured, included in “prediction” (e. g. Gondolf)

Main “spins” put on dv data l 6) citation circles – author a cites b cites c cites a as “evidence” l Usually Dekeseredy-> Jaffe -> Dobash-> Gondolf-> Dekersedy (repeat) -> Kimmel l Paradigm journals now exist: i. e. Violence Against Women

The gender bias l Studies of police, psychologists (Follingstad et al 2004) and the general public (Sorensen et al 2005) all show that the same action is more likely to be seen as abusive if it is performed by a man rather than a woman e. g. asking a partner where he (or she) has been

Hence 1) Police (arrest stats) and what follows from arrest (e. g. court-mandated treatment groups) severely underestimate female aggression 2) if a male in a court-mandated “psychoeducational” group mentions that his partner is violent- he is accused of “victim blaming” l

Paradigm perseverance Research that still finds males more violent than females is always based on either court-mandated treatment or transition house samples l It routinely overlooks the fact that the criminal justice system disproportionately is called more by female victims, responds more aggressively with female victims, arrests males more frequently l

One Size Fits All l Police arrest, victim services, court policies and psycho-educational intervention are all based on the 5. 7% of married males and 7. 3% of cohabiting males who use severe violence against their wives -> this has become a stereotype for all forms on IPV But that is 5. 7 -7. 3 % of everyone who reported IPV- what about the other patterns? Female severe assault against a non-violent male partner (husband battering) is 2 X as frequent as the reverse (wife battering)

A New reality Check

Longitudinal Studies What’s coming next l Female IPV in younger samples is increasing (eg Whittaker 2007) national sample of 18 -28 year olds l Female IPV is predictable from their developmental trajectory, starting at Grade 1 (Serbin 2006) l

Higher Incidence of female violence in younger populations is consistently found Ehrensaft- New York sample l Capaldi- Oregon youth survey l Douglas & Straus (2003) dating violence survey in 37 countries l Moffitt, Magdol – Dunedin, New Zealand study l l All studies find that personality disturbance is more important predictor than gender

Personality Disorders Now found to be the main predictor of IPV l Ehrensaft, Moffitt (NEM) l Longitudinal studies (Capaldi, Serbin) l

Concordia Longitudinal project l l Started in 1976, followed children into adulthood Aggressive children of both sexes had lower IQ’s and academic achievement than comparison controls (n = 4109). Both were more physically aggressive during play. Girls aggression was associated with a preference for male partners who were also aggressive.

Concordia Study l l As they approached adolescence, the aggressive girls had 1) elevated rates of smoking, alcohol and illicit drug use and 2) “continue(d) to seek out behaviorally compatible peer groups, probably comprised of boys and girls with similar aggressive or “predelinquent” behavioral styles” (p. 268). 3)The aggressive group had elevated levels of depression and anxiety disorder by late teens.

Concordia Longitudinal project l l 4) When they married their children had higher health risks (p. 272) and 5) the Aggressive girls had become Aggressive mothers, exhibiting maternal childhood aggression and having children who had more visits to the ER, specifically for treatment of injuries. indicates that these women will be select aggressive men and contribute to the intracouple aggression. Called “assortative mating”

Was the Concordia study an anomaly? Capaldi (Oregon) l Moffitt, Magdol ( Dunedin: New Zealand) l Ehrensaft ( New York State) l Morse ( US National Sample) l All found longitudinal peer cohort data where female IPV exceeded male IPV l Female adult IPV was predictable from earlier factors ( violence, personality disorder) l

The conclusions Females commit more IPV than males, especially in younger populations (Archer, Douglas & Straus) l Females are hurt more often but are not the exclusive victims ( Archer, others) l Males are not more controlling ( Felson & Outlaw) l DV does not escalate if left unchecked (Feld & Straus 1992) l

The conclusions 2 l l l IPV has many patterns (bilateral, female perpetrated) paradigm of male domination is a stereotype that does not represent the majority of IPV cases ( Stets and Straus found that 5. 7% of IPV was male severe- female none used once, about 2/3 of all 1 time IPV is repeated, therefore about 3. 8% fits the stereotype) The Canadian Social Survey found 4. 2% were male perpetrated and instrumental (over 5 years)

Personality Disorders 2 Males convicted of spouse assault have high incidences of personality disorder l Typically antisocial or borderline l l Females convicted of spouse assault have same personality disorder peaks (Moffitt, Magdol, Henning)

Borderline PD in Males Dutton (1994 -2006) Borderline PD, insecure attachment, trauma exposure lead to IPV in large subsample of males in treatment for spouse assault l Mauricio et al confirmed this finding l

Borderline PD in Females Henning , Jones and Holford (2003): l Women in IPV treatment 5 times as likely as males to have BPD l Hughes et al (2007) in sample of 103 court mandated women, borderline features were associated with IPV perpetration l

NEM (Moffitt et al 2001) Longitudinal study of peer cohort in new Zealand (the Dunedin sample) l Similarities in anti social behavior for girls and boys outweighed the differences l l Source: Sex Differences in Anti-social Behavior l T. Moffitt, A. Caspi, M. Rutter, P. Silva. Cambridge U Press

NEM (Negative Emotionality) l Moffitt et al found that “Negative Emotionality” predicted abusiveness in both genders. l Negative Emotionality was measured by 49 true false items from the Multidimensional Personality Questionairre (Tellengen & Waller 2001). These in turn measured: 1) reactions to stress “I often get irritated at little annoyances”. 2) experience of emotion “ Sometimes I feel strong emotions like anxiety or anger without knowing why” 3) expectations of others “ Most people stay friendly only so long as it is to their advantage” 4) attitudes towards using aggression “When someone hurts me, I try to get even” l l l

NEM l Negative emotionality measured in women at age 15 predicted their use of violence towards an intimate other at age 21 regardless of whether that person fought back or not.

ASPD in Mothers ASPD in mothers is correlated with children’s being maltreated l Julia Kim-Cohen at Yale has published data on this relationship l Lisa Serbin in the Concordia longitudinal study has found these high risk mothers are predictable from early adolescence l CD girls become mothers whose children have more emergency room visits l

Personality Disorders Now found to be the main predictor of IPV l Ehrensaft, Moffitt l Longitudinal studies (Capaldi, Serbin) l Also relevant in custody – some developing research showing PD (NPD, Borderline PD, ASPD) in either parent related to externalizing and internalizing behaviors in children l

Our Current View of IPV l Based on groupthink, group polarization along gender group lines l Males are the “outgroup” -> violent, controlling, warlike l Females are the ingroup -> gentle, nonviolent

Group Polarization: Attitudes Shift to extreme in “like minded” groups l l 1) interaction produces generation of new ideas supportive of group norm -> in groups where ideas are received passively, polarization is less 2) people attempt to present themselves favorably to others, generate persuasive arguments in socially desirable direction 3) Group schema: a complex of ideas, information shifts to the extreme, becomes dogma, ideology-> this then becomes the new social reality -> informational influence is provided from within this framework 4) groupthink – when group must develop position under perceived threat

Groupthink (Janis 1972) l “ a mode of thinking that people engage in when they are deeply involved in a cohesive ingroup, when members strivings for unanimity override their motivation to realistically appraise alternative courses of action” l (Wrightsman & Deaux, p. 463) l -> studied failed/successful US foreign policy decisions ( e. g. bay of Pigs, Cuban missile Crisis)

Groupthink l l l l Occurs when: 1)highly cohesive group (selected for unanimity) 2)insulated from independent judgments 3)perceived threat (or injustice) Symptoms: 1)illusion of invulnerability -> leads to excessive optimism/risk taking 2)collective rationalizing of warnings that might temper position 3)unquestioned belief in groups moral superiority 4) negative stereotype of outgroup making negotiation unfeasible (see Corvo 2002). 5) Direct pressure on dissenters from group ideology 6) Self-censorship of deviation from apparent consensus 7) Shared illusion of unanimity 8) Emergence of self-appointed “mindguards” to protect group from adverse information.

Groupthink l l When an activist group with a predetermined direction confers in isolation from dissenting views, 1) status is gained from taking more extreme positions (in the pre-determined direction) 2) people with strong needs for dominance will advance more extreme positions in order to gain status, power and control of the group 3) these traits will then be projected onto the outgroup ( “battering is all about power and control”)

In Domestic Violence l l Activist group formed the Paradigm explanation for DV Then through legal (lawsuits against police forces who failed to arrest violent men) And political influence (sold dv as a woman’s right issue at a time when no one was against women’s rights) Took over county dv councils, set agenda for court intervention, judges training, APA/ABA view of dv

The perception of Male Action l The perception of male action is fundamentally different- it is seen as more dangerous and abusive even when the act is identical

Sorenson & Taylor (2005) Sorenson & Taylor implemented a random digit dialed survey in four languages of 3, 769 adults in the Los Angeles area. l They were presented with five vignettes in which characteristics of the victim, assailant and incident were experimentally manipulated. l

Sorenson and Taylor 2 l The vignette variables (assailants motive, type or intensity of abuse and whether alcohol was involved, presence of weapons, presence of children, frequency of abuse) and respondent characteristics were examined using a multivariate log regression.

Sorenson & Taylor (2005) 3 l l l Judgments about women’s violence against male intimates (vs. the opposite) were less harsh and took contextual factors into account. This was true for both male and female respondents. Across vignettes male violence was seen as more likely to be illegal, that the police should be called, that the assailant should be arrested, should serve jail or prison time, that a restraining order should be issued

Sorenson & Taylor (2005) 4 l l l It is important to note that while some of the abuse types were physical, others were psychological describing control or humiliation. Physical abuse (slap, force sex) were more likely to be perceived as illegal by women when the assailant was a man. To test the hypothesis that social norms about female abuse are less clear, Sorenson & Taylor examined the standard deviation of the residuals (a measure of response variability); it was 36% greater when the assailant was a woman.

Perception of crime by gender l l Not only the general public but professional psychologists rate aggression as less serious when it is performed by females and even when it is psychological aggression, according to a study by Follingstad and her colleagues (2004). Two scenarios describing the context and psychologically abusive behaviors with the genders reversed were given to 449 clinicians (56% male), median age 52.

Follingstad et al l Psychologists rated male perpetrated behavior as more abusive and severe than the wife’s use of the same actions. Contextual factors (frequency/ intent/perception of recipient) did not affect this tendency. As Follingstad et al concluded, ‘the stereotypical association between physical aggression and males appears to extend to an association of psychological abuse and males” (p. 447).

Ferguson & Nagy (2004) When gender is manipulated in a dv scenario (X comes home late is accused by Y of infidelity) Y punches X who has facial bruise next day l Perceived as more criminal when Y (perpetrator) is a male l Effect occurred across European. American, African-American and Hispanic groups l Effect occurred for both male and female respondents (university students) l

Do we not bleed? Pimlott-Kubiak and Cortina (2003) l Used data from the VAWS l Found that amount of exposure to trauma determined long term negative effects l Gender did not predict amount of negative effects. l l Source: Journal of Consulting and Clinical Psychology 2003, 71, 3, 528 -539.

Female Dangerousness Coontz, Lidz & Mulvey (1994) found that clinical predictions of dangerousness made in psychiatric emergency rooms consistently underestimated female dangerousness. l Predictions that a male would not be violent were correct 70% of the time, but for females, they were correct only 55% of the time. l

Reform 1) drop gender as a central organizing worldview (the demographic du jour) l 2) focus on best available evidence for risk to adults and children from intimate others l 3) ensure professionals do not have erroneous mindsets influencing decision making (e. g. arrest, custody) l 4) develop awareness of groupthink for all policy makers l

References l Dutton, D. G. (2006) Rethinking domestic violence. UBC Press: Vancouver, BC. l Johnson, M. P. (1995). Patriarchal terrorism and common couple violence: Two forms of violence against women. Journal of Marriage and the Family, 57, 283 -294. l Mac. Kinnon CA. Toward a feminist theory of the state. Harvard University Press. : Cambridge , Massachusetts, 1989.