Types of Domestic Violence
Research Evidence
Michael P. Johnson, Ph.D.
Sociology, Women's Studies, and
African & African American Studies
Penn State
Photos from Donna Ferrato, Living with
the Enemy. New York: Aperture, 1991
Catholic Family Services
of Peel Dufferin
June 3, 2010
McKeesport, PA
 The Continuing Gender Debate
 Anti-feminist politics and conflicting data
 Explaining the ostensible contradictions
A
Control-based Typology of Partner Violence
 The
three major types (plus one or two)
 Gender differences and sampling biases
 Dramatic Differences Among the Types
 Violence severity, frequency, mutuality, and escalation
 Health consequences
 Relationship consequences
 Miscellaneous other major differences
 Preview of Policy Implications
Screening, Primary prevention/education, Intervention
with perpetrators, Intervention for survivors, Law enforcement
issues, Custody and access issues
The Anti-feminist Backlash
 The
Men’s Project. February 2009. Submission to
Ontario Ministry of the Attorney General
 …the
Ontario Government may be in violation of their
obligations… [because] the existing network of shelters for
victims of family violence exclude men….
 Pittsburgh
Post Gazette July 26, 2009
 Feminist
ideologues ignore research that shows domestic
violence is just as often started by women as by men
 Globe
 Men
and Mail July 27, 2002 (Web site)
as likely to suffer spousal abuse, Statscan says.
General Surveys Indicate That
Women Are as Violent as Men
Heterosexual intimate partner violence
by gender
Data Source
Men
Women
U.S., NFVS, 1975—the beginning
U.S., NSFH, 1988
North Carolina, 8th & 9th Grade, 1994
U. of Maine, students, 1997
New Zealand, young adults, 2002
Canada, GSS, 2004
51%
53%
35%
39%
39%
54%
49%
47%
65%
61%
61%
46%
But Agency Studies Indicate That
Men Are the Batterers
Heterosexual intimate partner violence
by gender
Data Source
Men
Women
Cleveland, Divorce Court, 1966
Ontario, Family Court, 1982
92%
94%
8%
6%
Santa Barbara, CA, Police, 1983
94%
6%
U.K., Emergency Rooms, 1988
U.S., FBI, 1996-2001
83%
75%
17%
25%
Canada, Spousal Homicide, 1995-2005
82%
18%
A Small Theory
that
Reconciles the Contradiction
There is more than one type of partner violence
 The different types are differently gendered
 Both major sampling plans are biased

 General
survey studies are biased toward
situationally-provoked violence, which women
are as likely to perpetrate as are men
 Agency studies are biased toward coercive
controlling violence, perpetrated almost entirely
by men
 The Continuing Gender Debate
 Anti-feminist politics and conflicting data
 Explaining the ostensible contradictions
A
Control-based Typology of Partner Violence
 The
three major types (plus one or two)
 Gender differences and sampling biases
 Dramatic Differences Among the Types
 Violence severity, frequency, mutuality, and escalation
 Health consequences
 Relationship consequences
 Miscellaneous other major differences
 Preview of Policy Implications
Screening, Primary prevention/education, Intervention
with perpetrators, Intervention for survivors, Law enforcement
issues, Custody and access issues
Intimate Terrorism
Coercive Control
Violent Resistance
Resisting the Intimate Terrorist
Situational Couple Violence
Situationally-provoked Violence
Separation-instigated Violence
No History of Violence or Control
Mutual Violent Control
Two Intimate Terrorists
Domestic Violence/Intimate Terrorism
Two major subtypes: (a) Emotionally dependent; (b) Antisocial
Coercive Control Scale
Thinking about your husband [yourself], would you say he [you]…
 is jealous or possessive?
 tries to provoke arguments?
 tries to limit your contact with family and friends?
 insists on knowing who you are with at all times?
 calls you names or puts you down in front of others?
 makes you feel inadequate?
 shouts or swears at you?
 frightens you?
 prevents you from knowing about or having access to the family
income even when you ask?
*These are items from the 1995 National Violence Against Women Survey (Tjaden & Thoennes, 1998). They were
adapted from the Canadian Violence Against Women Survey (Holly Johnson, 1996).
Gender Symmetry/Asymmetry
by Type of Violence
(1970s Pittsburgh: Violent husbands and wives)
Husbands
Intimate terrorism
97%
Wives
3%
N
97
Violent resistance
4%
96%
77
Situational couple violence
56%
44%
146
2000s Britain: IT 87% male; VR 10% male; SCV 45% male
The Biases of Major Sampling Plans
(Violent men: Pittsburgh*)
General
Sample
(n = 37)
Court
Sample
(n = 34)
Shelter
Sample
(n = 50)
Intimate terrorism
14%
68%
78%
Violent resistance
0%
0%
2%
Situational couple violence
86%
29%
18%
*The pattern is essentially the same for the British research.
 The Continuing Gender Debate
 Anti-feminist politics and conflicting data
 Explaining the ostensible contradictions
A
Control-based Typology of Partner Violence
 The
three major types (plus one or two)
 Gender differences and sampling biases
 Dramatic Differences Among the Types
 Violence severity, frequency, mutuality, and escalation
 Health consequences
 Relationship consequences
 Miscellaneous other major differences
 Preview of Policy Implications
Screening, Primary prevention/education, Intervention
with perpetrators, Intervention for survivors, Law enforcement
issues, Custody and access issues
Pittsburgh data
Mixed sample
76% severe
75% escalated
29% mutual
1/25
couples
28% severe
28% escalated
69% mutual
1/8
couples
British data
Mixed sample
43% severe
78% escalated
15% mutual
13% severe
20% escalated
87% mutual
Canadian GSS 1999
Previous partner
41% frequent violence
56% feared for life
8% frequent violence
17% feared for life
Canadian GSS 2004
Previous/current
partner
57% frequent violence
60% feared for life
8% frequent violence
9% feared for life
Women’s Health Outcomes
by Type of Male Violence
Any Injury
Severe injury
General health
Pittsburgh
SCV
56%
IT
94%
***
U.S., NVAW
13%
32%
***
Pittsburgh
28%
76%
***
U.S., NVAW
2%
5%
*
Good to Very
Good
Fair to Good
*
37%
79%
***
Chicago
Post-traumatic stress+ U.S., NVAW
+ Percent above the median for female victims of partner violence
*.05 **.01 ***.001
Relationship Outcomes
by Type of Male Violence
Low marital happiness Pittsburgh
Situational
Couple
Intimate
Violence Terrorism
13%
50%
***
Left more than once
Pittsburgh
26%
74%
***
Rarely a good time
Sex often unpleasant
U.S., NVAW
Pittsburgh
Pittsburgh
7%
3%
9%
29%
20%
23%
***
***
***
***.001
Need to Re-assess Everything
Various studies by Various Social Scientists
 Intergenerational
“transmission”
 SCV
d = .11; IT d = .35
 SCV odds ratio = 2.40; IT odds ratio = 7.51
 Marriage
 SCV
b = -.62; IT b = .58
 Gender
traditionalism or hostility toward women
 Traditionalism:
SCV d = -.14; IT d = .80
 Hostility: non-viol., SCV, IT, IT = 154, 153, 135, 131
 Gender,
frequency, severity, escalation, mutuality,
impact on victim, impact on children, etc.
 The Continuing Gender Debate
 Anti-feminist politics and conflicting data
 Explaining the ostensible contradictions
A
Control-based Typology of Partner Violence
 The
three major types (plus one or two)
 Gender differences and sampling biases
 Dramatic Differences Among the Types
 Violence severity, frequency, mutuality, and escalation
 Health consequences
 Relationship consequences
 Miscellaneous other major differences
 Preview of Policy Implications
Screening, Primary prevention/education, Intervention with
perpetrators, Intervention for survivors, Law enforcement
issues, Custody and access issues
Preview
of Policy Implications
 Screening
 Primary
prevention/education
 Intervention
with perpetrators
 Intervention
for survivors
 Law
enforcement
 Custody
and access issues
Different types of partner violence have…
 Different
causes
 Different developmental trajectories
 Different effects
 Different successful intervention strategies
We make big mistakes if we don’t
make big distinctions.
Support Your Local Women’s Shelter
Safety
Support
Information
Advocacy
Photos from Donna Ferrato, Living with
the Enemy. New York: Aperture, 1991
Philadelphia, PA shelter
Screening/Triage
 Different
models for different clients
 To screen we need information on control
and violence for both members
 Safety first!
 Initially
assume the worst (intimate terrorism)
 If SCV seems likely, try individual application
of other approaches
 If SCV and safety become clear, move to
couple approaches with protections in place
Primary Prevention/Education
You’re the experts
 Intimate
terrorism
 Equality
 Violent
and respect
resistance
 Dangers
of violent resistance
 Safety planning
 Entrapment/escape issues
 Situational
 Sources
couple violence
of conflict
 Anger management tactics
 Communication
 Substance abuse
Intervention with Perpetrators
Hold them all accountable in the criminal justice system
to provide an essential motivation for change
 Intimate terrorism
 Control-focused
education
 Perhaps different tactics for sub-types
 Violent
resistance
 Alternatives
to violence/Safety planning
 Neutralize entrapment
 Situational
 Sources
couple violence
of conflict
 Anger management
 Communication counseling
 Substance abuse rehab
Intervention with Perpetrators
Outcomes of Duluth-type
Batterer Intervention Program
(Thirteen Months Post-adjudication)
SCV
Dependent
IT
Antisocial
IT
Completed
Program
77%
38%
9%
Re-arrest
18%
38%
46%
Repeat
Violence
55%
62%
88%
Eckhardt et al., 2008
Differential Success of Intervention
Strategies by IT Sub-type
(Percent non-violent two years after completing treatment)
Dependent
Antisocial
Feminist cognitive-behavioral
48%
65%
Process-psychodynamic
67%
49%
Adapted from Saunders (1996)
Intervention for Survivors
 Intimate
terrorism
 Long-term
support
 Alternatives to violent resistance
 Empowerment to leave
 Transitional support
 Situational
 Source
couple violence
of conflict
 Anger management
 Communication counseling
 Substance abuse rehab
Custody and Access Issues
 Separation-instigated
violence
 Manipulative accusations
 Resources for thorough evaluation
 Custody/access options
 Joint
custody/Co-parenting
 Parallel parenting, minimal couple contact
 Supervised exchanges
 Supervised access
 No contact
 Pittsburgh,
1978 (Frieze)
Married women from shelters and courts, matched with
married women living on the same block (n=272)
86% White; 14% Black
Data on self and husband, reported by wives
Incident data on most violent incident
 United States, 1995-96 (NVAW, Tjaden & Thoennes)
National random sample; subsample=4967 married women
83% White; 10% Black; 8% Hispanic (all races)
Data on current husbands, reported by wife
Incident data on most recent incident
 Chicago, 1995 (Lloyd)
Random sample of women in a poor neighborhood (n=596)
5% White; 54% Black; 41% Hispanic
Data on male partners, reported by female partner
No incident data
Pittsburgh Control Scale (High>2.74; 40%m; 10%w)
When you and your husband go places together, who decides
where you will go?
 If you disagree [about people you like], which people do the two of
you spend more time with?
 Does your husband know where you are when you are not
together?
 Are there places you might like to go but don’t because you feel
your husband wouldn’t want you to? How often does this
happen?
 Do you generally do what your husband asks you to do?
 Who decides how the family money will be spent in terms of major
expenses?
 [How often} does he try to get what he wants by doing any of the
following?…emotionally withdraws?
 …restricts your freedom?
 …stops having sex with you?
 …threatens to leave you?
 Has your husband ever pressured you to have sexual relations?

Pittsburgh: Other Items
“Has your husband ever gotten angry and threatened to use
physical force with you?” followed by the item that is actually
used: Has he ever actually slapped or pushed you or used other
physical force with you?
 Can you estimate how many times, in total, he was violent with
you?
 Did he become more violent over time?
 How badly were you hurt [the time your husband was most violent
with you]? Frieze codes: severe, severe superficial, severe
trauma, and extreme permanent.
 Were you afraid he would be violent again? Already “very
frightened” at the first violent incident.
 How would you rate the happiness of your marriage on a scale
from 1-Not at all to 10-Very happy? Low=1-4, 32%
 Is sex ever unpleasant for you?
 Do you and your husband have a good time when you go out
together?

Chicago Items
Items: In the past 12 months, when you’ve had
an argument, how often did your husband/boyfriend…
 Control
 …say
something to spite you?
 ...insult you, swear at you, or call you out of your name?
 ...accuse you of being with another man?
 ...try to control your every move?
 ...withhold money, make you ask for money or take yours?
 …threaten you with a knife or gun?
 ...threaten to kill you?
 ...threaten to hurt your family or friends?
Pittsburgh-cutoff
67% severe
72% escalated
37% mutual
1/25
couples
29% severe
29% escalated
74% mutual
1/8
couples
NVAWS Control Scale
(High = 3 or more)
“Thinking about your current husband, would you say he is jealous
or possessive?”
 “…tries to limit your contact with family and friends?”
 “…insists on knowing who you are with at all times?”
 “…calls you names or puts you down in front of others?”
 “…makes you feel inadequate?”
 “…shouts or swears at you?”
 “…prevents you from knowing about or having access to the
family income even when you ask?”
---------------------------------------------------------------------------------- “…insists on changing residences even when you don’t want or
need to?”
 “…prevents you from working outside the home?”

Control Scale
“Thinking about your current husband, would you say he
is jealous or possessive?”
 “…tries
to limit your contact with family and friends?”
 “…insists on knowing who you are with at all times?”
 “…calls you names or puts you down in front of others?”
 “…makes you feel inadequate?”
 “…shouts or swears at you?”
 “…prevents you from knowing about or having access
to the family income even when you ask?”
NVAWS
 The Great Gender Debate
 Distinguishing among types of partner violence resolves it
A
Control-based Typology of Partner Violence
 The
three major types
 Gender differences and sampling biases
 Other differences
 Implications for Research and Theory
 Everything we “know” has to be re-assessed
 Need a standard operationalization
 Tricky sampling problems
 Need for differentiated theory
 Implications for Intervention
 Screening/triage
 Intervention with perpetrators
 Intervention for survivors
 Custody and access issues
Table 2: Ex-Spouse Violence by Gender
Violence Type
Intimate Terrorism
Situational Couple
Non-violent
(n)
Violence
Ex-husband
22.0%
7.4%
70.5%
(2413)
Ex-wife
5.4%
3.9%
90.7%
(2051)
 The Great Gender Debate
 Distinguishing among types of partner violence resolves it
A
Control-based Typology of Partner Violence
 The
three major types
 Gender differences and sampling biases
 Other differences
 Implications for Research and Theory
 Everything we “know” has to be re-assessed
 Need a standard operationalization
 Tricky sampling problems
 Need for differentiated theory
 Implications for Intervention
 Screening/triage
 Intervention with perpetrators
 Intervention for survivors
 Custody and access issues
Need a Standard Operationalization
 Problems
with cluster analysis
 Extremely
sensitive to sample
 Not comparable across studies

Need a standard operationalization
 NVAWS
items
 Tolman: Psychological Maltreatment of
Women Inventory
 Graham-Kevan & Archer: Controlling
Behaviors Scale
 Dutton & Goodman: Coercive control
Need for Differentiated Theory
 Intimate
terrorism
 Coercive
control theory
 Gender theory
 Theories of paternalism
 Violent
Resistance
 Coping
 Entrapment
 Situational
 Family
couple violence
conflict theory
 Communication
 Anger management
 Substance abuse
 The Great Gender Debate
 Distinguishing among types of partner violence resolves it
A
Control-based Typology of Partner Violence
 The
three major types
 Gender differences and sampling biases
 Other differences
 Implications for Research and Theory
 Everything we “know” has to be re-assessed
 Need a standard operationalization
 Tricky sampling problems
 Need for differentiated theory
 Implications for Intervention
 Screening/triage
 Intervention with perpetrators
 Intervention for survivors
 Custody and access issues
A
Control-based Typology of Partner Violence
 The
three major types (plus one or two)
 Gender differences and sampling biases
 Some other basic differences
 We need to re-assess everything we thought we knew
 Implications for Intervention
 Screening/triage
 Primary prevention/education
 Intervention with perpetrators
 Intervention for survivors
 Custody and access issues
Sampling Problem: General Surveys
Probability of Violence
Current Spouses-NVAWS
0.14
0.12
Men
Women
0.1
0.08
0.06
0.04
0.02
0
0
Felson & Outlaw, 2007
1
2
Control Score
3
4
Sampling Solution: General Surveys
Probability of Violence
Former Spouses-NVAWS
0.14
0.12
Men
Women
0.1
0.08
0.06
0.04
0.02
0
0
Felson & Outlaw, 2007
1
2
Control Score
3
4
Spousal Violence Type by Gender
General Survey Data
Intimate Terrorism
Situational Couple Violence
Non-violent
(n)
Husband
0.7%
3.9%
95.3%
(4846)
Wife
0.5%
1.7%
97.9%
(5126)
Ex-husband
22.0%
7.4%
70.5%
(2413)
Ex-wife
5.4%
3.9%
90.7%
(2051)
Data from NVAWS, Johnson, Leone, & Xu, 2008
Types of Domestic Violence
Research Evidence
The Continuing Gender Debate
A Control-based Typology of Partner
Violence
Dramatic Differences Among the Types
Preview of Policy Implications