|
1. A difficult field
2. First Family Violence Program (FVP) Evaluation
3. Current 2004 research
4. A 6-month report on the 2004 project
Abstract:
Outcome research on programs for men who use violence against women in relationships is difficult and fraught with controversy. How can a small program with limited resources meet the need of the community to be assured that the agency makes a worthwhile contribution? Alayne Hamilton will discuss some of the research dilemmas in the field and how the Family Violence Project charted a course into these rough waters. The 2001 results of the original data-gathering project will be distributed and discussed. Alayne has launched the ship of research again this year, and will describe how the new data-gathering has been informed by the experience of the last voyage. She will share recent results from the first 6 months of this study and welcome questions and discussion.
Author:
Alayne Hamilton is the executive director of the Victoria Family Violence Project, and a long-time ACAM/ERA board member.
1. A difficult field
Reference: Gondolf, Edward W., Evaluating batterer counseling programs: A difficult task showing some effects and implications. Aggression and Violent Behavior 9 (2004) 605-631
Issues:
Control groups • Classic random assignment to treatment and control groups raises ethical issues of withholding treatment when others are in danger and treatment is required by court order. • Men assigned to treatment may drop out. • If drop-outs are used as a control, these men may be very different from men who complete. • Control groups are beyond the capacity of small working agencies.
Confounding variables • Is it the treatment or something else that happened concurrently? • Most studies use subjects who have been arrested and charged. • Women who are supported in the process may get more power in the relationship. • Programs are different and change during the evaluation. • Counsellor characteristics may be more important than content.
Conflicts based on belief and goals • Does faith in men’s treatment increase risk to women? • Is leaving the answer (short vs. long term prevention)? • Will needed dollars be diverted from victim services to offender treatment? • Can men change? Evidence vs. strongly held values and beliefs.
Definitions of success • Physical assault or use of coercion and non-physical abuse? • Will other forms of abuse increase if physical assault decreases? • Short or long term effects? • Who is the judge of change? • Are we trying to save a family?
Results:
Gondolf reports that in the 40 published academic evaluations 50 to 80% of program completers are non-violent at the end of 6-month to 1-year periods, based on partner reports. All these studies are dismissed on technicalities: no control groups, selection bias, low response rates, short follow-up periods, no calculation of size of effect attributable to the program – "circumstantial evidence." Calculation of effect size results in huge differences, but averaged out it is low. Even if effect size is 5%, this would translate to 42,000 women/year not re-assaulted. Cost effectiveness is also an issue; group programs are inexpensive compared to jail. Meta-analysis results in "“no conclusive evidence that most programs are effective or that any programs are highly effective."
However, as Gondolf notes,
"The most surprising finding in our multi site evaluation is that the vast majority of men referred to batterer counseling appear to stop their abusive behaviour and reduce their abuse in general."
2. First Family Violence Program (FVP) evaluation
Beginning issues in 1997:
"At first it seems so simple. But what does 'work' mean? Are we trying to stop physical assaults or also verbal and psychological abuse and control, those assaults on a woman’s self esteem, dignity, and freedom? What if physical abuse has ended, but the woman is still living in fear, controlled by intimidation and threats? And does improvement count? What if a man who used to break bones now breaks cups? If a man who was frequently violent before treatment stops for several years, then on one occasion chooses to push his wife, has treatment failed? Are we justified in demanding higher standards of non-abuse for program completers than is the norm among other men in the community?"
We decided not to give up. We cannot prove that our program works, but we can find indicators of success. We will do our own data gathering with the good of clients as the highest goal.
Outcome goals:
What we learned:
(See FVP Annual Report 2000/2001: Research Component)
Piles of data and pages of analysis end up as one or two sentences of useful results. Women partners provide the most conservative and convincing reports. It is possible to demonstrate improvement among men who complete on all three indicators of success. It is important to follow up at six months and at least another year. It becomes extremely difficult to contact people three years after treatment. Physical abuse and psychological abuse increase AND decrease together. Longer attendance (in Phase II) is correlated with partners reports of decreased psychological abuse. The program content (the Tool Chest) is related to the success indicators.
3. Current 2004 research
The same three indicators are used. The goals have not changed, so the results will be comparable. Women partners who have at least weekly contact with the man rate his success. Data are gathered 3 times: at the end of Phase I, 6 months after, and 18 months after completion.
Changes and additions
Program satisfaction and accessibility data are gathered at the end of the program. Do those who do not identify with mainstream culture feel we were respectful and responsive to their needs? May not gather data from men at 6 month and one year later follow-ups. Add parent rating of children’s safety, emotional and physical. Additional questions for those referred with outstanding charges; eg women’s satisfaction with referral, outcome of treatment and outcome in court. Intervening variables: replaced use of tools taught in program to ratings of "Behavioural indicators of Change in Abusive Men", which is used as the basis of MCFD reports.
4. A 6-month report on the 2004 project
Interviews:
Program satisfaction:
Sample women’s comments:
"At FVP I felt really listened to and given the opportunity to get things off my chest. I felt supported."
"I thought they were highly respectful of confidentiality."
"I appreciated the follow-up calls."
"You guys understand the patterns – that is helpful."
(About 1/3 of the women attended women’s groups.)
Indicators of change:
|
|
Women
|
Men
|
He accepts responsibility for his behaviour.
|
|
Much more
|
6
|
11
|
|
More
|
8
|
8
|
|
Unchanged
|
4
|
1
|
|
Less responsible
|
1
|
0
|
He uses what he learned to avoid violence, abuse and control.
|
|
Much more
|
4
|
10
|
|
More
|
8
|
8
|
|
Unchanged
|
3
|
3
|
|
Uses it to control
|
3
|
0
|
He has empathy and understands the impact of abuse on her.
|
|
Much more
|
1
|
11
|
|
More
|
9
|
4
|
|
Unchanged
|
4
|
3
|
|
Less empathy
|
3
|
0
|
He has empathy for his children and understands the impact of abuse on them.
|
|
Much more
|
1
|
9
|
|
More
|
6
|
7
|
|
Unchanged
|
6
|
3
|
|
Less
|
1
|
0
|
He limits his use of alcohol and drugs to prevent abuse.
|
|
Always limits
|
4
|
7
|
|
Sometimes limits
|
1
|
2
|
|
Use is unchanged
|
3
|
4
|
|
Use is worse
|
0
|
0
|
|
Use of physical violence.
|
|
None
|
78%
|
95%
|
|
Use of psychological abuse.
|
|
Less
|
58%
|
89%
|
|
More
|
21%
|
|
Safety
(89% of women have at least weekly contact)
87.5% of women (in contact) say they feel safer as a result of his attendance.
67% of women (with children) say the children are safer (physically and emotionally).
Positive comments:
"He has learned to talk to me."
"I feel safer because I left."
"When he gets angry he takes a time-out."
"Children are not hearing threats."
"He is learning his behaviour is wrong."
"He stopped shouting at us. He can stop himself."
"This service is needed in society. It’s not OK that you have to live in a violent situation."
October 20, 2004
|