A Group Design Paper Submitted by: Kathleen Cooper

The following paper presents a group design using Pretest - Posttest Comparison Group to evaluate the impact of a spouse abuse intervention. The diagram below illustrates this design:

" The APA (American Psychological Association) Task Force on Violence and the Family (APA, 1996) defined domestic violence as a pattern of abusive behaviors including a wide range of physical, sexual and psychological maltreatment used by one person in an intimate relationship against another to gain power unfairly or maintain that person’s misuse of power, control and authority." (Lenore Walker January 1999).

There are 30 subjects in this experiment who will be divided up into 2 groups; an experimental group and a control group. These subjects are all males who range in age from 21 - 40 years old, who are cohabiting with their partner and have been charged with being physically abusive toward their partner. They come from the City of Buffalo and have been referred by the Family Court judge for court mandated enrollment in an anger management program for male batterers herein after called a ‘batterers program’ so not to be confused with the terminology in the literature.

At Court, the men were given the name of a Non-Profit Agency that offers State accredited programs and they were expected to initiate contact with the Agency to register for one of their male batterers programs. This agency is a multi-service agency that offers programs for substance abuse; Batterers program; as well as partnerships/linkages to other social service agencies. I.e. Women’s shelters, Rehab programs. The court department will forward to this agency a list of male abusers whom they have referred to their program. This will enable the agency to know who has made initial contact to enroll in their program and who has not. When contact is made by the abuser, they will be given a date and time to come in.

These men will then come to the agency for their first appointment and complete the Physical Abuse of Partner Scales (PAPS). The Research Clinician will then tabulate the scores of the abusers and choose 30 who have scored within a certain range and are of a good cross section of ages in the target group and divide them evenly into two groups. One group will receive treatment which is the Batterers Program and the other group (the control group), will be placed on a ‘Waiting List’ for the twelve-week period of the course, at which time both groups will again be administered the PAPS, to see if there is any change in the recorded scores. The group who will be on the Waiting List or the control group will be given pamphlets and literature about physical abuse until they are enrolled in a batterers group at the end of the twelve weeks.

At the time of orientation / information session, if the male abuser discloses a drinking / drug problem, or if information has been forwarded from the Court concerning the involvement of alcohol in the abuse incident, the clinical researcher will speak to him about it and assist him in a referral to a program within the agency where this issue can be addressed more appropriately. Literature and a referral are given to the client. Due to studies which indicate physical abusers will use alcohol as a reason for the violent incident or other excuses, "I was drunk---; I don’t remember," the client (abuser) is referred to another branch specializing in treating this concurrently or upon completion of the batterers program. The focus is on accepting responsibility for their behavior. The participants are also expected to pay $100.00 for the sessions in advance of them starting the course. The idea is for them to be committed to this program and a recognition that once a group has begun other people cannot join a group that is already in session who have made group connections / trust. Prior to the start of this program by the abuser the clinical researcher and co-leader will meet with the victim to explain the program to her, inform her of the services in the community to support her when needed, crisis numbers and a warning that this program does not ensure the abuser will not become physically abusive while enrolled in these sessions.

DEPENDENT VARIABLE

The dependent variable in this study is the physical abuse which will be measured during the Pretest, during treatment and at Posttest. The Pretest and Posttest measurement is the ‘Physical Abuse Partner Scale (PAPS).The purpose of the Physical Abuse of Partner Scale is to measure physical abuse that clients report they have inflicted on a spouse or partner. The PAPS is a twenty-five item instrument that is designed to measure the degree or magnitude of perceived physical abuse that clients report they have inflicted on a spouse or partner. The PAPS was developed for use with heterosexual or homosexual couples who are dating or living together as married or unmarried couples. This scale is one of the few to examine perceptions of the abuser as to the amount of abuse he or she perceives as inflicting. As such, it can be very useful as a device for tracking the abuser’s perceptions over time during an intervention program. Another advantage of the PAPS is that it is one of several instruments of the WALMYR Assessment Scales package reproduced here, all of which are administered and scored the same way. Like most WALMYR Assessment Scales instruments, the PAPS is scored by summing the scores, subtracting the number of completed items, multiplying this figure by 100, and dividing by the number of items completed times 6. This will produce a range from 0 - 100 with higher scores indicating greater magnitude or severity of problems.

Reliability: The PAPS has excellent internal consistency, with an alpha in excess of .90. Data on stability are not available.

Validity: The PAPS is reported as having good content and factorial validity, plus beginning evidence of construct validity. (Corcoran and Fischer,1994).The scale is attached at the end of this paper, before the References page.

During the treatment phase for the first group, they will not only have twice weekly sessions on anger management / Batterer program, but the men will have been told during the first session that they are required to keep a daily ‘anger journal’. This book will be provided to the group members. In their daily anger journal, the men will be encouraged to share their feelings about how their time/situations at home have been going since the last session. The total number of incidents reported by the men on two items will be tracked and recorded at the bottom of the page. The men will hand in their anger journal at the beginning of each session. The group leader will discuss/go over these anger logs with the group members individually noting their comments and tallying scores each group session. The following sample of a daily anger journal record sheet and the graph of reported incidents will demonstrate the way in which the researcher and the male abuser will visually track anger and violence episodes.

The group session dates are scheduled on a Monday and a Thursday allowing space between groups and the weekend.

Date: ___________

Write down how things have been since we had our last session:

Mon. Tue. Wed Thurs. Fri. Sat Sun

1. I felt I was provoked into becoming angry_____(Fill in the number of times since last session)

2. I was so angry I did something physical _____(Fill in the number of times since last session)

TOTALS:

Graph of Incidents:

10

9

8

7

6

5

4

3

2

1________________________________________________

MTWTFSS MTWTFSS MTWTFSS MTWTFSS

Wk.1 Wk. 2 Wk. 3 Wk. 4

* A red dot indicates the number of times I acted physically when I was provoked into anger.

* A blue dot indicates the number of times I felt I was provoked into becoming angry.

A chart the same as the one above will be done for each week of the entire 12 week program. The information gathered from these recordings will create opportunity for further study .I.e., Are certain days of the week worse than others? * This graph will provide the clinical researcher or co-leader and the male abuser with a visual tracking of his anger and violence. **For the purposes of this study, only the red dots or physical acts will be recorded to determine if the intervention worked.

The literature on Anger Management groups or Batterers Groups also supports the use of "Anger Journals" as a way of tracking / providing intervention to the subjects. David Myer (1992, p.500), indicates that "cognitive behavioral or psycho educational programs involve the learning of well-established behavioral practices: self-monitoring (identifying his violent and controlling behavior and its antecedents, keeping logs, using a checklist . . . " (Browne, 19997). These studies reflect the use of "anger journals" and are not only used in some cases to track violent incidents but for discussion with the group for further education. These journals, as explained at the abusers orientation session will be confidential and used for learning and to visually track progress. I believe the daily anger journal provides an outlet to express feelings, an opportunity to learn about and from their feelings. The resulting graph, tracks two areas over the 12 weeks and totals the scores. This gives the participant (abuser) a visual sense of their progress and possible days of vulnerability. This will also show numerically the abusers attendance due to the reports at the bottom of the journal, i.e., name, day, # of incidents. Although the journal’s content validity could be questioned, the attendance at classes confidence in the confidentiality agreements, sharing of experiences / feelings and being involved in a well-planned Anger Management group (Batterers Group), hopefully will encourage honesty in their recording of incidents. The more reliable and valid scale that could be used is the (PASPH), Partner Abuse Scale-to be completed by the victim. Although this scale has been proven reliable and valid, concerns for the victim’s safety again overrode the decision to use it.

INTERVENTION (INDEPENDENT VARIABLE: TREATMENT VS. WAITING LIST)

The subjects in group 1, the treatment group will attend an ‘Anger Management Group (Batterers Group), based in part on the nationally recognized Duluth Model, a curriculum developed by Michael Payman and Ellen Pearce of Minnesota. Many programs throughout Wisconsin and around the country use this curriculum or portions of it. In Duluth the 24 week educational program examines eight themes in three week segments. The eight themes are from the power and control and equality wheels. The themes are: Intimidation; emotional abuse; isolation; minimizing; denying; and blaming; using children; using male privilege; economic abuse and coercion and threats. During Week one of the three week segment, participants define the theme and analyze abusive behavior from videotaped vignettes. In Week 2, the focus is on understanding participant personal patterns of abuse and control and Week 3, is used to explore and practice nonviolent alternatives and behaviors. Supporters of an educational model, such as the Duluth curriculum, believe highly structured groups avoid collusion with batterers, hold abusers accountable and keep the focus away from victim blaming while men examine their power and control issues. (Brandl, 1990).

The Anger management classes will be held two times per week for a twelve-week period of 90 minute sessions (24 sessions). The training will be provided by the clinical researcher and another accredited anger management counselor. The program will begin with an orientation session. The entire group of fifteen men chosen for the treatment group will have the program goals and expectations laid down for them. Group members will discuss confidentiality, what it means and that their journals are kept in confidence as is all information shared between group members. The content of the sessions will focus on the Duluth Model and stress the accountability of the abuser; issues of respect; nonviolent / non-threatening behaviors; partnership; negotiation and fairness. Interventions to be focused on include: -awareness and acceptance of self as a batterer; knowledge of batterers’ use of power and control tactics; skill development to assist in reducing sexual, social and psychological patterns of power and control; and strategies for using alternative, non-controlling behaviors.

Everyone in the treatment group (experimental condition) will receive the exact same treatment because anyone missing the session would receive instruction from the same instructors prior to the next group session; if they miss more than one session, they will not be allowed to return to the same group. They would have to re-enroll in another group.

The research clinician in this study is female, has an MSW , is a certified family therapist and is a certified ‘Batterers Treatment Program’ instructor. She has five years experience in working with adults and children who are victims of domestic violence as well as two years experience working with an anger management program for male batterers. Her co-leader is a male with a BSW and who is also certified in the Batterers Treatment Program. He has three years experience in the field of domestic abuse and is currently enrolled in an MSW program.

HYPOTHESIS

My one-tailed hypothesis for this experiment is: Heterosexual couples--intimately involved and cohabiting for at least six months-- in which the male partner attended 24 weeks of anger management group workshops based on the Duluth model will demonstrate a reduced number of physical abuse incidents in their relationship 12 weeks after attending anger management (Batterers ) group sessions compared to a similar group in which the male partners have not attended an anger management (Batterers) group. Empirical evidence that supports this hypothesis begins with a study by Gondolf (Oct.1996, p.5) in which he states "the model batterers program appears to contribute to the cessation of assault at least in the short term." "The majority of women indicate their ‘quality of life’ has improved and that they feel ‘very safe’." "Due to numerous methodological problems, researchers’ best estimates of ‘success’ in ending physical violence range from 53% to 85% for those who complete the program. Groups are significantly less successful in ending controlling and abusive behavior. A lower percentage of success occurred with lengthier follow-up and reports from battered women." (Brandl, 1990, p33). Edelson (1995) reviewed ‘Batterer Program Evaluations’ and indicated that these summaries reveal a consistent finding that in varying programs, using various methods of intervention, a large proportion of men stopped their physically abusive behavior subsequent to involvement in the programs. Reports ranged from 53% to 85%. He further cites a study by the Domestic Abuse Project where the results indicated that men batterers involved in a structured educational program or combined education and group processing were reported nonviolent by their women partners during six and eighteen month follow-up periods (ranged from 62.1% to 64.7%) men in these groups achieved more stable outcomes than did those randomly assigned to a less structured self-help program.

CONSIDERATION OF THE THREATS TO INTERNAL VALIDITY

History: This was going on for both groups so it has been controlled for.

Maturation: A possible ‘cooling off’ period effect could work on abusers due to twelve week sessions from beginning to end, however this is the same for the treatment and control group.

Testing As test is given twice there is the threat for both groups of this impacting on their responses due to familiarity of the test. Hopefully the twelve weeks between Pre and Posttest will ‘day to day life ‘ going on in between, for both groups will lessen the threat of multiple testing.

Instrumentation: The administration of Partner Physical Abuse Scale - PPAS has demonstrated validity and reliability. The "Anger Journals" hold this as a threat to internal validity due to control analysis. Hopefully the design helps to overcome this somewhat.

Statistical Regression: Subjects were not chosen by extreme score for groups rather tried to make them equal.

Selection: Selection is a threat as randomization was not used. Groups were chosen based on their referrals from the court and where specific age ranges were used.

Experimental Attrition: Due to up front costs to cover the program and permission for one missed group session, this threat has been controlled.

Practitioner Interaction: Curriculum model styled / formatted to reduce this threat.

CONSIDERATION OF THE THREATS TO EXTERNAL VALIDITY

Reactive Effects of Testing: I don’t feel this is a threat to the experiment. The test is the Scale where the abuser himself scores items on the Physical Abuse of Partner Scale.

Unrepresentative Samples: The same for citizens of Western New York except for age clustering.

Reactive Settings: The clinical researcher and co-leader are familiar, due to group sessions, with male abusers and therefore this represents a threat to the external validity of this design.

Multiple Treatment Interference: The PAPS was administered Pre and Post and could be a threat to external validity; however it is hoped that time/space in between testing controls this somewhat.

Practitioner Interaction: Due to the tight control on the sessions and two leaders with fifteen participants in the group this should be controlled as opposed to a more informal group with one leader and fewer participants.

Researchers can use systematic replication in any city/state where similar legislation has been passed, recognizing physical abuse as a crime where the abuser would be charged and incarcerated. Even in areas that do not yet have this legislation, and the social mores of that area do NOT condone and DO condemn physical abuse, I believe this study could be replicated.

Direct replication would also be viable and enable greater generalization of results by using different client populations; For example: Clients who are inmates in a prison or correctional facility or clients from another country that can speak and understand English but have different value systems regarding women in society, who are having difficulty adjusting to our culture and its laws and value system. I believe these studies would enable greater generalization of results.

References

1. Brandl, Bonnie (August 1990). Programs for Batterers: A Discussion Paper (Pp.1-38). Departments of Health and Social Services, Wisconsin. (Lycos)

2. Corcoran, K. and Fischer, J. (1994). Measures for Clinical Practice. (Vol.1) New York: Free Press.

3. Edleson, J.L. (1995). Do Batterers Programs Work? (Pp. 1-7) University of Minnesota and Domestic Abuse Project Inc. (Lycos).

4. Gondolf, E.W. (April 1996). Characteristics of Batters in a Multi-site evaluation of Batterer Intervention Systems (Pp.1-22) (Lycos).

5. Gondolf, E.W. (October 1996 - Revised April 1997). Multi-site Evaluation of Batterer Intervention Systems (Pp.1-13) (Lycos).

6. Myers, D.L. (1995). Eliminating the Battering of Women By Men : Some ConsiderationsFor Behavior Analysis. Journal of Applied Behavior Analysis 28, 493-507.

7. Walker, L.E. ( 1999 January). Psychology and Domestic Violence Around the World. American Psychologist, 54 (1) 21-29.