|
|
Section
14 Question
14 found at the bottom of this page 1. Female partners' reports of post program abusive behaviors have been considered the most valid indicator of perpetrator change (Johnson and Kanzler, 1993). Although numerous researchers have included women in their study designs, only a few have asked them whether they believed the programs were effective. For the most part, women have responded positively. For example, Saunders and Hanusa's (1986) follow-up with 45 female partners of men who had completed a program for woman abuse found that of the 34 who remained in contact with their partners, 68% stated that they believed that decreases in violence were due to the program. 48% of the women reported that their partners showed an increased expression of feelings. As part of an evaluation study of the 'Diversion of Violent Emotions' program in Phoenix, Johnson and Kanzler (1993) followed-up with 515 spouses from a sample of 687 men who completed the program between June 1987 and June 1988. A total of 78% of these women responded positively to the program, 15% responded negatively, and 7% would not comment. Women still living with the perpetrator had the most positive perception of the program's effectiveness. Meredith and Burns (1990) found that approximately 50% of the women interviewed reported that their partners' participation in treatment affected their decision to continue in the relationship, and approximately 80% of the women reported feeling safer with their partners in a program. In Dobash et al.'s (1996) study, women whose partners had been sentenced to a reeducation program vs those sentenced more traditionally, were more likely to report that they were "happy, more relaxed and less frightened than before the intervention"' (p. 3), and that their partners' emotional states and sense of personal responsibility had improved. These women also reported that the attempts to restrict their lives had decreased. In Gondolf's (1997a) multisite evaluation, nearly 70% of female partners were contacted for the full follow-up (every 3 months for 15 months). The findings indicated that at the 15-month follow-up, 66% of women reported being "better off"' than before treatment (primarily those who had not been reassaulted), and 12% reported being "worse off." A total of 72% reported feeling "very safe" at each follow-up period, and 64% felt that it was unlikely their partners would hit them again in the near future. However, these women also reported that their partners were controlling (45%), verbally abusive (70%) and threatening (43%). 2.
Conceptual and methodological issues Despite advances in batterer program evaluation research, conceptual and methodological challenges persist. The lack of control groups, low response rates, high attrition levels, and short-term follow-up periods have been repeatedly discussed (Burns; Eisikovits; Gondolf; Gondolf; Tolman and Tolman). When some of these issues have been addressed by examining multiple sites, recruiting a large sample, and controlling for factors such as relationship status of the batterer and the abused woman, key outcomes used to compare interventions have not varied significantly between programs (Gondolf and Gondolf). Researchers have commented that the treatment itself may not be the factor leading to change (Tolman & Edleson, 1995), that other variables not included in statistical models may be operating (Shepard, 1992), and that the modest relationships found between independent and dependent variables suggest that "many cases will not fit the overall pattern" described (Cadsky et al., 1996, p. 60). As well, concerns have been raised about the paucity of research into the relationship of batterer programs to other interventions and community linkages (Gondolf and Gondolf). Researchers continue to struggle with the question of how success is defined (Eisikovits; Gondolf; Tolman and Tolman), which is related to what is being measured and what kind of change can be deemed valuable. For example, violence was initially defined as physical assault and the Conflict Tactics Scale (CTS) was the primary tool used to measure the phenomenon. The CTS was developed by Straus (1979) to study violence within families, and has been used extensively. The failure of the CTS to capture the severity and range (e.g. sexual, psychological) of abuse experienced was identified (Gondolf and Saunders) and, consequently, modified versions have been used. These versions have also been critiqued since changes to the scale affect internal reliability and consistency (Römkens, 1997). Most of the studies conducted have relied on the CTS and other inventories to measure abusive behaviors, men's attitudes or traits assumed to be correlated with violence such as anger, depression, and attitudes towards women, and responses to the batterer programs. While the numbers generated from these instruments provide some useful information, they present a certain perspective of the phenomenon being studied. What are these tools measuring? What kind of information is being missed? Furthermore, it has been noted that a statistically significant finding may have no practical significance (Edleson and Gondolf). Although the purpose of batterer programs is to assist women who are abused by their partners, there has been only a cursory examination of their concerns and experiences regarding these treatment modalities. While earlier studies relied on self-reports from male batterers, the limitations of this approach were quickly identified when discrepancies were found between men's and women's reports of violence (Eisikovits; Edleson; Gondolf; Meredith; Palmer; Petrik and Tolman). Subsequently, much of the research has been based on data collected from both the women and their partners. However, the majority of the data collected has been through their completion of the CTS (or modified versions). A few studies have asked women about their perceptions of, and experiences following their partners' involvement in, batterer programs. Yet these findings have been limited to brief responses such as feeling "safer" and "better off." While studies have found that the majority of women interviewed had positive responses to the batterer programs, which is promising, it is unclear how these findings fit with the statistical results collected through the CTS regarding ongoing abusive behaviors. Concerns have been raised that women's reports may underestimate violence out of fear of their husbands, due to a strategy of denial used to cope with the abusive situation, or because of fear of involvement with the criminal justice system (Dutton and Tolman). Furthermore, what do "safer" and "better off" mean? How are the reported changes manifested in the realities of women's lives? What types of changes are desired and valued by women? What outcomes do women define as successful? What are women's expectations of batterer programs, and how do they compare to their experiences? Do women affect men's participation in the programs and if so, how? Do women want the legal system to mandate their partners to treatment? In recent years, researchers have called for a greater focus on the needs and knowledge of women, including their experience of safety (Edleson; Gondolf and Tolman). However, the response to date has been limited. Research is needed that provides for an exploration of the meaning and significance of varied outcomes as defined by women themselves, and which enables a more comprehensive exploration of wider processes. While progress has been made in this respect, such as the recognition that a reduction in physical abuse is not always accompanied by a comparable decrease in other forms of violence, strategies are needed that provide for a more thorough appreciation of the complexity and multidimensionality of abuse. Quantitative studies need to be complemented by qualitative studies that might be able to address some of their limitations described, thereby providing a more comprehensive picture of the value and effectiveness of batterer programs. 3. A qualitative, woman-centered approach Research questions should address the primary goal of batterer programs: the safety of women. Questionnaires that have asked about intimate violence, even those that included verbal and emotional abuse, have not addressed the full range of abusive behaviors that may be occurring in a relationship. Research and experience have demonstrated the multifaceted nature of the harms perpetrated by men. Women may be emotionally, psychologically, verbally, socially, financially, ritually, physically, sexually, and religiously abused (Martin & Younger-Lewis, 1997). Batterer program evaluation research needs to study the diverse types of abuse that may be occurring in a relationship and their changes over time, as well as the meaning and implications of these issues within the context of women's everyday lives. The
sole qualitative research study on women's experiences with batterer programs
that was located at the time of this review demonstrates the type of knowledge
about women's safety that can be gained through in-depth interviews with female
partners. Austin and Dankwort (1999) interviewed 25 women whose partners had completed
a batterer program administered by a women's shelter in a Canadian urban setting.
The program lasted for eight weeks and was based on a sociocultural, feminist
analysis of battering. Program goals were to make men responsible for their behavior
and to enhance women's safety. Counselors provided women with information about
the program and their partners' progress, and met with women interested in resuming
their relationship. The women interviewed were mostly white, aged 18-57, with
the majority being in their mid-30s. At the time they were interviewed, 16 women
were living with, and nine were separated from, their partners. Through the use
of open-ended questions, women were asked about their experiences with the program
and their perceptions of change in their partners. They were also asked what they
believed was responsible for the differences they described. Austin and Dankwort (1999) found that the majority of women experienced positive changes following their partners' completion of the program. The women's accounts revealed a complex, subjectively defined notion of safety; there were different reasons as to why women felt safer, different degrees to which safety was perceived, and different interpretations of safety. For example, women reported feeling safer, but feared that the abuse would never completely cease. One woman felt safer although her partner continued to be abusive. The batterer program affected change in various ways. In addition to helping the men decrease their violent behaviors, women appreciated having a respite from fear and from their sole responsibility for providing support to their partners. Furthermore, women experienced increased well-being, due to the emotional support, validation, and knowledge about abuse received through the program. Importantly, this study also found the batterer program was only one factor that affected change in the women and in their relationships, and that women were uncertain of the degree to which the benefits experienced were associated with their partners attending the program. This study has increased our knowledge of how, according to the women themselves, success is defined, with the accompanying contradictions, and the varying factors that contribute to "positive changes." It demonstrates the importance of revealing the context within which change is reported. As the authors noted, safety exists on a continuum, and may be subjectively defined by each woman. Furthermore, the exploratory nature of this study provided data about the benefits of the program that would not have been captured through quantifying changes in abusive behaviors. For example, the women talked about the value of feeling validated, and their enhanced sense of self-worth, as well as being more informed about the range of what constitutes abusive behavior. By exploring the varied factors contributing to change, the authors raised the question of the relative contribution of batterer programs to changes experienced by women. This research has confirmed the need to study multiple types of abuse, and reinforces concerns that programs may provide a false sense of security to female partners. Further qualitative, woman-centered research is needed to address the gaps in Austin and Dankwort's (1999) study, and to explore other pertinent questions. Austin and Dankwort (1999) interviewed white women whose partners had completed the program, of whom only three were court-referred. Sampling to include women from varied ethnic and cultural backgrounds is needed, as well as women whose partners have been court-mandated, or who did not complete the program. As highlighted in their study, additional research on other factors affecting change and on alternative interventions that may better address women's needs is also required. Examples of other questions that remain to be addressed through a qualitative approach are: What types of interventions do women desire? Are women sufficiently included in the batterer program process? What issues do women want addressed in these programs? What social factors and interventions, other than the programs themselves, influence men's behaviors? What types of changes occur over time? What are the experiences of women whose partners do not complete treatment? From the women's perspective, who may benefit from the program? How do women's social, cultural, and religious backgrounds affect their perceptions of, and needs for, batterer programs? There are various limitations and issues that investigators must consider in qualitative research. For example, these studies may be more labor intensive (Murphy & O'Leary, 1994) and the detailed nature of data collection will likely limit samples to a manageable size. In-depth exploration on sensitive topics such as partner assault can raise ethical concerns such as the emotional demands that interviews may exact from both participants and researchers (Cowles, 1988). Field texts, including what is told, and the meaning of what is told, are shaped by the relationship between the researcher and the participant (Clandinin and Holstein). Thus, researchers must critically reflect on the interview process and how it influences the research findings. Each research approach has its strengths and limitations,
and thus qualitative methods should be viewed as one component of a larger research
agenda to assess the value of batterer programs to women. This agenda should integrate
the expertise and experience of individuals working in the field as well as of
women who have been abused. Collaborations between the academy and community are
challenging because of the different priorities and orientations, but the advantages
of researchers, policy makers, practitioners, advocates, and survivors working
together have been well articulated (Gilfus; Gondolf and Levin). As Levin (1999)
notes: Evaluation
and Program Planning, Volume 24, Issue 3 , August 2001, Pages 297-305 J. Goldman
and J. Du Mont Personal Reflection Exercise #2 |
