Attitudinal Change in Participants of Partner Assault Response (PAR) Programs: A Pilot Project


2. Methodology

2.1 Participants

Participants were recruited from Changing Ways, a non-governmental organization in London, Ontario. Established in 1982, Changing Ways is one of the most well-established batterer intervention programs in Ontario and Canada. The Changing Ways program provides intervention to clients from Early Intervention, Coordinated Prosecution and voluntary referral processes.

All men who completed intake for the Changing Ways program between January and March 2004 were invited to participate in the current study. Men who expressed interest in participating in the study and had basic proficiency in reading and writing English provided written consent and completed all measures. In total, pre-intervention data was collected from 41 men out of 131 who completed intake during this period. This translates into a 31.1% consent rate, which was considerably lower than expected. Investigation into the reasons for this low consent rate found that difficulties were mostly due to differences among the contract counsellors who were completing intake assessments. Only three of the eight counsellors were inviting men to participate in research. Other counsellors were not offering this invitation. Rate of consent for those men invited to participate was approximately 80%. With an 80% consent rate, concerns about self-selection bias are no stronger for the current study than for the majority of research done in this area. In addition, further investigation showed that the demographic profile of men who participated in the current research was consistent with the characteristics of clients generally presenting at Changing Ways.

2.1.1 Demographic characteristics: Age, ethnicity, income and employment, and relationship status

Men ranged in age from 19 to 52 years, with an average age of 34.05 years (SD[2] = 9.17). Of the men for whom ethnicity data was available (n=28), 86% (n=24) indicated that they identified themselves as belonging to a particular ethnic group. Fifty-four percent of these men (n=13) identified themselves as Canadian, 25% (n=6) identified as North American Indian, Latino, Irish, Polish, Albanian, or Arabic, and 21% (n=5) identified with more than one ethnic group. The employment status among the men was variable, with 35% (n=14) working days, 3% working nights (n=1), 38% (n = 15) working shift work and 25% (n=10) reporting unemployment. Of those men for whom annual income was available, 39% (n=12) reported that they earned under $10,000, 35 % (n=11) earned between $10, 000 and $29, 999 per year, and 26% (n=8) earned over $30, 000 per year.

In terms of the men's current relationship status, 46% (n=19) were separated, 22% (n=9) were married, 15% (n=6) were living common-law, 7% (n=3) were planning to divorce, 5% (n=2) were dating and 5% (n=2) were single. The average length of the men's current relationship was 5.73 years (range = 3 months to 20 years, SD = 5.52). Of those men who were separated from their partner, the average length of separation was 11 months (range = 4 weeks to 2 years, SD = 8.71). Approximately one-third of these men (n=6) indicated that they planned to reconcile with their partner. Of the remaining 13 men, 12 indicated that they did not plan to reconcile and one did not answer the question.

2.1.2 Referral information

Sixty three percent of the men were referred to the Changing Ways program through probation services (n=26), 17 percent through Early Intervention Court (n=7), and 20 percent through voluntary referral (n=8). The majority of the men were first time Changing Ways clients, with one fifth (n=8) of the men having been previously enrolled in the program.

2.1.3 Past assaults and alcohol use

Twenty-four percent (n=9) of participants reported a history of any past assaults, with the number of assaults ranging from 1 to 3 (x =1.33[3], SD =0.71).

With regards to alcohol intake, 38 % (n=15) reported that they did not drink, 30% (n=12) reported that they have less then one drink per month, 20% (n=8) reported one drink weekly, 10% (n=4) reported one drink daily and 3% (n=1) reported more than one drink daily. Rates of past assault and alcohol use are both higher than for the population overall, confirming that the sample of men referred to batterer intervention is at-risk for difficulties in a variety of life domains.

2.1.4 Post-Assessment Participants

Following program completion, post-assessment data was collected from 14 of the 41 men (34%; 11 men referred through probation; 3 men through voluntary referral). Reasons for not completing post-assessment were variable. Thirteen of the men (31.7%) were ineligible for post-assessment because they did not complete the program (9 men referred through probation; 4 men through voluntary referral). Of the remaining 14 men, 8 men (19.5%) could not be located (4 men referred through EIC; 3 men referred through probation; 1 man through voluntary referral), 4 men (9.8%) had incomplete files because their counsellor was not available to complete the counsellor rating form (2 men referred through EIC; 2 men referred through probation, 1 refused to complete the post-group questionnaire (referred through EIC), and 1 man (2.4%) was incarcerated and subsequently unable to complete the questionnaire (referred through probation).

The demographic characteristics of the post-assessment group (n=14) were similar to that of the original group (n=41). The men ranged in age from 19 to 51 years with an average age of 35.43 years (SD = 10.75). Of the men for whom relationship status information was available, 31% (n=4) were living common-law, 23% (n=3) were married, 23% (n=3) were single, 15% (n=2) were divorced, and 8% (n=1) were separated. Of the men who were separated or divorced, none indicated that they wanted to reconcile with their partner. Twenty-one percent (n=3) of the men reported that they were in a relationship with a new partner.

2.2 Measures

Four newly developed measures were used in the current study. In addition to these measures, participating men gave consent to have information gathered from their program files (see Appendix A for Letter of Consent). Details about each measure are provided below.

2.2.1 Information form

The information gathered from men's program files included demographic characteristics, information regarding men's referral status (i.e., Early Intervention, Coordinated Prosecution or Voluntary), self-reported abusive behaviour, counsellor judgements of men's success, men's reports of drunkenness, and men's history of violence (see Appendix B for the Research Information Form).

2.2.2 Attitudes

A 78-item self-report attitude measure, entitled the Abuse-Related Attitudes Assessment (ARAA), was developed for the current study. To assist in measure development, a large number of attitudinal measures available within the literature were reviewed. Items deemed relevant to the current study were incorporated into the newly developed measure in either original or modified form. Particular measures that helped inform the study included: the Inventory of Beliefs About Wife Beating by Saunders, Lynch, Grayson, and Linz (1987)(3 items); the Ambivalent Sexism Inventory by Glick and Fiske (1996)(3 items); the Beliefs about Wife Beating by Haj-Yahia (1998)(2 items); the Attitudes Towards Women Scale (ATWS) by Spence and Helmreich (1978)(3 items); the Sex Role Attitudes scale by Burt (1980)(3 items); and the Attitudes Towards Correctional Programming (ACT) Scale by Baxter (year)(4 items). The newly developed measure was also informed by questions from the research of: C.A. Caesar (1996)(3 items); Swim, Aikin, Hall, & Hunter (1995)(4 items); R. Serin (1988)(1 item); Budman et al. (1987) (3 items); Ward (1984)(2 items); and A.W. Pearson's (2002) assessment and refinement of Jehn's Intragroup conflict Scale (2 items). Forty-five original, or modified, items were also included to ensure all relevant attitudinal areas were tapped by the new measure.

The resulting attitudinal measure captured three attitude domains. The first domain was men's attitudes towards abuse itself. Items in this domain tapped men's responsibility for their abusive behaviour (e.g.,"I need to be the one to end my use of violence"), denial of relationship difficulties (e.g., "I have never been annoyed when my partner expressed ideas very different than my own"), blame of their partner (e.g., "My partner often brings up conflicts that have already been resolved") and justification for abusive behaviour (e.g., "When people have been drinking, they cannot be held responsible for their actions"). The second attitude domain was men's hostility towards women (e.g., "Women are sly and manipulate men") and sexism (e.g., "Many women have a quality of purity that few men possess"). Finally, items were included to assess men's attitudes towards intervention in general (e.g., "I don't think that I am going to gain anything from participating in this group") and towards their counsellors (e.g., "I expect that counsellors here will act as if they are better than their clients") (see Appendix C for final questionnaire).

2.2.3 Attitudes towards referral incident

In addition to this general questionnaire, items were created to assess men's attitudes towards the incident that resulted in their referral to the Changing Ways program. Men rated their attitudes on 16 items tapping their view of this incident (e.g., "Do you believe that you are guilty of the offence"; "My partner manipulated the situation to get me in trouble with others") (see Appendix D for final questionnaire).

2.2.4 Knowledge of abusive behaviour

An attempt was also made to assess change in men's knowledge that would be expected to result from participation in a batterer intervention program. One dimension targeted in intervention is men's definition of abusive behaviour. For this questionnaire, men were presented with 22 behaviours and were asked to judge if they were "hurtful/controlling" "not hurtful/controlling" or whether the impact of the behaviour "depends." Answers were marked either correct or incorrect on the basis of typical PAR program teaching (see Appendix E for final questionnaire). Items on this questionnaire were partially derived from an existing self-report form used at Changing Ways.

2.2.5 Knowledge of abuse-supporting self-talk

Finally, men were asked to examine a series of common cognitions (e.g., "I am the only person who can make me mad or keep me calm") and asked to judge if these cognitions were "likely to lead to healthy behaviour," "likely to lead to hurtful/controlling behaviour" or "depends." Items on this questionnaire tapped dimensions commonly targeted in the Changing Ways intervention program. Again, answers were marked either correct or incorrect according to typical PAR program teaching (see Appendix F for final questionnaire).

2.3 Limitations

There are a number of limitations to the current study that should be taken into account when considering results. First, creation of these scales was based on data collected from a modest number of participants. Given the modest numbers, the internal consistency and reliability found is quite good; however, additional analyses are recommended once more data is available. Second, with only a small number of participants completing pre- and post-intervention assessment, no conclusions are possible about changes that may or may not occur as a result of PAR intervention. Analyses of change from pre- to post-intervention are included to help judge the utility of created measures, not to examine program efficacy. Finally, for all measures, additional research is needed to determine if change in specified attitude domain lead to expected change in abusive behaviour.