Community Involvement among Residents of Second-Order Change Recovery Homes

Abstract

The present study investigated the process of second-order change among a group of individuals recovering from substance abuse problems. Data were collected from 56 individuals who were current or past members of Oxford Houses, which are democratically operated recovery homes that have no professional staff and where there is no limit on length of stay. We collected data on individual and house demographics, per week involvement in the community, house involvement in the community, and types of community involvement while residing in the Oxford House. Findings reveal a significant positive relationship between the length of time living in an Oxford House and level of participant involvement in the community. Participants reported multiple factors that increased their community involvement and reported the type of impact that their involvement had on their neighborhoods. Findings from the present study indicate that not only do residents help themselves stay abstinent by living in the Oxford Houses, but residents report that they also make important contributions to their neighborhoods and communities.

Each year, a significant number of individuals re-enter society after substance abuse treatment and yet after treatment many return to former high-risk environments or stressful family situations. Returning to these settings without a network of people to support abstinence increases chances of a relapse (Montgomery, Miller & Tonigan, 1993). As a consequence, substance abuse recidivism following treatment is high, and this is true for both for men and women (Jason, Olson, & Foli, 2008; Jason, Ferrari, Davis, & Olson, 2006). In addition, under modern managed care, private and public sector inpatient substance abuse facilities have reduced their services dramatically. There is a need for alternative models of delivery of services to those with substance abuse, and also to determine their effectiveness and impact on the communities in which they are located.

Traditional treatment approaches might be characterized as involving first order change as the outcomes such as abstinence often do not endure when individuals return to pre-treatment contexts. In this way, first-order interventions can exacerbate the crises they were originally set out to resolve (Watzlawick Weakland, & Fisch, 1974). In contrast, second-order change is more transformative and involves efforts to influence the individual, his or her social network, and all other components of the environment that can contribute positively to a particular problem like substance abuse. If individuals with substance abuse finish brief detoxification or treatment programs and return to the same environments that contributed to their substance abuse, and if appropriate environmental supports for their recovery are not available, the probabilities of relapse are high. Much of the funding in the addictions field is unfortunately invested in these types of first-order interventions. Clearly, more research needs to be directed at understanding approaches that might lead to more permanent and comprehensive forms of substance abuse treatment outcomes (Olson, Jason, d’Arlach, Ferrari, Alvarez, Davis, et al., 2002).

According to the transtheoretical model (Prochaska, Johnson, & Lee, 1998), social liberation involves a therapeutic focus moving away from one’s own problems to a broader awareness of the social issues that surround the challenge (Olson, Jason, Ferrari, & Hutcheson, 2005). As one example of social liberation, emphasis is placed on helping others who are facing similar problems, and this emphasis both contributes to the well being of the individual helping as well others. For example, living in a recovery home and working with neighborhoods on a block crime reduction program allows the individuals in recovery to feel like they are contributing to the larger good of their community. Social and community interventions with these types of features might be better able to promote second-order change.

The Oxford House Model

One form of aftercare recovery that might promote this type of structural change is the Oxford House model. The Oxford House is a recovery home that offers a sober environment for individuals recovering from alcohol and/or substance abuse. Oxford Houses are run in a democratic fashion with members of the house (typically 7 to 12 individuals) holding elected positions and making decisions on a majority-rule basis (Oxford House Manual, 2004). This direct responsibility includes holding an official position in the house (President, Secretary, Chore Coordinator, etc.), helping to maintain the house by performing weekly chores, consistently paying their portion of the house rent, and maintaining an environment that is conducive to recovery. Unlike many treatment programs, the Oxford House does not staff professionals to facilitate recovery. Rather, the members of each house are responsible for promoting sobriety and, if necessary, expelling house members who relapse or do not comply with house rules. The Oxford House is also self supporting, as each house member is responsible for his or her portion of the house rent and expenses. There are no time limits for how long residents can stay in Oxford Houses, and this also is in contrast with most treatment programs and halfway houses that have limits on how long individuals can stay within those programs. The Oxford House might create a type of second-order change for individuals in recovery, where those in recovery are taking direct responsibility for their sobriety.

In 1988, a loan fund was established under the United States Anti-Drug Abuse Act to enable Oxford Houses to borrow money from the federal government, up to $4,000 per house, for initial resources including rent and security deposit (P.Malloy, personal communication, June 30, 2005). This fund has helped the Oxford House organization to grown rapidly from a single Maryland house in 1975 to over 1,300 houses in 40 states within the United States. Over 9,500 individuals currently live in Oxford Houses across the US. A number of Oxford Houses have now opened in and around Melbourne, Australia. Ferrari, Jason, Blake, Davis, and Olson (2006) compared 55 Oxford Houses in the United States to six Oxford Houses in Australia and they found that the houses were established in safe areas where community resources were accessible.

In Partnership with Oxford House: An Action Research Approach

The work that the authors have conducted with Oxford House typifies an action research perspective, one focusing on developing practical knowledge on issues of pressing concern using participatory processes (Reason & Bradbury, 2001). In 1991, the first author saw Paul Molloy, the founder of Oxford Houses, on a television news broadcast in the United States called 60 Minutes. Intrigued by the description of these houses, he contacted Mr. Molloy and out of that initial conversation grew a long-term collaborative partnership between a university-based research team and a grassroots community-based organization. Mr. Molloy was enthusiastic about the first author’s interest in examining Oxford House, and he felt that having an independent program evaluator would be integral to providing credibility to the program.

Shortly after the first author contacted Mr. Molloy, Oxford House decided to establish Oxford Houses in the Midwest. In 1992, the first Oxford House representative, named Bill, was sent from Oxford House, Inc. to Chicago in order to begin the establishment of Oxford Houses in Illinois. Although the Illinois Department of Alcohol and Substance Abuse (DASA) had awarded money from the state’s revolving fund to support the opening of the first house, there were funding complications at DASA that left the representative without necessary housing and financial support. Somewhat discouraged, Bill found temporary lodging at a local shelter where consequently he was robbed of all his personal belongings. Frustrated and dejected, he was on the verge of leaving Chicago and abandoning his task all together. Congruent with an action-oriented agenda, our research team provided Bill with free accommodations, first at the home of one of the members of the research team and then at the DePaul University priests’ residence, so Bill could proceed with his venture. For several months, we also provided him with office space, a telephone, and other resources to facilitate his efforts. Because of this joint effort, Bill was able to successfully establish the first Oxford House home in Illinois. The home was located near the university and graciously named the “DePaul House.”

Over the next 7 years, the DePaul University research team conducted pilot studies and continued collaborative work with the local and national organization. As an example, we jointly wrote a grant proposal to a local foundation to provide funds to hire a recruiter to open two Oxford House homes for women and children in the area. The funded grant was jointly administered by both the DePaul University research team and the Illinois Oxford House organization. DePaul researchers also talked to reporters when members of the press wrote articles about Oxford House. Finally, the research team supplied some of their preliminary research findings to the Oxford House organization during a Supreme Court lawsuit against an Oxford House home in the state of Washington. The suit, based on a zoning law that prohibited more than five unrelated people from living in one dwelling, was representative of some communities’ unwillingness to support Oxford House for fear of reducing their property values. Fortunately, the suit against Oxford House was defeated, and the positive precedent the case set has had an important impact on other Oxford Houses, similar residences, and other halfway houses.

During this time, a team of researchers at DePaul University began seeking federal external funding to support larger and more sophisticated research studies on the process of communal living within Oxford House. The research team submitted multiple federal proposals, but members of a scientific review committee recommended that our team needed to evaluate the effectiveness of Oxford House through a randomized outcome study. The research team was hesitant to advance a methodology that could potentially upset the natural process of self-selection that occurs within Oxford House. That is, members of each Oxford House interview, discuss, and vote on whether an applicant should be admitted as a resident in their house. This democratic process is an important cornerstone to the Oxford House approach to recovery, and the researchers did not want to disrupt that process; as doing so would fundamentally change the structure of how Oxford House operates.

When this dilemma was presented to Mr. Molloy, he said he would support a random assignment design. After extensive discussion between DePaul University and the Oxford House organization on the strengths and possible difficulties with adopting this design, we finally developed a protocol that accommodated random assignment within Oxford House’s democratic system of selection. In the proposed study, individuals finishing substance abuse treatment would be randomly assigned to either an Oxford House or usual aftercare, with follow-up assessments at two years. After years of continued effort to obtain external funding, DePaul researchers finally secured their first Oxford House focused National Institues of Health (NIH) grant. We later learned that in the late 1980s, another independent research group had approached Mr. Molloy with a request to do a randomized outcome study of Oxford Houses. Mr. Molloy had refused this request, as he had no established relationship with the investigators. Certainly, having a prior collaborative relationship with Oxford House helped the DePaul researchers gain the approval of Mr. Molloy, who was able to provide the organizational support and technical expertise for a rigorous outcome study.

Findings from the NIH funded empirical evaluations by the DePaul University research team of the Oxford House have recently been reported. In the randomized study that occurred in Illinois, Jason, Olson, Ferrari, and Lo Sasso (2006) assigned 150 individuals discharged from short term substance abuse treatment randomly to either an Oxford House or “usual care,” which consisted of customary aftercare services. At the 24 month follow up, those assigned to the Oxford House condition had lower substance use (31.3% vs. 64.8%), higher monthly income ($989.40 vs. $440.00), and lower incarceration rates (3% vs. 9%). In a second NIH funded evaluation of Oxford House, Jason, Davis, and Ferrari, and Anderson (2007) followed a national sample of 897 Oxford House residents, and found that after 12 months in the house, substance use was relatively low, income was significantly higher, and that receiving support from other residents significantly increased self-efficacy and reduced the probability of relapse. By involving participants in the design of both of these research projects, by actively discussing the topics to be evaluated and the methods to collect data, the researchers gained a greater appreciation of the culture and unique needs of the community. In addition, the researchers close association with the Oxford House organization facilitated their efforts to obtain funding at NIH.

Recovery Houses and Community Involvement

Although the Oxford House organization has positively affected the outcomes of the residents of these homes, it was unclear whether these homes had other effects at the local and neighborhood level. As an example of this work with other organizations, Zemore and Kaskutas (2004) studied community involvement among recovering alcoholics/drug abusers. Zemore and Kaskutas distinguished between recovery related helping of other alcoholic/substance abusers and non-recovery related helping in the neighborhood and community (e.g., raking leaves, volunteering at a health fair). Many early recovery efforts focus on recovery related areas---helping peers in recovery as 12 step work and only in later recovery do efforts extend to the community such as volunteering with civic groups.

Our research team and the Oxford House central organization were frequently asked by policy officials, media, and community members from towns where Oxford Houses were being introduced what were the effects of Oxford Houses on the surrounding communities. As one way to approach this issue, the Oxford House organization, as well as our research team were most interested in collecting information to document the level of resident participation in their neighborhoods and communities. The present study explored whether the enduring second-order change that occurs among Oxford House residents has ripple effects on their participation in their neighborhood activities. Areas of focus of the present study include determining the types of neighborhood involvement Oxford House residents participate in and measuring factors that lead to neighborhood involvement. Anectodally, we had been informed that those with longer stays in Oxford Houses began to become more active in giving back to their communities. We wanted to explore this hypothesis, so we investigated whether increased length of time within an Oxford House would be associated with increased residents’ neighborhood involvement.

Method

Participant Recruitment

On October 13^th to 16^th, 2005, Oxford House Inc. held their annual Oxford House World Convention in Alexandria, Virginia, United States. The annual convention provides an opportunity for Oxford House residents and others associated with the Oxford House (leadership, alumni, family, etc.) to hold elections, attend presentations, reflect on recovery, and attend social events. Fifty-six participants of the convention took part in this investigation by completing the Neighborhood Involvement Survey (see below). There were about 300 individuals who attended this conference. Twenty-four of the participants who completed this survey attended a presentation led by researchers from DePaul University and were asked to complete the measure. The other 32 participants were a convenience sample recruited throughout the convention within common areas of the convention location.

Action Research Methods

In the present study, the members of the DePaul research team were active participants in the research process rather than objective observers. In addition, the action research was collaborative and participatory, as the hypotheses and methods were developed in collaboration with the participants. According to Reason and Bradbury (2001), action research should enhance actionability, have practical value, employ adequate methods, and be sustainable. We believe that this type of research has a greater chance of ultimately producing positive desirable changes for the key stakeholders.

We involved key stakeholders at an early stage, and throughout our collaboration, we wanted to be sure that all parties felt engaged and that the issues being explored were of importance. As indicated in the introduction, Mr. Molloy was initially interested in determining whether Oxford House was successful in helping residents maintain their abstinence over time. Our collaborative studies reviewed in the introduction were able to clarify this goal, and those studies indicated that the Oxford House approach was extremely effective in promoting abstinence over time. As we discussed these findings, other issues needing attention emerged, such as whether or not an Oxford House had an effect on the larger community.

This question had practical value, as Mr. Molloy and his attorneys frequently have to deal with town officials that bring law suits against his organization in an attempt to stop Oxford Houses from being located in residential communities. Mr. Molloy and the Oxford House organization were most interested in finding out whether the Oxford Houses contributed positively to their neighborhoods, and if so how. The answers to these types of questions were of extreme interest to both the Oxford House organization and the DePaul University research team. If we found that the Oxford House members were positive contributing members of their neighborhoods, these findings could be have practical outcomes in providing towns and neighborhood groups this type of information, thus increasing the chances that towns would be more willing to accept the opening of these houses in neighborhoods.

It was also of importance to develop adequate methods to investigate the community impact. The authors initially spent months talking to residents of Oxford Houses and hearing their stories about their involvement in their communities. It was only after the DePaul University research team and the Oxford House residents had a very clear idea of what types of questions might best tap the experiences of community involvement that an instrument was constructed. The particular items were mutually generated and the Oxford House community provided constant feedback and reflection. It was also ultimately decided to employ both qualitative and quantitative ways of gathering information, and to collect the data in person at the annual Oxford House conference rather than more impersonally over the telephone.

Finally, action research also needs sustainability, and at the outset we developed an infrastructure to involve stakeholders. Our infrastructure is somewhat informal, but it did involve regular meetings, phone calls, and planning sessions, where the DePaul University research team and members of the Oxford House organization discussed goals, methods, and collaborative projects. This infrastructure has allowed our collaborative work to continue for the past 15 years, and the current study is one example of this process. As another example, each year the DePaul University research team presents findings at the Oxford House World conference and solicits suggestions among members for issues needing exploration. At one prior conference, an Oxford House resident approached the first author and stated that it would be very important to assess the issue of tolerance, as living in a house with individuals of different races, sexual orientations, and economic backgrounds might lead residents to become more tolerant over time, and if this occurred, it might be an outcome as important as any changes in substance abuse. This question was brought back to the research team and became the basis for a current collaborative research effort investigating changes in tolerance over time among the residents. As another example, the lawyers working for the Oxford House organization recently approached our research team with a question involving whether Oxford Houses with seven or fewer members are as successful as those with larger numbers. Several towns have recently developed ordinances to preclude a certain number of non-related members of a household, and the Oxford House organization is challenging these ordinances. We have recently written expert opinions arguing that larger households (8 to 15 residents) are more successful (i.e., in terms of abstinence outcomes, maintaining the solvency of the homes, etc.) than smaller households (7 or fewer), and such data could have important implications for ongoing court cases.

Materials

The Neighborhood Involvement Survey consists of 19 items and it was developed to measure neighborhood involvement among Oxford House residents. This is primarily a quantitative instrument which involved residents in the development of questions. The data gathered from the open-ended questions are qualitative. All items and issues were discussed and developed with members of the Oxford House organization and the DePaul University research team. Four items were related to individual level demographics, four pertained to house level demographics, three focused on community involvement, one involved number of hours worked in a typical week, and one requested information about a specific anti-drug movement. One item inquired twenty-two various community activities and allowed participants to check those activities in which they were involved. It also gave participants a chance to indicate other types of involvement that were not one of the 22 listed. The list of activities was also cross checked with Kurtz and Fisher’s (2003) “Kinds of Community service mentioned by AA and NA respondents.”

The remaining five items were open-ended questions that allowed participants to share their perspectives on issues around community involvement, including: What do you feel are the biggest problems in your community? Do you think living in the Oxford House increased your likelihood of involvement in your neighborhood? If you answered “Yes,” how do you think that living in an Oxford House increased your neighborhood involvement? What motivated you to initially get involved in the community? What do you think is the biggest challenge to getting involved in the community? Of all your community involvement, which are you most proud of and why? How did this change the community?

Data Analysis Procedure

Quantitative data including individual and house demographics, per week involvement in the community, house involvement in the community, and types of community involvement while residing in the Oxford House, were analyzed using SPSS 11.0. Atlas.ti was used to categorize responses to the five open-ended questions that participants answered at the end of the survey.

Results

Of the 56 respondents, 66% were men and 34% women. The age of the participants ranged from 19 years old to 69 years old, with a mean age of 40.7 years. Of the participants who reported their ethnicity, 72.2%, reported being White, while 20.37% were African American, 5.6% were multi-racial, and 1.9% were Hispanic or Latino. Forty-four of the participants were current Oxford House members (77.2%), while one was a member of an alumni Oxford House, ten were alumni not living in an Oxford House, and one individual did not report his or her house status.

While the participants who completed the survey represented various individual level demographics, they also represented house-level diversity. Sixteen participants resided in eastern states, 16 participants resided in Western states, 13 in Midwestern states, ten in Southern states, and one participant did not report his or her state of residence. The majority of participants, 51.8%, reported that their Oxford House was located in a suburban setting, while 35.7% reported an urban setting.

Amount and Type of Community Involvement

When asked about individual level involvement in the community, on a scale of one to five (1 = not involved, 3 = somewhat involved, and 5 = very involved), the mean response was 2.9. Each participant also reported perceptions of his or her Oxford House’s level of community involvement as 2.6.

Participants reported participating in the community about 10.6 hours per month. Participants also reported the activities in which they were typically involved. Among the 56 participants, the majority of participants were involved in activities around their recovery. Sixty-three percent were involved in mentoring others in recovery. Forty-four percent of the sample was involved in administering and running support groups. Neighborhood involvement around recovery also came in the form of educating the community; 56% were involved in educating the community about the Oxford House, while 36% were involved in educating the community on recovery in general.

Involvement around recovery also included involvement in large community initiatives, as 39% of participants reported involvement in informing or advising agencies or local leaders and 32% reported involvement in community anti-drug campaigns. For some, this involvement also included speaking at political events (16%), and attending community meetings (30%), and public hearings and forums (21%). Other general community activities reported by participants included working with youth (32%), fundraising (30%), and volunteering time with community organizations (23%).

Length of House Status and Housemate Influence on Community Involvement

Correlations were used to determine the relationship between factors occurring in the house and participants’ involvement in the community. Among Oxford House Members and Alumni House Members, there was a significant positive correlation between the length of time living in an Oxford House and participant involvement in the community (rho = 0.32, N = 49, p = .03). Among Oxford House Members and Alumni House Members, there was a significant correlation between number of housemates involved in the community and participant involvement (rho = .67, N = 31, p < .001).

Linear regressions were computed that had participant involvement as a criterion variable and length of current Oxford House status as the predictor. In this regression, only current house members and alumni house members were included (n = 45). This model explained 15.2% of the variance [F (1,28) = 5.00, p < .33)]. In a second model with only current members or alumni house members, participant involvement again served as a dependent variable, and with number of housemates involved in the community and length of current Oxford House residency as predictor variables, 44.1% of the variance was explained [F (2,27) = 10.64, p < .001)].

Factors that increased community involvement

When asked, “Do you think living in the Oxford House increased your likelihood of involvement in your neighborhood,” 48 of 57 participants answered, “yes.” On a follow up question, asking, “How do you think living in an Oxford house increased your neighborhood involvement,” participants listed many different reasons.

Seven of 48 participants said that the Oxford House helped them become more aware of the need for community involvement. The following quotations reflect this category: “Made me more aware of community issues, social issues,” “By making me more aware of giving back to Oxford House and the community at large,” and “Informed of what is within the community conscious.”

Six of 48 participants cited increased responsibility as a factor that promoted their neighborhood involvement, by saying: “By teaching me how to be a responsible neighbor and participating in the neighborhood,” “Gave me a sense of responsibility and self-worth, as well as gratitude,” and “Commitment to myself and others.”

Five of the 48 participants reported that it was the Oxford House culture/traditions that increased their neighborhood involvement. Participants wrote the following: “The model encourages involvement,” “That being involved can promote the growth of Oxford house as a whole and I am willing to do that any way possible,” and “News, house members.”

While awareness, increased responsibility, and Oxford House culture/traditions were most cited as how the Oxford House increased neighborhood involvement, participants also cited that Oxford House gave them a more formal role and support for community involvement (4 participants), an increased perspective and empathy for others (4 participants), and the opportunity to meet and talk to diverse people, a factor that increases neighborhood involvement.

Impact on the Neighborhood

Participants were asked to consider all their forms of neighborhood involvement, and to list which ones they were most proud of and why. As a follow up to this question, participants were asked how they felt this involvement changed the neighborhood. One of the changes noted by participants was around the decrease in drug abuse for themselves and others, as well as the decrease in crime, as 4 of the 22 participants noted. Examples included: “Starting Oxford Houses and forming chapters…More people in recovery,” [Helping others] get sober…Less crime,” and “[Leading] NA and AA meetings…Helping us stay clean.”

The second theme that emerged was around the impact neighborhood involvement had on youth, which 4 of 22 participants noted. Several participants wrote the following: “Addressing the youth…[It] is helping by giving the youth options to drugs and gangs,” “Working at a local youth shelter…showing the teens that not all men are mean and abusive; and some are in fact, loving,” and “School fundraisers…it helped the children want more for the area around them.”

While reducing drugs/crime and helping youth were cited most frequently, other categories that emerged were neighborhood involvement making the community a better place to live (2 participants), involvement improving housing/transportation (1 participant), and involvement raising money for educational funding (1 participant).

Discussion

As millions of individuals exit treatment programs each year and re-enter community life, successful re-entry into the community becomes an issue for entire communities. Findings from the present study indicate that not only do residents help themselves stay abstinent by living in the Oxford Houses, but that residents report making important contributions to their neighborhoods and communities. The significant positive relationship between length of time living in an Oxford House and level of involvement in the community suggests the importance of time in the process of change. According to the transtheoretical model of change, social liberation focuses moving away from one’s own challenges to a broader awareness of the social issues that surround the challenge. Others including Freire (1998) and Watts, Williams, and Jagers (2003) have also extensively written about this social liberation process. While the process of social liberation may begin in traditional forms of treatment that are often limited to 30 days or less, true second-order change and social liberation strengthens with time. Most of the community involvement that the participants reported suggests the social liberation process occurring on multiple levels. On the individual level, social liberation occurred in the form of mentoring others in recovery; on the organizational level, it occurred in the form of helping other entities run and administer support groups; and on the community level, it occurred in the form of educating the community on recovery in general.

While time is a crucial factor for second-order change and social liberation to occur, the findings of the present study also suggest the important roles that housemates played in bringing about change and liberation. As the number of housemates involved in the community increased, participant involvement in the community also increased. Participants cited both internal changes that occurred while living in the Oxford House, which subsequently increased their community involvement (i.e., being more aware, gaining an increased sense of responsibility, and gaining an increased sense of empathy), as well as external changes that contributed to increases in community involvement (i.e., exposure to the Oxford House cultures and traditions, a role and support for community involvement, and the chance to meet and talk to diverse people). While these changes that lead to community involvement may begin in treatment, they seem to come to fruition upon reintegration within a positive and supportive community.

The present study identifies some of the important factors that contribute to second-order change and social liberation among individuals in recovery at Oxford House. It is important to understand this on a deeper level and future research might explore type of helping more specifically, as well as tracking the community involvement of participants over time. In addition, future research on second-order change and social liberation among those in recovery should also include the perspectives of non-recovering community members for a better understanding of how change and social liberation take place throughout communities, and preliminary data indicate that neighbors of Oxford Houses feel very positive about these recovery homes (Jason, Roberts, & Olson, 2005).

Limitations

There are several limitations in this study. Sample bias is an issue as only those Oxford House residents who attended the conference were eligible. Therefore, generalization of the findings cannot be attributed to the overall population of Oxford House residents. It is certainly possible that the participants who chose to participate in the study might have been more involved than residents who do not attend these types of conferences. It is also possible that the content and/or title of the presentation may have influenced the sample and how they responded to the questionnaire. The quantitative portion of the study had a small sample size, and future studies should include larger samples. In addition, there is a need to use a scale indicating extent of involvement to provide more data. Variables that might potentially be important but were not measured include civic engagement and occupation prior to treatment. Finally, it is important to confirm the self reports of contributions of Oxford House members to their communities by interviewing neighbors or others who are independent of the Oxford House organization.

Conclusions

Over the course of 15 years, the relationship between the DePaul University research team and Mr. Molloy has developed into a long-term collaborative partnership. Kelly (1990) considers this type of collaborative endeavor a discovery process as community partners and researchers work together to define the intervention and assessment activity. The DePaul research team has successfully collaborated with Oxford House over the years, and has developed some important guidelines that have helped this partnership succeed. Some of these guidelines include building trust, sharing resources, and focusing on community strengths rather than weaknesses. Over the years of collaborative interaction, the research team has welcomed the expertise of the citizen recruiters and made many substantial revisions based upon their opinions. One of our original study proposals, for example, stipulated that interviews to Oxford House members across the country would be conducted via telephone. However, based on feedback from members of Oxford House, we were informed that residents were more willing to participate and would respond more openly to personal methods of data collection. Our research team then revised the proposed methodology and instead collected the data in person. The Oxford House members also constantly helped us think about ways of sensitizing the interviews. We simplified questions when collecting these data, and it was apparent that small changes helped us obtain data that was not compromised by jargon used by researchers. Actively involving the Oxford House members in the implementation of the study also helped them feel a central part of the collaborative research process (Davis, Olson, Jason, Alvarez, & Ferrari, 2006).

Citizen participation might enhance ways of understanding a variety of community problems (Jason et al., 2003), such as the social problems of drug and alcohol addiction. The Oxford House organization has developed an innovative and inexpensive way to deal with the high rates of recidivism following treatment. Oxford House represents a promising citizen initiated innovation involving community living that offers an empowerment orientation and findings from the present study suggest that his approach may have an enduring positive influence at the local level.