Self-efficacy and sense of community among adults recovering from substance abuse.
Previous researchers found that abstinence-specific self-efficacy has been predictive of the likelihood of relapse for individuals in recovery from substance abuse. Oxford Houses are sober living homes that provide mutual support and are resident funded and operated. The relationships between house level sense of community and abstinence-specific self-efficacy were examined in the present study with a sample of 754 adults living in 139 Oxford Houses nationwide in a multilevel model. A significant positive relationship between house level sense of community and self-efficacy was observed. These findings have both research and therapeutic implications.

Article Type:
Self-efficacy (Psychology) (Health aspects)
Substance abuse (Care and treatment)
Substance abuse (Psychological aspects)
Stevens, Edward B.
Jason, Leonard A.
Ferrari, Joseph R.
Hunter, Bronwyn
Pub Date:
Name: North American Journal of Psychology Publisher: North American Journal of Psychology Audience: Academic Format: Magazine/Journal Subject: Education; Psychology and mental health Copyright: COPYRIGHT 2010 North American Journal of Psychology ISSN: 1527-7143
Date: June, 2010 Source Volume: 12 Source Issue: 2
Product Code: 8000143 Alcohol & Drug Abuse Programs NAICS Code: 62142 Outpatient Mental Health and Substance Abuse Centers SIC Code: 8093 Specialty outpatient clinics, not elsewhere classified
Geographic Scope: United States Geographic Code: 1USA United States
Accession Number:
Full Text:
Maintaining abstinence for individuals in recovery from substance abuse is a challenging task. Post treatment relapse rates vary across studies and definitions of relapse (Walitzer & Dearing, 2006) but often exceed 50% within 12 months of treatment. For example, Jason, Olson, Ferrari, Majer, et al. (2007) found approximately 65% of individuals exiting substance abuse treatment did not remain abstinent two years following the end of substance abuse treatment. Among a sample of over 2,200 participants, Ilgen, McKellar, and Tiet (2005) found a 69% relapse rate at a one year follow-up. In addition, Witkiewitz and Marlatt (2007) surveyed studies and found relapse rate definitions ranging from no substance use up to no greater than 50% of pre-treatment levels. These findings underscore the difficulty of individuals remaining abstinent following substance abuse treatment.

Bandura (1977) theorized the importance of self efficacy, which is the belief one can successfully perform a task or a particular behavior change. Individuals in recovery from substance use often engage in a number of behavioral changes, with the most critical being maintaining an absence of drug or alcohol use. Thus, their self efficacy to remain clean and sober is important to the recovery process (Bandura, 1999). Research on the relationship between substance abuse and self efficacy has supported the significance of self efficacy in the recovery process (Moos, 2007).

Abstinence-specific self-efficacy, or confidence associated with future drug and/or alcohol use, predicted relapse and recovery in former substance abusers (e.g. Ilgen, McKellar & Moos, 2007; Jason, Davis, Ferrari & Anderson, 2007; Jason, Olson et al., 2007; Trucco, Connery, Griffen & Greenfield, 2007). Ilgen et al. (2005) found 100% confidence in ability to abstain from substance use to be the strongest predictor of abstinence at one year follow up. In addition, many studies demonstrated that individuals who leave substance abuse treatment programs with high abstinence specific self efficacy have lower relapse rates than individuals who leave treatment with low abstinence specific self efficacy (Ilgen, McKellar & Moos, 2007; Ilgen, Tiet, Finney & Moos, 2006; Moos & Moos, 2006). Further, research found high self efficacy upon completion of treatment to influence both short term (1 year) and long term (16 years) recovery outcomes (McKellar, Ilgen, Moos & Moos, 2008).

Sarason (1974) defined psychological sense of community as a construct that focuses on the ability of a community or social network to promote psychological well being and quality of life for its members. McMillan and Chavis (1986) elaborated on this definition and provided a theoretical framework for sense of community. The framework identified four components fundamental for a sense of community: membership, influence, integration and needs fulfillment, and a shared emotional connection among members (McMillan & Chavis, 1986). Bishop, Chertok and Jason (1997) investigated sense of community within a population in substance abuse recovery. They proposed that three factors reflect sense of community for persons residing in recovery facilities: mission, reciprocal responsibility, and disharmony. Mission refers to the shared goals and ideals by group members; reciprocal responsibility refers to shared responsibilities and mutual support; and disharmony refers to discord among group members (Bishop et al., 1997; Graham, Jason, Ferrari & Davis, 2009).

Past research has suggested a relationship between psychological sense of community and subjective well-being (such as happiness; Davidson & Cotter, 1991). For individuals in recovery, sense of community has been researched in sober living environments; however, the relationship between sense of community and self-efficacy has not been studied (Bishop et al., 1997; Curtis, Jason, Olson & Ferrari, 2005; d'Arlach, Olson, Jason & Ferrari, 2006; Graham et al., 2009; Olson et al., 2003). Hill (1996) recommended that research with the sense of community construct move beyond the individual level of analysis and be analyzed as a group level characteristic to better capture possible environmental effects. In other words, how characteristics of recovery homes might influence individuals' abstinence self-efficacy.

In an effort to move beyond the individual level of analysis, the present study examined the role of abstinence self-efficacy and house level sense of community among adult men and women residing in a sample of 139 abstinent residences located across the USA--Oxford House. Oxford Houses are independently run, democratic, sober living residences for people with a past history of substance abuse. The main requirement for admission to an Oxford House is the desire to abstain from drugs and alcohol (Oxford House, Inc., 2008). The houses are fully self-sufficient and members support the house by paying rent, completing chores, and participating in weekly business meetings (Oxford House, Inc., 2008). The Oxford House model has been evaluated as a successful aftercare model for individuals with a history of substance abuse (Davis & Jason, 2005; Jason, Davis et al., 2007). Research has demonstrated that the Oxford House model promotes abstinence from drugs and alcohol (Davis & Jason, 2005; Jason, Olson et al., 2007) and increases self-efficacy among its members (Jason, Davis et al., 2007).

An Oxford House member's likelihood of relapse is significantly related to their length of stay in the house (Bishop, Jason, Ferrari & Huang, 1998; Davis & Jason, 2005; Jason, Davis, et al., 2007). Research found significant differences in outcomes for individuals who stay longer than six months in an Oxford House as compared to those who stay for less than six months (Bishop et al., 1998; Davis & Jason, 2005; Jason, Davis, et al., 2007). Length of stay, therefore, has important predictive properties in evaluating an Oxford House resident's recovery trajectory.

Previous research has examined sense of community among Oxford House residents (see Ferrari et al., 2002). These studies investigated sense of community among Oxford House members including women and children's houses (d'Arlach et al., 2006), women with trauma histories (Olson et al., 2003), and women with disordered eating (Curtis et al., 2005). Research clearly demonstrated the impact of psychological sense of community to Oxford House members; however, few studies have examined the specific factors that comprise sense of community (mission, reciprocal responsibility, disharmony) that contribute to this effect.

Graham and colleagues (2009) found that houses with greater age and income differences among members had higher levels of house harmony. In addition to house harmony, reciprocal responsibility, or mutual support among house members, may influence the success of Oxford House members. Thus, it is important to investigate how the specific factors that comprise psychological sense of community are related to outcomes for Oxford House members.

Prior research on Oxford Houses established these dwellings as safe, sober environments for those in recovery from substance abuse (Davis & Jason, 2005; Jason, Olson, et al., 2007). Findings indicated that residence in Oxford Houses were associated with significantly increased abstinence self efficacy (Jason, Davis, et al, 2007), abstinence from drugs and alcohol use (Davis & Jason, 2005; Jason, Olson, et al., 2007), helping behaviors (Viola, Ferrari, Davis & Jason, 2009) and individual house members' sense of community (Curtis et al., 2005; d'Arlach, et al., 2006; Ferrari et al., 2002; Olson et al., 2003). Nevertheless, no published study has investigated the predictive relationships between an individual's abstinence specific self-efficacy and a house level sense of community for Oxford House members. Specifically, the present study examined whether house level sense of community predicted abstinence-specific self-efficacy after controlling for duration of Oxford House residency.



This study analyzed baseline data from a National Institute of Drug Abuse (NIDA) funded, national, longitudinal study in which an initial convenience sample of 897 Oxford House members from 214 Oxford Houses participated (Jason, Davis, et al., 2007). Individuals were recruited through several methods, including a direct solicitation to House Presidents within the Oxford House system (approximately 91% of sample) and a solicitation at the Oxford House national convention. Participants were fully briefed on the study aims prior to obtaining consent. A nominal sum was paid for participating. Individuals were surveyed at baseline and thereafter, at 4 month intervals for one year. Baseline refers to the initial interview for participants who were living in an Oxford House. For the subsequent waves, participants completed the surveys whether or not they were still living at Oxford House.

Of the initial participants, 754 individuals were residents in Oxford Houses where a participation rate of more than two house members permitted an aggregated house level measure of sense of community. A person can come to Oxford House regardless of how recently they've completed substance abuse treatment. The mean age was 38.4 years (SD = 9.1 years, range =18.3 to 69.1 years) and females were 34% of the sample. At baseline, the mean length of stay in an Oxford House was 10.8 months (SD = 15.5 months, range = 0 to 122 months). The majority of the participants were European American (59%) with African Americans representing the second largest racial/ethnic segment (34%). Overall, 139 houses met the criteria for inclusion (degrees of freedom greater than or equal to 2, and reliability was assessed by intra-house correlation of individual scores) in the analysis, for an average house size of 5.4 individuals (range 2-12; SD = 2.2). On average, at baseline, the participants had been abstinent for 1.6 years (SD = 2.4 years).


The Alcohol Abstinence Self-Efficacy Scale (AASE; DiClemente, Carbonari, Montgomery & Hughes, 1994) is a 20-item Likert scale (1 = not at all; 5 = extremely true) that evaluates an individual's confidence to remain abstinent from alcohol in situations that elicit drinking cues. The scale was duplicated to include substance use other than alcohol, such that the term 'alcohol' was replaced with 'drug' to measure efficacy to remain abstinent from drug use. Participants provided data on both the alcohol and drug measures, and the lower alcohol or drug self-efficacy summary measure was used in order to capture the substance in which they felt least confident. Respondents reported a mean abstinence self-efficacy score of 80.7 (SD=21.2) at baseline. Higher scores suggest a greater degree of self-efficacy. Reliability for the current sample indicated a Cronbach's alpha value of .98.

The Perceived Sense of Community Scale (Bishop, et al., 1997) is a 30-item Likert scale (1 = not at all true; 5 = completely true) that measures perceived sense of community as indicated through three factors: Mission, Reciprocal Responsibility, and Disharmony. The subscale used in the current study was the Reciprocal Responsibility subscale (12 items), which was chosen because it evaluates shared responsibility and mutual support among group members. Examples of items from the Reciprocal Responsibility subscale include: "Members know they can get help from the group if they need it" and "The group makes you feel good for helping." The scores on the Reciprocal Responsibility subscale of the PSOC for individual house members were averaged to create an aggregate house level reciprocal responsibility score. For the 139 Oxford Houses in the sample, the mean score equaled 4.00 (SD=0.39), and the overall sample reliability (a) for the Reciprocal Responsibility subscale was .94.


Hierarchical linear modeling (HLM) was used to examine the relationship of house reciprocal responsibility and an individual assessment of abstinence-specific self-efficacy. Hierarchical linear modeling is a more flexible type of linear regression procedure that allows one to measure the relationship of observations that cluster in some systematic way (e.g., houses, neighborhoods). A covariate was utilized to control for the varying prior length of residencies in an Oxford House. This cross-sectional study had individuals who had spent from a single day to over 10 years in residence at Oxford Houses. The time covariate utilized the natural log of time in OH as a linear predictor of self-efficacy.

The dependent variable was the abstinence self-efficacy, or AASE score. The individual characteristic of Oxford House members, length of time in Oxford House, was placed in the equation (at Level 1). A house level variable was the PSOC Reciprocal Responsibility (RR), and according to HLM, this was placed at Level 2. Of particular interest was how house level Reciprocal Responsibility (Level 2 house variable) influenced participants ratings on the AASE (Dependent Variable) while controlling for length of time lived in Oxford House (OHTime; Level 1). The final HLM equation was:

AASE = [b.sub.0] (=[G.sub.00] + [G.sub.01] RR+ u) + [b.sub.1] OHTime + r

where RR = Reciprocal Responsibility, and OHTime = Time in Residency.

Results of the HLM analysis are displayed below in Table 1. Table 1 indicates that both length of time and reciprocal responsibility were significant predictors of AASE scores.

As shown in Figure 1, with greater length of time living in an Oxford House, average self-efficacy scores increased. Findings of interest suggested an aggregated house level measure of mutual support (Reciprocal Responsibility) was positively and significantly related to higher self-efficacy scores. For the purpose of visually presenting the magnitude of the relationship, predicted self-efficacy was calculated using the 25th percentile (= -.224) and 75th percentile (=.266) of centered RR scores. A one-point change (on a five point scale) in Reciprocal Responsibility was associated with an 8-point change in self-efficacy. The difference between the 25th and 75th percentile was 3.7 AASE points.


The findings of the current study suggest a positive relationship between a house level measure of sense of community and an individual's abstinence-specific self-efficacy. Specifically, houses with higher levels of Reciprocal Responsibility, or mutual support among members, predicted incremental average house member's self-efficacy scores. These findings are of importance given findings that previous research has demonstrated that higher self-efficacy scores predicted a lower likelihood of relapse for individuals in recovery from substance abuse (Ilgen et al., 2005; Jason, Davis, et al., 2007; Jason, Olson, et al., 2007; Trucco et al., 2007). These results suggest a positive relationship between sense of community characteristics of an Oxford House and an individual's prospects for higher self-efficacy and recovery.


Previous research discussed the presence of sense of community in Oxford Houses (Curtis et al., 2005; d'Arlach et al., 2006; Ferrari et al., 2002; Olson et al., 2003); however, little research has tested and quantified the effects of house level sense of community levels on an individuals' self efficacy. The relationship of house level interconnectedness and reciprocity on individual house members may have implications for house composition and operations. This finding may suggest that living in environments that promote mutual responsibilities may assist individuals in recovery to feel more confident in their recovery process and in their ability to remain clean and sober. In addition, this relationship may indicate access to resources, both physical and informational, that strengthen a person's self-confidence. For example, the democractic self-governing aspects of Oxford Houses may both foster increased self-confidence as well as participation in a mutual self-help organization.

As expected, length of time in Oxford House was a significant covariate of abstinence self-efficacy. Thus, longer time lived in an Oxford House predicted higher abstinence self efficacy. Previous research highlighted the importance of length of stay in Oxford House, as individuals who stay for six months or more tend to have better outcomes, such as lower relapse and recidivism rates, than individuals who stay for less than six months (Bishop et al., 1998; Davis & Jason, 2005; Jason, Davis, et al., 2007). The dynamic of this change process, however, is not well understood and it would be beneficial to further examine the processes by which self-efficacy, on average, increases over time within the Oxford House system. This research might possibly have significant implications for post-treatment recovery protocols. For example, if a person being released from a substance abuse treatment setting has lower self-efficacy scores, it might be prudent to refer this person to an abstinence-specific recovery home that has higher sense of community or mutual help resources.

Although these results provide some evidence for a house level relationship of reciprocal responsibility and abstinence-specific self-efficacy, the study had several limitations. First, the sample was a convenience sample, and may not be representative of all Oxford Houses and members. By being an Oxford House-only sample, the results may not generalize to similar types of facilities. Second, the design and analysis was correlational in nature, thus limiting any inferences relating to causality or relationship direction. Another limitation is that the study focused on self-reported abstinence self-efficacy (an attitude) not on actual abstinence (a behavior). Perhaps future studies could examine whether house-level community relates to actual abstinence. Finally, scores on the AASE outcome variable were negatively skewed. These design and data limitations, however, do not undermine the basic findings of this study.

Overall, this study found a significant relationship between a house level sense of community score and an individual's abstinence-specific self-efficacy within the Oxford House system. The specific positive relationship measured an 8 point change (scale 0-100) in abstinence-specific self-efficacy for every one point change (scale 1 to 5) in house level sense of community. On average, individuals who reside in safe and sober, democratically operated aftercare dwellings with a higher sense of community seem to have higher abstinence self -efficacy scores.


Bandura, A. (1977). Toward a unifying theory of behavioral change. Psychological Review, 84(2), 191-215.

Bandura, A. (1999). A sociocognitive analysis of substance abuse: An agentic perspective. Psychological Science, 10(3), 214-217.

Bishop, P. D., Chertok, F. & Jason, L. A. (1997). Measuring sense of community: Beyond local boundaries. The Journal of Primary Prevention,18(2), 193-212.

Bishop, P. D., Jason, L. A., Ferrari, J. R. & Huang, C. (1998). A survival analysis of communal- living, self-help, addiction recovery participants. American Journal of Community Psychology, 26(6), 803-821.

Curtis, C. E., Jason, L. A., Olson, B. D. & Ferrari, J. R. (2005) Disordered eating, trauma, and sense of community: Examining women in substance abuse recovery homes. Women & Health, 41(4) ,87-100.

d'Arlach, L., Olson, B. D., Jason, L. A. & Ferrari, J. R. (2006). Chapter 11: Children, women, and substance abuse: A look at recovery in a communal setting. Journal of Prevention and Intervention in the Community, 31(1), 121-131.

Davidson, W. B. & Cotter, P. R. (1991). The relationship between sense of community and subjective well being: A first look. American Journal of Community Psychology, 19, 246-253.

Davis, M. I. & Jason, L. A. (2005). Sex differences in social support and self efficacy within a recovery community. American Journal of Community Psychology, 36(3/4), 259-274.

DiClemente, C. C., Carbonari, J. P., Montgomery, R. P. G. & Hughes, S. O. (1994). The alcohol abstinence self-efficacy scale. Journal of Studies on Alcohol, 55(2), 141-148.

Ferrari, J. R., Jason L. A., Olson, B. D., Davis, M. I., & Alvarez, J. (2002). Sense of community among Oxford House residents recovering from substance abuse: Making a house a home. In: Fischer A, ed. Psychological Sense of Community. New York, NY: Kluger/Plenum Inc.

Graham, B. C., Jason, L. A., Ferrari, J. R. & Davis, M. I. (2009). Sense of community within Oxford House recovery housing: Impact of resident age and income. Journal of Groups in Addiction and Recovery, 41, 62-70.

Hill, J.L. (1996). Psychological sense of community: Suggestions for future research. Journal of Community Psychology, 24, 431-438.

Ilgen, M., McKellar, J. & Moos, R. (2007). Personal and treatment related predictors of abstinence self efficacy. Journal of Studies on Alcohol and Drugs, 68(1), 126-132.

Ilgen, M., McKellar, J. & Tiet, Q. (2005). Abstinence self-efficacy and abstinence 1 year after substance use disorder treatment. Journal of Consulting and Clinical Psychology, 73(6), 1175-1180.

Ilgen, M., Tiet, Q., Finney, J. & Moos, R. H. (2006). Self efficacy, therapeutic alliance, and alcohol use disorder outcomes. Journal of Studies on Alcohol and Drugs, 67(3), 465-472.

Jason, L.A., Olson, B.D., Ferrari, J.R., Majer, J.M., Alvarez, J., and Stout, J. (2007). An examination of main and interactive effects of substance abuse recovery housing on multiple indicators of adjustment, Addiction,102, 1114-1121.

Jason, L.A., Davis, M.I., Ferrari, J.R. & Anderson, E. (2007). The need for substance abuse after-care: Longitudinal analysis of Oxford House, Addictive Behaviors, 32, 803-818.

McKellar, J., Ilgen, M. A., Moos, B. S. & Moos, R. H. (2008). Predictors of changes in alcohol related self efficacy over 16 years. Journal of Substance Abuse Treatment, 35, 148-155.

McMillan, D.W., & Chavis, D.M. (1986). Sense of community: A definition and theory. American Journal of Community Psychology, 14(1), 6-23.

Moos, R. H. (2007). Theory-based active ingredients of effective treatments for substance use disorders. Drug and Alcohol Dependence, 88(2-3), 109-121.

Moos, R. H. & Moos, B. S. (2006). Rates and predictors of relapse after natural and treated remission from alcohol use disorders. Addiction, 101(2), 212-222.

Olson, B. D., Curtis, C. E., Jason, L. A., Ferrari, J. R., Horin, E. V., Davis, M. I., Flynn, A. M., & Alvarez, J. (2003). Physical and sexual trauma, psychiatric symptoms, and sense of community among women in recovery. Journal of Prevention and Intervention in the Community, 26(1), 67-80.

Oxford House, Inc. (2008). Manual, 2008. Available at http://www. Accessed on November 15, 2009.

Sarason, S.B. (1974). The psychological sense of community: Prospects for a community psychology. San Francisco: Jossey-Bass.

Trucco, E. M., Connnery, H. S., Griffen, M. L. & Greenfield, S. F. (2007). The relationship of self-esteem and self-efficacy to treatment outcomes of alcohol-dependent men and women. The American Journal on Addictions, 16, 85-92.

Viola, J. J., Ferrari, J. R., Davis, M. I., & Jason, L. A. (2009). Measuring In-Group and Out-Group Helping in Communal Living: Helping and Substance Abuse Recovery. Journal of Groups in Addiction & Recovery, 4(1/2), 110-128.

Walitzer, K. S., & Dearing, R. L. (2006). Gender differences in alcohol and substance use relapse. Clinical Psychology Review, 26(2), 128-148.

Witkiewitz, K., & Marlatt, G. A. (2007). Modeling the complexity of post-treatment drinking: It's a rocky road to relapse. Clinical Psychology Review, 27(6), 724-738.

Author's Notes: Funding for this study made possible in part through the National Institute on Drug Abuse grants DA16037 and DA13231. The authors express gratitude to Meg Davis for supervising data collection.

Edward B. Stevens, Leonard A. Jason, Joseph R. Ferrari, & Bronwyn Hunter

DePaul University

Author info: Correspondence should be sent to: Leonard A. Jason, at the DePaul University, Center for Community Research, 990 W. Fullerton Ave., Chicago, IL, 60614:
TABLE 1 HLM Predictors of AASE Scores

Fixed Effect          Coefficient    S.E.    t-value

Level-1 Predictors
  Intercept              44.83       9.31    4.81 *
  Length of Time         2.94        0.57    5.17 *
Level-2 Predictors
Responsibility           8.20        2.00    4.09 *

* denotes significance at .001 level.
Gale Copyright:
Copyright 2010 Gale, Cengage Learning. All rights reserved.