Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
J Groups Addict Recover. Author manuscript; available in PMC 2010 August 4.
Published in final edited form as:
J Groups Addict Recover. 2009 January; 4(1-2): 110–128.
doi:  10.1080/15560350802712488
PMCID: PMC2916197

Measuring In-Group and Out-Group Helping in Communal Living: Helping and Substance Abuse Recovery


With a national U.S. sample of communal-living residents in substance abuse recovery, the tendency to help members inside and/or outside their community was examined. Study 1 (n = 670) developed of the Communal Living In-Group Helping Scale to distinguish helping directed toward housemates vs. others. Study 2 (n = 419) used this communal helping measure and a general altruism scale to explore gender, ethnicity, and 12-Step sponsorship related to in-group (housemates) and out-group (others in the community) behaviors. Results revealed significant sex differences and significantly higher helping for both men and women was reported among 12-Step sponsors along two dimensions. Implications focused on gender-related differences in social helping interactions and in-group formation in recovery communities.

Keywords: helping, in-group, sex differences, recovery, 12-step sponsorship, communal living settings

Numerous research reports indicate several positive long-term psychological and physical health effects associated with helping others (see Brown, Nesse, Vinokur, & Smith, 2003; Smith, Fernengel, Holcroft, Gerald, & Marien, 1994). For example, benefits from helping others include tranquility, improved self-worth, greater optimism, raised self-esteem, as well as decreased depression and helplessness (Luks, 1992). People who frequently engage in helping activities, such as volunteer work or mentoring, experience better perceived physical health and live longer in relation to others who do not perform community service (Andrews, 1990; Moen, Dempster-McClain & Williams, 1992). Helping also affords persons the ability to develop ties and sustain connections with others in their community.

One important application of helping behavior that may have “real-world” relevance is within the field of substance abuse intervention. The effects of the helping processes on the helpers have been rarely studied (Campbell & Campbell, 2000), but the small body of literature that does exist focuses on the topic of persons in recovery providing assistance to peers with similar histories of substance abuse (members of their in-group). Kahn and Fua (1992) found high rates of continued sobriety among people in recovery from substance abuse who served as substance abuse counselors. In learning to be substance abuse counselors, participants gained skills that enabled them to be effective, socially useful, valued by society, and earn a living by their efforts. Kahn and Fua also reported that participants experienced an increase in self-esteem and self-concept. More recently, Zemore, Kaskutas, and Ammon (2004) reported that helping others by sharing experiences, explaining how to get help, and giving advice on housing and employment emerged along with 12-Step involvement as an important predictor of successful substance abuse recovery. Furthermore, among individuals still drinking at follow-up, helping during treatment predicted a lower probability of binge drinking (Zemore, Kaskutas, & Ammon, 2004).

However, the motivation and likelihood for helping to occur may differ if the recipients are considered either in-group or out-group members (Gaertner & Dovidio, 2000). In-group bias, a well-known phenomenon in social psychology (Schroeder, Penner, Dovidio & Piliavin, 1995; Sherif, Harvey, White, Hood, & Sherif, 1961), was defined by Turner, Brown, and Tajfel (1979) as any instance of favoritism whether unfair or unjustifiable, in any manner such as perception, behavior, attitude, or preference. In short, in-group favoritism is simply favoring one’s own “kind” without substantial reasoning. Researchers have found that persons tend to favor members of his/her group in order to foster a social identity (Tajfel & Turner, 1979). Triandis (1994) suggests that establishing group boundaries enables individuals to identify themselves as belonging to either the in-group or out-group and creates security and interdependence among in-group members. In addition, research has shown that in-group members are more likely to be generous, forgiving, and prosocial toward other members of their in-group (Gaertner & Dovidio, 2000).

Research with persons in recovery has not explored the distinction between in-group and out-group helping, or potential sex differences in helping. The distinction between helping behavior in general and helping in recovery may be particularly relevant to the study of recovery processes given that in-group identification seems to be an integral aspect of recovery (Morgenstern & McCrady, 1993). Early in 12-Step substance abuse recovery programs, for instance, individuals are encouraged to personally identify themselves as alcoholics/addicts and seek support and help from members of the group of alcoholics/drug addicts (Alcoholics Anonymous, 1995). The perception that one is accepted by members of an in-group may result in greater optimism about the future, a lessening of the effects of stress, and may increase perceived sense of community (Bishop, Chertok & Jason, 1997; Ferrari, Jason, Olson, Davis, & Alvarez, 2002). Furthermore, persons who feel a part of a group that shares the experience of recovery may display increased levels of empathy and helping behavior (Roberts, Salem, Rappaport, Toro, Luke, & Seidman, 1999).

With regard to helping behavior, in-group/out-group dynamics may affect men and women in different ways. Sex and gender role differences related to helping have been reported in areas such as empathy and prosocial behavior (Skoe, Cumberland, Eisenberg, Hansen, & Perry, 2002; Jaffee & Hyde, 2000). For instance, women typically score higher on self-report indexes of empathy and prosocial personality (e.g., Eisenberg & Fabes, 1998), but men score significantly higher than women on self-reported frequencies of actions (Penner, Fritzsche, Craiger, & Freifeld, 1995). Eagly and Crowly (1986) also reported that in studies of adults’ helping behavior in brief encounters with strangers (out-group members), men provided more instrumental acts of helping than women (e.g., carrying heavy packages, helping people with car troubles), and women received more of such help than men. Women compared to men provide more emotional (in-group) support, especially in close relationships (e.g., Zahn-Waxler, Cole, & Barrett, 1991).

One appropriate population with which to develop a measure and then study the links between helping and in-group/out-group affiliation among individuals in recovery may be Oxford House. Oxford Houses (OH) are self-supported, self-governed, communal-living recovery houses for men and women recovering from alcohol and drug addiction without professional staff involvement (Oxford House Inc., 1998). Each same sex setting utilizes a community-based, social support approach to abstinence from drugs and alcohol where the residents monitor each other’s abstinence recovery. At present, over 1200 Oxford Houses in the US, Canada, and Australia are located in middle class neighborhoods as rented, single-family dwellings that residents of each house self-govern (see Ferrari, Jason, Sasser, Davis, & Olson, 2006). OH residents share the experience of addiction to substances and related hardships (e.g., stigma, housing, family, and other problems associated with substance abuse); share the goal of maintaining abstinence; and spend numerous hours a week resolving conflicts, doing chores, bookkeeping, and interviewing potential residents.

Oxford Houses have been compared with therapeutic communities and found to be similar in their aims (Ferrari et al., 2004). However, OH residents were found to have more personal freedom and self-governance. While neither setting allows for self-destructive behaviors by residents, OH does permit residents personal liberties that are agreed upon by housemates in the form of house rules whereas therapeutic communities do not allow residents to make their own rules (Ferrari et al.). A recent randomized trial exploring the effectiveness of OH found that participants assigned to a communal living Oxford House compared to usual care condition had significantly less substance use and criminal involvement, and significantly better employment outcomes after two years. These findings suggest that there are significant public policy benefits for these types of lower cost, residential, non-medical, community-based care options for individuals with substance abuse problems (Jason, Olson, & Ferrari, 2006). Previous research on sense of community suggests that OH residents experience strong in-group cohesion (Bishop, Jason, Ferrari, & Huang, 1998; Ferrari et al., 2002). The shared experiences of OH members are likely to increase reciprocal responsibility and in-group identification.

While all Oxford House members are encouraged to engage in a reciprocal helping process with in-group members, a sub-population of house members who are also 12-Step sponsors may provide even more help than the typical resident. In their role as sponsor they are expected to provide social support for their sponsee to abstain from substance use. Huselid, Self and Gutierres (1991) reported that the amount and helpfulness of support from an Alcoholics Anonymous sponsor predicted successful completion of a halfway-house recovery program for women. Sponsors may benefit themselves as well. The helper-therapy principle originated by Riessman (1965) asserts that the act of helping others who face similar struggles may have therapeutic outcomes for the person offering the assistance (Wallston, Katahn, & Please, 1983). A key element of this principle is that individuals within a group have a shared experience and identify with others who have suffered in similar ways (Riessman, 1965). Acting as a sponsor might facilitate the learning of important interpersonal competencies. This learning process may lead helpers to experience greater feelings of independence and social usefulness, an increased sense of control, and more willingness to receive help (Riessman, 1990).

Despite the important applied implications of studying in-group helping among persons recovering from substance abuse, there exists no published measure of in-group/out-group helping with men and women in recovery. Although a reliable and valid personality measure of helping behavior in general, the Self-Report Altruism Scale (SRAS) (Rushton, Chrisjohn, & Fekken, 1981), taps helping directed at strangers and acquaintances (out-group members), there are no similar measures for in-group helping among persons in recovery from substance abuse or any mutual-support network. In Study 1 we developed a new in-group helping scale for residential mutual help groups. In Study 2 we utilized this new measure as well as the SRAS to assess helping among men and women in recovery from substance abuse to further understand in-group as well as out-group helping among communal-living group members. This second study examined potential differences between men and women as well as sponsors versus non-sponsors who are OH residents.

Study 1

Study 1 focused on the development of a new self-report inventory, called the Communal Living In-Group Helping Scale, a 10-item measure designed specifically for persons residing in communal-living settings for recovery, such as OH. The scale ascertained the frequency and types of helping behaviors expressed by communal-living group residents. The 10 new items are not reworded items from the SRAS; instead, the new items were written in the same style and with the same parameters as the SRAS. Because the SRAS does not address helping towards in-group members (i.e. housemates/OH members) it is not specific enough to explore in-group helping among and between people in recovery. After constructing and piloting this new scale, we conducted a factor analysis on the items to determine the psychometric properties including internal consistency and factor structure.



After the researchers completed the appropriate training in ethical treatment of human subjects and the completed the university institutional review board processes, participants were recruited through a larger National Institute on Drug Abuse (NIDA) funded study of Oxford Houses from across the United States clustered in Washington/Oregon, Texas, Illinois, Pennsylvania/New Jersey, and North Carolina (see Jason, Ferrari, Davis, & Olson, 2006). Only participants who had lived in OH for at least four months were included in the present study (Mdn = 11 months; range = 4 to 125 months). Previous research found that these same sex dwellings (private, single family homes) housed about 7 employed adults who may or not seek professional help to sustain abstinence, and that residents usually stay for around 12 months before living independently away from their OH setting (Jason et al., 2006, provides an overview and assessment of the OH model of recovery).

A total of 670 residents (representing 194 different U.S. Oxford Houses) fully completed the surveys. The sample was composed of 451 men (M = 39.40 years old; SD = 9.6) and 219 women (M = 36.10 years old; SD = 8.8), ranging from 18 to 67 years of age, which is reflective of the overall national gender and age composition of OH members with two men’s houses for every women’s house (Oxford House Inc, personal communication, September, 1999). Fifty-nine percent of the participants were European American, 33% were African American, 4 % were Latino and 4 % were of other ethnic backgrounds (similar to the population of Oxford House members across the country). On average, participants reported slightly over 12 years (M = 12.43, SD = 2.98) of education (high school diploma or GED), had been paid for work for approximately 15 (M = 15.3, SD = 11.04) days during the last month, and almost one out of every three participants (M = 0.32, SD = 0.71) had one or more dependents to sustain financially. Participants reported an average length of daily poly-substance use (i.e., alcohol and drugs) for nearly a dozen years (M = 11.85 years, SD = 10.42), a history of criminal charges, such as shoplifting/vandalism (47%) and assault (22%) and spending over a month of time incarcerated prior to entering an OH (65%). These characteristics are representative of the profile of most US residents living in an OH (see Jason, Ferrari, Dvorchak, Groessl, & Malloy, 1997).

Scale development

The Communal Living In-group Helping Scale was developed as a supplement to the Self-Report Altruism Scale (SRAS; Rushton, Crisjohn & Fekken, 1981), a valid and reliable uni-dimensional measure of helping behaviors directed toward strangers and acquaintances. Similar to the SRAS, the Communal Living In-group Helping Scale asked participants to rate on a 5-point scale (from 1= never; to 5= very often) the frequency with which they engaged in various helping behaviors. However, the SRAS does not tap into helping directed toward in-group members (e.g., close friends, family, or others in recovery). We created new items meant to supplement the SRAS and increase the relevance of the scale to in-group residents by including questions concerning helping behaviors directed at housemates. The Communal Living In-group Helping Scale includes questions on specific-goal related helping (i.e., helping housemates remain abstinent from drugs or alcohol) as well as questions about receiving help from housemates to tap into the reciprocal nature of in-group helping behavior. In addition, the new measure is unique in that it includes questions concerning individuals’ perceptions of the influence that their communal living setting may have had on their participation in helping behaviors.

In order to create the SRAS supplement, the authors initially constructed a pool of 25 items. A team of 15 OH researchers then evaluated the items. According to their suggestions five items were eliminated and several wording changes were made to the remaining 20 items. Next, in accordance with Jason, Fennell, Klein, Fricano, and Halpert (1999), each of the 20 items was rated by persons experienced working with the population of interest. Eight OH alumni (4 adult men, 4 adult women) scored each item on a 5-point scale (1= definitely do not agree, 5= very strongly agree) in terms of how: this item is “understandable,” this item is “relevant to Oxford House residents,” and this item will not be “misinterpreted.” The 10 items that received the highest ratings as the most understandable and relevant to OH members, and least likely to be misinterpreted were selected for inclusion in this measure. Wording changes occurred as needed and were returned to the raters for input concerning the clarity of revisions.

Sample items for this new measure included, “I have helped a new fellow housemate to get settled at the house and learn the tasks involved” and “I have helped members of the house to get around (bus/taxi fare, ride in car).” A reliability analysis to measure the internal consistency of the 10-items developed for the present study revealed a Cronbach’s alpha of 0.83 (M = 31.76, SD = 6.93: scale range = 10 to 50). See Appendix A for the full version of the Communal Living In-group Helping Scale.

Appendix A
The Communal Living In-group Helping Scale


Participant recruitment for the national study included letters and phone calls to house presidents, and house visits by members of the study’s research team. Additional participants were recruited at the Oxford House national convention. After completing his or her surveys, each participant received a $15 check as a token of appreciation for participating. The surveys utilized for this study were pilot tested and took from 40 to 90 minutes to complete.


Preliminary analyses revealed no significant difference among demographic items (i.e., age, sex, ethnicity, income, level of education, length of residence at Oxford House, alcohol use, drug use) based on the recruitment method. The demographic profiles of participants, described in the participants section above, were consistent with those reported in previous studies (see Ferrari et al., 2002; Davis & Jason, 2004).

An exploratory factor analysis of the Communal Living In-group Helping Scale was conducted. Maximum likelihood analysis revealed a three-factor solution with eigenvalues greater than one, explaining 64% of the common variance. A varimax rotation was performed in order to obtain orthogonal factor loadings. Table 1 presents the varimax factor structure loadings. Using a criterion of 0.40 or greater for factor loadings, the first factor contained five items (alpha = 0.78; M score = 15.29; SD = 4.12) and was titled Help Giving, focusing on whether residents helped other housemates (e.g., “I helped fellow housemates get settled and learn tasks involved.”) This factor concerned how often in the previous six months a respondent participated in specific helping behaviors as opposed to receiving help. Factor 2 was called Oxford Influence and consisted of two items (alpha = 0.75; M score = 8.31; SD = 1.74) and related to participant perceptions regarding how their helping behavior in general has changed because of living in the house, regardless of the target of their helping behavior. (e.g., “Since living in Oxford House I have become a more helpful person in general.”). The third factor was labeled Abstinence Help and contained two items (alpha = 0.76; M score = 6.85; SD = 2.12) that focused on the reciprocal help to remain abstinent from drugs and alcohol that participants give to, and receive from, their housemates (e.g., “My housemates have actively helped me maintain abstinence/sobriety.”). It reflects participant perceptions of how much help they have received from other housemates and how much help they have given to other housemates related to abstinence. One additional item (“My housemates have helped me get a job or pay rent.”) did not load highly on any of the three factors (load = 0.26) and, therefore, excluded from further analyses. Furthermore, while these three factors were significantly inter-correlated none of the coefficients suggest they are measuring the same construct (r =.23 –.53, p <.01). This outcome suggests an initial index on the validity of separate aspects related to mutual support group helping behaviors. It also should be noted that we conducted an additional factor analysis separately for men and for women. The factor structures that emerged were the same across gender.

Table 1
Varimax-Rotated Factor Loadings of “Oxford House Relevant Helping” Items.


Study 1 was a first step in establishing reasonable psychometric properties of the Communal Living In-group Helping Scale. In Study 1, we determined that the scale tapped into three discreet types of participants’ perceived in-group helping rates; namely, Help Giving, Oxford Influenced helping, and reciprocal Abstinence Help. The three factors in this scale were subsequently used to explore how the Oxford House environment may differentially affect the helping behavior and recovery process of individuals who choose to move into an Oxford House, looking for a safe, supportive home to maintain their abstinence from drugs and alcohol. Furthermore, since Oxford Houses tend to be run similarly to other communal living recovery settings (e.g. therapeutic Communities, see Ferrari et al., 2004) the Communal Living In-group Helping Scale may be a useful tool for examining in-group helping in various recovery home settings. One limitation of Study 1 is that it did not include a test-retest sequence, and thus test-retest reliability is not yet available for this scale. Future research with this scale ought to explore the reliability of the measure over time. Furthermore study one does not provide information concerning convergent or divergent validity.

Study 2

Recent research with OH residents suggests that social support plays a different role in women’s recovery than it does in men’s (see Davis & Jason, 2005). These findings underscore a need to explore factors that may unfold during and differentially impact the process of recovery for women and men. As noted earlier, some research suggests that among the general population women tend to provide more help than men do to their close friends or family. Furthermore women tend to help more in environments where they feel safe, whereas men tend to provide more help in public and more help directed toward strangers than do women (Eagley & Crowley, 1986; Zahn-Waxler et al. 1991).

Because African Americans often report less satisfaction with substance abuse interventions (Wells, Klap, Koike, & Sherbourne, 2001) and have lower rates of retention in substance abuse treatment (Mertens & Weisner, 2000) it is also important to explore whether there are differences in in-group helping rates between ethnic groups. However, there is no prior published research suggesting differences in helping rates between European-Americans and African-Americans. Therefore, in our second study we examined ethnic differences between European and African-American OH residents on out-group helping (using the SRAS) and in-group helping (measured by factors of the Communal In-group Helping Scale). Study 2 also tested for differences in helping tendencies toward in-group and out-group members between participants who were versus were not 12-Step sponsors. Research has indicated that sponsorship in mutual help affiliations may lead to better outcomes and build skills among sponsors (Huselid, Self, & Gutierres, 1991; Crape, Latkin & Knowlton, 2002).

Therefore, Study 2 utilized the newly developed Communal Living In-group Helping Scale to explore potential gender, ethnicity, and 12-Step sponsorship differences in helping behaviors among OH residents. In order to explore the relationship between gender and helping behaviors, participant scores from the three factors of Communal Living In-group Helping Scale as well as a measure of out-group help (SRAS) were collected. Based upon the recent literature suggesting sex differences in recovery and context dependent helping (Davis & Jason, 2005; Skoe et al, 2002) we predicted that women would report significantly higher scores on the Communal Living In-group Helping Scale whereas men would report significantly higher scores on our measure of out-group directed helping (as determined by the SRAS). We predicted that sponsors would report higher rates of helping than non-sponsors across each of the dependent variables, while we explored ethnic difference and helping behaviors related to in- and out-groups.



Study 2 included a subset of participants from Study 1. Out of 670 residents included in our first study, only 459 individuals completed both surveys for Study 2 and had similar history of poly substance abuse (i.e., had abused both alcohol and at least one other drug) to be included. Also, 40 additional participants were excluded from Study 2 based upon their ethnic background because of the very small sample sizes per group (i.e., Native American [n=2], Asian/Pacific Islander [n=4], Latino/a [n=22], “other” [n=12]). Consequently, the second study included 419 persons who identified themselves as either African American (n = 159) or Caucasian (n = 260) and who self-reported a history of both alcohol and drug abuse. The sample included 291 men (181 Caucasian, 110 African American: M = 39.94 years old, SD = 9.1) and 128 women (79 Caucasian, 49 African American: M = 36.75 years old, SD = 8.8). Similar to Study 1, these participants reported an average of 12 years of education (high school diploma or GED), had been paid for work for approximately 15 days during the last month, and more than a third of the sample had dependents to sustain financially (M = 0.38, SD = 0.86).

Psychometric measures and procedure

Demographic information (age, sex, ethnicity, level of education, income, number of dependents) included items from the 5th Edition of the Addiction Severity Index-lite (ASI; McLellan, Kushner, Metzger, Peters, Smith, Grissom, et al., 1992). Participants also completed the self-deception enhancement (SDE) subscale of the Balanced Inventory of Desirable Responding (Paulhus, 1988). This 20-item subscale examined the potential for inadvertently deceptive answers to self-reports of questions regarding helping behavior. Participants rated their agreement with each statement along a 7-point scale (1= not true; 7= very true). Paulhus (1988) reported mean scores ranging from 7.5–7.6 (SD = 3.2) and 6.8–7.3 (SD = 3.1) for males and females respectively, with good internal consistency (Cronbach’s alpha ≥ 0.68) and test-retest correlation over a 5-week period of 0.69. The current samples revealed a somewhat lower means across men (M = 5.34, SD = 3.14) and women (M = 4.85, SD = 2.76).

All participants completed the 20-item Self-Report Altruism Scale (SRAS; Rushton, Chrisjohn, & Fekken, 1981) along with the Communal Living In-group Helping Scale (developed in study 1). Participants rated on a 5-point scale (1= never; 5= very often) the frequency with which they engaged in various helping behaviors. Rushton and colleagues (1981) reported respondents’ mean scores ranged from 52.01–57.11, (SD range = 8.89–11.70) and that the SRAS was internally consistent, with a Cronbach’s alpha of ≥ 0.78. The directions for the SRAS were slightly altered in the current study to ask participants to answer each question in reference to the “last 6 months” rather than their lifetime. P.J. Rushton (personal communication, January, 2002) and R. Johnson (personal communication, January, 2002) agreed that for the present study, it would be appropriate to add a phrase regarding a six-month time frame. Scores on the SRAS with the new directions retained their internal consistency with the current sample, yielding a Cronbach’s alpha of 0.85, but did show lower means across men (M = 42.11, SD = 10.87) and women (M = 40.19, SD = 9.96). Participants’ scores on the Help Giving, Oxford Influence, and Abstinence Help factors of the Communal Living In-group Helping Scale from Study 1 were also utilized in the analyses below.


Four Pearson product-moment correlation coefficients were obtained to examine the relationship between the helping measures and the SDE scale. Significant positive correlations were found between the scores on the SDE and Oxford Influence Helping scores (r =.19, p <.01), and between the SDE scores and Help Giving scores (r =.12, p <.01). Consequently, even though these coefficients were rather low in magnitude all successive inferential analyses concerning the helping measures included SDE scores as a covariate.

Next, a two-way MANCOVA was employed using a 2 (ethnic race: Caucasian vs. African American) x 2 (gender: men vs. women) design, with the SRAS and three factors of the new communal helping measures as dependent variables and social desirable tendencies as the covariate. The multivariate test showed that SDE was a significant covariate F (4,411) = 4.66, p <.01. The between-subjects test revealed no significant effect of SDE on SRAS scores or Abstinent Help scores; however, there was a significant main effect for both Help Giving, F(4,411) = 6.35, p =.01, and Oxford Influence, F(4,411) = 15.64, p <.01. The MANCOVA omnibus test revealed no significant effects of race on any of the helping measures.


Even after accounting for the effect of socially desirable reporting (as measured by the SDE), there was a significant main effect (omnibus) for the helping measures, F(4, 411) = 8.00, p <.01. With regards to in-group related helping, female compared to male residents reported that they provided more help to housemates over the past six months as assessed on the helping scales (Help-Giving), F(1, 411) = 6.58, p =.01. Compared to males, female residents reported that OH had more positively influenced them to become helpful people in general and that they had done more to help others maintain their abstinence as a result of OH (Oxford Influence), F(1, 411) = 11.20, p <.01. Compared to males, female residents also reported more reciprocal abstinence help occurring in their houses (Abstinence Help), F(1, 411) = 6.17, p =.01. In contrast, with respect to out-group related helping, F(1, 411) = 4.73, p =.03, men reported greater rates of helping strangers and acquaintances that did not live in OH than women.


An additional item of interest explored in the present study was whether or not serving as a 12-Step program sponsor (e.g., Alcoholics Anonymous or Narcotics Anonymous) had an impact on helping behaviors. A sub-sample of 85 participants reported being sponsors at the time of the survey. In mutual-help organizations, 12-step sponsors are required to provide support to individuals whom they sponsor. Thus, sponsors may have the skills and knowledge of how to help their housemates remain abstinent and this behavior may generalize from the individuals whom they sponsor to their housemates. Also, individuals who decide to sponsor others through the recovery process may be more helpful people in general than individuals who do not volunteer to become 12-Step Sponsors.

A 2 (sponsorship: sponsor vs. non-sponsors) x 2 (gender: men vs. women) MANCOVA omnibus test, controlling for SDE scores, revealed that there was a significant main effect of sponsorship, F(4, 406) = 3.58, p <.01. The between-subjects portion of the MANCOVA test revealed that even after accounting for socially desirable responding, there were significant main effects of sponsorship for three of the four measures of helping. As expected, SRAS scores were significantly higher among sponsors than non-sponsors, F (1, 406) = 7.54, p <.01. Help-Giving was significantly greater among sponsors than non-sponsors, F (1, 406) = 11.36, p <.01 and Oxford Influence scores were also significantly higher among sponsors than non-sponsors, F (1, 406) = 7.35, p <.01 (see Table 2).

Table 2
Mean Scores on the Self-Report Scale Scores

The between-subjects tests also revealed significant sex main effects that replicated the findings previously discussed. Specifically, females compared to males reported higher rates of helping on the three factors of OH relevant helping measure (i.e., Help-Giving, Oxford Influence, & Abstinence Help), with a reverse effect for the out-group helping measure (i.e., SRAS) in which men reported higher scores that women.


Significant sex differences were a major finding for Study 2. More specifically, controlling for the effect of social desirability, women compared to men reported providing more help to housemates over the past six months (Help-Giving), were more likely to report that they helped others maintain their abstinence as a result of OH (Oxford Influence), and reported engaging in more reciprocal help related to abstinence in their houses (Abstinence Help). In contrast, men reported greater rates of helping strangers and acquaintances who did not live in OH than women.

This pattern of women providing more help within the home and carrying out similar amounts or less help outside of the home are consistent with the traditional gender role ascribed within American society (Yoder, 2003). It has been noted that as women’s comfort level increases, so does their helpfulness. Eagley and Crowley (1986) report that when women are in safer-feeling environments existing gender differences in helping rates, even with strangers, virtually disappear. In relation to recovery, women in OH have reported strong appreciation for the safe and supportive environment of Oxford Houses (Dvorchak, Grams, Tate, & Jason, 1995) and have reported psychological sense of community both when they enter the homes and after being there for sometime (d’Arlach, Curtis, Ferrari, Olson, & Jason, in press). An increased comfort may account for the greater helping tendency reported by women’s Oxford Houses. This sex difference did not emerge for helping strangers or acquaintances outside of the Oxford Houses, consistent with past research that has found men engage in more helping activities than women when helping a stranger (e.g. Oswald, 2000; Hope, Jackson, Howard & Avis, 1988).

The sex differences that emerged in Study 2 were consistent across the different helping factors. Women reported that they were providing and receiving a variety of help in their Oxford Houses. Study 2 adds to prior research on social support by Davis and Jason (2005) using a different sample of OH residents which suggests that a social-support based model of recovery may be consistent with traits that are traditionally considered feminine, such as nurturing, sharing, and cooperation. Together, these studies argue for the importance of recognizing the differential processes occurring for women in addiction recovery. While more research in this area is needed prior to developing specific suggestions for promoting clinical support to women in recovery, these findings support practices that increase women’s comfort providing and receiving support from their housemates during the recovery process. Consistent mutual helping behaviors in and around their homes may increase the likelihood of more consistent and lasting abstinence from drugs and alcohol for women.

Another interesting finding from Study 2 was the significant impact of 12-step sponsorship on helping. Controlling for socially desirable responding, participants who served as 12-step program sponsors scored significantly higher than non-sponsors on the SRAS, the Help-Giving, and the Oxford Influence measures. A plausible explanation for this outcome of the present study may be generalization of helping behavior (Nemeroff & Karoly, 1991). That is, because 12-step sponsors are required to provide support to individuals whom they sponsor, they already have the skills and knowledge of how to help their housemates remain abstinent. These skills and behaviors may generalize from the individuals whom they sponsor to their housemates. An alternative explanation is that individuals who decide to sponsor others through the recovery process may be more helpful people in general than individuals who do not volunteer to become 12-step sponsors.

A priori assumptions concerning the physical and psychological health benefits of helping for the helper, based on community and health psychology literature (Brown et al., 2003; Schwartz & Sendor, 2000; Riessman, 1965; Wallston, et al., 1983), guided the focus of the present studies toward exploring whether living in OH may serve as catalyst for residents to become more helpful. However, both studies presented here were only a first step toward answering this question. Because of the nature of Study 2’s design (i.e., using a cross-sectional method), causal statements cannot be made about the effects that emerged over time. Further research using a longitudinal design might be beneficial to understand changes over time. A repeated measures design also might control potential confounding variables, such as individual differences in altruistic tendencies (e.g., self-concept and self-esteem factors) among OH residents.

General Conclusion

Along with a longitudinal design and additional personality variables that might impact aspects of altruistic tendencies of residents, we suggest that future research attempt to incorporate actual observation of behaviors into the methodology of “real-world” in-group/out-group studies. Together, these methods would triangulate data and attain the most accurate information concerning the frequency of specific helping behaviors.

As more information is gained with regard to building a sense of community and the in-group/out-group influences on helping behavior, female Oxford Houses might be seen as models for recovery homes promoting great potential for success. In addition, consistent with the helper-therapy principle, recovery models that encourage individuals to become 12-step sponsors may benefit the recovery of the sponsor as well as the “sponsee” (Crape, Latkin, Laris, and Knowlton, 2002). Therefore, we also suggest that 12-step sponsorship be studied further in its own right for its benefits to the recovery process. Nevertheless, the present studies contribute to a better understanding of altruism and support group processes by developing a brief, reliable and valid self-report instrument. Furthermore, Studies 1 and 2 add useful information on both of helping behaviors in situations of substance abuse addiction recovery and of the processes involved in in- and out-groups associated in social support settings.

Table 3
Mean Scores of Helping Measure and Subscales by Twelve-Step Sponsorship and Participant Sex


Funding was made possible in part through National Institute on Drug Abuse (NIDA) grants #5F31DA16037 and # R01DA13231. Portions of this project come from the first author’s Masters Thesis under the supervision of the second author, and were presented at the 2004 annual meeting of the Midwest Psychological Association.

Contributor Information

Judah J. Viola, DePaul University.

Joseph R. Ferrari, DePaul University.

Margaret I. Davis, Dickinson College.

Leonard A. Jason, DePaul University.


  • Alcoholics Anonymous. Alcoholics anonymous. 3. New York: Alcoholics Anonymous World Services, Inc; 1995.
  • Andrews HF. Helping and Health: The relationship between volunteer activity and health-related outcomes. Advances (the Journal of Mind-body health) 1990;7(1):25–34.
  • Bishop PD, Chertok, Jason LA. Measuring sense of community: Beyond local boundaries. Journal of Primary Prevention. 1997;18:193–212.
  • Bishop PD, Jason LA, Ferrari JR, Huang CF. A survival analysis of communal living, self-help, addiction recovery participants. American Journal of Community Psychology. 1998;26:803–821. [PubMed]
  • Brown SL, Nesse RM, Vinokur AD, Smith DM. Providing social support may be more beneficial than receiving it: Results from a prospective study of mortality. Psychological Science. 2003;14(4):320–327. [PubMed]
  • Campbell DE, Campbell TA. The mentoring relationship: Differing perceptions of benefits. College Student Journal. 2000;34(4):516–523.
  • Crape BL, Latkin CA, Laris AS, Knowlton AR. The effects of sponsorship in 12-step treatment of injection drug users. Drug & Alcohol Dependence. 2002;65(3):291–301. [PubMed]
  • d’Arlach L, Curtis CE, Ferrari JR, Olson BD, Jason LA. Substance-abusing women and their children: A cost-effective treatment option. Journal of Social Work Practice in the Addictions (in press)
  • Davis MI, Jason LA. Sex differences in social support and self-efficacy for recovering from substance abuse. American Journal of Community Psychology. 2005;36(3–4):259–274. [PubMed]
  • Dvorchak PA, Grams G, Tate L, Jason LA. Pregnant and postpartum women in recovery: Barriers to treatment and the role of oxford house in the continuation of care. Alcoholism Treatment Quarterly. 1995;13(3):97–107.
  • Eagly AH, Crowley M. Gender and helping behavior: A meta analytic review of the social psychology literature. Psychological Bulletin. 1986;100:283–308.
  • Eisenberg N, Fabes RA. Prosocial development. In: Damon W, editor. Handbook of child psychology. 5. Vol. 3. New York: Wiley; 1998. pp. 701–778.
  • Ferrari JR, Jason LA, Davis MI, Olson BD, Alvarez J. Similarities and differences in governance among residents in drug and/or alcohol misuse: Self vs. staff rules and regulations. Therapeutic Communities: The International Journal for Therapeutic and Supportive Organizations. 2004;25:179–192.
  • Ferrari JR, Jason LA, Olson BO, Davis MI, Alvarez J. Sense of community among Oxford House residents recovering from substance abuse. In: Fisher AT, Sonn CC, Bishop BJ, editors. Psychological Sense of community: Research applications and implications. New York: Klewer/Plenum; 2002. pp. 109–122.
  • Ferrari JR, Jason LA, Sasser K, Davis MI, Olson BD. Creating a home to promote recovery: The physical environment of Oxford House. Journal of Prevention & Intervention in the Community. 2006;31:27–40. [PubMed]
  • Gaertner SL, Dovidio JF. Reducing Intergroup Bias: The common ingroup identity model. Psychology Press/Taylor & Francis; Philadelphia PA: 2000. p. 212.
  • Hope SL, Jackson HF, Avis MJ. Helping: The effects of sex differences and locus of causality. British Journal of Social Psychology. 1988;27(3):209–219.
  • Huselid RF, Self EA, Gutierres SE. Predictors of successful completion of a halfway-house program for chemically-dependent women. American Journal of Drug & Alcohol Abuse. 1991;17(1):89–101. [PubMed]
  • Jaffee S, Hyde JS. Gender differences in moral orientation: A meta-analysis. Psychological Bulletin. 2000;126:703–726. [PubMed]
  • Jason LA, Fennell PA, Klein S, Fricano G, Halpert J. An investigation of the different phases of the CFS illness. Journal of Chronic Fatigue Syndrome. 1999;5(3/4):35–54.
  • Jason LA, Ferrari JR, Davis MI, Olson BD, editors. Oxford House: A decade of collaboration. New York: Haworth Press; 2006.
  • Jason LA, Ferrari JR, Dvorchak PA, Groessl EJ, Malloy JP. The characteristics of alcoholics in self-help residential treatment settings: A multi-site study of Oxford House. Alcoholism Treatment Quarterly. 1997;15:53–63.
  • Jason LA, Olson B, Ferrari JR. An evaluation of communal housing settings for substance abuse recovery. American Journal of Public Health. 2006;91:1727–1729. [PubMed]
  • Johnson RC, Danko GP, Darvill TJ, Nagoshi CT. “Docility” versus reciprocity as an explanation for evolutionary selection for altruistic behavior. Personality and Individual Differences. 1992;13(3):263–267.
  • Johnson RC, Danko GP, Darvill TJ, Bochner S, Bowers JK, Yua-Huang H, et al. Cross-cultural assessment of altruism and its correlates. Personality and Individual Differences. 1989;10(8):855–868.
  • Kahn MW, Fua C. Counselor training as a treatment for alcoholism: The helper therapy principle in action. The International Journal of Social Psychiatry. 1992;38(3):208–214. [PubMed]
  • Luks A. The Healing Power of Doing Good. Columbine, New York: Fawcett; 1992.
  • McLellan AT, Kushner H, Metzger D, Peters R, Smith I, Grissom G, Pettinati H, Argeriou M. The fifth edition of the Addiction Severity Index. Journal of Substance Abuse Treatment. 1992;9:199–213. [PubMed]
  • Mertens JR, Weisner CM. Predictors of substance abuse treatment retention among women and men in an HMO. Alcoholism: Clinical and Experimental Research. 2000;24:1525–1533. [PubMed]
  • Moen P, Dempster-McClain D, Williams RM. Successful aging: A life-course perspective on Women’s multiple roles and health. American Journal of Sociology. 1992;97(6):1612–1638.
  • Morgenstern J, McCrady BS. Cognitive processes and change in disease-model treatment. In: McCrady BS, Miller WR, editors. Research on alcoholics anonymous: Opportunities and alternatives. Rutgers Center of Alcohol Studies; New Brunswick, NJ: 1993. pp. 153–164.
  • Nemeroff CJ, Karoly P. Operant methods. In: Kanfer FH, Goldstein AP, editors. Helping people change: A textbook of methods. New York: Pergamon Press; 1991.
  • Oxford House, Inc. Oxford House Manual. Silver Springs; MD: 1998.
  • Oswald PA. Subtle sex bias in empathy and helping behavior. Psychological Reports. 2000;87(2):545–551. [PubMed]
  • Paulhus DL. Assessing self deception and impression management in self reports: The balanced inventory of desirable responding. Department of Psychology, University of British Columbia; Vancouver, B.C., Canada V6T 1Y7: 1988. Manual available from the author at the.
  • Penner LA, Fritzsche BA, Craiger JP, Freifeld TR. Measuring the prosocial personality. In: Butcher J, Spielberger CD, editors. Advances in personality assessment. Vol. 10. Hillsdale, NJ: Erlbaum; 1995. pp. 147–163.
  • Riessman F. The “Helper” Therapy principle. Social Work. 1965;10:27–32.
  • Riessman F. Restructuring help: A human services paradigm for the 1990s. American Journal of Community Psychology. 1990;18(2):221–230.
  • Roberts LJ, Salem D, Rappaport J, Toro PA, Luke DA, Seidman E. Giving and Receiving Help: Interpersonal transactions in mutual-help meetings and psychosocial adjustment of members. American Journal of Community Psychology. 1999;27(6):841–868. [PubMed]
  • Rushton PJ, Chrisjohn R, Fekken C. The altruistic personality and the self-report altruism scale. Personality and Individual Differences. 1981;12 (4):293–302.
  • Schroeder DA, Penner L, Dovidio JF, Piliavin JA. The Psychology of Helping and Altruism: Problems and Puzzles. New York: McGraw-Hill; 1995.
  • Schwartz C, Sendor M. Helping others helps oneself: Response shift effects in peer support. In: Schmaling K, editor. Adaptation to changing health: Response shift in quality of life research. Washington DC: American Psychological Association; 2000. pp. 43–70.
  • Sherif M, Harvey OJ, White BJ, Hood WR, Sherif CW. Intergroup Conflict and Cooperation: The Robbers Cave Experiment. Norman OK: University of Oklahoma Book Exchange; 1961.
  • Skoe EA, Cumberland A, Eisenberg N, Hansen K, Perry J. The Influences of Sex and Gender-Role Identity on Moral Cognition and Prosocial Personality Traits. Sex Roles. 2002;9/10(46):295–309.
  • Smith C, Fernengel K, Holcroft C, Gerald K, Marien L. Meta-analysis of the associations between social support and health outcomes. Annals of Behavioral Medicine. 1994;16:352–362.
  • Tajfel H, Turner JC. An integrative theory of intergroup conflict. In: Austin WG, Worchel S, editors. The social psychology of intergroup relations. Monterey, CA: Brooks/Cole; 1979. pp. 33–47.
  • Triandis HC. Culture and Social Behavior. New York: McGraw-Hill; 1994.
  • Turner JC, Brown RJ, Tajfel H. Social comparison and group interest in in- group favoritism. European Journal of Social Psychology. 1979;9:187–204.
  • Wallston MM, Katahn M, Please J. The helper therapy principle applied to weight management specialists. Journal of Community Psychology. 1983;11:58–66. [PubMed]
  • Wells K, Klap R, Koike A, Sherbourne C. Ethnic disparities in unmet need for alcoholism, drug abuse and mental health care. American Journal of Psychiatry. 2001;158:2027–2032. [PubMed]
  • Yoder JD. Women and gender: Transforming psychology. 2. Upper Saddle River, New Jersey: Prentice Hall; 2003.
  • Zahn-Waxler C, Cole PM, Barrett KC. Guilt and empathy: Sex differences and implications for the development of depression. In: Garber J, Doge KA, editors. The development of emotion regulation and dysregulation. New York: Cambridge University Press; 1991. pp. 243–272.
  • Zemore SE, Kaskutas LA, Ammon LN. In 12-step groups, helping helps the helper. Addiction. 2004;99(8):1015–1023. [PubMed]