Relational-Cultural Therapy Intervention With Incarcerated Adolescents

Abstract

The authors implemented a small series (N = 4) single-case research design to assess the effectiveness of a nine-session Relational-Cultural Therapy (RCT) intervention with adolescent females incarcerated in a youth detention facility. Analysis of participants’ scores on the Relational Health Inventory using the Percentage Of Data Points Exceeding the Median procedure yielded treatment effects indicating that RCT may be effective for promoting relational empowerment and engagement with others; however, treatment was not associated with a greater amount of relational authenticity with others. Discussion of systemic variables that limit relational development during incarceration is provided.

Keywords

counseling outcome relational-cultural therapy relational health incarcerated adolescents females single-case research design RHI

The National Center for Juvenile Justice has estimated that approximately 97,000 juveniles aged 18 and younger were housed in residential correctional facilities within the United States (Snyder & Sickmund, 2006). Of these youth, 15% were female juvenile offenders, a population that has increased in the frequency and severity of criminal offences during the past decade (Snyder, 2004; Snyder & Sickmund, 2006). The number of female offenders in custody increased 52% from 1991 to 2003, with most female offenders being 15 to 16 years of age (Snyder & Sickmund, 2006). With the female adolescent population increasing within the correctional system, researchers have begun examining the effects of juvenile incarceration. These researchers have detected ubiquitous trends for decreased educational attainment (Holman & Zeidenberg, 2006), restricted occupational opportunities (Glaser, Calhoun, Bates, & Bradshaw, 2003), increased mental health symptoms (Corneau & Lanctôt, 2004), and social skills deficits (Gagnon & Barber, 2010) following release when compared to nonadjudicated peers. Furthermore, despite the differential experience of females and males while incarcerated, both groups often receive similar rehabilitative programming with little desired effect on recidivism or social adjustment among females (Covington, 2007). Thus, it is a prudent task for counselors to implement and evaluate counseling programs that promote the social and relational development of adolescent females.

The National Mental Health Association (2004) has endorsed several therapeutic intervention strategies for use with adolescents involved with justice systems, each emphasizing prosocial interactions with peers, family members, and significant others. Of these, one widely endorsed counseling intervention is family-focused therapy, which incorporates active participation of the family in developing treatment, educational, and aftercare plans. Cognitive–behavioral therapies addressing relational and social skill deficits, coping skill development, and self-management are also recommended. Both of the aforementioned approaches are highly structured, rigorous, and organized around the development of social functioning through positive behavioral change (Altschuler, 1998). In a meta-analysis conducted by Lipsey, Wilson, and Cothern (2000), interpersonal skills and family-oriented programming consistently yielded statistically significant declines in recidivism rates for institutionalized youth offenders. In this review, the most successful programs often included the use of multiple counseling services from community programs that offered advocacy, counseling, educational, vocational supports in conjunction with behavioral programs such as stress management training, anger management, and cognitive mediation (Lipsey, Wilson, & Cothern, 2000).

Despite progressive strides reported by these researchers, little can be found regarding evidence-based practices that are effective in other areas of social and relational functioning such as engaging in growth-fostering relationships. Additionally, although the gender gap between incarcerated male and female offenders has begun to close, a marked discrepancy exists in the representation of females in the juvenile justice system research. As a consequence, the data available addressing effective interventions with female adolescents is ambiguous, resulting from the practice of grouping female youth within the same standards for rehabilitation as those of males. Therefore, more investigations are needed to extend and evaluate counseling interventions such as Relational-Cultural Therapy (RCT; Jordan, 2010) that are sensitive to the development of women.

RCT Groups

Originally founded on the writings of Jean Baker Miller in Toward a New Psychology of Women (1976), RCT adheres to the core beliefs that all personal growth occurs during and throughout meaningful relationships. Jordan (2010) later noted that all people desire connection and that mutual empathy and empowerment are needed for relationships to promote growth. Proponents of RCT posit that growth-fostering relationships are characterized by increases in vivacity ability to take action, clarity about relationships between self and others, a sense of worth, and the yearning to form more meaningful relationships(Miller, 1986). Miller (1990) proposed that when relationships are associated with marginalization, individuals experience a relational disconnection recognizable by diminished energy, disempowerment, confusion, decreased sense of self-worth, and turning away from relationships. Although all relationships include disconnections, growth-fostering relationships utilize and transform disconnections, through mutual effort, into strengthened connections (Jordan, 2010).

Group practice of relational-cultural theory explores and implements the core tenets of RCT with the goal of promoting growth-fostering relationships, which increases creativity, clarity, and relational awareness/intelligence (Jordan & Dooley, 2001). Comstock, Duffey, and St. George (2002) presented an RCT framework for approaching group counseling. The stages of the relational movement throughout the group are outlined as: (a) establishing safety through supported vulnerability, (b) flexibility in making relational choices, (c) experiencing empowerment and conflict, and (d) the development of relational confidence and relational resilience. This model has been proposed as useful for working with men and women experiencing grief, sexual and physical trauma, addictions, divorce, and eating disorders (Comstock, Duffey, & St. George, 2002). Jordan and Dooley (2001) presented an alternative model specifically for working with mental health agency, school, and prison populations. This approach is highly structured and explicitly provides an eight session manual for counselors that includes experiential activities to facilitate learning. Although these outlines for group intervention are promising, there is a paucity of research available assessing the utility of this model with diverse populations such as incarcerated female adolescents.

RCT groups may have distinct utility within the detainment setting given that within the juvenile justice system traditional views of independence and self-reliance are markedly intensified. Covington (2007) outlined the harmful impact of the traditional detainment environment for the relational development of females by noting:

Tragically, current correctional settings often recreate women’s relationships of disconnection and violation on a systemic level. Our criminal justice system, which is based on power and control, reflects the dominant/subordinate model of our patriarchal society. It is a microcosm of the larger social system. Relationships in correctional setting are based on ranking people, with women and girls at the lowest rung of the ladder. (p. 12)

Therefore, it may be conjectured that the juvenile correctional system does not promote rehabilitation or development of critical social and relational skills by virtue of systemic design. Given this supposition, Covington (1998) has advocated for women to be provided with opportunities for change, growth, and healing based on relationships and mutuality while incarcerated. As such, counselors implementing RCT programming may be able to support this initiative by providing relational development opportunities that prepares females to connect with one another and the community-at-large after serving their sentences. Based on these two assumptions, this study examined the following research questions: Is an RCT group counseling intervention effective for increasing the relational health of incarcerated adolescent females as measured by scores on the Relational Health Indices? (Liang et al., 2002); and what relational domains are more resistant to development while in the correctional setting?

Method

Participants

Five adolescent females were solicited to participate in group therapy based on the RCT program developed by Jordan and Dooley (2001); one declined to participate. The four participants electing to complete the group were Hispanic (n = 2; 50%), Caucasian (n = 1; 25%), and African American/Black (n = 1; 25%). Participants were 16 (n = 3; 75%) and 17 (n = 1; 25%) years of age who were incarcerated for offenses including family violence, drug-related, assault, absconding, and violation of probation. Three participants resided within lower socioeconomic status (SES) households (75%) and one was from a middle SES household (25%); two reported living with a single parent. Pseudonyms were self-selected by participants to protect their identities during data collection, analyses, and reporting.

Participant 1

Maegan was a 16-year-old Caucasian female of Scottish-American descent. She is from a traditional nuclear family with SES in the middle range. Maegan was academically on track, verbal during interactions with group, and reported that detainment was associated with assault and family violence.

Participant 2

Lauren was a 17-year-old Hispanic female born and raised in a traditional nuclear family with a lower SES background. Lauren was academically behind one grade, participated in groups moderately in the beginning, but became highly engaged within latter sessions, and reported a history of assault, family violence, and narcotics-related misdemeanors.

Participant 3

Crystal was a 16-year-old African American female raised in a single parent household with a lower SES background. Crystal was academically behind one grade, actively engaged and highly verbal during group meetings, although distracting and encouraging of group to get off topic at times, and reported a history of assault charges.

Participant 4

Samantha was a 16-year-old Hispanic female raised in a single parent household with a lower SES background. Samantha was academically on track, did not participate in group process often, but intermittently conveyed understanding of content, and reported history of absconding from her family home and repeated violations of probation terms.

Instrumentation

Relational health indices

The Relational Health Indices (Liang et al., 2002) was developed to detect the degree to which individuals are engaging in growth fostering relationships with peers, mentors, and their community. Several researchers have used the Relational Health Inventory (RHI) to investigate the influence of environmental and interpersonal variables on the relational health (Frey, Beasley, & Miller, 2006; LaBrie et al., 2008; Liang & West, 2011; Liang, Tracy, Kenny, Brogan, & Gatha, 2010). The RHI (Liang et al., 2002) is a 37-item self-report questionnaire that yields three composite subscales and can be calculated to measure the relational quality with Peers, Mentors, and the Community. The response format for the RHI is a 5-point Likert-type scale ranging from 0 (never) to 4 (always) and requires participants to assess variables associated with relationships. Higher scores on the RHI represent higher quality of relational health. Three subscales can be calculated to assess relational domains of authenticity, empowerment, and engagement within each of the composite domains. Authenticity was defined as “the process of acquiring knowledge of self and other and feeling free to be genuine in the context of the relationship” (Liang et al., 2002, p. 26). Empowerment was defined as “the experience of feeling personally strengthened and inspired to take action,” whereas engagement was described as the “perceived mutual involvement, commitment, attunement to the relationship” (Liang et al., 2002, p. 26). Frey, Beasley, and Newman (2005) reported high Cronbach’s α coefficients for the subscales ranging from .86 to .91 indicating strong reliability of the items. Liang, Tracy, Kenny, Brogan, and Gatha (2010) detected similarly robust reliability and retention of the factor structure of the RHI with a sample of adolescents students (N = 188).

Setting

During the present study, each of the participants was incarcerated in a military-style detention center for a period ranging from 8 to 10 months. The facility, located in a central southern state of the United States is based on discipline, education, and counseling. Participants and their peers complete regular physical training activities, go to school with the encouragement to complete their General Education Diploma, and attend regular programming associated with life skills development and problem solving. Counseling services for adolescents detained at this facility are contracted through community partners such as the mental health agency that provided the RCT groups in this study. In this particular facility, males and females are separated and under 24-hr observation by same sex staff that implement facility rules and standards for behavior including protocols for interpersonal communication, academic task completion, and completing scheduled activities. Noncompliance with facility policies is associated with noncorporal punishments including physical training, increased cleaning duties, and isolation.

Treatment

Participants received nine sessions of group RCT using the manual presented by Jordan and Dooley (2001). The RCT treatment manual consists of eight structured sessions completed in approximately one and a half to two hours. Each group session is led by two cocounselors and composed of a relational check-in, creative moments, clarity in connections, and closing connections/relational impact. During the relational check-in each group member, including the counselors, makes a statement about their current status emotionally, cognitively, or spiritually and what impact their state-of-being may have on the group process. Next, the creative moments portion of the session educates group members about relational concepts (e.g., relational images, mutual empathy, experiencing disconnection, etc.) and how one is shaped by personal experiences. Group members are supported and guided to contribute to the groups understanding of these concepts and how they are expressed in their lives. Following, the clarity in connections section, the group provides concept-related experiential activities to facilitate participants’ deeper understanding of the content through personal expression, mutual sharing, and collaborative discussion. Finally, the closing connections/relational impact component provides group members an opportunity to discuss their emotional, cognitive, and spiritual status following the education and experiential activities, how the group has affected them, and whether any new questions or perspectives have emerged.

Prior to providing treatment, the authors reviewed the content of the RCT treatment manual and made revisions to content if necessary to promote developmental and cultural fit for the population being served. The counselors (second and third authors) made an initial round of accommodations to the content independently and collaborated with their supervisor (first author) to assure that the interventions were within the intended scope of the sessions’ purpose. Additionally, one session was added to the curriculum addressing the expression of empathy to individuals from different backgrounds. The activity was based on a participatory visual analysis task developed by Lenz and Sangganajanavanich (in press) which has been demonstrated to promote empathic responses among students in one study. This additional session followed the format of all previous and subsequent sessions and was integrated into a discussion regarding mutual empathy.

Procedure

As part of an agreement with a local juvenile justice detention facility in a central southern state of the United States, a community mental health agency provided pro bono group counseling services as part of the facility’s enrichment programming. We implemented an A-B single-case research design to assess the effectiveness of an RCT group therapy intervention for improving three domains of relational health (i.e., authenticity, empowerment, and engagement) among adolescent females that were incarcerated. This design was selected to accommodate the limited time frame (12 weeks) to complete the group intervention and was regarded as a reasonable strategy for evaluating the RCT program among individual participants. Participants were educated about group structure, content, process, and rationale during an orientation provided by two masters-level counselors completing their final internship experience through a counseling-oriented fellowship program (second and third authors) and were provided the choice to participate in RCT weekly groups. As an alternative to RCT programming, all participants were provided the choice to use group time to complete school work or have leisure time to assure voluntary participation. After electing to participate in RCT groups, participants chose a pseudonym to be used when completing the RHI protocols and completed an initial RHI to establish a baseline for relational health.

Three baseline measurements of the RHI (A) were completed during a 2-week period prior to starting the RCT groups. During the treatment phase (B), participants met weekly for nine sessions on Friday afternoons for one and a half hours and participated in an RCT group intervention; participants had no contact with the group leaders outside of group time. During RCT group meetings, the group leaders maintained several controls presented by Jordan and Dooley (2001) including group structure and process (i.e., relational check-in, creative moments, clarity in connections, and closing connections/relational impact), consistency of group facilitators, and time. At the conclusion of each group meeting, participants completed the RHI including their pseudonym. Inventories were collected, sealed in an envelope, labeled by date and session number, and stored in a secure location until the conclusion of data collection.

Data Analysis

The effectiveness of an RCT group intervention was evaluated for each participant using the Percentage of Data Exceeding the Median procedure (PEM; Ma, 2006). This procedure was implemented in favor of other nonoverlap methods such as the Percentage of Nonoverlapping Data procedure (Scruggs, Mastropieri, & Casto, 1987) due to the existence of baseline data covering a considerable range of scores without becoming stable in some instances. Additionally, PEM has been demonstrated as a robust measure of treatment effect for within-subjects research designs (e.g., AB), especially when the number of data points in treatment phases is less than 10 (Ma, 2009).

To calculate PEM, the intended change was identified, in this case it was hoped that participant scores on the RHI would increase during the 9 weeks of RCT group involvement. Next, all data points were graphed on a semilogarithmic chart and the median data point in the baseline condition was identified. Following this, a trend line was drawn extending from the median datum point in the baseline phase through the treatment phase. The baseline and treatment measurements within each figure are denoted on the abscissa of each graph as B1 through B3 and T1 through T9, respectively. The number of data points above the median line in the treatment phase were counted and divided by total number of data points in the treatment condition. The resultant effect size coefficient was interpreted using the guidelines presented by Scruggs and Mastropieri (2001) for evaluating treatment effectiveness in single-case research designs. According to these guidelines, treatments that yield an effect size coefficient in the range of .90–1 are regarded as very effective, .70–.89 as effective, .50–.69 as debatable, and scores less than .50 are indicative of ineffective interventions. In addition to interpreting the PEM statistic, each participant’s graphical data were analyzed using visual trend analysis.

Results

Participant 1

Figure 1 presents the subscale ratings for Maegan on the RHI illustrating that an RCT intervention was effective for promoting relational empowerment, but ineffective for promoting relational authenticity and engagement during the 9-week intervention. Evaluation of the PEM statistic for authenticity (.22) indicated that only two scores recorded during the treatment condition were above the baseline median (31) demonstrating treatment ineffectiveness for this domain. A trend analysis of Maegan’s authenticity subscale scores demonstrated that both these data points were recorded shortly after the treatment phase began, with a deep recess during the middle sessions, and an upward trend toward the baseline during the second half of the RCT group. Maegan’s PEM statistic for the empowerment subscale (1) represents all data points in the treatment phase exceeding the baseline median (25) and that RCT was very effective for promoting relational empowerment. Trend analysis of this subscale revealed that all but one score was within 5 points of the baseline median indicating sustained improvement throughout the RCT group sessions. Finally, the PEM statistic for the engagement subscale (.22) demonstrated that 2 points in the treatment phase exceeded the baseline median (25); thus, the RCT group was ineffective for improving relational engagement for Maegan. The trend analysis for this subscale illustrates four data points at the baseline median with the final two observations exceeding the value.

Figure 1.

Ratings for Maegan on the authenticity, empowerment, and engagement subscales.

Participant 2

Figure 2 presents the subscale ratings for Lauren on the RHI illustrating that an RCT intervention was effective for promoting relational empowerment, debatable for promoting engagement, and ineffective for promoting authenticity during the 9-week intervention. Evaluation of the PEM statistic for authenticity (.22) indicated that only two scores recorded during the treatment condition were above the baseline median (29) supporting the conclusion that treatment was ineffective for improving this domain. A trend analysis of Lauren’s authenticity subscale scores demonstrated that data points were decreased as far as 6 points from the baseline median and abruptly increased above the PEM line following the final two group sessions. Lauren’s PEM statistic for the empowerment subscale (.88) represents all but one data points in the treatment phase exceeding the baseline median (31) and that RCT was effective to very effective for promoting relational empowerment. Trend analysis of this subscale revealed that all scores above the PEM line were within a 5-point range indicating a sustained level of improvement throughout the RCT group sessions. Finally, the PEM statistic for the engagement subscale (.55) resulted from 5 points in the treatment phase exceeding the baseline median (32) indicating a debatable effect of the RCT group for improving relational engagement for Lauren. The trend analysis for this subscale illustrates an 11-point range separating the lowest and highest scores for this subscale with the final 2 points notably above the PEM line connoting considerable change following the RCT intervention.

Figure 2.

Ratings for Lauren on the authenticity, empowerment, and engagement subscales.

Participant 3

Figure 3 presents the subscale ratings for Crystal on the RHI illustrating that an RCT intervention was ineffective for promoting relational authenticity, empowerment, or engagement during the 9-week intervention. Evaluation of the PEM statistic for authenticity (.22) indicated that only two scores recorded during the treatment condition were above the baseline median (32) indicating that treatment was ineffective for improving this domain. A trend analysis of Crystal’s authenticity subscale scores demonstrated that data points were predominately decreased by 6 points and as much as 13 points from the baseline median with no points above the median following the fourth session. Crystal’s PEM statistic for the empowerment subscale (.22) represents all but only two data points in the treatment phase exceeding the baseline median (34) and that RCT was ineffective for promoting relational empowerment. Trend analysis of this subscale revealed that the scores above the PEM line were marginal elevations, whereas decreases in this domain deviated from the baseline median by as much as 8 points. Finally, the PEM statistic for the engagement subscale (.11) resulted from one elevated point in the treatment phase that exceeding the baseline median (28) indicating that participation in the RCT group was ineffective for improving relational engagement for Crystal. The trend analysis for this subscale illustrates only moderate variability in the data with the majority of data points residing within a 6-point range under the PEM line indicating little change in this domain associated with the RCT intervention.

Figure 3.

Ratings for Crystal on the authenticity, empowerment, and engagement subscales.

Participant 4

Figure 4 presents the subscale ratings for Samantha on the RHI illustrating that an RCT intervention was ineffective for promoting relational authenticity and empowerment, but effective for promoting engagement during the 9-week intervention. Evaluation of the PEM statistic for authenticity (.42) indicated that four scores recorded during the treatment condition were above the baseline median (32) supporting the conclusion that treatment was ineffective for improving this domain. A trend analysis of Samantha’s authenticity subscale scores demonstrated that data points steadily increased from below the baseline median to above it throughout participation in the group. Samantha’s PEM statistic for the empowerment subscale (.22) reflects two data points in the treatment phase exceeding the baseline median (43) and that RCT was ineffective for promoting relational empowerment. Trend analysis of this subscale illustrates that all scores were grouped very close to the PEM line and did not exceed a 6-point range indicating very little magnitude of change during the treatment phase. Finally, the PEM statistic for the engagement subscale (.77) resulted from 7 points in the treatment phase exceeding the baseline median (37) indicating effectiveness of the RCT group for improving relational engagement for Samantha. The trend analysis for this subscale illustrates a progressive increase is subscale scores above the baseline median within a 6-point range.

Figure 4.

Ratings for Samantha on the authenticity, empowerment, and engagement subscales.

Discussion

The results of this study found that three of the four participants reported a noteworthy amount of change in at least one domain of relational health as measured by the RHI. Maegan’s level of relational empowerment increased during the treatment phase; similar results were noted for Lauren as indicated by consistent increase of scores on the empowerment subscale. In both cases, an RCT group intervention was within the range of effective to very effective for increasing participant feelings of being personally strengthened and inspired to take action within relationships. This change is especially poignant when considering the supposition by Covington (2007) that correctional facility milieu commonly represses the development of social and relational skills needed for growth fostering relationships. It is reasonable to surmise that if counselors are able to inculcate adolescent females to desire meaningful engagement with others while detained, the trends of social interaction may be positively influenced when released. Liang, Tracy, Taylor, Williams, and Jordan (2002) suggested that developing the energy for involvement with others was requisite for creating a sense of belonging with peers, mentors, and the community at-large.

While these two participants’ ratings on the empowerment subscale improved, two participants reported scores on the RHI engagement subscale that indicate that an RCT group demonstrated debatable to effective results during treatment. For Samantha, the PEM scores indicate that her scores maintained sustained improvement over time. For Lauren, increases in activities associated with meaningful engagement behaviors with others were intermittently reported for several weeks at a time without consistent or sustained development. Liang et al. (2002) suggested that when individuals are mutually engaged with others, associated closeness and empathy with others may mediate self-esteem, self-actualization, and satisfactions with relationships. This increase in relations with others based in mutuality is a hopeful indication of the effectiveness of counselors within the correctional setting purported by Covington (1998, 2007) as characterized by disproportionate power hierarchies, ubiquitous control, and compliance with subordination.

Finally, our results did not yield any treatment effect in perceived authenticity among any of the participants. This finding may be attributed to the participants’ developmental experiences with important others that have resulted in the reported acts of drug abuse, running away from home and family violence. It is possible that although some participants reported being inspired to move toward (empowerment) and take meaningful action within relationships (engagement), interpersonal histories may have inhibited willingness to be freely genuine and open with others. Conversely, while working within the system as mental health professionals, we frequently questioned how effective this RCT intervention would be based upon the supposition that the interpersonal environment of a correctional does not promote honest, authentic expression of feelings, thoughts, and values; instead, premium behaviors include nondisclosure, compliance with authority, and uniform values. This observation is commensurate with Covington’s (1998, 2007) description of correctional facilities as being interpersonally austere and disparaging environments for meaningful, positive self-expression. It may be possible that within other contexts, through long-term interventions, or in the absence of a correctional staff member that a safe disposition regarding self-disclosure and relational exploration with others can be promoted.

Limitations

Several caveats are noted regarding the results of our investigation related to design, participant selection, and data analysis. Foremost, although the AB approach to single-case research designs is an accepted evaluation method, time constraints to complete therapeutic programming did not provide opportunity to implement an ABA design that would have allowed provided stronger internal validity for the RCT intervention and reliable determinations of lasting change. Next, our number of baseline measurements (N = 3) is regarded as sufficient when implementing a single-case research design, however, more data collections may have allowed for the data to stabilize prior to the RCT intervention. Furthermore, our results would have been regarded as more robust if participants had been randomly assigned to the RCT group and another group completed measures while receiving alternative treatment. As noted, by Parker and Hagan-Burke (2007), despite their usefulness, effect sizes reflect amount of change, not cause of change; the latter is contingent upon strong research designs with considerable internal validity. Future studies implementing an alternative treatment group will also be capable of applying more rigorous data analysis procedures such as the Success Rate Difference, Relative Success Rate, and Relative Success Rate Improvement procedures (see Parker and Hagan-Burke) used in medical trials.

Conclusions

There has been considerable discussion concerning the differential experiences of female offenders within correctional institutions. Most prominently, the rehabilitative structure of detainment facilities is generally characterized by a system that promotes subordination, individualism, and isolation as a means of thriving. Covington (1998, 2007) has suggested that this system is not conducive to promoting the social and relational rehabilitation that may be beneficial for females. We have provided some initial support for the use of RCT groups for promoting relational development empowerment and engagement among incarcerated adolescent females; similar findings were not noted for relational authenticity. Given the interpersonal austerity of many correctional systems, we regard RCT as a promising intervention for creating positive relational movement between adolescent females; however, it is undeniable that this burgeoning approach to counseling needs to be investigated further. We are hopeful that scientist–practitioners will evaluate similar practices in other correctional facilities to ascertain the relevancy for this approach and substantiate RCT as an evidence-supported treatment.

Footnotes

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Altschuler

. (1998). Intermediate sanctions and community treatment for serious and violent juvenile offenders. In Loeber

Farrington

(Eds.), Serious and violent juvenile offenders (pp. 367–388). Thousand Oaks, CA: Sage.

Comstock

Duffey

St. George

. (2002). The relational-cultural model: A framework for group process. Journal for Specialists in Group Work, 27, 254–272.

Corneau

Lanctôt

(2004). Mental health outcomes of adjudicated males and females: The aftermath of juvenile delinquency and problem behavior. Criminal Behaviour and Mental Health, 14, 251–262.

Covington

(1998). Women in prison: Approaches in the treatment of our most invisible population. Women and Therapy Journal, 21, 141–155.

Covington

(2007). The relational theory of women’s psychological development: Implications for the criminal justice system. In Zaplin

(Ed.), Female offenders: Critical perspectives and effective interventions (2nd ed.). Gaithersburg, MD: Aspen Publishers.

Frey

L. L.

Beesley

Miller

(2006). Relational health, attachment, and psychological distress in college women and men. Psychology of Women, 30, 303–311.

Frey

L. L.

Beesley

Newman

J. L.

(2005). The relational health indices: A reanalysis of a measure of relational quality. Measurement and Evaluation in Counseling and Development, 38, 153–163.

Gagnon

J. C.

Barber

(2010). Characteristics of and services provided to youth in secure Acare facilities. Behavioral Disorders, 36, 7–19.

Glaser

B. A.

Calhoun

G. B.

Bates

J. M.

Bradshaw

C. P.

(2003). Self-reported career interests among adjudicated male adolescents: A pilot study. Journal of Additions and Offender Counseling, 23, 73–82.

10.

Holman

Ziedenberg

(2006). The dangers of detention: The impact of incarcerating youth in detention and other secure facilities. Washington, DC: Justice Policy Institute.

11.

Jordan

J. V.

(2010). Relational-cultural therapy. Washington, DC: American Psychological Association.

12.

Jordan

Dooley

(2001). Relational practice in action: A group manual. Wellesley, MA: Stone Center Publications.

13.

LaBrie

J. W.

Thompson

A. D.

Ferraiolo

Garcia

Huchting

Shelesky

(2008). The differential impact of relational health on alcohol consumption and consequences in first year college women. Addictive Behaviors, 33, 266–278.

14.

Lenz

A. S.

Sangganjanavanich

V. F.

(in press). Evidence for the utility of a Photovoice task as an empathic skill acquisition strategy among counselors in training. Journal of Humanistic Counseling.

15.

Liang

West

(2011). Relational health, alexithymia, and psychological distress in college women: Testing a mediator model. Journal of Orthopsychiatry, 81, 246–254. doi:10.1111/j.1939-0025.2011.01093.x

16.

Liang

Tracy

A. J.

Kenny

M. E.

Brogan

Gatha

(2010). The relational health indices for youth: An examination of the reliability and validity aspects. Measurement and Evaluation in Counseling and Development, 42, 255–274. doi:10.1177/0748175609354596

17.

Lipsey

Wilson

Cothern

(1998). Effective intervention for serious juvenile offenders. In Loeber

Farrington

(Eds.), Serious and violent juvenile offenders. Thousand Oaks, CA: Sage.

18.

H. H.

(2006). An alternative method for quantitative synthesis of single-subject research: Percentage of data points exceeding the median. Behavior Modification, 30, 598–617.

19.

H. H.

(2009). The effectiveness of intervention on the behavior of individuals with autism: A meta-analysis using percentage of data points exceeding the median of baseline phase (PEM). Behavior Modification, 33, 339–359. doi:10.1177/0145445509333173

20.

Miller

J. B.

(1976). Toward a new psychology of women. Boston, MA: Beacon Press.

21.

Miller

J. B.

(1986). What do we mean by relationships? Work in Progress No. 22. Wellesley, MA: Stone Center, Working Paper Series.

22.

Miller

J. B.

(1990). Connections, disconnections, and violations. Work in Progress No. 33. Wellesley, MA: Stone Center, Working Paper Series.

23.

National Mental Health Association. (2004). Mental health treatment for youth in the juvenile justice system: A compendium of promising practices. Washington, DC: NMHA.

24.

Parker

R. I.

Hagan-Burke

(2007). Useful effect size interpretations for single case research. Behavior Therapy, 38, 95–105.

25.

Scruggs

T. E.

Mastropieri

M. A.

(2001). How to summarize single-participant research: Ideas and applications. Exceptionality, 9, 227–244. doi:10.1207/S15327035EX0904_5

26.

Scruggs

T. E.

Mastropieri

M. A.

Castro

(1987). The quantitative synthesis of single-subject research: Methodology and validation. Remedial and Special Education, 8, 24–33.

27.

Snyder

(2004). Juvenile arrests 2002. Washington, DC: U.S. Department of Justice, Office of Juvenile Justice and Delinquency Prevention.

28.

Snyder

Sickmund

(2006). Juvenile offenders and victims: 2006 national report. Washington, DC: U.S. Department of Justice, Office of Juvenile Justice and Delinquency Prevention.