Girls fighting trouble: Re-storying young lives

Kelley, Patricia

PRACTICE: JUSTICE AND JUVENILES

Abstract

This study assessed the usefulness of narrative group therapy as part of a summer program for female adolescents who had broken the law, and it sought to gain greater understanding of this population. Mainly qualitative methods were used because there is little research on narrative approaches and an exploratory study was most appropriate. Other methods used, include interviews, pretesting and posttesting, and reviews of court records 6 months after beginning. Various means of assessment showed that narrative approaches were useful in helping these young women examine their lives, cope with problems, explore options, and assess possibilities. While several problem themes were isolated, solution and coping themes were also found, which were brought forth and highlighted in the sessions. These adolescents experienced the most improvement in the area of coping. Selfesteem and control were found to be important coping devices and were strong throughout.

STUDIES AIMED AT UNDERSTANDING adolescents who have broken the law and at assessing ways to work with adolescents are important. These youth commit higher rates of most criminal acts than any other age group and they consume many of the resources of the youth services system (Henggeler, Melton, & Smith, 1992). Furthermore, while adults commit most violent crime, the growth rate is in the juvenile population (Snyder, Sickmund, & Poe-Yamagata, 1996). Studying ways to work with the younger adolescents is especially important for prevention reasons, and because studies have found that younger children are increasingly committing violent crimes (Snyder et al., 1996). Not only are youth committing crimes at younger ages, they are also more likely to be female (Snyder, 2000; Stahl, 1998). Violent crime committed by young females doubled between 1985 and 1994 (Snyder et al., 1996) and the number of delinquency cases involving females increased by 76% between 1987 and 1996 (Stahl, 1999).

This study was designed to increase understanding of young females who had broken the law and to assess the usefulness of narrative approaches in working with them. The newer narrative approaches (Freedman & Combs, 1996; Kelley, 1996; Neimeyer, 1999; White & Epston, 1990) are more collaborative, which decreases resistance from adolescents that may lead to an impasse (Biever, McKenzie, Wales-North, & Gonzalez, 1995; Morgan, 1999; Smith & Nylund, 1997). These approaches are especially appropriate for work with young women who frequently have had narrow and restrictive stories about their lives and the identities created for them (Penn & Frankfurt, 1994). Adolescent lives are not completely storied (Zimmerman & Dickerson, 1994). Narrative therapy helps youth identify, challenge, and broaden the stories around which they have organized their lives.

Therapists have found narrative therapy to be useful and it has been touted as an important new tool for working with adolescents and for gaining a better understanding of them as individuals. (Attanucci, 1993; Biever et al, 1995; Burnette, 1995, Holcomb, 1994; Morgan, 1999; Smith and Nylund, 1997; Zimmerman and Dickerson, 1994). But naarative therapy has not been systematically studied for its effectiveness. Since research in this area is sparse, smaller, exploratory studies may help us more fully understand this population and assess these newer approaches; approaches that are already embraced by many youth workers.

Theoretical Considerations

Narrative therapy, as used in this study, was developed by White and Epston of Australia and New Zealand respectively (White & Epston, 1990). It falls under the rubric of postmodern theory and focuses on meanings people attribute to life events and to the stories around which their lives are organized. People's views about themselves are storied and co-constructed in interaction with other people and with societal institutions, and they often involve internalized views of gender, ethnicity, and power. Similarities to the deconstructivist movement in literary criticism can be seen, where stories are deconstructed for meaning; and to social constructionism in psychology, where reality is viewed as constructed in the minds of the individual in interaction with others and society. In narrative therapy, clients' views are carefully listened to, and clients are invited to examine their problem stories to assess other truths in their lives, which may involve strength and coping. Problems are externalized, that is, viewed as not intrinsic to the person, so client and therapist work together to fight the effects of the problems. In this collaborative approach, clients are viewed as experts in their own lives and are invited to explore other views of reality, which may also be true. This approach is consistent with the principles of empowerment theory (Lee, 1996) and with some aspects of strengths-based approaches (Saleebey, 1997).

Methodology

Research Questions and Methods of Assessment

This was an exploratory study, which was appropriate given the nature of the questions to be addressed, the exploratory nature of narrative work itself, and the lack of previous research on narrative therapy. Qualitative methodology, which focuses on understanding a phenomena more than explaining it (Charmaz, 1990), was used here as the chief method since it fits with the aims of narrative therapy and with our questions. Other methods were used to gain additional perspectives.

Two broad questions guided the study. The first question has two parts: "What are the ideas and experiences of these young women who have broken the law, and what is the relationship between their lived experiences and the stories that they develop about themselves?" This question was addressed through qualitative analysis of transcripts of the audiotaped sessions (daily 1-hour sessions over 8 weeks). Themes were isolated by the three investigators by reading the transcripts independently, agreeing upon themes through discussion, independently coding the transcripts by themes, and meeting again to agree on the coding and hearing each others' perspectives. The data were then entered into Hyperesearch, a qualitative software program that pulled out the material coded by the researchers, essentially making "index cards" of the material under each theme by date. The material was then broken into subthemes and assessed for differences in earlier and later transcripts. The participants' creative artwork was also assessed by the three investigators through discussion of the artwork in context of the themes.

The second question was: "Are narrative approaches, where participants are challenged to assess stories around which they have organized their lives, useful in helping them identify and mobilize strengths they have to fight the effects of the problems they face?" This question was also addressed by qualitative analysis, assessing transcripts for changes over time. Other methods were also used to address this question, including (a) interviews with the participants before and after the program by professor Kelley, the principal investigator, who was not involved in the treatment, (b) therapists' review of court records for arrests 6 months after the program ended, and (c) administration of pretests and posttests using existing instruments. While the number of participants (eight) was appropriate for qualitative study, it was small for quantitative study. When assessing approaches and when reporting to a court and an agency, however, quantitative testing is useful as a secondary method of analysis because progress can be more clearly shown through numbers and charts, and the two methods can complement and explain each other. Three standardized instruments were chosen in areas found to be important in adolescents: self-esteem (Harter, 1990; Rosenberg, 1989), locus of control (Harter, 1990; Steinberg, 1990), and coping, which was isolated as a major theme in the pilot group. Since a goal of narrative therapy is to broaden life stories and views of self and to empower people through choice, we believed that increasing self-esteem, coping, and control were reasonable hoped-for outcomes.

The standardized instruments were the Rosenberg SelfEsteem Scale (1989), which was designed to assess selfesteem in adolescents; the Norwicki-Strickland Locus of Control Scale (N-SLCS) (1973), which was designed to measure whether a youth believes that reinforcement comes by chance or fate or through one's own behavior; and the Coping Response Inventory-Youth Form (CRI-Youth) (Moos, 1993), which aims to measure different types of coping skills used in stressful situations. Two nonstandardized tools were also used before and after the group experience: the Problem Oriented Instruments for Teens (POSIT) (Rahdert, 1991), which suggests which domains might be problematic for youth, and the Automatic Thoughts Questionnaire (Clear Thinking Workshop, 1995), a series of open-ended questions assessing how persons interpret events. This tool was chosen because it had been used in the previous years, and we believed that a comparison might be helpful. For this small number, each individual was assessed as a single subject. We also assessed group differences, pre- and post-group, using a nonparametric test, the Wilcoxon, noted for its effectiveness with small sample sizes (Pett, 1997).

The Setting and Process

Because this study took place in a small university town in a rural area, special care is taken in referring to the agency and the participants. The County Juvenile Court was interested in developing programs for girls, since most of the programs designed for mixed groups were found to be more useful for boys. The court asked local agencies to develop programs specifically for girls. The agency discussed here is a facility with programs for a wide range of youth who have varying interests. The agency's mission is to develop youth through engagement in community work, through fostering creativity in the arts, and through showing unconditional positive regard to all youth. In 1997, this agency received referrals from the court to develop a cognitive-based summer program for young females who had broken the law. Agreement by the girls to join this program served as a diversion from arrest or a more serious sentence. This first group, in 1997, was based on a cognitive educational program with specified sessions, the Clear Thinking Workshops, which used a "pre" and "post" questionnaire, Automatic Thoughts (Clear Thinking, 1995) to assess progress. The two agency therapists assigned to this project found that five of the eight referred girls attended the sessions and improved on the questionnaire in that their answers were judged to demonstrate increased ability to think before acting, but there was not a lot of involvement or discussion of themselves by the participants.

In 1998, the therapists wished to change the focus of the group experience, and one of the authors of this paper, Patricia Kelley, of the University of Iowa, was searching for a place to assess narrative approaches with young people. Professor Kelley had found this approach useful with women facing chronic pain (Kelley & Clifford, 1997) and believed that it might be especially helpful for adolescents who are just developing their stories, especially females. The therapists and the professor agreed to develop this idea. A pilot project was designed in 1998 to incorporate narrative approaches into the existing cognitive approaches used in 1997. Only three girls were referred that summer, but the results of that project set the stage for the research program of 1999, reported here. Both of the therapist coinvestigators had studied narrative therapy with Professor Kelley in graduate school, which ensured some agreement as to what constitutes narrative therapy. The therapy and research aspects of the project were kept as separate but cooperating partnerships.

Population

The population comprised young women who had broken the law and were referred to this group by the juvenile court. Based on the previous 2 years experience (the 1997 cognitive group and the 1998 pilot narrative group), the research project began in 1999 with 10 youths referred to the project, with 8 agreeing to participate in the group and the study. While participating in the research was not mandatory for treatment, all of those who stayed in the program agreed to participate in the research, for which they received a small honorarium. The participants understood that they were to be coinvestigators in learning more about themselves and in helping the investigators understand the beliefs and experiences of young females in trouble with the law in today's society. All female adolescents referred by the court were accepted, since they represented the population we wished to study. The principal investigator explained the research aspect of the program to the participants before the program began, and did not see them again until the post-- program interviews. They all understood that the group discussions would be audiotaped, but they could request that the recorder be turned off or erased for specific portions of the discussion. Videotaping was not used for reasons of confidentiality. Since the participants were all minors, both they and a parent or guardian signed permission forms. Confidentiality and voluntary participation were made explicit and the project was approved by the University of Iowa Protection of Human Subjects Committee.

The research group ranged in age from 13 to 17, with five of eight members representing ethnic diversity (African American, Hispanic, Asian American, and multiracial). Since the youth referred by the court in the 2 previous years were not ethnically diverse, and the therapists had already been selected by the agency and by the principal investigator, ethnic matching of the therapists was not possible. The program was planned based on the 2 preceding years, and the referrals were not known to the agency until the week before the program began. The 1999 participants were older and had more serious offenses, including theft and assault, than those referred in previous years. For reasons of confidentiality, the court mandated that the principal investigator could not have access to the court records or know the individual offenses. The 1999 group also had a wider range of family situations, income levels, and peer groups among its members than did the previous groups.

Interventions

The program was for 3 hours daily, 5 days a week, for 8 weeks. Usually the days began with an hour of group discussion, which was audiotaped for analysis, except for those days when there was a special activity or field trip. Following the discussions was a creative activity of the group's choosing. This was designed to increase self-reflection on the participant's part, and to tell others about themselves.

In the group sessions, participants examined the dominant stories around which they had organized their lives, their personal, social, cultural, and gender stories. They were then challenged to assess other aspects of themselves, often involving strengths and coping skills, to broaden their self-views and see more alternatives. The narrative concepts developed by White and Epston (1990) that guided the group interventions were:

1. Externalization, where they were encouraged to separate themselves from the problems they discussed, so they could assess them more objectively;

2. Mapping the problem's domain, where they discussed the effects of the identified problems on their lives;

3. Unique outcomes, where they searched for those times when they had effectively challenged these problems;

4. Problem-dominated story, where they discussed how the stories around which they had organized their lives did not take into account all of their traits or experiences; and

5. Spreading the news, where others were told about the newly discovered aspects of themselves and the changes they decided to make.

Creative projects after group discussions were used to aid in-group development and to reinforce ideas discussed in the group. Showing their projects to each other was a way of "spreading the news" of change. Throughout the program, the participants were encouraged to challenge their thinking (cognitive) and the meanings (narrative) about events in their lives, and to increase their sense of empowerment to cope with and challenge adversity.

Findings

Research Question I: Understanding

One aim of the study was to increase an awareness and understanding of the beliefs, experiences, and ideas of young females in trouble with the law; a perspective important for social workers involved with youth. Using narrative therapy, these adolescents were invited to explore these areas, thus increasing their self-understanding. At the same time, the process helped further educate the therapists about these matters.

In the pilot group, several themes had been identified, which helped the therapists plan for this group. Themes identified in the pilot group of 1998 included abandonment, accompanied by a lack of trust in adults; lack of stability, with everything seeming transitory-houses, parents, and schools; violence, with all of the three girls being the victims of assault; negativism, especially in the beginning of the group; anomie, as they felt marginalized from the main stream of their peers; and being tough, as a way of coping with adversity they faced. They did not admit weakness,

which deprived them of the opportunity to be reflective. They were "parentified" children, who had taken on child care and housekeeping tasks beyond their years. They had a live-for-the-moment focus, but they believed in the possibility of control, as they talked about having choices, such as quitting smoking, attending Alcoholic Anonymous to quit drinking, choosing when and with whom to have sexual relationships. Power was an important issue for these girls. The dependence-independence struggle faced by most adolescents seemed more important to them. Privileging male relationships over others was noted, even when the relationships were hurtful. Not "getting launched," as they took on adult roles and activities without the preparation, was also noted. For their creative project they chose to make a beautiful dollhouse as a group, with dolls inside that represented themselves in this group setting.

In this group, it was decided not to superimpose the themes from the pilot group, but to use the same process to identify themes. Because of the larger size and more negative and defensive discussions, the process was more complex. Within each theme several subthemes were identified, often with two sides of the same theme representing more and less socially acceptable responses. For example, with "control," there were attempts to control and manipulate others and attempts to control themselves and their own behaviors, demonstrating different sides of the same theme. The themes identified in this group are noted here.

Anger. A great deal of anger and justification of anger was found in the transcripts and the subthemes included overt hostility, aggression, sarcasm, and frustration. Examples of overt expression of anger and frustration included, "It makes me feel better to complain, if I don't I'm gonna get frustrated and snap." A different girl on a different day said, "I don't care who you are, if you get on my bad side I will snap at you, and I am like that most of the time." In another matter, one participant yelled, "I don't give a damn if you mind-I ain't talking about s- [meaning anything], so don't even ask me to talk or reflect. OK?" There were also sarcastic themes throughout, where the leader would ask a question and get a nonresponse or a joke. For example, when asked where she was born, one member responded, "I don't know," and when asked when she was born said "Sometime in the 80s."

Pride and identity. This theme had two sides to it, with pride manifesting itself in identity and self-esteem in some cases, but in excessive bragging at other times. The girls identified with their peer groups and their cultural groups, but also wanted to maintain individuality. One member said, "I get sick of this whole group process thing; I think we need to be individuals, too." In an example of positive pride one girl noted, "I think I'm unique. My boss says I'm energetic and animated, whatever that means, and my friends say I am honest and funny." This girl had a problem with self-identity, however, as she went on to say, "But, what I think of myself comes from what others say-I don't know about myself unless it's somebody else's opinion." The author-therapists noted that while the girls took a certain pride in their identity and their actions, at the same time they questioned these things.

Control and coping. This was a major theme. The participants tried to control others as a coping device, but also aimed to take control of themselves. Often they found that their control tactics got them what they wanted. One said, "I complain to the people at the gas station to sell me cigarettes and they do." Another said, "I took the master bedroom when I complained and complained. See, it works." They tried to control the sessions as in, "OK, you can tape now." This controlling others can be viewed as a less positive and more manipulative side of control. They also talked with pride about their self-control, which was considered to be the more positive side of this theme, such as "I haven't smoked weed for 8 months." Another said, "I learned to think about things first," and another said, "I am more even tempered with my parents now." One of the more interesting comments about control was one member saying she liked being in a detention shelter because she "liked structure."

Coping with the many problems they faced was a major theme of these participants and was also a major focus of the study. An example of coping came in a discussion about death, when one said, "Like if we let everybody that died get to us then we ..." and a second member broke in and said, "might as well be dead ourselves," and the first member said, "Yeah." A third member went on to say, "You can't think about it all the time because you'll be feeling bad for yourself," and the first girl again said, "You'll drive yourself crazy."

Loss. Loss was another theme, which was isolated, and there were four subthemes: concrete loss, interpersonal loss, intrapersonal loss, and resolution of loss. The girls spoke of losing friends through moving, and family and friends through death, which were both concrete and interpersonal losses. One member said, "Like when people in my family die-you know they wasn't meant to die-they just got killed." When asked who died, she named several people. Another member who had grown up in a large city said, "I grew up with like nine girls [peer group], three of them is still living. Two of them got killed trying to give themselves abortions." A stark example of intrapersonal loss was when one participant noted by the facilitators to have a weak sense of self said, "I see myself as a plain person-there are empty people and I am just one of them." Resolution of loss was noted as in the statement, "Yeah, it was like about how we were going to change something instead of what's already been done. Our past lives or whatever-it's like moving towards." Another member told of getting into a lot of trouble in high school and missing out on a lot as she did drugs with her peers, and she added, "Once I got out of the shelter, I just stayed clean."

Chaos and violence. The girls had all experienced chaos and violence in their worlds, and it was an identified theme. They discussed the stress and chaos in their lives and also discussed being both the victims and perpetrators of violence. One member said, "I think that's why my dad's in a nut ward prison [laughter]-that is a mental hospital. That's where I am going to be next; I'm going crazy right now just sitting here." Another talked of moving often. When asked how many times, she said, "I couldn't say 'cause it's a lot." When asked, "More than twelve?" she said "Uh huh." When asked to talk about their world, one said, "How do you spell chaos?" Another said,

Teens are faced with lots of crazy decisions every day. We were raised with TV and lots of violence that's not supposed to be accepted but it is on TV We have lots of choices about drugs, and the multimedia all around us forces us to starve ourselves to be an accepted weight. Oh and to change our features to look good. I heard about teenagers getting breast implants over spring break so they can look good at prom-everybody wants to look perfect. Then everybody is on some kind of drug-for asthma or depression or something. People are worried about getting AIDS or getting pregnant, people are unfaithful. It's hard to find the right one. We are in single-parent families and have had to deal with death and divorce, and we are confused about how to raise our own kids. A lot of kids try to act like adults. Aa lot of us are really stressed-out.

These adolescents have been victims of violence, either directly or through exposure to it. One said, "In grade school I used to get beat up all the time; it just sucked there. There was one guy shot outside of our apartment; he just laid there." Another girl said,

There was this guy who somebody just rode up and shot him in the face; his face blew all over my uncle's face and my uncle went crazy. I won't even go near him. My mom's sister got raped and killed in Chicago. My grandpa did not get killed-he died of cancer, but my daddy got killed when I was somebody robbed him and shot him in the head five times. Confusion and choice. This was another theme isolated from the text. It seems connected to the chaotic and violent world theme. One participant noted, "I confuse all those around me because I never know what I'm feeling. I especially confuse myself because I can never tell if I want to be alone with people or whatever." Others added that there is a lot of confusion but they try to cover it up. After a discussion about choices, when asked if there are too many choices, one girl said, "Uh huh, all the time."

Support from others. This theme is another with different aspects. When one member talked about running away and breaking probation, another tried to talk her out of it with the following statements in response to comments inbetween.

No, you're not.... That's really stupid.... You'll have to re-do probation.... Don't even think about it.. They'll put you in a lock-up place where you can't even walk out the [expletive deleted] door Sorry for my language but ...

Most of these girls received support from others which sustained them, and was probably what helped them to function as well as they did. One said, "My grandma fixes me breakfast every morning." Another said, "My mom is my mom and dad. She takes care of me." Several in the group spoke of liking school because of the friends they had. On the other hand, they sometimes gave or received support that was not helpful, as in, "Oh, I had really cool roommates [in the shelter]. That's how I got into trouble in the first place."

Connection to family. This theme related to support from others. All of the positive connections to family references were noted in the last 2 weeks of the 8-week program. One member reflected,

I think that's what got me through probation, having something I want for the future. It's really supportive if your parents tell you they have been through the same thing and they survived through it and that you will too.

Another said, "I respect my mother for making that decision [to give birth to a child as a young person and place the child for adoption] when she was young, but my brother found her when he was 16."

One group member seemed proud to say that she is allowed to visit her mother. Another member of this group said, as she was doing artwork, "Pink is for my sister because she is friendly and stuff. Brown is for my dad because I'm mad at him right now, but my mom and I are getting along better." Research suggests the importance of emotional bonds to parents continuing throughout adolescence (Hauser & Bowlds, 1990) and of developing a sense of autonomy within the context of continuing family relationships (Steinberg, 1990). Our findings here support that premise, in their emphasis on the importance of family.

Trust. Trust, and lack of it, was a major theme for this group. While the development of trust is a major task in any group process, it was especially important in this group because they have had trouble trusting anybody. In the beginning there was mainly lack of trust as in, "Why do you record these conversations?" and "I think a lot of people trust others more than you [therapists]," and "When you ask me to share with the group it makes me mad," and "I've been in groups before but they never taped before." Later in the session they talked about developing trust as in, "I have never really trusted adults-I have never had people there for me consistently, so I thought that [what her parents did] was really cool-I trust them now and I respect them a whole lot, too." Another said, "We were ready to start another path and start trusting people, 'cause him and I just got out of treatment and were on probation and we were happy to find somebody to trust."

Privileged world. We found that group members had created a privileged world for themselves in which they excluded adults and other peers, used their own language and jokes, and did not like others using their language. They also shared antics and behaviors that got them into trouble in some bragging ways, especially at first. One said that outside the group, "I'm a lot livelier-I'm giddy all the time." Another said, "I could bring up a lot of stuff but I'm not going to-not like for grown-up people." They discussed their music and liking clubs in other places, not in the current "boring" town and state. When an adult used a word that they thought was theirs, one said, "Hold it right there-don't use my words-those are kids' words and you're not a kid, it don't sound right-don't try it." This theme tied in with the trust theme and as trust increased, they let others into their world more, including parents, teachers, and the group therapists.

Hope. The last theme that was isolated was hope, and it only showed up near the end of the sessions. One group member made up a story to go with a painting and said, "Oh, at first he was raised by idiots and jerks, but he was adopted when he was young. That was good, he had a permanent place to grow up." Another said in regard to getting off drugs, "My parents know me a lot better now. It's like I'm alive and I have a lot better relationships than ever before. I think I used drugs because I was lonely and scared to let anybody close to me." Another asked about social work as a job and said, "It seems like it would be a cool job if you were interested in helping people." This look toward the future and to the job world seemed hopeful. Another said, "You have to believe in having something at the end of whatever it is you're going through."

In this group there were some themes that had also been noted in the pilot narrative group of 1998, which is interesting since the groups were quite different. In both groups the members had moved often and had many difficult transitions. In the pilot group the members had moved mainly around the local town and surrounding countryside, whereas in this group most had moved from state to state. It is not known if the different kinds of moves had different kinds of effects, but for both groups the frequent moves had been disruptive.

Both groups had members who had experienced violence and rejection. In the pilot group it was often within the family and peer group (one had been abandoned by her mother), whereas in this group it was also social and cultural violence.

Both groups felt rejected and marginalized from the larger peer group, and in this group, members also felt cut off from cultural customs. For example, "In Chicago we had block parties and barbecues, and it was fun and I didn't waste time stealing cars and stuff, but here it is so boring. It is sad they won't let us barbecue in our apartment complex, and Black people like to barbecue."

The girls in both groups had stories of loss, chaos, and abuse, but they also had stories showing strength, coping, and caring. These subjects were highlighted in the discussions, helping them see themselves in broader terms that were less restrictive and offered hope.

Research Question II: How useful is the narrative approach with this population?

Qualitative analysis was also used in assessing the transcribed text for changes over time. The themes were pulled out of the transcripts and dated by the computer program, Hyperesearch, which helped to isolate changes. Under the theme chaotic and violent world theme, for example, the talk of violence was all in the early sessions, when the girls were more likely to relate direct memories and events, whereas in later sessions they spent more time reflecting on the meaning and effects of the events. Chaos was a theme throughout the transcripts. Changes were noted in the control and coping theme. In early sessions there was emphasis on being tough, and using anger to control others; during the middle sessions they discussed ways to cope with difficulties; and in later sessions they placed more emphasis on compromise and change, and on discussing past ways of coping. The pattern of developing compromise and accommodation does fit the pattern of group process in general, but this change seemed more than a group process phenomenon in that it generalized to outside behaviors, as noted in discussions as well as the post-program interviews with Professor Kelley. All instances under the connection to family theme were in later sessions. Under the trust theme, we noted lack of trust more in early sessions with more trust evident in later sessions, including examples from outside the group context, too. As for the hope theme, all expressions of hope were in later sessions, too. Many of these themes, such as the trust and hope themes, and changes are reported in group process literature (Garvin, 1997), but the fact that the positive changes over time were carried outside the group suggested that the narrative group approach had been useful for these adolescents.

Individual interviews with group members at the end of the summer by the principal investigator, who had not been involved in the treatment, was another method of assessing the usefulness of this approach. In addition to asking about personal change, the participants were asked for their ideas about the group program, and suggestions for it. They were considered co-investigators in the study; a participatory approach consistent with qualitative research (Tutty, Rothery, & Grinnell, 1996) as well as narrative therapy. In addition, Professor Kelley, as the principal investigator, summarized her impressions from the transcripts and asked for the participants' views of the accuracy of the summaries, something often important in qualitative research (Lincoln & Guba, 1985).

Even though the participants were assured that their individual comments would not be shared with the group leaders, they all responded positively about the group experience and about the leaders to whom they had complained. One suggestion was for fewer group hours since many had jobs or summer school. Some suggested more structure and control in the group, while others wanted less structure where they could talk about whatever they wanted. They all liked the field trips and art projects. Two members said they were too sophisticated and had been in too many groups; that this project would serve younger girls better, which is interesting because the project was designed for a slightly younger population.

They all named positive personal outcomes. One said that she had learned to stop taking her frustrations out on other people, especially her sister, and that she now handles her own problems better. She also noted it was helpful getting other people's ideas. Another said she spends more time with her parents now, and also more time alone. She used to be "obsessed" with her friends. Now they are just one part of her life. She and her parents are happy with this outcome. Another said it was a new and positive experience for her to have others listen to her--especially adults. This is an especially important statement, since at the beginning she complained about having to talk to adults. She added, "From now on, I will make decisions based on what is good for me." She noted that group members stopped trying to prove they were "badasses" in the group and just became people. Another member said she learned responsibility, noting that just "getting here every day and on time and following orders" was good for her. She added that she learned to listen to and respect people different from herself, saying, "I am grunge and heavy metal and they are rap." She added that she learned about herself, too, including that she is a good communicator. One member said she learned to analyze and solve problems better. She said the group helped her develop creativity and helped her see that she is a normal person, not just a "big problem." Another believed she was too old for the group, but thinks that others learned a lot. She suggested more structure since she felt some of girls got a bit out of hand. Another said she learned to pay attention to others, something she had never been good at before. She said it carried over to her home life and was getting along better with her sister as a result. Another girl said that she liked talking to and listening to the other girls; that she learned from them. She liked adventure activities and art projects best.

Different from the pilot project members, who liked telling their stories and talking about their lives, these group members found it intrusive and didn't want others knowing about them. They expressed anger more openly and were more hesitant in group discussions, but really opened up after the art projects began. They enjoyed telling their stories when they were connected with art projects of their choosing. They chose individual projects, which they shared with the group and discussed. This was different from the pilot project members who chose a group project. The larger size, older age, more group experience, and the diversity of this group are possible causes for these differences.

Pretests and posttests were used to gain additional perspectives and for reporting to the agencies and courts. It is interesting to note that the members of this group showed comparable or greater improvement on standardized tests than those in the pilot project, even though there was less group time and more resistance to the experience.

The authors believe that narrative therapy might increase self-esteem, since it helps people to find new aspects of themselves, often involving strengths. Because these adolescents had been marginalized and rejected by many adults and peers, the authors thought that self-esteem might be a problem for the girls. They also note that it was interesting to find that the participants showed high self-esteem at the beginning and at the end. On the Rosenberg Self-Esteem Inventory, there are 40 possible points on the 10-item, 4-- point scale, with higher scores indicating higher self-esteem. This test was designed for adolescents (Frank-Stromborg, 1988; Rosenberg, 1989). Generally, scores of 1 or 2 indicate lower self-esteem and scores of 3 or 4 to represent higher self-esteem; thus a total score over 30 indicates good self-esteem. These adolescents had a mean score of 31.4 at pretest, and 32.4 at posttest, which was not a significant change. Five participants' scores went up and three went down, but none showed very low self-esteem before or after the project (see Figure 1). This finding was also true in the pilot project the previous year. The qualitative analysis of the data suggested some rationale for this. At the beginning of both groups, the girls showed bragging and bravado, which could indicate false pride and inflated self-esteem, and by the end they were more reflective and realistic, suggesting a more accurate appraisal of themselves.

The Norwicki-Strickland Locus of Control Scale (NSLCS) (1973), was developed in a series of studies involving 1,000 males and females from 3rd through 12th grades representing all socioeconomic levels. Higher scores represent greater external locus of control, so lowering scores represents improvement. Norms established means ranging from 11.01 to 18.80, with internal locus of control developing with age. In our group, the mean scores went from 14.9 at pretest to 13.7 at posttest, suggesting some improvement, but the change was not significant (see Figure 2). The two members who showed the greatest improvement had the highest initial scores-the only two who were outside the normal range at the beginning. We had thought that locus of control might be more external for these adolescents because so many things that were out of their control had happened to them, but this did not appear to be true for most of them.

Significant differences in coping were noted between pretests and posttests on the Coping Response Inventory (CRI-Youth Form) (Moos, 1993). This measure was added for this group since qualitative data in the pilot group suggested coping was a key variable. This inventory has 8 scales with four point items measuring different types of coping; four scales measure approach-coping and four scales measure avoidance-coping. Higher scores represent increased coping. In the four approach-coping scales, measuring adaptive coping mechanisms, the mean scores improved on all scales and significantly in two scales (see Figure 3). With a small number, the Wilcoxon test shows significance only when all participants improve, which they did on the Logical Analysis and Positive Reappraisal scales. On the Logical Analysis scale, the normed mean from field tests is 9, and the means for these participants went from 5.75 to 8.75, which was significant at the .042 level. For the Positive Reappraisal scale, the normed mean is 8 and our participants went from 4 to 10, which was significant at the .043 level. On the Seeking Guidance and Problem Solving scales, there was overall improvement of the means scores, but the differences were not significant on the Wilcoxon. The participants also increased on the four avoidance coping scales. While avoidance coping is usually seen as less adaptive than approach coping, such coping mechanisms might be useful for these adolescents. For example, the one scale that showed a significant change, Acceptance and Resignation, might be a necessity to cope with the many real situations they face. The normed mean here is 6.75, and these participants went from 6.75 to 10, as they came to learn that they could not change some things. This difference was significant at the .043 level. They also used more cognitive avoidance, seeking alternative rewards, and emotional discharge as coping mechanisms. The fact that the mean scores of participants showed an increase on every scale suggests that they were more able to identify the methods they use to cope in situations they encountered. If they are better able to choose their coping strategies, it can be viewed as a positive outcome.

The Problem Oriented Instrument for Youth (POSIT) (Rahdert, 1991) measures the extent to which youth are having problems in 10 domains. Lower scores demonstrate improvement and a higher score demonstrates more problems in that area. The scoring system is based on clinical judgment, and has not yet been field-tested. Here, the participants improved on all scales except in the area of family relationships, where they stayed the same (see Figure 4). It is interesting to note that family relationships is an area where they reported improvement in interviews. None of the mean changes were significant, however. It should be noted that all of the mean scores exceeded the cutoff points designated by the test designers at pretesting and posttesting, with the exceptions of physical health and leisure at posttest. These high scores indicate that the participants were at risk in almost every area both before and after the group. In the areas of mental health and substance abuse, substantial problems were suggested.

The Automatic Thoughts Questionnaire (Clear Thinking Workshops, 1995) did not show a great deal of change. This is a series of open-ended questions assessing how persons interpret events, and no norms are established. While there was a slight improvement here in that their answers reflected increased ability to think before acting, the respondents as a group expressed contempt for the questions through jokes. It was included because it had been used with the earlier cognitive group, and in that group the members were considered to have made improvement. This test is part of a planned workshop, used in the earlier cognitive group, where specific activities fostering the learning of these specific skills are taught.

It is also interesting to note differences and similarities between this group and the two earlier groups, the cognitive-only group of 1997 and the narrative pilot group of 1998. While these are not comparison groups, they were used to inform and change how this group was designed. The cognitive group began with eight adolescents and ended with five, and it resembled this group more than the pilot group did in that the members were a bit older and had been involved with more frequent and more serious illegal activities. The co-investigators noted that the girls did not become as involved with the process, or with each other in the cognitive group as they did in the 2 years of the narrative group experience. In the narrative groups, the members were more involved in examining their lives, challenging the power of the problems they faced, and in acknowledging and celebrating the times they could triumph over the problems. The group process was more intense than in the cognitive group, as they tested their power and learned to negotiate and compromise. They also assessed using their power to change directions in their lives; they tested limits and learned the power of choosing when to cross the line.

A review of the arrest records gave credence to the coinvestigator's assessment of the differences between the narrative groups and the cognitive only group. We looked at firther offending by the participants who completed the programs at 6 months after the groups ended. In the cognitive group of 1997, three out of five (60%) had reoffended within 6 months; in the pilot group none of the three reoffended within 6 months, and in this group, two out of eight (25%) had reoffended within 6 months. This review tentatively supports the qualitative and quantitative findings suggesting the possible usefulness of narrative approaches with this population.

Summary and Discussion

This project assessed the usefulness of narrative group therapy as part of a summer program for female adolescents who had broken the law, and sought to gain more understanding of this population. In narrative therapy, clients and therapists work collaboratively to discuss the problem, helping clients to view the problem as separate from themselves, like a character in a story, and to then change, re-story, and rewrite their relationship to it. The clients' expertise on their own lives is respected, as alternative narratives are discovered and celebrated. Such an approach seemed useful for this population, and there were anecdotal reports of success with young people, but with little research to support the enthusiasm. Exploratory research to gain greater understanding of adolescent offenders and to assess narrative approaches for work with them appeared to be useful at this point.

Qualitative methods of addressing the two questions is appropriate since discussion, story telling, and looking for different views of a situation are a part of narrative therapy. They also give a qualitative researcher a rich view of the participants' lives and of changes made in the group members. Since it was a time-limited program, and since accountability was important with public funding, other methods of assessment, including pretests and posttests were also considered useful. The approaches complemented each other, adding richness and explanations to the findings. For example, self-esteem was not significantly changed as a result of this group, as shown on pretests and posttests, and in fact was higher than average at the start. Qualitative analysis helped to explain this finding. Participants' discussion early in both years' groups was fill of bravado ("Hey, I'm great"), whereas they were more reflective and realistic in later sessions. While bravado can be an important defense, the increased reflection can demonstrate decreased need for the defense.

Overall, on various means of assessment, the authors found narrative approaches useful in helping these girls examine their lives and to assess their future possibilities. The authors also found that the more "intervention" the girls had had previously, the less open they were to storytelling. It seemed a further invasion of privacy into their lives. Through the use of artwork and creative projects, however, the girls opened up a great deal and were willing to examine their lives. The qualitative analysis helped the authors better understand this population. Violence, anger, loss, premature adulthood with little preparation, chronic transition, and lack of trust were among the themes extracted from the transcripts; and one test (POSIT) showed a high degree of mental health problems in this group. Many have been shunned by mainstream society, and they have developed toughness and a sense of a privileged world to sustain themselves. Yet every problem theme had a solution or coping theme within it, which was brought forth and highlighted in these sessions. For example, their need to be in control of others was an expression of the same need for control, which helped them manage their own anger and behavior. The girls provided many examples of this. Coping was one of their strengths and both qualitative analysis and test scores found this to be the the most improved area during treatment. The importance of family relationships for these adolescents was also noted.

Two important findings, supported by both qualitative and quantitative data, were that self esteem was higher and locus of control more internal than anticipated at the start and finish. These girls faced many problems in their lives, and they did have a higher than average number of mental health problems, but feeling in charge and feeling OK about themselves were, for them, very important coping mechanisms. These findings are important for future therapy and research with this population.

References

Attanucci, J. S. (1993). Timely characterization of mother-daughter and family-school relations: Narrative understandings of adolescence. Journal of Narrative and Life History, 3(1), 99-116.

Biever, J. L., McKenzie, KC, Wales-North, M., & Cortez Gonzalez, R. (1995) Stories and solutions in psychotherapy with adolescents. Adolescence, 30(118), 491-499.

Burnette, M. (1995). Secrets: Essentialism in narrative therapy. Journal of Child and Youth Care, 10(3), 1-10.

Charmaz, K (1990). "Discovering" chronic illness: Using grounded theory. Social Science Medicine, 30, 1161-1172.

Clear Thinking Workshops. (1995). Automatic thoughts. Iowa City, IA: River Lights Publishing.

Frank-Stromborg, M. (Ed.). (1988). Instruments for clinical nursing research. Norwalk, CT: Appleton & Lange.

Freedman, J., & Combs, G. (1996). Narrative therapy: The social construction of preferred realities. New York: WW. Norton.

Garvin, C. (1997). Contemporary group work (3rd ed.). Boston: Allyn and Bacon.

Harter, S. (1990). Self and identity development. In S. Feldman & G. Elliott (Eds.). At the threshold: The developing adolescent (pp. 352-387). Cambridge, MA: Harvard University Press.

Hauser, S., & Bowlds, M. (1990). Stress, coping and adaptation. In S. Feldman & G. Elliot (Eds.), At the threshold: The developing adolescent (pp. 255-276). Cambridge, MA: Harvard University Press.

Henggeler, S. W., Melton, G. B., & Smith, L. A. (1992). Family preservation using multisystemic therapy: An effective alternative to incarcerating serious juvenile offenders. Journal of Consulting and Clinical Psychology, 60(6), 953-961.

Holcomb, R W (1994). Incorporating new treatment perspectives into a residential, adolescent drug and alcohol program. Journal of Child and Youth Care, 9(2), 43-50.

Kelley, P. (1996). Narrative theory and social work treatment. In F. J. Turner (Ed.), Social Work Treatment (4th ed.) (pp. 461-479). New York: The Free Press.

Kelley, P., & Clifford, P. (1997). Coping with chronic pain: Assessing narrative approaches. Social Work, 42(3), 266-277.

Lee, J. (1996). The empowerment approach to social work practice. In F. J. Turner (Ed.), Social Work Treatment (4th ed.) (pp.218-249). New York: The Free Press.

Lincoln, Y. S., & Guba, E. G. (1985). Naturalistic inquiry. Newbury Park, CA: Sage.

Moos, R H. (1993) Coping Responses Inventory-youth form. Odessa, FL: Psychological Assessment Resources, Inc.

Morgan, A- (Ed.) (1999). Once upon a time: Narrative therapy with children and their families Adelaide, South Australia: Dulwich Centre Publications.

Neimeyer, R A. (1999). Narrative strategies in grief therapy. Journal of Constructivist Psychology, 12, 65-68.

Norwicki, S., & Strickland, B. R (1973). A locus of control scale for children. Journal of Consulting and Clinical Psychology, 40, 148-154.

Penn, P., & Frankfurt, M. (1994). Creating a participant text: Writing, multiple voices, narrative multiplicity. Family Process, 33(3), 217-232.

Pett, M. (1997). Nonparametric statistics for health care research: Statistics for small samples and unusual distributions. Thousand Oaks, CA; Sage.

Rahdert, E. (Ed.) (1991). The adolescent assessment/preferred system manual. Washington, DC: National Institute of Drug Abuse, U.S. Department of Health and Human Services.

Rosenberg, M. (1989). Society and the adolescent self-image. Middletown, CT: Wesleyan University Press.

Saleebey, D. (1997). The strengths perspective in social work practice (2nd ed). White Plains, NY: Longman.

Smith, C., & Nylund, D. (Eds.). (1997). Narrative therapies with children and adolescents. New York: Guilford.

Snyder, H. N., Sickmund, M., & Poe-Yamagata, E. (1996). Juvenile offenders and victims 1996 Update on violence. Washington, DC: Office of Juvenile Justice and Delinquency Prevention.

Snyder, H. N. (2000). Juvenile arrests 1999. Juvenile Justice Bulletin, December, 1-11. Washington, DC: Office of Justice Programs.

Stahl, A. (1998). Person offenses in juvenile court, 1986-1995. OJJDP Fact Sheet, 77. Washington, DC: Office of Justice Programs.

Stahl, A. (1999). Delinquency cases in juvenile courts, 1996. OJJDP Fact Sheet, 109. Washington, DC: Office of Justice Programs.

Steinberg, L (1990). Independence in the family: Autonomy, conflict, and harmony in the parent-adolescent relationship. In S. Feldman & G. Elliot (Eds.), At the threshold: The developing adolescent (pp. 255-276). Cambridge, MA: Harvard University Press.

Tutty, L. M., Rothery, M. A., & Grinnell, FL M. (1996). Qualitative Research for Social Workers. Needham Heights, MA: Allyn & Bacon.

White, M., & Epston, D. (1990). Narrative means to therapeutic ends. New York: W.W. Norton.

Zimmerman, J. L., & Dickerson, V. C, (1994). Using a narrative metaphor: Implications for theory and clinical practice. Family Process, 33(3), 233-245.

Patricia Kelley is professor, School of Social Work 308 North Hall, University of Iowa, Iowa City, IA 52242; e-mail: patria-kelley@uiowa.edu. Lou Blankenburg is doctoral student in social work University of Iowa, 1634 Wilson Street Iowa City, IA 52245. She was formerly with United Action for Youth of Iowa City,

Judith McRoberts is research associate, National Resource Center for Family Based Practice, W 206 OH, University of Iowa, Iowa City, IA 52242. She was formerly with United Action for Youth of Iowa City.

Authors' notes: The authors thank and acknowledge grant support provided by the Lois and Samuel Silberman Fund of New York City, and the Obermann Center for Advanced Studies Spelman Rockefeller (CASSPR) Grant, University of Iowa. We also thank the United Action for Youth of Iowa City and the Johnson County (Iowa) for their cooperation and support of this program. Leslie Schnoll, now a clinical social worker and therapist at Alternative Homes for Youth in Greeley, Colorado is thanked for her invaluable work on this project when she was an MSW student at the University of Iowa.

Manuscript received: March 21, 2001

Revised: February 25, 2002

Accepted: April 30, 2002

Copyright Families in Society Sep-Dec 2002
Provided by ProQuest Information and Learning Company. All rights Reserved