Effectiveness of Solution-Focused Brief Therapy in a School Setting

Written By: Sam Savage

Published Date: February 6, 2008 Last Edited: February 6, 2008

By Franklin, Cynthia Moore, Kelly; Hopson, Laura

This study evaluated the effectiveness of solution-focused brief therapy with children who have classroom-related behavior problems within a school setting. Five to seven sessions of solution-focused brief therapy services were provided to 67 children, identified by school faculty and staff as needing assistance in solving behavior problems. Teacher inservice training and three to four consultation meetings were also provided. Externalizing and Internalizing scores from both the Youth Self-Report and Teacher Report Forms of the Child Behavior Checklist were used as outcome measures. Outcomes were evaluated by using a pretest/posttest follow-up design with a comparison group. Effect sizes and improved percentage scores were calculated. Findings provide support that solution-focused brief therapy was effective in improving internalizing and externalizing behavior problems. KEY WORDS: brief therapy; effectiveness; intervention; middle school; solution-focused

School social workers are increasingly called upon to demonstrate the effectiveness of their interventions. Reliably improving student outcomes requires the use of interventions that have research support and evaluation of interventions on an ongoing basis in school settings (Franklin & Hopson, 2004).This article describes a study that evaluated the effectiveness of solution-focused brief therapy (SFBT) with middle school students.

SFBT, an approach that first originated at the Family Therapy Center in Milwaukee, Wisconsin,by its cofounders Steve DeShazer (DeShazer, 1985, 1988, 1991, 1994; DeShazer et al., 1986) and Insoo Kim Berg (Berg, 1994; Berg & DeJong, 1996; Berg & Miller, 1992) and colleagues (for example, Miller, Hubble, & Duncan, 1996; Walter & Peller, 1992), has been successfully applied to change child and adolescent behavior problems. Using a multiple baseline design, Conoley et al. (2003) found, for example, that three families experienced positive outcomes for children’s aggressive and oppositional behavior after four or five sessions of SFBT.

Only a handful of studies have explored the effectiveness of SFBT in school settings (Franklin, Biever, Moore, Clemons, & Scamardo, 2001; Geil, 1998; LaFountain & Gardner, 1996; Littrell, Malia, & Vanderwood, 1995). However, SFBT is an approach that has demonstrated promising outcomes in these studies (Franklin & Streeter, 2003; Krai, 1995; Metcalf, 2002; Murphy, 1997; Sklare, 1997;Webb, 1999). Because SFBT takes a social construction approach and encourages engaging teachers, parents, and others involved with the child in treatment, it is consistent with the intervention characteristics that are successful in school settings (Roans & Hoagwood, 2000). Both Metcalf (2002) and Murphy (1997) used clinical case studies to demonstrate that SFBT is a useful approach in schools (Franklin et al.,2001;LaFountain & Garner, 1996;Littrell et al., 1995; Newsome, 2004).

Research on SFBT in school settings demonstrates promising outcomes in increasing self-esteem and positive attitudes. Littrell and associates (1995) evaluated SFBT in comparison with two other approaches. High school students (N =61) were randomly assigned to one of three groups: a problem-focused brief counseling approach with a task, a problemfocused brief counseling approach without a task, and a solution-focused brief counseling approach with a task. The results indicate that all students experienced positive outcomes in goal attainment and alleviating concerns and in decreasing the intensity of undesired feelings. However, the three brief- counseling approaches did not differ in their effectiveness.

In more quasi-experimental designs, several researchers found statistically significant positive outcomes (LaFountain & Garner, 1996; Newsome, 2004; Springer, Lynch, & Rubin, 2000). LaFountain and Garner (1996) investigated the impact of solution-focused groups on schoolage children (N- 311) and practitioners. They found significant between-group differences on three subscales of the Index of Personality Characteristics (Nonacademic, Perception, and Acting In), suggesting that the experimental group had higher self- esteem, more positive attitudes and feelings about themselves, and more appropriate ways of coping with emotions. Springer et al. (2000) determined that children receiving SFBT made significant pre- and posttreatment improvement on the Hare Self-Esteem Scale, whereas scores for those receiving no treatment were unchanged; the effect size of .57 was moderate.

SFBT has also been associated with positive academic and behavioral outcomes. Students in a quasi-experimental, pre- and posttest design with SFBT (N = 6) were found to have increased their posttest grade point average (GPA) scores if in the SFBT group compared with those in the comparison group when using pre-GPA scores as the covariate (Newsome, 2004). School attendance, however, showed no difference between the two groups. Franklin and associates (2001) in a single-case design study with fifth and sixth graders found that five out of seven (71 percent) of the cases of special education students with classroom behavior problems improved per the teacher’s reports. Franklin and Streeter (2004) compared attitudes of students attending a school in which all staff were trained in SFBT techniques with attitudes of similar students attending a more traditional high school. Student responses to the School Success Profile indicated that students attending the solution-focused school rated school satisfaction and teacher support as assets, whereas the comparison group rated them as risk factors. The solution-focused group also earned significantly more credits than did students in the comparison group.

Although these studies show promise for the solution-focused interventions in schools, replication of effectiveness studies that use larger sample sizes and improved research designs is needed to further substantiate these findings and to build an evidenced-based SFBT therapy for education. The majority of the studies were quasi- experimental designs that did not use random assignment because schools often do not permit assigning students randomly to treatment conditions. In addition, it may be important to combine individual and group-level intervention with organizational interventions to achieve positive outcomes for students to strengthen treatment outcomes. Franklin and associates (2001) suggested from the results of their single-case study that SFBT intervention might be more effective in a school if the teachers and staffare trained in the model. These authors stated their belief that in order to maximize the model’s effectiveness in a school setting, the entire school culture, norms, and practices would need to change and follow the strengths and empowerment orientation of the solution-focused model.This is consistent with research demonstrating that multicomponent interventions that work to change the school climate as well as teacher and parent interactions with the child are useful in school settings (Rones & Hoagwood, 2000).

PURPOSE OF THE STUDY

The purpose of this study is to build on the growing body of quasi-experimental research on the effectiveness of SFBT in school settings.This study focuses on the effectiveness of an SFBT approach with middle school children identified as having school-related behavior problems. Previous studies found that SFBT resulted in positive behavior changes for students and was promising as an intervention with students experiencing academic and behavior difficulties (Franklin & Streeter, 2003; Krai, 1995; Metcalf, 2002; Murphy, 1997;Sklare, 1997;Webb, 1999). This study builds on this previous research by investigating the impact of a multicomponent SFBT approach with behavior problems in students.

METHOD

Participants

Sixty-seven students who exhibited schoolrelated behavior problems from the Schertz Cibolo Universal Unified Independent School Districts were recruited for participation in the study. Students were enrolled in the Dobie Intermediate School and the Wilder Intermediate School and ranged in age from 10 to 12 years (fifth and sixth graders). Principals and teachers purposively selected students in each school who had received more than one behavioral referral from a classroom teacher. The behavioral referral is an incident report that requires disciplinary action and referral for pupil services. Common reasons for behavioral referrals include inattentiveness, tardiness, school phobia, difficulty completing tasks, and social problems that affect school performance.

Thirty children were in the experimental group and 29 were in the comparison group (see Table 1 for demographic characteristics of the sample).There were 15 boys and 15 girls in the experimental group and 25 boys and four girls in the comparison group. A chi-square analysis indicated a significant difference between groups on gender, chi^sup 2^(1, N= 59) 8.85, p = .003. The groups did not differ significantly in age, ethnicity, grades in school, or participation in special education.

With alpha = .05 there were no significant differences between the experimental and comparison groups on the pretest for the children’s Externalizing or Internalizing scores as reported by both the Teacher Report Form (TRF) and the Youth Self-Report (YSR) of the Child Behavior Checklist (CBCL) (Achenbach, 1991;Achenbach & Edelbrock, 1983,1986) (see Table 2). Research Design

Research participants in the SFBT outcome study were evaluated with a pretest-posttest follow-up design with a comparison group. Children who attended Dobie Intermediate School were assigned to the experimental group, and children who attended Wilder Intermediate School comprised the comparison group. A quasi-experimental design in which experimental and comparison conditions were located at two different schools was selected over a traditional randomized experimental design to reduce the possibility of comparison group contamination through interaction between children in the two conditions. Because children are in several classes a day, there was also the possibility that a teacher of a student in the comparison group would also be a teacher of a student in the experimental group, leaving the study open to treatment contamination.

Procedures

Two schools (Dobie Intermediate School and Wilder Intermediate School) were selected to participate in the study, located in the Schertz Cibolo Universal Unified Independent School District in Cibolo,Texas. Cibolo is a rural community now transitioning to a suburban, bedroom community of San Antonio. Dobie and Wilder were selected as schools for this study because of their similar student populations.

Dobie Intermediate School was selected to receive the treatment because of greater availability of space for conducting sessions, and Wilder Intermediate School acted as the comparison condition. The principals, classroom teachers, and practitioners from Dobie and Wilder generated a list of eligible students for the study. Each student had three teachers. The two teachers who saw the child more frequently would fill out the CBCL.

The study included four practitioners who provided SFBT for this study. All four were masters-level practitioners with several years of post-master’s training, including extensive experience with children and families. Three of the four practitioners were concurrently enrolled as doctoral students during the study. The two principal investigators and one of the practitioners received training by Insoo Kim Berg and Steve DeShazer, co-developers of Brief Solution-Focused Therapy, at the Brief Family Therapy Center in Milwaukee where they attended a four-day advanced training.The other practitioners attended a one-day training by Insoo Kim Berg at the Texas Association for Marriage and Family Therapist Conference. All practitioners were clinically supervised by the principal investigators on a biweekly basis.

To evaluate practitioners’ adherence to solution-focused techniques and to ensure treatment integrity, the principal investigators videotaped and reviewed two sessions for each practitioner. The criteria for review were that each session must contain the following: the “miracle question,” exception finding questions, scaling questions, coping and motivational questions, a break, and a formulated task. All reviewed tapes met these requirements.

Pretesting was conducted after consents were obtained from the parents. Children completed the YSR, and teachers completed the TRF of the CBCL. Students were given the pretest YSR by the researcher and research assistant in small group administration of about four to five children. Makeup administrations were given until all children completed the test. As Achenbach (1991) recommended, students needing help received assistance with reading and interpretation of questions. Teachers were provided with the instruments in a folder with a due date.The folders with completed tests were collected several days after dissemination.

After the teacher training was completed and each student in the experimental group participated in an average of five sessions, students were given a posttest, administered in a similar manner as the pretest. Finally, a one-month follow-up test was administered.The students in the comparison group were given pretests, posttests, and a follow-up test without treatment. Incentives included movie passes for students who completed the follow-up and for teachers in the comparison group to compensate for their time. A longer follow-up was prohibited because of the end of the school year. The practitioners were prepared to give follow-up resources to any child who appeared to need further intervention. One child was given a referral because of a concern over possible alcohol abuse, but no other children appeared to need additional resources.

To assess comparability of treatment and comparison groups, we collected, in addition to the CBCL information, demographic and biopsychosocial data through interviews over the phone or in person with the parents.

Measures

Two standardized measures were used to evaluate outcomes: the Internalizing score and the Externalizing score of the YSR and TRF versions of the CBCL. The Internalizing score is defined as the sum of scores of the problem items of the Withdrawal, Somatic Complaints, and Anxious/ Depressed scales.The Withdrawal scale measures traits such as perceiving oneself or perceiving the student (depending on who is answering the questions) as shy, underactive, and sad. The Somatic Complaints scale includes behaviors such as having symptoms of headaches, stomachaches, and other physical complaints. The Anxious/Depressed scale includes affective states like feeling lonely, worthless, nervous, fearful and guilty, in addition to behaviors like self-injurious behavior, frequent crying, suicidal thoughts, and worrying (Achenbach & Edelbrock, 1986).

The Externalizing score is defined as the sum of scores on the problem items of the Delinquent and Aggressive Behavior scales. Delinquent behavior is measured by questions such as frequency of lying and cheating, running away, swearing, truancy, and involvement with alcohol and drugs. Aggressive behavior is evaluated through questions about arguing, being mean to others, being destructive, having a bad temper, and fighting (Achenbach Sc Edelbrock, 1986).

The CBCL was chosen for a measure of school-related behavior because it has wellestablished reliability and validity for use in schools and is considered one of the best instruments for measuring child behavioral disorders (Achenbach & Edelbrock, 1986; Husain & Cantwell, 1991). Achenbach and Edelbrock (1986) reported high reliability in their test manual. For the TRF, the one-week median test-retest Pearson correlation was .90, with negligible changes in mean scale scores. In addition, the TRF has high-content, construct, and criterion validity. The content validity of the TRF was evaluated in terms of whether its items were related to concerns about pupils’need for special help for behavioral and social- emotional problems.

Achenbach (1991) reported that for raw scores on theYSR competence scales, the mean sevenday test-retest reliability was r = .68 for 11- to 14-year-olds and r = .82 for 15- to 18-year-olds. Content validity is supported by the ability of most YSR items to discriminate significantly between demographically matched referred and nonreferred children. Criterion-related validity is supported by the ability of theYSR’s quantitative scale scores to discriminate between referred youths after demographic effects were partialled out.

Intervention

The intervention included a combination of individual sessions of solution-focused brief therapy with students plus teacher training, consultations, and collaborative meetings. see Table 3 for a description of the treatment components, and see Table 4 for the elements of the individual sessions. Students received five to seven individual SFBT sessions lasting 30 to 45 minutes each week (except in cases of illness or absence) with the practitioner. A treatment protocol was followed as a guide for implementing the SFBT model with children in the school (Franklin & Biever, 1996).

The methods specified in the treatment protocol are the ones followed most closely by Berg and DeShazer (Berg, 1994; DeShazer, 1985; 1988). The use of at least three main process sequences (the miracle question, scaling questions, and giving client compliments) in a session is necessary for an intervention to be considered solution-focused (personal communication with LK. Berg and S. DeShazer, Brief Family Therapy Center, Milwaukee, WI, January 2,1997).The miracle question provides a way of helping clients envision solutions to a problem. The practitioners ask something like, “Let’s suppose that overnight a miracle happened and the problem you are having with your teacher disappeared. But, you were sleeping and did not know that it happened. When you came to school the next day, what would be the first thing that you would notice?” Scaling questions ask clients to rate their problems on a scale of 1 to 10 and project their future progress (see Franklin & Moore, 1999, for a more detailed explanation of questioning techniques associated with SFBT). Every session followed the same process and used the same questioning techniques. These types of questions aim to help clients find strategies for improving problem behaviors by building on pre-existing strengths.

Teachers completed a four hour in-service training that educated the faculty, staff, and administration about the study, components of SFBT, and techniques to use in the school setting. Practitioners also spoke with teachers about the progress of the cases and provided teacher consultations for the cases. Each practitioner met with each participant’s teacher three to four times during the study.These meetings included an initial assessment meeting, formal sit-down meetings, casual discussions, and teacher referral and student information worksheets. Most teacher consultations were brief, lasting only 10 to 20 minutes a week (some teachers received more consultation, determined by their commitment and as required by the cases). Teacher referral forms were used to maintain communication with the practitioners when meetings were not possible.The form asked the teacher to list specific behaviors the child did well during the week.The child also completed a form that indicated helpful teacher behavior perceived by the student. In addition, one or two formal collaborative meetings were held with the teacher, practitioner, and student together to discuss successful behavior of both student and teacher. Data Analysis

In the interest of clinical relevance, all analyses were conducted with only those children who scored above the clinical cutoff point (t = 60, with scores of 60 to 63 considered borderline) on the pretest of either the Internalizing or Externalizing scales of theTRF or theYSR. Scores above the clinical cutoffindicate that the child’s behaviors fall into the problem range rather than the normal range. ESs and improved percentage scores were also calculated to corroborate the magnitude of changes between pretest and posttest scores. Multivariate analyses of variance (MANOVAs) with repeated measures (pretest, posttest, and follow-up) were conducted to assess the effect of a solution-focused approach on the children’s Internalizing and Externalizing scores as reported by the YSR and on the externalizing and internalizing behavior of children as reported by the teachers on the TRF of the CBCL.

RESULTS

The treatment and comparison groups were tested for equivalency on the pretest measures by using independent t tests. Because each group consisted of only those students who scored in the clinical range on the particular scale under comparison (clinical range: t >/ = 60), each sample size was different for each measure. For example, only students who scored in the clinical range on the scale measuring teacher reports of internalizing behaviors were included in the analysis for that scale. Because more students scored in the clinical range on the TRF-Externalizing than on the TRF- Internalizing, the analysis of the externalizing behaviors measure had a larger sample. It is important to note that the sample size for each scale does not add up to the total sample size because it is possible for a child to score in the clinical range on more than one scale. These children were included in more than one analysis.

Attrition

During the study one practitioner needed to take a leave of absence. The children she served received the intervention but not in enough time to be included in the study; therefore, four children in the experimental group needed to be dropped from the final data analysis. In addition, two children from the comparison group moved away in the middle of the study, resulting in two scores from the comparison group being dropped from the final data analysis.

Teacher Reports

Most children had at least two teachers who completed the TRF. The only exception was when a child was in only one class and therefore had only one teacher.The mean of the two scores comprised the teacher score for each child. see Table 5 for a summary of the MANOVA results of the externalizing and internalizing symptoms from the TRFs. Analysis of the externalizing symptoms of the TRF revealed significant differences between groups over time [F(2, 84) = 11.20, p

Analysis of the internalizing symptoms of the TRF shows a significant difference between groups over time [F(2, 42) = 4.60, p

Children’s Reports

Analysis of the externalizing symptoms from the YSR shows a significant difference between groups over time [F(2, 54) = 7.80, p

There were no significant group differences on Internalizing scores over time [F(2, 54) = 0.22, p = .90]. Scores for both groups improved to fall in the normal range. The patterns of change in self- report of internalizing behavior was not significantly different for the two groups, suggesting that the intervention had little impact on the pattern of change (ES = 0.08). Both groups started out with similar pretest scores and dropped at an equal rate. see Table 6 for a summary of the results of the Internalizing scores from the YSRs.

DISCUSSION

The findings of this study provide continued support that SFBT was effective in reducing classroom-related behavioral problems (Franklin et al., 2001; LaFountain & Garner, 1996). Students who participated in the experimental group showed significant improvement over students in the comparison group on teacher’s report of internalizing behavior, teacher’s report of externalizing behavior, and student’s report of externalizing behavior. The scores on these assessment measures moved from scores in the clinically significant or problem ranges to the normal range. The intervention failed to show significant effects on the children’s report of internalizing behavior because scores in both groups improved. All children included in the analyses scored above the clinical cutoff point at pretest. With the exception of the children’s report of internalizing behavior, only children in the experimental group scored below the clinical level at posttest and remained subclinical at follow-up.The comparison group maintained scores above the clinical cutoff point up through follow-up. All ESs, except those for the children’s report of internalizing behavior were large, falling between 8 percent to 16 percent variance of the dependent variable (Cohen, 1977; Rubin & Babbie, 1997).

Overall there was a high level of agreement between the adult and child observers of change. One interesting finding was the improvement in externalizing behaviors for children in the SFBT group only, whereas children in both groups improved in their reports of internalizing behaviors. Some extraneous factor may have caused students in both groups to perceive a change in their internalizing behaviors. In addition, students in the comparison group perceived a change in their internalizing behaviors, whereas their teachers did not. The discrepancy between the teachers’ perspectives and the students’ self-perceptions may indicate a need for teachers to re-evaluate the way they perceive and address some of the children’s more internal processing.

More research is needed to investigate the types of behavior for which this intervention is most beneficial. It would also be helpful to examine the components of the intervention to determine which are most responsible for positive outcomes. For example, it would be helpful to know whether the teacher training and consultation components are important for the effectiveness of the intervention or whether the individual sessions would be sufficient for positive outcomes. Future research should also examine factors that could moderate the effectiveness of the intervention, such as the relationship between participating students and teachers or the school climate.

Including the teachers in the training allowed for more cooperative intervention strategies like consultation, both formal and informal.Teacher consultation was a mandatory component of the treatment, allowing the teachers to have “coaching” on the model over time.This served two purposes. One was that the teachers were complimented and shown that they were making changes with the aim of helping them feel successful in their roles. Another was that teachers were able to collaborate with the practitioners on targeted behavioral changes and techniques for interventions.What is more important is that training teachers and administrative staff to use SFBT consistently with students provides a more cohesive approach to student-teacher interactions. Although it is possible that compliments to teachers may bias their ratings of students, collaborating with teachers and providing compliments are an important component in the SFBT intervention. By encouraging teachers to become active change agents, positive results may continue after official treatment has stopped between the child and the practitioner. It will be helpful for future research to determine whether SFBT has long-term effects on both teachers and students (for example, six-month, one-year, two-year, and five-year follow-ups).

It is also expected that an even greater effect might occur if greater emphasis is placed on transforming educational systems into solution-focused cultures (Franklin & Streeter, 2004).This would likely represent a shift in the educational paradigm for many current school systems. Study Limitations

A limitation of the small group administration of measures was that the children may not have felt completely comfortable being honest in their responses. As a result of lack of time and resources, however, group administration was necessary. In addition, because lack of randomization and possible unequal groups was a factor, threats to internal validity are possible and the results must be interpreted with caution.

Because of the limited number of girls in the study, gender may be a confounding variable. However, the t scores for the CBCL were normed for gender and age (Achenbach, 1991; Achenbach & Edelbrock, 1986), so the scores of the YSR and TRF should be comparable. In addition, the Externalizing scores remained the same even after the girls were dropped from the analysis. However, when controlling for girls, the teachers’ reports of internalizing behavior were not retained. It is difficult to make a definitive conclusion about this finding because once the girls were dropped from the analysis, the number of total participants dropped (from N= 23 to N = 15), greatly reducing statistical power.

Implications for Social Work: Practice and Policy

This study, along with the growing body of literature in this area, has great implications for social workers. School social workers need interventions that work quickly. By demonstrating that with an average of five sessions of this intervention children can move from clinical to nonclinical ranges on the CBCL strongly supports the idea that with purposeful short-term intervention significant change can happen rapidly. This is vital information for overworked, underfunded practitioners.

Using the consultation model presented in this study, the practitioners are working with the teachers and administrators as “coaches.” This allows a great deal of the intervention to be performed by the teacher, the individual who has the most contact with the student and as such has the greatest chance of affecting classroom behavior. With increased attention devoted to collaboration in children’s services, an intervention whose purpose is to empower all individuals involved is highly valuable. In addition, by encouraging teachers to become active agents of change along with social workers, positive results can continue long after official treatment has ended between the child and the social worker.

Empowering teachers to effect change is important given the newest No Child Left Behind Act, whose stated purpose is to “close the achievement gap with accountability, flexibility, and choice, so that no child is left behind.” (No Child Left Behind Act, 2001, p. 1) Teachers need to have ways to address the needs of all of their students, especially those students whose behavioral problems are interfering with academic progress.The results of this study suggest that SFBT, combined with teacher training and consultation, may be one way to successfully address behavioral problems that could inhibit learning, particularly more externalizing acting out types of behaviors.

Conclusion

The results of this study add strength to the argument that school-based helping professionals might benefit from training in brief, strengths-based approaches like SFBT Sklare (1997) pointed out that most practitioners are trained in long-term therapies but are expected to abbreviate these therapies to fit into the reality of heavy caseloads. Perhaps by recognizing that school interventions that are short-term in nature are effective, future clinicians can receive training more congruent with the realities and demands of their jobs. By demonstrating that a mean of five SFBT sessions delivered in a school setting with teacher consultations can move children from clinical to nonclinical ranges on the CBCL, the SFBT intervention strongly supports the idea that purposeful short-term intervention can make a difference.

By encouraging teachers to become active change agents, positive results may continue after official treatment has stopped between the child and the practitioner.

REFERENCES

Achenbach.T. M. (1991). Manual for theYouth Self-Report and 1991 Profile. Burlington,VT: Department of Psychiatry, University of Vermont.

Achenbach.T. M., & Edelbrock, C. (1983). Manual for the Child Behavior Checklist and Revised Child Behavior Profile. Burlington: Department of Psychiatry, University of Vermont.

Achenbach.T. M., & Edelbrock, C. S. (1986). Manual for Teacher’s Report Form and Teacher Version of the Child Behavior Profile. Burlington: Department of Psychiatry, University of Vermont.

Berg, I. K. (1994). Family-based services: A solution-focused approach. NewYork:W.W. Norton.

Berg, I. K., & Dejong, P. (1996). Solution-building conversations: Co-constructing a sense of competence with clients. Families in Society, 77, 376-390.

Berg, I. K., & Miller, S. D. (1992). Working with the problem drinker: A solution-focused approach. New York: Norton.

Cohen, J. (1977). Statistical power analysis for the behavioral sciences. New York: Academic Press.

Conoley, C. W., Graham,J. M., Neu,T, Craig, M. C, O’Pry, A., Cardin, S.A., Brossart, D. F., & Parker, R. 1. (2003). Solution- focused family therapy with three aggressive and oppositional- acting children: An N = 1 empirical study. Family Process, 42, 361- 374.

DeShazer, S. (1985). Keys to solution in brief therapy. New York: Norton.

DeShazer, S. (1988). Clues: Investigating solutions in brief therapy. New York: Norton.

DeShazer, S. (1991). Putting differences to work. New York: Norton.

DeShazer, S. (1994). Words were originally magic. New York: Norton.

DeShazer, S., Berg, I., Lipchik, E., Nunnally, E., Molnar, A., Gingerich, W., & Weiner-Davis, M. (1986). Brief therapy: Focused solution development. Family Process, 25, 207-222.

Franklin, O, & Biever.J. (1996). Treatment manual and research protocol for solution-focused therapy with learning challenged students in schools. Unpublished manuscript. University of Texas at Austin.

Franklin, C, & Hopson, L. (2004). Into the schools with evidence- based practices [Editorial]. Children & Schooh, 26, 67-70.

Franklin, C, & Moore, K. C. (1999). Solution-focused brief family therapy. In C. Franklin & C.Jordan (Eds.), Family practice: Brief systems methods for social work (pp. 105-142). Pacific Grove, CA: Brooks/Cole.

Franklin, C, & Streeter, C. L. (2003). Solution-focused accountability schools for the tiventy-first century: A training manual for Gonzalo Garza Independence High School. Austin,TX:The Hogg Foundation for Mental Health.

Franklin, C, BieverJ., Moore, K., Clemons, D, & Scamardo, M. (2001).The effectiveness of solutionfocused therapy with children in a school setting. Research on Social Work Practice, 11, 411-434.

Franklin, C, & Streeter, C. L. (2004). Solution-focused alternative schools: An evaluation of Gonzalo Garza Independence High School. Austin,TX: Hogg Foundation for Mental Health.

Geil, M. (1998). Solution-focused consultation:An alternative consultation model to manage student behavior and improve classroom environment. Unpublished doctoral dissertation. University of Northern Colorado.

Husain, S.A., & Cantwell, D. P. (1991). Fundamentals of child and adolescent psychopathology. Washington, DC: American Psychiatric Association.

Kral. K. (1995). Solutions for schools. Milwaukee, WI: Brief Family Therapy Center.

LaFountain, R. M., & Garner, N. E. (1996). Solutionfocused counseling groups:The results are in. Journal for Specialists in Group Work, 21, 128-143.

Littrell, J. M., MaliaJ. A., & Vanderwood, M. (1995). Single- session brief counseling in a high school. Journal if Counseling and Development, 73, 451-458.

Metcalf, L. (2002). Counseling toward solutions: A practical solution-focused program for working with students, teachers, and parents. San Francisco: Jossev-Bass.

Miller, S. D, Hubble, M. A., & Duncan, B. S. (Eds.). (1996). Handbook of solution-focused brief therapy. San Francisco: Jossey- Bass.

Murphy, J. J. (1997). Solution-focused counseling in middle and high schools. Alexandria, VA: American Counseling Association.

Newsome, S. (2004). Solution-focused brief therapy group work with at-risk junior high school students: Enhancing the bottom line. Research on Social Work Practice, 14, 336-343.

No Child Left Behind Act. (2001). 20 US.C. 6301 [section] 2002.

Roans, M., & Hoagwood, K. (2000). School-based mental health services: A research review. Clinical Child and Family Psychology Review, 3, 223-241.

Rubin, A.,& Babbie, E. R. (1997). Research methods for soda! work (3rd ed.). Pacific Grove, CA: Brooks/Cole.

Sklare, G. B. (1997). Brief counseling that works: A solutionfocused approach for school counselors. Thousand Oaks, CA: Sage Publications.

Springer, D.W., Lynch, C, & Rubin, A. (2000). Effects of a solution-focused mutual aid group for Hispanic children of incarcerated parents. Child and Adolescent Social Work, 17, 431- 442.

Walter,J. L., & PellerJ. E. (1992). Becoming solution-focused in brief therapy. New York: Brunner/Mazel.

Webb, W (1999). Solutioning: Solution-focused interventions for counselors. Philadelphia:Taylor & Francis.

Cynthia Franklin, PhD, is associate professor, and Kelly Moore, PhD, is a graduate, School of Social Work, University of Texas at Austin, 1925 San Jacinto, Austin, Texas 78712; e-mail: [email protected]. Laura Hopson, PhD, is assistant professor, School of Social Welfare, University at Albany, State University of New York.

Accepted January 12, 2007

Comments

comments