Using single-participant research to assess counseling approaches on children's off-task behavior.

Off-task behavior in today's students may pale in comparison to the seriousness of the violent student behavior reported almost daily in the media. However, Langdon (1997), in a Phi Delta Kappa Poll of Teachers' Attitudes toward the Public Schools, found that teachers reported two off-task behaviors as their most serious concerns. Schoolwork/homework assignments not completed was ranked first by 71% of the teachers followed by behavior that disrupts class which was checked 58% of the time by teachers in the survey. With off-task behavior being a major concern for teachers as negatively impacting learning, it would follow that research is needed to identify efficient interventions for remediating off-task behavior and enhancing student learning.

Reality therapy (Glasser, 1965, 1986; Meduna & Wigle, 1997) and solution-focused brief counseling (Mostert, Johnson, & Mostert, 1997; Sklare, 1997) are popular, short-term, counseling interventions school counselors use in counseling students for a variety of problems. Neither reality therapy (RT) nor solution-focused brief counseling (SFBC) have broad research support in the elementary school counseling literature for working with off-task student behavior. However, some related literature does exist supporting both counseling methods.

Edens and Smryl (1994) found that RT was an effective intervention in reducing disruptive behavior in middle school physical education classes, which are commonly known to teachers as some of the most difficult classes to manage. MacDonald (1989) offered a model of teaching RT control theory (now referred to as choice theory) to elementary school children using the analogy of controlling their behavior in the same manner they are able to control their shadows. By teaching children that their control over events in their life was much like the control that they are able to exert over their own shadows, children were able to conceptualize the concepts of RT. McDonald concluded that the language of RT may be modified for and practiced by elementary school children. Glasser (1998) developed choice theory to support the practice of reality therapy and to provide a theory to explain why and how reality therapy works with students and adults as well. Reisberg-Pollack (1985) recommended that certain phrases and concepts be repeated on a daily basis to help children develop a working RT vocabulary of self-talk to be used in times of crises or when they were tempted to misbehave. Hart-Hester, Heuchert, and Whittier (1989) found that disruptive behavior of children ages 9 to 11 significantly decreased after they were taught the principles of RT. They concluded that when children were taught to become independent and to rely on their own resources rather than those of the teacher and other students, they exhibited more responsibility and self-control. However, while the practice of RT is widespread, very little data on its efficacy with elementary school students are reported in literature, particularly as it is used with off-task behavior.

Solution-focused brief counseling (SFBC), as described by deShazer and Molnar (1964), was developed in the mid-1960s, but has only recently became a popular intervention for school counselors (Sklare, 1997). SFBC is an attractive intervention for school counselors because of its utility, focus on current behavior, and relative short-term duration (Mostert et al., 1997; Sklare). Downing and Harrison (1992) advocated incorporating SFBC into the elementary school because the model fits the "time-conscious atmosphere" that is beginning to define modern school counseling. LaFountain, Garner, and Boldosser (1995) advised school counselors to create solution-focused counseling groups from their current caseloads in order to free time for parent education and the other necessities of the job. They noted that the solutions approach in these groups may not carry the "stigma" effect that students might experience from individual, problem-centered counseling. LaFountain et al. noted that students should also benefit from learning about what works well for other group members in solving problems. Mostert et al. and Sklare reported that the SFBC methods provide students with the confidence they often lack, and that counselors, themselves, reported feeling more effective when using SFBC. However, even more so than was the case for RT, very little data are reported in the literature on using SFBC with off-task student behavior in the elementary school. Therefore, the purpose of the current study was to research the efficacy of RT and SFBC with elementary school students engaging in off-task behavior. A secondary purpose of the study was to demonstrate the utility of a single-participant design in conducting counseling research.

Method

Participants

Participants in this study were two children recruited from a suburban elementary school. The participants were selected based on the school counselor's recommendation, teacher reports, and willingness to participate. Child one was an 8-year-old African-American male who was in third grade at the time of the study. Child two was a 9-year-old Caucasian female enrolled in the fourth grade.

Child one came from a two-parent, intact family of lower to middle socioeconomic status. He was the second youngest of five children who ranged in age from 6 to 17. In initial interviews, he reported that his father had recently lost his job and did not spend a significant amount of time in the home. The mother seemed to be the primary caregiver as well as the breadwinner in the family. This child reported a close relationship with his mother and younger sibling but did not seem to possess the same sort of relationship with any of the other family members. An investigation of test data revealed that he achieved average scores in ability and achievement in all areas in the first grade. The child's math scores dropped one grade level in the second grade, but scores in all other areas remained in the average range. The same test score pattern was repeated in third grade. His teacher reported the child's presenting problems as exhibiting a variety of off-task behaviors such as daydreaming and doodling, a lack of interaction with the classroom teacher, and a failure to complete and turn in schoolwork. He possessed a pleasant disposition, a willingness to cooperate, good communication skills, and an eagerness for attention.

Child two was also a member of a two-parent, intact family of middle socioeconomic status. The child was the oldest in a two-sibling dyad, having a younger brother of age six. She reported a close relationship with both parents, but an unstable relationship with her younger brother. An investigation of test data revealed above average scores in ability and achievement in grades one through four. The above average scores showed capability in all subjects, with the highest scores being in reading and math. The teacher reported the child's presenting problems as a tendency to daydream and talk to friends in class rather than focusing on work and a general failure to complete and turn in assignments. The counselor found the child to possess a pleasant disposition, a willingness to cooperate, and general enthusiasm to participate in counseling.

Setting

The study took place in a suburban elementary school mainly serving families of middle socioeconomic status. All sessions with the counselor took place in a private office and were generally held in the morning while the children's classes were in music, art, or computer skills.

Procedure

Permission forms allowing the counselor to work with and videotape each child were sent home, and parents of both children signed them. Following the initial intake interviews, the counseling approach assignments were determined by drawing names from a box--solution-focused brief counseling was used with child one; reality therapy was used with child two. The counselor worked with each child over a 2-month period meeting with each child for one session, once a week. Sessions typically lasted between 20 and 40 minutes. The counselor, who is the senior author, was completing her master's degree in counseling at The University of Tennessee. She was enrolled in the second author's seminar in reality therapy and solution-focused brief counseling at the time the study was completed.

Specific goals related to classroom behavior and assignment completion were developed in cooperation with the children and their teachers. Goal attainment scales were used to track the students' progress relative to their baseline rates on completion of their assignments.

Goal-attainment scaling (Kiresuk, Smith, & Cardillo, 1994) is a system for tracking student progress toward meeting measurable counseling goals. Goal-attainment scaling requires setting five levels of outcome, ranging from a "-2" for the worst possible outcome to a "+2" for the best possible outcome for each counseling goal. The goals are weighted by the student on a scale from 10 to 30 depending on their relative importance to the student. A goal with a weight of 30 would be described as being three times more important for the student than a goal with a weight of 10. Kiresuk and Sherman (1968) developed a formula for measuring the significance level of student progress from the point of intake (baseline) to termination of counseling. They adopted a conventional T score scale with the mean set at 50 and standard deviation set at 10 for their goal attainment scale. Scores of 50 or better represent successful outcomes and are indications of at least one level of improvement on the goal attainment scale. Weekly progress is also graphed for each goal according to one of the five levels attained for that week. The goal attainment scales for each child are presented in Tables 1 and 2.

An AB design was used with goal attainment scaling which included repeated behavior measurements in a time-series format over the 8 weeks of the study. Baseline data were gathered during the first 3 weeks of the study followed by 5 weeks of the treatment phase. The baseline data became the intake or pre-treatment levels of performance on the goal attainment scale as noted in Tables 1 and 2. Weekly performance results during treatment were recorded on the goal attainment scale and calculated for each child. Initial T scores were calculated for each child's goal attainment scale at intake (baseline) and final T scores were calculated following termination. The following formula (Kiresuk et al., 1994) was used:

(T) = 50 + 10[SIGMA] [W.sub.i][X.sub.i]/ [square root of ((.7[SIGMA] [W.sup.2.sub.i]) + .3([SIGMA] [([W.sub.i]).sup.2])

[W.sub.i] = Weight value of each goal (10 to 30) [X.sub.i] = Scale values for each goal (-2 to +2)

According to Thompson and Rudolph (2000), goal attainment scaling has the advantage of providing numerical data and a method for analyzing that data for statistically significant differences or lack of same on the student's progress. Kiresuk et al. (1994) provided T score tables for those who wish to avoid calculating the scores.

Interventions

Child one received SFBC as described by Sklare (1997) and Thompson and Rudolph (2000). Child two received RT as described by Glasser (1965, 1998, 2000) and Thompson and Rudolph. During the 3-week baseline period, the counselor met with the students once a week for the purposes of constructing their goal attainment scales, building rapport, and collecting data on their current level of performance on each of their goals. The baseline period was followed by the 5-week counseling phase of the study.

The following steps were used by the counselor in her solution-focused brief counseling approach with child one:

1. Focused on solutions rather than problems.

2. Used scaling to determine how much of the problem had been solved. For example, the counselor asked, "On a scale of 1 to 10 where 10 is completely solved and 1 is completely not solved, which number is your problem today?" If the student reported that it was a four, the counselor asked the student, "What keeps it from being a one?" The counselor's objective was to help the student do more of the things that were currently going well for him.

3. Used the miracle question to determine what the student would be doing differently if the problem suddenly were solved overnight.

4. Focused on exceptions when the miracle happened a little bit.

5. Used positive blame, "How in the world did you make that happen?"

6. Aimed for a 10% improvement each week. The student was asked, "What would have to happen to have you go from a 5 to 6 on a 10 point scale where 10 is what it would be like if the miracle happened?"

7. Flagged the minefield. The student was asked, "What are the things that could stop you from going from 5 to 6 this week?" or "What things could blow up your plan?"

8. Reinforced the things the student was already doing to make the miracle happen.

9. Inquired about the student's motivation to work on solving the problem. The student was asked, "On a scale of 1 to 10 where 1 represents not willing to do anything to solve the problem to 10 which represents doing whatever it takes to solve the problem, what number are you today?"

10. Closed each session with a written note from the counselor to the student bridging the students assets and strengths to the tasks to be accomplished during the coming week.

The following reality therapy steps were used by the counselor in her reality therapy counseling approach with child two:

1. Focused on what the student was doing rather than on why she was doing it.

2. Had the student evaluate how what she was doing was helping or hurting in getting what she wanted.

3. Asked the student if she would be willing to stop doing things that were not helping her get what she wanted.

4. Asked the student to brainstorm about things she might like to try instead to solve the problem or get what she wanted.

5. Had the student commit to trying some of the new behaviors for one week and report on how well they worked.

6. Focused on logical consequences rather than punishment when the student's commitments were not kept.

7. Focused on her responsibility for meeting her own needs.

8. Focused on the fact that the only behavior the student could control was her own and that she had a choice in what behaviors she choose.

Each child received weekly counseling for a period of 5 weeks. Counseling sessions were audio- or videotaped. Both authors listened to or viewed the tapes and compared them to the counselor's written case notes to ensure that the steps of RT and SFBC were followed correctly with the student for which each method was assigned.

Results

The results for child one, who received SFBC, are presented in Table 3. Child one presented an initial T score of 26.95 (2.3 SD below the mean) during the baseline period. His T score improved to 65.96 (1.6 SD above the mean) following treatment. He showed a general improving trend, with a minor setback during the fourth session. He progressed from turning in no assignments or just one or two assignments to submitting all weekly assignments in three of his four school subjects. As noted in Table 1, he moved from completing two to completing four weekly assignments in the fourth subject.

The results for child two, who received RT, are presented in Table 4. She showed a general trend of improvement with each counseling session. Her T scores ranged from a low of 23.25 (intake score, 2.67 SD below the mean) to a high of 75.15 (post treatment score, 2.5 SD above the mean). She progressed from no assignments completed to all assignments completed in three of her four subjects. As shown in Table 2, she moved from completing one to two assignments to completing three weekly assignments in a journal she was writing for one class.

Post treatment scores of 50 or better are considered to be significantly successful outcomes (Kiresuk & Sherman, 1968). In addition, scores exceeding one standard deviation above the mean of 50 are considered extremely successful outcomes.

In summary, the results offer support to both methods of counseling for working with off-task behavior. The initial baseline scores of both children were well below the mean score of 50 for goal attainment scaling. Each method enabled the children to set specific goals, track their progress, and experience a positive change in behavior and in their classroom performance. While child two's final T score (75.15) for RT was nearly one standard deviation higher than the final score of child one's T score (65.96) for SFBC, both treatments were successful. The school counselor conducted a follow-up interview with the students' teachers during the fall semester of the following year. Both children were continuing to maintain the goals they had attained during the study.

Discussion

The purpose of this study was to utilize single-participant research to access the efficacy of solution-focused brief counseling and reality therapy with elementary school students engaging in off-task behavior. Significant positive changes in the on-task behaviors of the students resulted from both approaches. Both students responded well to the counseling approach they received.

The study was limited to two participants who each received five counseling sessions. This limitation raises the question of the reliability and validity of using single-participant designs in counseling research. However, Morgan and Morgan (2001) made a very strong case for using single-participant research for evaluating and researching counseling effectiveness. They view single-participant research as featuring a design power and flexibility well suited to both science and applied research. Noting that Pavlov, Piaget, Ebbinghaus, and Skinner used single-participant research in their work, Morgan and Morgan listed several advantages of the method. First, the method provides a means for counselors to document their effectiveness with individual students with individual problems not lending themselves to large group research designs. Second, participants can serve as their own control with repeated measures of their target behavior being taken during baseline, treatment, and follow-up periods. Third, single-participant designs are best suited to projects in which the unfolding of certain target behaviors in the student is the focus of the counseling intervention. Fourth, interventions can be compared on the same student and across different settings such as school and home. In addition, the incorporation of goal attainment scaling, with its before and after treatment data comparisons, strengthens the single-participant design for counseling research.

A second limitation of the study would be the difficulty in identifying which steps of both counseling approaches made the most impact on the student. In fact, the counseling relationship, itself, may have made the difference in the students' performance. However, no changes in the two students' baseline rate of off-task behaviors (failure to complete assignments) were found during the 3-week period when baseline data were gathered and rapport was being established with the students prior to the 5-week treatment period. Regarding the question of which step made the most impact on the students, both students made immediate improvement in completing their assignments with the introduction of the treatment phase. There were no indications that any one step in either method was more influential than another in motivating the students to complete their assignments. It is the authors' belief that both RT and SFBC are systems with steps that build on each other as counselors work with their students in meeting the goals their students are striving to attain. This is not to say that breakthroughs in counseling are never keyed by a particular step in the process. Glasser (1965) noted that people never change their behavior unless they have concluded that what they are currently doing is either not working or even hurting them in getting what they want. The evaluation of one's own behavior is a key step in RT. de Shazer and Molnar (1964) cited the miracle question in SFBC as often being the breakthrough step in bringing about change in their clients. It could be concluded that the most important step in both counseling methods was simply getting started.

Three implications are offered which school counselors might draw from the research. First, both RT and SFBC are short-term counseling interventions which can be used to provide some comparatively rapid relief to students who have problems with off-task behavior in completing their school assignments. Second, the research offers school counselors a single-participant design for evaluating and documenting the impact of counseling. Finally, goal attainment scaling was described as an additional way to evaluate how well counseling is working for counselors in helping students meet their goals. School counselors may wish to graph each week's goal attainment results to provide a visual aid in helping students to evaluate their progress. Some elementary school counselors use "thermometers" drawn on white boards with erasable markers to chart goal attainment progress on a daily or weekly basis. In fact, further research could be focused on the goal attainment scale, by itself, as the treatment variable in working with off-task behavior.

Table 1. Goal Attainment Scaling for Child One

     Spelling                   English
X    W = 30                     W = 20

-2   Completes 0                Completes 0
     assignments *              assignments

-1   Completes 1 to 4           Completes 1 or 2
     assignments *              assignments *

0    Completes 1/2 of           Completes 3
     assignments (5)            assignments

+1   Completes 6 to 8           Completes 4
     assignments [check]        assignments [check]

+2   Completes all              Completes all
     assignments                5 assignments

     Math                       Handwriting
X    W = 20                     W = 1

-2   Completes 0                Daydreams, doesn't
     assignments                start assignment

-1   Completes 1 or 2           Starts assignment,
     assignments *              but completes 1/4 *

0    Completes 3                Completes 1/2
     assignments                of assignment

+1   Completes 4                Completes 3/4
     assignments                of assignments

+2   Completes all              Completes all
     5 assignments [check]      assignments [check]

Note.

X = scale value.

W = weight value.

* = intake level of behavior (baseline)

[check] = termination level of behavior.
Table 2. Goal Attainment Scaling for Child Two

     Reading                    Math
X    W = 30                     W = 20

-2   Daydreams, doesn't         Daydreams, doesn't
     finish any work *          finish any assignments *

-1   Completes 1 question       Completes 2
     or 1/4 of assignment       assignments

0    Completes 1/2              Completes 3
     of assignment              assignments

+1   Completes 4 questions or   Completes 4
     1/2 of assignment          assignments

+2   Completes entire           Completes all
     assignment [check]         5 assignments [check]

     Composition                English/History
X    W = 10                     W = 30

-2   Does not write             Daydreams, does not
     any entries                complete any assignments *

-1   Writes 1                   Completes 1
     entry *                    assignment

0    Writes 2 or 3              Completes 2 or 3
     entries [check]            assignments

+1   Writes 4                   Completes 4
     entries                    assignments

+2   Writes all 5               Completes work [check]
     journal entries

Note.

X = scale value.

W = weight value.

* = intake level of behavior (baseline)

[check] = termination level of behavior.
Table 3. Weekly Results for Child One Receiving
SFBC

Week   z score   T score

1       -2.30     26.96
2       -1.95     30.49
3        0.71     57.09
4       -0.20     48.23
5        1.60     65.96
Table 4. Weekly Results for Child Two Receiving
RT

Week   z score   T score

1       -2.67     23.25
2       -2.20     27.97
3        1.10     61.01
4       -2.04     70.45
5        2.50     75.15

Editor's Note: Dr. Charles Thompson was recognized as the 2001 Writer of the Year by the American School Counselor Association. He received the award at the Annual Conference in Portland, OR. The following was presented in the Recognition Celebration Program:

Dr. Thompson contributed to various publications including six books and more than 100 professional articles. For 20 years, he served as the editor of the Idea Section of Elementary School Guidance and Counseling. His most recent publication, Counseling Children (5th ed.), is being translated into Chinese and Norwegian. Dr. Thompson was an early pioneer in the development of state funding for elementary and middle school counselors in Tennessee. His background includes teacher, basketball coach, school counselor, teacher's aide and resource in his grandchildren's class, psychologist, and counselor educator. Dr. Thompson has a distinguished record as a contributor to the professional literature in school counseling.

Dr. Thompson--Congratulations and thank you for your contributions to the profession.

References

de Shazer, G., & Molnar, A. (1964). Useful interventions in brief family therapy. Journal of Marital and Family Therapy, 10, 297-304.

Downing, J., & Harrison, T. (1992). Solutions and school counseling. The School Counselor, 39, 327-332.

Edens, R. M., & Smryl, T. (1994). Reducing disruptive classroom behaviors in physical education: A pilot study. Journal of Reality Therapy, 13(1), 40-44.

Glasser, W. (1965). Reality therapy. New York: Harper & Row.

Glasser, W. (1986). Control theory in the classroom. New York: Harper & Row.

Glasser, W. (1998). Choice theory: A new psychology of personal freedom. New York: Harper Collins.

Glasser, W. (2000). Reality therapy in action. New York: Harper Collins.

Hart-Hester, S., Heuchert, C., & Whittier, K. (1989). The effects of teaching reality therapy techniques to elementary students to help change behaviors. Journal of Reality Therapy, 8(2), 13-18.

Kiresuk, T., & Sherman, R. (1968). Goal attainment scaling: A general method for evaluating comprehensive community mental health programs. Community Mental Health, 4, 443-453.

Kiresuk, T., Smith, A., & Cardillo, J. (1994). Goal attainment scaling: Applications, theory, and measurement. Hillsdale, NJ: Erlbaum.

LaFountain, R., Garner, N., & Boldosser, S. (1995). Solution-focused counseling groups for children and adolescents. Journal of Systemic Therapies, 14, 39-51.

Langdon, C. (1997). The fourth Phi Delta Kappa poll of teachers' attitudes toward the public schools. Phi Delta Kappan, 79, 212-221.

MacDonald, A. (1989). Me and my shadow: Teaching "control theory" in elementary school. Journal of Reality Therapy, 8(2), 30-32.

Meduna, M. J., & Wigle, S. E. (1997). Do they work and what are the consequences of their use? Two questions to ask about external motivators. International Journal of Reality Therapy, 17(1), 42-45.

Morgan, D. L., & Morgan, R. K. (2001). Single-participant research design: Bringing science to managed care. American Psychologist, 56, 119-127.

Mostert, D. L., Johnson, E., & Mostert, M. P. (1997). The utility of solution-focused, brief counseling in schools: Potential from an initial study. Professional School Counseling, 1(1), 21-24.

Reisberg-Pollack, Z. (1985). The practical application of control theory in the classroom. Journal of Reality Therapy, 5(1), 19-21.

Sklare, G. B. (1997). Brief counseling that works: A solution-focused approach for school counselors. Thousand Oaks, CA: Corwin.

Thompson, C. L., & Rudolph, L. B. (2000). Counseling children. Pacific Grove, CA: Wadsworth/Brooks/Cole.

Jamie L. Yarbrough is a doctoral student in School Psychology, and

Charles L. Thompson, Ph.D., is a professor of Counselor Education. Both are with the Educational Psychology and Counseling Department, The University of Tennessee, Knoxville.