ABSTRACT:
Three adolescents (ages 14-17) with emotional and behavioral disorders displayed chronic disruptive behavior in their self-contained classrooms at a self-contained alternative school. A descriptive functional behavioral assessment was conducted for each student. Data from file review, structured interviews, and direct observations were used to identify the functions of their disruptive behaviors. Then, function-based interventions were systematically constructed for each student and implemented for an extended period (nearly 6 weeks) within the most problematic situation in their classrooms. The interventions improved each student's behavior and the effects maintained during follow-up and generalized to instruction in a nonintervention classroom. Social validity data comparing the interventions to baseline practices revealed the function-based intervention had moderately higher social validity among teachers and substantially higher social validity among students.
Alternative schools began to emerge in the 1960s with a goal of providing education to students who could not "succeed" in regular school settings (Raywid, 1999). The National Center for Educational Statistics (NCES) reported there were 3,850 public alternative schools in the 1997-1998 academic year (NCES, 2002). By 2002, the number had nearly tripled to at least 11,000 public alternative schools (Kleiner, Porch, & Farris, 2002). Chronic behavioral difficulty was the primary reason for placement in an alternative setting (Kingery, 2000, 2001; Kleiner et al., 2002; Skiba & Peterson, 1999).
The number of students with emotional and behavioral disorders (EBD) who are educated in a separate facility is three times higher than that of students with other disabilities (U.S. Department of Education, 2006). Increasing numbers of these students are being placed in alternative schools (Cox, 1999; Franklin, 1992; Gregg, 1999; Lehr, Moreau, Lange, & Lanners, 2004; Tobin & Sprague, 2000). Currently, more than 20% of students with EBD are placed outside of their mainstream school, and greater than 52% spend at least 60% of their time outside of a regular educational setting.
Before placing a student with a disability in an alternative educational setting, federal law (Individuals with Disabilities Education Act, 1997; Individuals with Disabilities Education Improvement Act, 2004) requires that a functional behavioral assessment (FBA) must be conducted. Functional behavioral assessment is a method for gathering information that identifies the function of a problem behavior and the events that predict when it will occur (Sugai et al., 2000). Specifically, the FBA identifies (a) the antecedent conditions present when the behavior occurs and when it does not occur and (b) the consequences that maintain the behavior. Function-based intervention refers to the development of behavior change strategies, based on the FBA data, which directly address the problem behavior's function.
Considerable research (see reviews by Fox, Conroy, & Heckaman, 1998; Fox & Gable, 2004; Heckaman, Conroy, Fox, & Chait, 2000; Kern, Hilt, & Gresham, 2004; Lane, Karlberg, & Shepcaro, 2009; Sasso, Conroy, Stichter, & Fox, 2001; Wood, Umbreit, Liaupsin, & Gresham, 2007) has pointed to the effectiveness and durability of function-based intervention. These studies have been conducted with students and adults of all ages, with individuals with various disabilities, in a variety of settings, and with various problematic behaviors. Ironically, only one study (Turton, Umbreit, Liaupsin, & Bartley, 2007) has reported on the effectiveness of function-based intervention in supporting adolescents with EBD once they are in an alternative setting. Turton and colleagues conducted an FBA and then used a function-based intervention package to reduce a student's use of profanity and increase her use of appropriate social responses during instructional activities. Social validity ratings completed by multiple classroom staff, the student, and a student peer were positive.
The purposes of this study were to further examine the efficacy of function-based intervention to support adolescents with EBD placed in an alternative educational setting, to expand the data collected to include both generalization and maintenance probes, and to assess the social validity of the interventions. The study was conducted in two phases. In Phase 1, descriptive FBAs were conducted for each student. Each FBA included file review, structured interviews, and direct observations to identify the functions of target behaviors. In Phase 2, function-based interventions were systematically constructed for each participant and then implemented in the classroom during the activity the teacher identified as being the most problematic. Data were collected daily for nearly 6 weeks (29 sessions) and for 3 days during a single week 3 weeks after intervention concluded. Generalization probes were conducted weekly in nonintervention classes once each student's intervention began and during the maintenance sessions.
General Method
Participants and Setting
The setting was a self-contained Pre-K-12 alternative day school program for students with EBD, mild or moderate disabilities, or autism. Severe challenging behavior was the common denominator underlying each student placement. The school's stated mission was to foster development of appropriate behavioral coping skills, relevant academic knowledge, and proper motivation to facilitate a successful return to a comprehensive campus. All classrooms covered the academic areas typical for the grade level. The difficulty level of the work was often modified because most students were behind academically, presumably because of their behavioral problems, which interfered with the delivery of instruction. Each classroom was staffed by a teacher and an instructional aide.
The school employed a school-wide system in which students earned points for meeting behavioral expectations. Points were exchangeable for special privileges such as participating in "Fun Friday." Once students met behavioral criteria determined by their individualized education plan and the appropriate level of points, they were considered ready to begin phasing back into a comprehensive neighborhood school campus. This was done slowly, one subject (e.g., math), day (e.g., Mondays), or specific teacher at a time, beginning with mainstreammg within the alternative school (moving from being in one class all day to different classes according to the subject), and then progressing to mainstreaming to classes at the home school, until the student was fully transitioned.
Participants were three White boys and their respective teachers (two White women and one White man). Tahir, age 1 5, attended a 9th-grade class of 22 taught by Ms. A, who had 1 year of teaching experience. Cain, age 17, attended a 12th-grade class of 8 taught by Mr. B, who had 5 years of teaching experience. James, age 1 4, attended an 8th-grade class of 7 taught by Ms. C, who had 8 years of teaching experience. All three teachers had degrees in special education and were certified to teach students with EBD. Their preparation had included coursework on FBA and managing challenging behavior in classrooms.
These participants were selected through a series of phases. First, the first author met with all teachers and staff who worked at the high school level to identify all students who remained resistant to the school's point system and had received an average of at least one office referral per week. Second, each teacher identified his or her most difficult student. Finally, the teachers and staff met again to determine, from their perspective, the feasibility of participation; this process resulted in identifying for this study the only three high school students who were not participating in any level of mainstreaming.
Behavioral Definitions
All three students exhibited behaviors that were considered to be disruptive to their learning or that of others. The topography of each student's behavior varied. "Disruptive behavior" was defined globally as a group of behaviors that interfered with a student's access and his peers' access to instruction. For Tahir, disruptive behaviors included making inappropriate comments, particularly sexual comments, about himself or others. For Cain, disruptive behavior included refusing to comply with directions (e.g., saying "No" or just continuing with a different activity), throwing books or chairs, banging on desks or walls, and screaming profanities. For James, disruptive behaviors included using profanity and making verbal threats. For all three students, the replacement behavior was on-task, defined as sitting in his seats, engaging in the assigned task, and asking for help appropriately (raising his hand, waiting, stating what he needed, e.g., "I don't understand how to do this math problem").
Phase 1: FBA
Procedure
A descriptive FBA was conducted for each student to identify the antecedent conditions that set the occasion for his target behaviors and the consequences that maintained these target behaviors. Data were collected via file review, staff and student interviews, and direct observation in the classroom. These data were then analyzed to identify the function(s) of each student's target behaviors and the most critical situations in which interventions should occur.
File Review
A comprehensive file review was conducted for each student. Specific information reviewed included prior record of discipline referrals and suspensions, daily points earned for positive interactions, time spent out of class in time-out or in-school suspension), anecdotal comments from current or prior teachers, school attendance history, and informal and formal academic assessment results.
Staff Interviews
Structured interviews were completed with the classroom teacher and instructional aide in each class. Interviews followed the Preliminary Functional Assessment Survey (Dunlap et al., 1993), a 22-item survey developed to solicit important information about functional relationships between the environment and student behavior. The survey also solicits information about medical conditions that may affect behavioral regulation, the impact of distal antecedent events (e.g., missed breakfast, conflict with family or peer), and rough estimates of the frequency and duration of the behavior.
Structured Student Interview
The Student Assisted Functional Assessment Interview (Kern, Dunlap, Clarke, & Childs, 1994) was conducted with each student. This instrument solicits information about what the student believes triggers and maintains his inappropriate behaviors, when he has the most and the least problems in school, what he thinks causes these problems, and how he thinks the situation could be changed for the better. The student also rates preferences for specific subjects and his own skill level relative to the work assigned. Finally, the child responds to questions about working conditions, teacher presentation, and interest level in the work.
Structured Observations
Antecedent-behavior-consequence (ABC) data (Bijou, Peterson, & AuIt, 1968) were collected individually for each participant on four occasions. Each observation lasted approximately 20 min and occurred in the student's classroom during naturally occurring activities in which the target behavior was most likely to occur.
Identification of Function
The function(s) of each student's disruptive behavior was identified by analyzing the FBA data using the Function Matrix (Umbreit, Ferro, Liaupsin, & Lane, 2007), a six-celled visual tool that organizes information into two columns identifying positive or negative reinforcement and three rows identifying specific types of consequences. The tool prompts users first to determine if the student is gaining access to something (positive reinforcement), escaping/avoiding something (negative reinforcement), or both. The user then identifies more specifically whether the student is gaining or escaping attention, tangibles/activities, or sensory consequences. The Function Matrix allows the identification of multiple functions for a single target behavior. In this study, the classroom teacher and first author jointly identified function by reviewing all of the interview results and ABC data with respect to the Function Matrix.
Results
Results of the file reviews, interviews, direct observations, and the Function Matrix analyses are presented for each student.
Tahir
File review indicated that Tahir's teachers believed his poor academic achievement was not due to ability but rather to the disruptive behaviors he displayed. At the end of the previous school year, he had started the mainstreaming process in which students gradually move from teacher to teacher and subject to subject. During the summer, a traumatic event triggered a relapse of inappropriate behaviors, which led the team to reconsider placement and start him back within the self-contained classroom.
The teacher and aide were most concerned about Tahir's inappropriate comments that got other students involved in a back and forth argument or led a peer to become aggressive. Tahir also made false accusations about staff that required administrative investigations. They believed his sexual references made it easier to get a reaction from others. Tahir rarely earned enough points to participate in Fun Friday and spent much of each day in the classroom time-out area or in a study carrel away from class activity. Identified reinforcers included one-on-one time with a staff member, computer time, or a visit to other classrooms.
Tahir reported that his classwork was sometimes challenging, but usually not and that he was only sometimes rewarded for good behavior. He felt he could work better with someone assisting him, like the full-time aide he had in his previous school.
During the four ABC data collection sessions, Tahir was on-task an average of 26% (range = 0%-40%). Instances of disruptive behavior occurred when (a) a new person entered the room, (b) the teacher asked the class a question, or (c) another student was called to the office or called on to answer a question. When these antecedents occurred, Tahir would (a) get out of his seat, (b) make an inappropriate comment to a peer or staff member, and (c) make repeated comments if he did not get a reaction. In every case, he gained teacher or peer attention.
Using the Function Matrix, the teacher and first author jointly decided that the staff interviews and ABC data indicated Tahir made inappropriate remarks to gain attention from peers and teachers.
Cain
File review indicated Cain's challenging behaviors had increased over the 8 weeks prior to the beginning of the study. During his 5 years at the school, he had improved in some academic areas. Cain often had episodes in which he "lashed out," he claimed, because others were talking about him. He spent considerable time either in the classroom time-out area or in in-school suspension, either by self-initiation or teacher request.
Staff were most concerned about Cain's refusal to follow directions and his violent outbursts. Typical approaches to addressing these behaviors had been to ignore them or to suggest that he go to the time-out area or take a walk. Staff believed his behavioral problems were increasing both in intensity and frequency, and that his target behaviors always occurred when he was challenged in any way or lost points on his point sheet. Reported reinforcers included listening to music, using the computer, and eating candy or food during class time.
Cain was happy to be interviewed. Once questioning began, he would answer about four questions before speaking on a different, often unrelated topic. Cain reported his work was sometimes challenging and sometimes easy, and that classes were occasionally too long. He stated he always received help when requested and did better when he worked together with someone. He also said he was distracted by classroom activity such as side conversations and the activities of the other students.
During the ABC data collections sessions, Cain was on task an average of 1 8% (range = 5%-40%). Disruptive behavior occurred when the teacher (a) began instruction (e.g., "Please turn to chapter 7''), (b) told Cain to get back to work, or (c) told Cain he might not receive his points for appropriate behavior during the class session. When these antecedents occurred, Cain would get out of his seat, engage in an alternate activity, bang on the wall, use profanity, or accuse the teacher of picking on him. In every case, he avoided further engagement with the assigned task and usually also gained attention from both staff and peers.
Using the Function Matrix, the teacher and first author jointly determined that Cain's disruptive behaviors occurred consistently following teacher directives and enabled him to both escape the task at hand and gain attention from staff and peers.
James
File review indicated that James was in his first year at the school. At the time of the study, he attended Ms. C's homeroom class. He had previously thrown furniture and threatened self-harm but had shifted to using profanity and making verbal threats to others. James lived at home with his mother and little brother and reportedly displayed similar behavioral problems at home.
Staff described several behaviors of concern, including refusal to comply with explicit directives and putting his head down and closing his eyes during instruction. They were most concerned about his use of profanity and threats, which were most likely when the teacher held him accountable for his actions or insisted on task completion.
James reported that he did not receive help when he requested it, that class periods were too long, and that no one noticed when he did a good job or awarded him the appropriate amount of points. He thought math was interesting, but the work was not challenging enough. James' favorite activities at school were using the computer and talking with friends.
During ABC data collection, James was on-task an average of 7% (range = 0%-20%). Disruptive behavior occurred when the teacher (a) started instruction, (b) asked James to answer a question, or (c) told James he might not earn points for appropriate behavior and would have to complete his assignment after class or during lunch. When these antecedents occurred, James would get up and walk around the room, accuse the teacher of lying, and make verbal threats to anyone nearby. In every case, he avoided further engagement with the assigned task and received attention from peers or staff.
Using the Function Matrix, the teacher and first author jointly determined that the interview and ABC data indicated James engaged in disruptive behavior to simultaneously avoid doing his work and gain attention. This occurred when the teacher requested he do a specific task or redirected him with a reminder of the consequences.
Phase 2: Function-Based Intervention
In Phase 2, the FBA data were used to design a function-based intervention for each participant (Tahir, Cain, and James). The resulting interventions were then implemented for several weeks (29 sessions) during each student's most problematic class.
Procedure
Function-based interventions for each participant were developed using the systematic process described by Umbreit et al. (2007). This process begins intervention development by posing two questions: (a) Can the individual perform the replacement behavior? and (b) Do the antecedent conditions represent effective practice? The answers to these questions lead to four possible outcomes. Each outcome identifies which of three intervention methods, individually or in combination, is appropriate for a given situation.
If the individual cannot perform the replacement behavior but the antecedent conditions represent effective practice, then Method 1 : Teach the Replacement Behavior is used. If the individual can perform the replacement behavior, but the antecedent conditions do not represent effective practice, then Method 2: Improve the Environment is used. If the answer to both questions is No, then both methods must be applied. Finally, if the answer to both questions is Yes, then Method 3: Adjust the Contingencies is used.
Each intervention method has three common components: antecedents are adjusted to increase the likelihood of the replacement behavior, and reinforcement is provided when the replacement behavior occurs and is withheld (extinction) when the target behavior occurs. The intervention methods differ in the ways specific antecedent and consequent variables are manipulated to address the presenting problems.
Intervention Design
In every case, the teacher, student, and first author jointly developed the intervention. Teachers helped to answer the key questions, suggested specific antecedent changes, and helped to identify reinforcement strategies that were functionally appropriate, yet also feasible and practical within their classrooms. Students participated by reviewing and approving their proposed intervention prior to implementation.
Tahir
Tahir could perform the replacement behavior - he could remain seated and engage in the assigned tasks for extended periods when he had the teacher's attention (proximity) or when asked a question. When considering classroom practices, the teacher's expectations were clearly conveyed and reinforcement was available for meeting them, the classroom routines were well established, the lessons and materials were appropriate for the subject, and the classroom arrangement was appropriate. However, the classroom schedule and instructional pacing often set the occasion for disruptive behavior. Specifically, unstructured time within lessons allowed Tahir to listen to others' conversations and make passing comments. Using this information, the first author and teacher determined that Method 2 (Improve the Environment) was most appropriate for improving Tahir's behavior.
At the start of each class, students were supposed to spend a few minutes preparing for the upcoming instructional activity. This created a situation in which Tahir's disruptive behavior was likely to occur and be reinforced. To correct this problem, Tahir was given a specific task to help the teacher get ready for class (e.g., getting and passing out handouts or setting up a DVD to be used for instruction). Instruction began immediately afterward. Because attention was the function of Tahir's behavior, attention (e.g., notes, comments, proximity, free time to socialize, lunch with teacher) was provided when he engaged in the replacement behavior and withheld (after brief redirection) whenever the target behavior occurred.
Cain
Cain lacked the social skills needed to interact appropriately with others. Although many of the classroom practices were appropriate, he often had difficulty finding the right page in his book and worked better with a partner, but rarely had the opportunity. Therefore, the first author and teacher decided that Methods 1 (Teach the Replacement Behavior) and 2 (Improve the Environment) were both needed. .Table 2 lists the intervention components they developed.
To develop more appropriate social skills, Cain met two to three times per week with the school counselor to develop and practice these skills. To improve the antecedent conditions, the book was already opened to the right page before the lesson and a partner (initially the teacher, and later a peer) worked with Cain after he was on-task for 15 min. When the replacement behavior occurred, he received attention (e.g., praise) and then a break, that is, free time to listen to music. These reinforcers addressed the attention and escape/activity functions, respectively. When the target behavior occurred, Cain was briefly redirected and the task demand was maintained. These extinction procedures addressed the same functions.
James
James frequently completed assignments accurately and on time and often showed sustained attention to assignments. In addition, the behavioral expectations were clear and reinforced, the classroom arrangement, routines, lessons, and materials were appropriate, and instructional delivery was steady and at the proper instructional level. Because James could perform the replacement behavior and the antecedent conditions reflected effective practice, the first author and teacher determined that Method 3 (Adjust the Contingencies) should be used for intervention.
To make it more likely the replacement behavior would occur, James was given a reminder of the new contingencies for appropriate behavior at the start of class. When the replacement behavior occurred, he received attention (e.g., praise) and, after 1 5 min on-task, one-to-one support. These consequences addressed the attention function. James also received a break via a "choice" card he could exchange for various preferred activities. This addressed the escape/activities function. Instances of the target behavior received two extinction procedures - brief redirection (to address the attention function) and continuation of the task at hand (to address the escape function).
Behavioral Definitions and Measurement
The behavioral definitions used in Phase 1 were used again in Phase 2. Data on the replacement behavior (on-task) and treatment integrity were collected for every session. Each session lasted 20 min and was conducted during the activity each teacher had identified as the most problematic. Generalization probes lasting 15 min were also conducted once per week during similar activities in a separate class.
On-task behavior was measured using a 30s whole-interval recording method. Treatment integrity data were collected using the same 30s whole-interval procedure. Specifically, at the end of each interval, each observer scored a "+" if all of the required intervention components were correctly implemented throughout the entire interval. If staff failed to implement any required part of the intervention at any point during an interval, it was scored as a "-." For more information on the collection of treatment integrity via whole-interval recording, see Umbreit et al. (2007) or Wood, Umbreit, Liaupsin, and Gresham (2007).
Design
A multiple baseline design across students (Kazdin, 1982) was used. All participants began in baseline at the same time after the FBAs were completed. However, the individualized interventions were introduced at different points in time for each student, that is, on Day 5 for Tahir, Day 10 for Cain, and Day 15 for James. Data were collected 5 days per week for nearly 6 weeks (29 sessions total) in Tahir's and James's social studies classes and in Cain's language arts class (i.e., those identified by each teacher as the most problematic). Generalization probes in other classes (language arts for Tahir and science for both Cain and James) were collected weekly once intervention began for each student. Follow-up data, including generalization probes, were collected for three sessions for Cain and Tahir, and two for James (due to absence) during a single week, 3 weeks after intervention concluded.
Interobserver Agreement (IO A)
Interobserver agreement data were collected by having a second observer independently record data on the replacement behavior and treatment integrity. Interobserver agreement data were collected in each condition for each student. Depending on the student, IOA data were collected for 33%50% of the baseline sessions, 38%-71 % of the intervention sessions, 33%-100% of the maintenance sessions, and 33%-100% of the generalization probes.
Interobserver agreement was assessed using the exact interval-by-interval method (Kazdin, 1982). Each interval scored identically was considered an agreement. Interobserver agreement was calculated by dividing the number of agreements by the total number of intervals and multiplying by 100%.
For the replacement behavior, IOA averaged 95% (range = 80%-100%) for Tahir, 90% (range = 75%-100%) for Cain, and 90% (range = 80%-1 00%) for James. For treatment integrity, IOA averaged 94% (range = 90%100%) for Tahir, 95% (range = 90%-100%) for Cain, and 94% (range = 85%-100%) for James.
Social Validity
Each teacher and student assessed social validity before and after intervention. The first administration occurred at the start of baseline and assessed the conditions in effect during baseline. The second, conducted at the end of the intervention phase, assessed the functionbased intervention. On each occasion, each teacher independently completed the Behavior Intervention Rating Profile (BIRP; Martens, Witt, Elliott, & Darveaux, 1985). The BIRP includes six questions that address whether an intervention targets an important goal, is warranted and reasonable within the classroom, and is likely to or did improve behavior. Each item is rated on a 6-point Likert-type scale ranging from 1 (strongly disagree) to 6 (strongly agree). Scores range from 6 to 36, with higher scores indicating higher social validity. Each student independently completed the Children's Intervention Rating Profile (CIRP; Witt & Elliott, 1985), a 7-item instrument that addresses whether the intervention was fair, helpful, acceptable, appropriate with other students, and likely to cause problems with the target student's friends. Each item is rated on a 6-point Likert-type scale ranging from 1 (strongly disagree) to 6 (strongly agree). Scores range from 7 to 42, with higher scores indicating higher social validity.
Results
Figure 1 shows that each student's on-task levels were low during baseline, increased during intervention, maintained during the follow-up period, and generalized to the secondary class setting. Specifically, Tahir's on-task behavior averaged 28% (range = 10%-50%) during baseline, increased to an average of 93% (range = 70%-100%) during intervention, and maintained at an average of 97% (range = 90%-100%) during follow-up. During generalization probes, the average was 92% (range = 80%-100%). There were no overlapping data points between the baseline and intervention conditions.
Cain averaged 24% (range = 0%-60%) during baseline compared to 90% (range = 75%-100%) during intervention, 97% (range = 90%-100%) during follow-up, and 94% (range = 75%-100%) during the generalization probes. There were no overlapping data points between baseline and intervention.
James averaged 11% (range = 0%-37%) during baseline compared to 86% (range = 48%-100%) during intervention, 95% (range = 90%-100%) during follow-up, and 100% during the generalization probes. There were no overlapping data points.
Treatment integrity (see Figure 1) during baseline was 0% for all teachers, indicating the interventions were not implemented. During intervention, Ms. A averaged 94% (range = 90%-100%), Ms. B 91% (range = 80%100%), and Ms. C 94% (range = 85%100%). During follow-up, teachers independently maintained these levels at 87%, 97%, and 89%, respectively.
The teachers' social validity ratings on the BIRP averaged 30.3 (of 36; range = 28-33) for the baseline conditions and 32.7 (range = SI36) for the intervention. The students' ratings on the CIRP averaged 27 (of 42; range = 2430) for the baseline conditions and 36.7 (range = 35-40) for the intervention.
Discussion
Systematically constructed function-based interventions were implemented with three adolescents with EBD who were placed in an alternative educational program because of their challenging behavior. The interventions improved the behavior of each student, and the effects maintained during follow-up and generalized to instruction in a nonintervention classroom. Assessments conducted at the start of baseline and at the end of intervention revealed the function-based intervention had moderately higher social validity among teachers and substantially higher social validity among students.
This study contributes to the literature in several ways. First, it adds to the broader literature supporting the efficacy and social validity of function-based intervention. Second, it adds to the literature on the effectiveness of function-based intervention in supporting adolescents with EBD once they have been placed in an alternative educational setting. Third, it extends the work of Turton et al. (2007) by providing additional data on both maintenance and generalization. Finally, it provides additional data supporting the effectiveness of the systematic process for designing function-based interventions that was described by Umbreit et al. (2007). The findings here are consistent with previous studies examining this process (e.g., Lane, Rogers, et al., 2007; Lane, Weisenbach, Phillips, & Wehby, 2006; Lane, Weisenbach, Little, & Wehby, 2007; Liaupsin, Umbreit, Ferro, Urso, & Upreti, 2006; Nahgahgwon, Umbreit, Liaupsin, & Turton, in press; Stahr, Cushing, Lane, & Fox, 2006; Umbreit, Lane, & Dejud, 2004; Underwood, Umbreit, & Liaupsin, 2009; Wood et al., 2007).
All three students engaged in disruptive behaviors. Depending on the student, these behaviors served different functions and required different intervention methods. Nonetheless, a positive outcome was obtained in each case. This finding adds support for the breadth and flexibility of the process of function-based intervention.
Peterson and El I ison (2005) reported a correlation between level of social validity and the likelihood of continuing to implement an intervention. In this study, teachers maintained high levels of treatment integrity during the follow-up phase, when implementation of the intervention was no longer required. The fact that each teacher independently continued to implement the function-based intervention when it was no longer required is the best indication of its social validity.
Certain limitations should be noted. First, the teachers who participated in this study had previous training in classroom management and FBA and previous experience teaching adolescents with EBD. This likely influenced their ability to contribute to intervention development. It also could have influenced how quickly they learned to implement the interventions with high levels of integrity. Different results might occur with less sophisticated staff.
Second, the target behavior - disruptive behavior - was not measured directly during intervention. In addition, the replacement behavior - on-task behavior - might best be considered an appropriate alternative behavior that is incompatible with disruptive behavior. Because a whole-interval method was used to measure on-task behavior, the level of disruptive behavior could never be higher than the level of off-task behavior. Nevertheless, a student might be off task without engaging in disruptive behavior. For this reason, it would be preferable in future research to simultaneously record on-task behavior using a whole-interval measure and disruptive behavior using a partial-interval measure. These methods would provide direct data on on-task, off-task, and disruptive behavior.
Third, the extinction procedure for all three students included brief redirection to address the attention function of their disruptive behaviors. Redirection, even if it is brief (e.g., "Do your work"), provides some attention, although a minimal amount compared to the amount their disruptive behaviors normally received. In addition, each teacher was concerned that completely ignoring the behavior might lead each student to escalate it. For these reasons, brief redirection was included as one of the extinction procedures for each student. It is possible that completely ignoring the behavior would have been equally effective.
Fourth, the interventions were designed collaboratively by a researcher (the first author), each teacher, and each student. Although collaboration between researcher and teacher has been reported previously (cf. Lane et al., 2007), participation by the involved student is a new element. Despite the collaboration, a consultant/expert model (cf. Bergan & Kratochwill, 1990) was clearly used in this research. Prior research (e.g., Scott et al., 2005) has suggested that, without expert support, practitioners typically return to the types of assessment and intervention methods with which they are most comfortable. Whether or not the teachers, individually or with their students, could have independently designed equally effective interventions is an area for future research.
Fifth, the improvement in pre- and postintervention social validity ratings was proportionally much greater for the students than for the teachers. Both groups gave high ratings to the function-based intervention. The difference in proportional change occurred because teachers gave much higher ratings to the baseline conditions than did the students directly affected by these practices. If both the teachers and the students had reassessed the baseline conditions after intervention, different results might have been obtained.
Sixth, generalization probes were conducted in another class setting once intervention started for each student. It would have been preferable to have conducted these probes during baseline, as well, but this was not possible. Tahir, Cain, and James were the only high school students in the school who received all of their instruction in their primary classroom. Once intervention began, each was able to earn enough points to begin the mainstreaming process within the school. The influence of this added reinforcer is unknown.
Finally, the interventions examined here were implemented during a limited portion of each student's school day. Nothing can be said about their effects on behavior at other times, when a different function of behavior might emerge and different intervention procedures might be required. Regardless, the success of these interventions raises questions about the potential of function-based support if it were provided throughout the school day. One must consider whether this approach, if implemented extensively, has the potential to prevent the need for alternative placement outside the mainstream.
Despite these limitations, the data reported here clearly point to the potential power of function-based intervention as a support for adolescents with significant behavioral challenges. The students studied here were the least successful among a group of students already placed in an alternative school because of their behavioral problems. Furthermore, the assessments and interventions targeted these students' most problematic situations in school. Nevertheless, their behavior improved, maintained, and generalized to a nonintervention classroom. Equally important, the interventions were deemed socially valid by those most directly involved - the students themselves - and were maintained by their teachers even after the study ended.
Future studies are needed to examine whether function-based support can prevent the need for alternative placement or, after placement, facilitate transition back to the mainstream. This effort will be aided by studies of function-based intervention extended throughout the school day and over longer periods of time. Studies are also needed to better document generalized effects by including data from multiple contexts during baseline, intervention, and follow-up. Research with staff less sophisticated than those who participated here may shed light on ways in which both staff and students can participate in intervention development more routinely and even more productively.
[Reference]
REFERENCES
Bergan, J. R., & Kratochwill, T. R. (1990). Behavioral consultation and therapy. Columbus, OH: Merrill.
Bijou, S. W., Peterson, R. F., & AuIt, M. H. (1968). A method to integrate descriptive and experimental field studies at the level of data and empirical concepts. Journal of Applied Behavior Analysis, 1, 175-191.
Cox, S. (1999). An assessment of an alternative education program for at-risk delinquent youth, Journal of Research in Crime & Delinquency, 36, 323-330.
Dunlap, G., Kern-Dunlap, L., dePerczel, M., Clarke, S., Wilson, D., Childs, K. E., & Falk, C. (1993). Preliminary functional assessment survey. Unpublished document, Division of Applied Research and Educational Services, University of South Florida, Tampa, FL.
Fox, J. J., & Gable, R. A. (2004). Functional behavioral assessment. In R. B. Rutherford, M. M. Quinn & S. R. Mathur (Eds.), Handbook of research in emotional and behavioral disorders (pp. 143-162). New York, NY: Guilford Press.
Fox, J. J., Conroy, M., & Heckaman, K. (1998). Research issues in functional assessment of the challenging behaviors of students with emotional and behavioral disorders. Behavioral Disorders, 24, 26-33.
Franklin, C. (1992). Alternative school programs for at-risk youths. Social Work in Education, 14(4), 239-251.
Gregg, S. (1999). Creating effective alternatives for disruptive students. Clearing House, 73(2), 107114.
Heckaman, K., Conroy, M., Fox, J. J., & Chait, A. (2000). Functional assessment-based intervention research on students with or at risk for emotional and behavioral disorders. Behavioral Disorders, 25, 196-210.
Individuals with Disabilities Education Act Amendments of 1997, 20 U.S.C. � 1401-1485.
Individuals with Disabilities Education Improvement Act of 2004, 20 U.S.C. � 1415 (k) (d) (iii).
Kazdin, A. E. (1982). Single-case research designs. New York, NY: Oxford University Press.
Kern, L., Dunlap, G., Clarke, S., & Childs, K. (1994). Student-assisted functional assessment interview. Diagnostique, 19, 29-39.
Kern, L., Hilt, A. M., & Gresham, F. (2004). An evaluation of the functional behavioral assessment process used with students with or at risk for emotional and behavioral disorders. Education and Treatment of Children, 27, 440-452.
Kingery, P. M. (2000). Suspension and expulsion: New directions. US Department of Justice, Office of Juvenile Justice and Delinquency Prevention Bulletin. Retrieved from www.ncjrs. gov
Kingery, P. M. (2001). Zero tolerance: The alternative is education. US Department of Justice, Office of Juvenile Justice and Delinquency Prevention Bulletin. Retrieved from www.ncjrs. gov
Kleiner, B., Porch, R. M., & Farris, E. (2002). Public alternative schools and programs for students at risk of education failure: 2000-01 (NCES 200204). U.S. Department of Education. Washington DC: National Center for Education Statistics.
Lane, K. L., Karlberg, J. R., & Shpecaro, J. C. (2009). An examination of the evidence base for function-based interventions for students with emotional and/or behavioral disorders attending middle and high schools. Exceptional Children, 75, 321-340.
Lane, K. L., Rogers, L. A., Parks, R. J., Weisenbach, J. L., Mau, A. C., Merwin, M. T., & Bergman, W. A. (2007). Function-based interventions for students nonresponsive to primary and secondary prevention efforts: Illustrations at the elementary and middle school levels. Journal of Emotional and Behavioral Disorders, 15, 169183.
Lane, K. L., Weisenbach, J. L., Little, M. A., Phillips, A., & Wehby, J. (2006). Illustrations of functionbased interventions implemented by general education teachers: Building capacity at the school site. Education and Treatment of Children, 29, 549-571.
Lane, K. L., Weisenbach, J. L., Phillips, A., & Wehby, J. (2007). Designing, implementing, and evaluating function-based interventions using a systematic, feasible approach. Behavioral Disorders, 32, 122-139.
Lehr, C., Moreau, R. A., Lange, C. M., & Lanners, E. (2004). Alternative schools: Findings from a national survey of states (Research report 2. Alternative Schools Research Project). Institute of Community Integration, College of Education and Human Development, University of Minnesota, Minneapolis, MN.
Liaupsin, C. J., Umbreit, J., Ferro, J., Urso, A., & Upreti, G. (2006). Improving academic engagement through systematic, function-based intervention. Education and Treatment of Children, 29, 573-591.
Martens, B. K., Witt, J. C., Elliot, S. N., & Darveaux, D. (1985). Teacher judgments concerning the acceptability of school based interventions. Professional Psychology: Research and Practice, 16, 191-198.
Nahgahgwon, K. N., Umbreit, J., Liaupsin, C. J., & Turton, A. M. (in press). Function-based planning for young children at risk for emotional and behavioral disorders. Education and Treatment of Children.
National Center for Education Statistics (NCES). (2002). Public alternative schools and programs for students at risk of education failure: 2000-01. Retrieved from http://nces.ed.gov/pubs2002/ 2002004.pdf
Petersen, P., & Ellison, D. (2005, November). The relationship between "Support Provider Buy-In" and intervention success. Paper presented at the 2005 Annual Conference of the Ontario Association of Behaviour Analysts (ONTABA), Toronto, ON.
Raywid, M. (1999). History and issues of alternative schools. Education Digest, 64, 49-51.
Sasso, C. M., Conroy, M. A., Stitchner, J. P., & Fox, J. J. (2001). Slowing down the bandwagon: The misapplication of functional assessment for students with emotional or behavioral disorders. Behavioral Disorders, 26, 281-296.
Scott, T. M., McIntyre, J., Liaupsin, C., Nelson, C. M., Conroy, M., & Payne, L. D. (2005). An examination of the relation between functional behavior assessment and selected intervention strategies with school-based teams. Journal of Positive Behavior Interventions, 7, 205-215.
Skiba, R., & Peterson, R. (1999). The dark side of zero tolerance: Can punishment lead to safe schools? Phi Delta Kappan, 80, 372-382.
Stahr, B., Cushing, D., Lane, K. L., & Fox, J. (2006). Efficacy of a function-based intervention to decrease off-task behavior exhibited by a student with attention deficit hyperactivity disorder. Journal of Positive Behavior Interventions, 8, 201-211.
Sugai, C., Horner, R. H., Dunlap, G., Hieneman, M., Lewis, T. J., Nelson, C. M., & Ruef, M. (2000). Applying positive behavior support and functional behavior assessment in schools. Journal of Positive Behavior Interventions, 2, 131-143.
Tobin, T., & Sprague, J. (2000). Alternative education strategies. Journal of Emotional and Behavioral Disorders, 8, 177-186.
Turton, A., Umbreit, J., Liaupsin, C., & Bartley, J. (2007). Function-based intervention for an adolescent with emotional and behavioral disorders in Bermuda: Moving across culture. Behavioral Disorders, 33, 23-32.
Umbreit, J., Ferro, J., Liaupsin, C., & Lane, K. L. (2007). Functional behavioral assessment and functionbased intervention: An effective, practical approach. Upper Saddle River, NJ: Prentice-Hall.
Umbreit, J., Lane, K. L., & Dejud, C. (2004). Improving classroom behavior by modifying task difficulty: The effects of increasing the difficulty of too-easy tasks. Journal of Positive Behavior Interventions, 6, 13-20.
Underwood, M. A., Umbreit, J., & Liaupsin, C. J. (2009). Efficacy of a systematic process for designing function-based interventions for adults in a community setting. Education and Training in Developmental Disabilities, 44, 25-38.
U.S. Department of Education National Center for Statistics. (2006). Dropout rates in the United States: 2004. Retrieved from http://nces.ed.gov/ pubs2007/2007024.pdf
Witt, J. C., & Elliott, S. N. (1985). Acceptable of classroom intervention strategies. In T. R. Kratchowill (Ed.), Advances in school psychology (vol. 4, pp. 251-288). Mahwah, NJ: Erlbaum.
Wood, B. K., Umbreit, J., Liaupsin, C. J., & Gresham, F. (2007). A treatment integrity analysis of function-based intervention. Education and Treatment of Children, 30, 105-120.
[Author Affiliation]
Amina M. Turton
University of Alaska, Anchorage
John Umbreit
University of Arizona
Sarup R. Mathur
Arizona State University
[Author Affiliation]
AUTHORS' NOTE
The work reported here was supported in part by U.S. Department of Education Grant H325D040019.
Address correspondence to Amina M. Turton, Assistant Professor, Counseling and Special Education, College of Education, University of Alaska Anchorage, Professional Studies Building, 206B, 3211 Providence Drive, Anchorage, AK 99508.
MANUSCRIPT
Initial Acceptance: 6/21/10
Final Acceptance: 7/12/10
