Participants and Setting

Participants were three middle school students (grades 6–8) and their respective classroom teachers. Participants were recruited from two public middle schools located in a southern US state. Both schools were specialized educational facilities designed to serve students with EBD. The schools served small student populations, with total enrollment ranging from 40 to 43 students. The percentage of students receiving free or reduced-price lunch was 67–70%, indicating a predominantly low-income population. Racial and ethnic compositions varied, with schools reporting 35–58% Black students, 20–25% Hispanic students, 16–40% White students, and a small percentage of students identifying as multiracial.

The researcher used a student screening questionnaire during the recruitment process and relied on teacher report to identify students who may benefit from participating in the study and determine if they met the inclusionary criteria. Inclusion criteria for student participants included:

(a)

Has not been responsive to the current behavior intervention strategies, as reported by their teacher,

(b)

Exhibits problem behavior that is likely maintained by socially mediated reinforcement, and

(c)

Engages in problem behavior that the teacher reports as interfering with academic work (e.g., aggression, screaming, disruption, off-task behavior).

Students would have been excluded from the study if they had an attendance rate of less than 80% of school days on average, though all nominated students met inclusionary criteria and were included in the study. Parent consent and student assent were attained prior to student participation.

One teacher was identified for each of the three student participants. Participating teachers reported having regular interaction with the participating student, willingness to partake in the PTR process as a team member and verified their ability to implement the PTR intervention plan during identified class periods. To avoid potential threats to internal validity, teachers were excluded from the study if they previously participated in a PTR or similar manualized, team-based FBA-BIP intervention process facilitated by an expert. No other school staff were recruited to serve as part of the PTR teams, only teachers and the primary researcher were included in the PTR team. Prior to the PTR process, all teachers reported that the only behavioral supports for participating students were class-wide point systems and occasional monthly rewards (e.g., ‘Fun Friday’) for good behavior. However, when the researcher asked participating students to explain the reward system and describe the behaviors required to earn points, students reported that they did not know how to earn points or that their access to rewards was contingent upon their performance.

Student Participants

Three male students participated in the study, each receiving special education services under the eligibility category of emotional disturbance (ED) with comorbidities of ADHD or oppositional defiant disorder. They were enrolled in grades 6 through 8 and represented varied racial and ethnic backgrounds. All students received instruction in a classroom with a 4:1 student–teacher ratio.

Dom was a 13-year-old, Black male student enrolled in eighth grade. According to teacher reports, Dom had difficulty actively participating in class activities. Although he would often remain seated with his head up and appear attentive, he frequently did not respond to questions or complete his work without prompting. At other times, he would disengage more noticeably, putting his head down, engaging in disruptive conversations with other students, and verbally refusing to participate in assigned activities. According to his teacher, he was earning failing to below-average grades in reading and math. Study sessions took place during his sixth-period Reading class, which included four students, including Dom.

Luke was an 11-year-old, White, Hispanic male student enrolled in sixth grade. On the basis of teacher reports, Luke frequently engaged in disruptive behavior in the classroom in the form of whining and refusing to work, which created distractions for other learners and limited the time the teacher was able to provide instruction. Luke’s teacher reported concerns about his academic performance in reading and in math. Study sessions took place during Luke’s second-period Language Arts class, a small group setting that included Luke and four other students.

Xavier was a 14-year-old, White, non-Hispanic male student enrolled in eighth grade. On the basis of teacher reports, Xavier frequently engaged in disruptive behavior in the classroom, typically characterized by inappropriate verbal language which distracted other students and comprised teacher instruction. His academic performance was below grade level in math and in reading. Study sessions took place during his third period Math class which contained four students, including Xavier.

Teacher Participants

Three teachers, two female and one male, participated in the study. They had between 14 and 20 years of experience working with students with disabilities and held at least a bachelor’s degree. Dom’s teacher was a White, non-Hispanic male with 14 years of experience teaching children with disabilities and held a Juris Doctor (JD). Luke’s teacher was a White, non-Hispanic female with 15 years of experience working with children with disabilities holding a bachelor’s degree in education. Xavier’s teacher was a Black female with 20 years of experience working with children with disabilities. She held a bachelor’s in funeral service and a master’s in divinity and ministry. Before transitioning to her current position at the special school, she had many years of experience working as a math teacher for high school students in low-income schools.

Materials

The PTR-SEC manual (Deenihan et al., 2022; Thie et al., 2024) was used during the study to guide the assessment and intervention process. The PTR-SEC manual includes indirect FBA questionnaires, rating scales, implementation fidelity checklists, and necessary tools for assessment and intervention development. The manual also includes a step-by-step guide to facilitate the process for team members. The research team utilized data sheets to collect data during direct observation sessions. A stopwatch was used during sessions in which student behaviors were recorded using partial interval data.

Measures and Data Collection

The primary dependent variables in this study were student problem behavior and alternative replacement behavior. Each PTR team selected and operationally defined a target problem behavior and a replacement behavior hypothesized to be a functional equivalent for their individual student. Other supplementary data, such as treatment fidelity and social validity, were also collected to further evaluate the PTR-SEC intervention process and outcomes.

Direct Observation of Problem Behavior and Replacement Behavior

Data were collected on problem behavior and replacement behavior for each participant via direct observation. Problem behavior and replacement behavior were identified and operationally defined for each student during the Goal Setting meeting at the beginning of the PTR-SEC process. The best fitting dimension of the behaviors to be measured (i.e., frequency, percentage of intervals, percentage of steps completed) were also identified during this meeting. The behaviors of interest were prioritized based on input from the team about which behaviors were most important to decrease, and which more appropriate alternative behavior(s) could be taught to replace the problem behavior. As such, replacement behaviors were selected on the basis of team priorities regarding the most socially appropriate alternative behavior that was meaningful and feasible within the classroom context. Although a replacement behavior that produced the same hypothesized reinforcer as the problem behavior was selected for each student, the team also identified other appropriate behaviors that produced reinforcers that could potentially compete with the reinforcer for problem behavior.

Dom’s target behavior selected for decrease was off-task behavior and his replacement behavior selected for increase was engagement. Off-task behavior was measured using a 10-s partial interval recording system and defined as engaging in one or more of the following responses during instructional time:

(a)

Refusing to participate by verbally or nonverbally indicating unwillingness, such as saying, “I’m not doing this,” pushing materials away,

(b)

Requiring more than one verbal prompt to begin or continue an assigned task,

(c)

Engaging in verbal exchanges with peers unrelated to the academic task,

(d)

Resting his head on the desk without other movement,

(e)

Directing attention toward others’ materials, conversations, or actions without teacher permission, or

(f)

Making comments or gestures that disrupted classroom activities, such as mimicking a classmate, making faces, or making unsolicited remarks.

Engagement was defined as independently performing a series of classroom behaviors related to participation and task completion without prompting from the teacher. Each sixth-period Reading class period offered a consistent and approximately equal number of opportunities for Dom to complete the steps included in the engagement definition. Percentage of engagement was calculated using the following checklist of steps:

(1)

Gathers appropriate materials from designated area,

(2)

Sits in assigned seat,

(3)

Begins and completes bell work,

(4)

Responds to teacher’s questions,

(5)

Asks two relevant questions or contributes two statements about the lesson,

(6)

Asks for help or assistance if needed,

(7)

Follows along with notes and class materials (e.g., head oriented toward teacher with appropriate materials in close proximity),

(8)

Volunteers for class activities at least once (e.g., reading aloud), and

(9)

Avoids side conversations during lesson.

If no opportunity was presented during sessions (e.g., there were no side conversations to avoid), the step was scored as N/A and did not contribute to the overall number of steps for that session. To calculate percentage of steps, the number of steps completed was divided by the total number of steps in the checklist and multiplied by 100.

Luke’s target behavior selected for decrease was whining and the replacement behavior selected for increase was self-advocacy. These behaviors were mutually exclusive. Whining was measured using frequency recording and was defined as any instance in which Luke engaged in one or more of the following for 5 s or more (a) prolonged, high-pitched vocalizations (e.g., “Ughhh,” “I don’t want to,” whining tone); (b) placing head on the desk, arms, or lap; (c) pushing materials away, turning body away from work, or verbally refusing to complete a task accompanied by whining vocalizations; or (e) soft, drawn-out vocalizations resembling a whimpering or moaning sound. The offset criterion for each instance of whining was the cessation of whining for 5 s.

Self-advocacy was also measured using frequency recording and was defined as: any independent vocal or nonvocal communication in which he appropriately requested help, expressed needs or preferences, sought accommodations, or refused a task appropriately without whining vocalizations (e.g., using words to request break in a normal tone of voice).

Xavier’s target behavior selected for decrease was verbal aggression and the replacement behavior selected for increase was following classroom procedures. Verbal aggression was defined as any instance in which he engaged in vocal behavior that met one or more of the following criteria:

(a)

Saying words that are commonly recognized as curse words or swear words (e.g., “damn,” “shit,” “fuck”),

(b)

Making statements that include name-calling, derogatory remarks, or negative comments about a person’s abilities, characteristics, or actions (e.g., “You’re dumb,” “You can’t do anything right,” “Nobody likes you”),

(c)

Vocal statements consisting of a phrase in which the literal meaning of the words is opposite to the intended message, paired with a change in vocal tone (e.g., exaggerated pitch or drawn-out speech) (e.g., saying “Wow, that was brilliant” immediately after someone makes a mistake), and

(d)

Speaking at a volume louder than typical classroom conversation and directed toward another person (e.g., yelling).

Following classroom procedures were defined as engaging in behaviors aligned with established classroom expectations. These included:

(a)

Entering the classroom without engaging in problem behavior;

(b)

Retrieving classroom materials without teacher prompting;

(c)

Remaining in assigned area during instruction and class activities; and

(d)

Requesting help or accommodations using appropriate communication without engaging in verbal aggression.

The researchers used 10-s partial interval recording for verbal aggression and 10-s whole interval recording for following classroom procedures during direct observations.

Individualized Behavior Rating Scale Tool (IBRST)

Teachers used the PTR data collection form called the Individualized Behavior Rating Scale Tool (IBRST; Dunlap et al., 2019) to rate student problem behavior and replacement behavior during targeted academic class periods. The IBRST is a rating form that combines direct observation with a 5-point Likert-type rating scale. It is designed to simplify data collection for teachers by eliminating the need for continuous behavior monitoring. The IBRST has been shown to have adequate inter-rater reliability and is validated for use in school-based settings (Iovannone et al., 2009). The IBRST allows teachers to rate their perception of student target behavior(s) and replacement behavior(s) for daily progress monitoring. Teachers use a unique, individualized rating scale to provide estimates of behavior at the end of the specified measurement period. For replacement behavior, a rating of 1 (bad day) indicates the student engaged in very low levels or did not engage in the replacement behavior during the observation period. Conversely, a rating of 5 (great day) indicates high levels of the replacement behavior. For problem behavior, the scale is reversed: a rating of 5 represents a bad day with high levels of problem behavior, and a rating of 1 signifies a great day with low levels of problem behavior. These teacher-collected IBRST data were used to supplement direct observation data. See Supplemental Material.

PTR-SEC Process Facilitation Implementation Fidelity

The primary researcher (first author) served as the PTR-SEC facilitator for this study. A trained research assistant (RA), recruited from a master’s program in Applied Behavior Analysis, assessed the facilitator's implementation of the PTR-SEC process by observing 50% of the audio-recorded team meetings. The RA used the 4-step PTR-SEC Facilitation Implementation Checklists (Sullivan et al., 2021) to ensure the facilitator implemented each step of the PTR process (i.e., teaming and goal setting, assessment, intervention development, evaluation) as outlined by the manual. In addition, the researcher conducted self-assessments using the same checklists for 100% of meetings. The percentage of correct implementation was calculated for each step as an overall fidelity average. On the basis of the self-assessment data, facilitation fidelity ranged from 95 to 100% for Dom, 100% for Luke, and 100% for Xavier across the steps. Interobserver agreement between the RA and the researcher averaged 100%, indicating high consistency in fidelity scoring.

Teacher Intervention Implementation Fidelity

Teacher implementation fidelity of intervention procedures was assessed during 50% of intervention and follow-up sessions using the PTR-SEC Intervention Implementation Fidelity Checklist developed for each teacher–student dyad. Copies of these fidelity checklists are available upon request. For each intervention session, the researcher collected data on the occurrence or nonoccurrence of each intervention step as outlined in the fidelity checklist. A total percentage of the steps completed was calculated as the fidelity score. Teachers were given copies of their checklists. Mean teacher fidelity was 93% (range 80–100%) for Dom’s teacher, 96% (range 90–100%) for Xavier’s teacher and 100% for Luke’s teacher.

Interobserver Agreement (IOA)

The research team calculated IOA for all student behaviors. Observer training was conducted prior to the study and consisted of researcher-implemented behavioral skills training (BST) to teach the operational definitions and data collection procedures. Training continued until the RA demonstrated proficiency collecting data (IOA reached 95% or higher) on three consecutive training scenarios individualized to each student before beginning data collection. IOA calculation methods varied depending on the type of data collection. For frequency data, total count agreement was used, calculated by dividing the lower frequency count by the higher frequency count and multiplying by 100. For interval data, IOA was calculated using exact interval agreement. The number of intervals with agreement was divided by the total number of intervals and multiplied by 100 to obtain a percentage of intervals with agreement. IOA for Dom was assessed for 30% of sessions (20% of baseline; and 40% of intervention) and mean agreement was 96.5% (range 93–98.3%) for off-task behavior and 100% for engagement. For Luke, IOA was assessed for 20% of sessions (29% of baseline and 13% of intervention) and was 100% for whining and self-advocacy. IOA for Xavier was assessed for 17% of sessions (13% of baseline and 25% of intervention) and mean agreement was 97.8% for verbal aggression (range 96–100%) and 97.8% for following classroom procedures (range 96–100%). Unfortunately, no assessments of IOA were conducted during follow-up sessions due to scheduling challenges.

Social Validity

Participating teachers and students received a social validity questionnaire to rate the acceptability and effectiveness of the PTR-SEC intervention process and outcomes. Teacher and student social validity questionnaires were a modified version of the Treatment Acceptability Rating Form-Revised (TARF-R; Reimers & Wacker, 1988), which have been utilized by several researchers evaluating PTR (e.g., Deenihan et al., 2022; DeJager and Filter, 2015; Dunlap et al., 2018). All items were rated on a 5-point Likert-scale with 1 = strongly disagree and 5 = strongly agree. Social validity questionnaires for teachers included 15 items. The questionnaire included items about the ease of implementation, contextual fit of the intervention procedures, attainment of student outcomes, and likelihood of using similar interventions and/or consultation models in the future. The teachers completed the social validity assessment independently and returned them to research staff.

Social validity questionnaires for students included five items. The primary investigator read the questions aloud to the students individually using developmentally appropriate language and providing examples as necessary to ensure understanding of each question. An example of the paraphrased presentation of questions can be found in Table 3. Items assessed their agreement with the goals, acceptance of the intervention procedures, enjoyment of participating in the process and/or intervention, and their perception of the effectiveness of the procedures (i.e., if they felt that the intervention helped them improve their class performance).

Experimental Design

This study utilized a concurrent multiple baseline design across participants consisting of baseline, PTR intervention, and follow-up phases to evaluate the effects of the PTR-SEC intervention on problem behavior and replacement behavior. The study team used a response-guided technique to determine the sequence and timing of intervention implementation. Specifically, phase changes were determined on the basis of three consecutive sessions of stable responding for both dependent measures across at least three consecutive sessions. Stable responding in each phase was indicated by an absence of changes in the level of behavior or an observed countertherapeutic trend (during baseline only) across three consecutive sessions. This approach aimed to reduce confounding variables, limit bias, and strengthen causal inferences, thus providing a robust evaluation of the intervention’s effectiveness while addressing real-world constraints and ethical obligations. Follow-up assessments were conducted 2 weeks after the intervention phase to evaluate the sustainability of the intervention effects over time.

Procedures

Team meetings were conducted separately for each of the four steps of the PTR-SEC process:

1.

Teaming and goal setting,

2.

PTR-SEC assessment,

3.

PTR-SEC intervention, and

4.

Evaluation.

Team meetings were led by the researcher (facilitator) and lasted approximately 10–30 min per meeting. The total meeting duration across all four steps was approximately 60 min, with step 1 taking the longest (M 27 min), followed by step 2 (M 11.7 min), step 3 (M 13 min), and step 4 (M 8.3 min). Meeting lengths varied slightly depending on the team.

Step 1: Teaming and Goal Setting

The first step of the PTR process was teaming and goal setting. Prior to the step 1 meeting, the facilitator asked each participating teacher if there were any other school personnel who would be able to join the student’s PTR team. Participating teachers reported the PTR team was sufficient with themselves and the primary researcher. One teacher reported that due to other staff duties and responsibilities it was unlikely others would be available to participate, a view that was consistent with the other teacher’s perspectives.

The purpose of step 1 is to identify and prioritize problem behavior to decrease and a replacement behavior to increase. Each team member filled out the student or teacher version of the PTR Goal Setting Form which is designed to identify and operationally define behaviors of interest based on the student’s current performance, goals, aspiration, and values. To complete the student version of the Goal Setting Form, the researcher met with the student individually at a location within the school during a time that was convenient for the student, using an interview format to ask questions and gather the necessary information (Thie et al., 2024). All three students were invited to take part in the PTR process. Luke was the only student who expressed interest in participating in each step of the PTR process; however, Xavier and Dom were still asked to provide input intermittently throughout the process. During step 1, Luke met with the researcher and his teacher to identify target behavior(s) and replacement behavior(s). He provided feedback on the behavioral definitions and confirmed he engaged in the identified behaviors, explaining the context in which they occurred.

The researcher also guided each team in developing an IBRST that was feasible for the teachers to use in monitoring the student progress. As previously described, the anchors for the rating systems were developed on the basis of the measurement dimension most appropriate given the identified behavior(s) of interest. The goal-setting step ensured consensus among team members regarding student goals and current performance levels (Dunlap et al., 2010).

Baseline

Baseline data were collected during the target class period after the step 1 meeting and prior to step 2. Direct observation data were collected on the problem and replacement behaviors exhibited by each student. Baseline data were collected over a period of approximately 4–8 weeks; each session lasted 30–40-min. Data collection began 5 min after the class start time for Luke and Dom and immediately after the class start time for Xavier. During baseline, teachers conducted class periods as usual which included the use of a point level system, and responding to problem behavior in their typical manner (e.g., verbal redirection, exclusionary time-out). The teacher also collected data on the targeted behaviors using the IBRST during all baseline sessions. The researcher did not provide feedback or instructions to teachers during baseline.

Step 2: Assessment

Before the step 2 meeting, the PTR-SEC FBA Questionnaire was completed by the teacher and returned to the researcher. The questionnaire includes 22 items consisting of checklist-style, open-ended, and multiple-choice questions. Examples of questions include: “Are there specific activities where the behavior is more likely to occur?” (prevent), “Does the behavior seem to be exhibited to gain peer attention?” (teach), and “What is the likelihood of the student’s problem behavior resulting in acknowledgment (e.g., reprimands, corrections) from teachers or other school staff?” (reinforce). The questionnaire was administered for the primary problem behavior identified in step 1 and addressed all components of the PTR model (i.e., prevent, teach, reinforce), including questions about antecedent events, potential function(s), and consequences associated with the target behavior.

The step 2 meeting focused on hypothesizing the function of problem behavior using information reported in the PTR-SEC FBA Questionnaire in addition to direct observation conducted by the researcher during baseline. The researcher synthesized the information reported regarding the antecedent variables likely evoking problem behavior, potential functions, and typical consequences that follow the problem behavior. The FBA process only included indirect assessments in the form of the PTR-SEC FBA Questionnaire and direct observation during baseline. Assessment results were then summarized into the PTR FBA Summary Table (Table 1).

Table 1 PTR-FBA summary tableStep 3: Intervention Planning and Training

After developing hypotheses about behavioral functions, the researcher convened the step 3 meeting to review the PTR-SEC Intervention Checklist, which included a variety of potential interventions for each PTR component based on the hypothesized function of the problem behavior. Per the PTR guidelines, the team is required to come to consensus on a minimum of one antecedent strategy that reduces the likelihood of problem behavior (prevent); one replacement behavior strategy that addresses the hypothesized behavioral function (teach); and two consequent intervention strategies: one to reinforce the performance of the replacement behavior as well as one that changes responses to the problem behavior (reinforce). The manual provided guidance on several factors, including considerations for intervention design, such as complexity, age appropriateness, resources required, and feasibility of implementation. Team members rated potential intervention components based on their feasibility and anticipated effectiveness. Through discussion, the team collectively selected intervention(s) from each category to address each PTR component and identified who, when, and how the intervention would be implemented. See Table 2 for the specific interventions selected for each student across PTR components.

Table 2 PTR-SEC intervention strategies included in each intervention plan

Once the team selected one intervention strategy for each component, they broke down its procedures into a detailed task analysis and created the Coaching/Intervention Training Checklist. The researcher met individually with each participating student after intervention development to ensure their willingness to participate in the planned procedures. This meeting was the only portion of the PTR process in which all students were directly involved. During these meetings, the researcher explained the intervention procedures in student-friendly terms and conducted an informal assessment of social validity by asking how the students felt about the selected intervention strategies and whether they would be willing to participate in the intervention. Additionally, the researcher asked each student for input on the types of reinforcement they would prefer to receive for engaging in the expected replacement behaviors. For example, Dom expressed that 3-min breaks were not worth earning at the end of class; therefore, the intervention was adjusted to include frequent check-ins to work toward accruing a 10-min break at the end of class with a preferred adult if he met the necessary criteria (i.e., 90–100% of steps completed). Student input was shared with teachers and incorporated into the PTR plan. These meetings were intended to ensure that the interventions were not only acceptable and feasible for teachers but also acceptable and motivating from the students’ perspectives.

After applying student feedback to refine the intervention procedures and the Coaching/Intervention Training Checklist, the researcher provided one-on-one training to each teacher on the implementation of the selected intervention strategies. The initial training required approximately 10–15 min and was scheduled at the teacher’s convenience. Researchers used BST to train teachers on each component of the intervention (i.e., prevent, teach and reinforce). Teachers were required to verbally describe the steps of intervention components with 100% accuracy before beginning role-play sessions. During modeling within role-play, the researcher played the role of the teacher while the teacher played the role of student. During teacher rehearsals within role-play, the researcher played the role of the student, while the teacher practiced implementing the intervention procedures. Teachers were required to implement the intervention components with 100% fidelity at least once during rehearsal to be considered competently trained. Once teachers felt confident implementing the intervention procedures and met the mastery criterion for training, they began intervention implementation.

Step 4: Implementation and Progress Monitoring

During step 4, the research team collected data on student behaviors during target class periods while the intervention was in place. The team met at least once (with availability to meet more as needed) during step 4 to review progress monitoring and discuss implementation fidelity and any barriers. The research team collected data on teacher implementation fidelity using the PTR-SEC Intervention Implementation Fidelity Checklist. The intervention phase lasted approximately 1–4 weeks across students, depending on individual schedules and progress.

The facilitator provided teachers in vivo coaching and performance feedback throughout the first two intervention sessions to ensure fidelity. Teachers completed the IBRST at the end of each session (same as baseline) to provide ratings of student behavior. After every intervention session, the researcher provided brief, 1-min performance feedback to the teacher about implementation fidelity. The feedback consisted of positive aspects of implementation along with brief instructions and role-play of any steps that were not observed or implemented incorrectly. Only one teacher (Dom’s teacher) required an additional 10-min booster training session after fidelity fell below 90% during one session.

During step 4, the facilitator graphed the direct observation and IBRST data immediately after each session to evaluate student progress and shared the data with the teacher. If the data indicated an increasing trend in problem behavior, the facilitator held a brief meeting with the teacher to review the data and collaboratively problem-solve. When errors in implementation were observed, team members discussed barriers impacting the teacher’s fidelity of implementation (e.g., necessary resources, available staff, time) and employed a system of continuous support, including consistent fidelity checks, feedback and coaching, and data-based decision-making to remediate performance. On the basis of the teacher’s performance and student’s responsiveness to interventions, the team determined whether additional coaching or modifications to the intervention were needed (Dunlap et al., 2019).

Dom was the only student who required modifications to his intervention. Owing to an increase in problem behavior during the second intervention session, the team reconvened to problem solve. The team decided to increase the number of opportunities for reinforcement by implementing check-ins every 10 min with the student as opposed to once at the end of class. During each 10-min interval, Dom received behavior-specific praise for demonstrating targeted engagement behaviors or was offered a 3-min break if he completed at least three out of ten items on his engagement checklist.

Follow-Up

One follow-up probe was conducted for each student. Follow-up probes occurred 2 weeks after the last intervention session. During these probes, the research team collected data on problem behavior and replacement behavior during the targeted class period in which the intervention was implemented during the PTR process. The researcher did not provide any feedback or instructions to the teachers during these sessions. Teacher implementation fidelity was assessed during follow-up. The researcher did not interfere or communicate with teachers or students about the intervention during these sessions.