Interfering behaviours can be defined as repetitive actions exhibited by students within the classroom setting. These behaviours can disrupt the learning environment and hinder positive social interactions among students and adults (Weston et al., Reference Weston, Rittenhouse-Cea, Gauert, Crandall, Radhakrishnan and Matson2023). Examples of interfering behaviour include shouting, aggression towards others, and breaking items within the classroom. Such behaviour can create significant disruption to effective instruction, reducing instructional time for both the student exhibiting these behaviours and their classmates (Sugai et al., Reference Sugai, Horner, Gresham, Shinn, Walker and Stoner2002). Moreover, teachers may face additional stressors and an increased workload (Agyapong et al., Reference Agyapong, Obuobi-Donkor, Burback and Wei2022). High occurrence of interfering behaviour in the classroom forces teachers to divert time and attention away from instruction to manage these behaviours. This change in focus interrupts the learning environment and places additional demands on teachers to document incidents, communicate with families or specialists, implement individualised behaviour plans, and maintain classroom safety. Therefore, evaluating effective procedures to support teachers in classroom behaviour management is of great importance.
Classroom Management in South Africa and Vietnam
Classroom behaviour management is often cited as one of the most significant issues teachers face in South Africa (Davids, Reference Davids2016). New teachers report they are not confident in designing and implementing classroom management procedures and programming to deliver appropriate consequences for behaviours that occur within their classrooms. They also report a need for more support from mentors to help increase their confidence in managing their classrooms (Mkhasibe & Mncube, Reference Mkhasibe and Mncube2020). Without sufficient training and confidence in effective behaviour strategies, teachers might resort to punitive approaches to manage student behaviour in the classroom.
Despite no longer being legal in South Africa, corporal punishment procedures are often used within the school systems (Mahlangu et al., Reference Mahlangu, Chirwa, Machisa, Sikweyiya, Shai and Jewkes2021; Statistics South Africa, 2023). In response to the prevalence of corporal punishment by teachers, the South African government prepared and distributed printed materials for teachers to offer practical classroom behaviour management guidelines (Department of Education, 2000, 2016). These materials focused on the roles of teachers in teaching and recognising prosocial behaviours to create a positive and nurturing learning environment. This positive approach to classroom management is termed positive behaviour support (PBS). The PBS approach places an emphasis on maintaining consistent rules, recognising positive actions with praise, and addressing misbehaviour (Kittelman et al., Reference Kittelman, McIntosh, Mercer, Nese, So and George2025). According to PBS, students’ misbehaviour should be addressed calmly through nonviolent consequences like verbal warnings or restorative tasks. In addition, teachers should make efforts to model respectful behaviour, involve students in creating classroom rules, and use clear communication to set expectations.
While these classroom practices are consistent with best practices for classroom management, no evidence exists that teachers and teachers-to-be receive competency-based training to implement these interventions. According to Dwarika (Reference Dwarika2020), PBS programs have been introduced within schools in South Africa and have been determined to be culturally suitable. Nonetheless, professional development initiatives to prepare teachers in South Africa to implement behaviour strategies are not sustainable or only limited to teachers within communities with more resources. According to Simonsen et al. (Reference Simonsen, Meyer, Plumb, Duble Moore and Sears2024), coaching and performance feedback are crucial for teachers to maintain and generalise effective behaviour strategies. However, many schools in South Africa lack the capacity to provide ongoing, hands-on professional development beyond initial training sessions (Dwarika, Reference Dwarika2020). These challenges are likely more pronounced in under-resourced communities, where systemic barriers hinder sustained implementation efforts.
Another country in need of professional development support is Vietnam, where limited training in student behaviour management poses challenges for effectively supporting teachers in the classroom. Educators and policymakers in Vietnam have recognised the importance of effective classroom behaviour management to improve students’ learning outcomes (Vu, Reference Vu2022). Regulations determined by the Ministry of Education and Training for the 2018 General Education Curriculum in Vietnam involved a reform of the basic education system (Nguyen et al., Reference Nguyen, Nguyen, Vu, Hoang and Nguyen2020). This reform is focused on the development of core competencies in students, including content knowledge, workforce preparation, and nurturing of future talents. With a focus on improving student outcomes, there has been increased interest among teachers’ preparatory programs in evaluating strategies to enhance teacher training, foster teaching competencies, and improve internal management processes.
Vietnamese government regulations have encouraged educators to seek innovative approaches to enhance the quality of education. Furthermore, numerous studies have been conducted on competency-based education, focusing on curriculum development and teaching methods (e.g., Pham et al., Reference Pham, Arthur-Kelly, Foggett and Lyons2023; Tran et al., Reference Tran, Nguyen and Hoang2021). However, there are still shortcomings in effectively managing classrooms and a lack of research on training in classroom behaviour management for teachers in Vietnamese schools (Ngo et al., Reference Ngo, Nguyen and Nguyen2022; Vu, Reference Vu2022). Teacher preparation must encompass knowledge of concepts and theories related to students’ behaviour and learning, as well as training, feedback, and coaching on evidence-based classroom behaviour management strategies.
Functional Behaviour Assessment as an Evidence-Based Strategy
Years of research have demonstrated the effectiveness of the functional approach in assessing and intervening in interfering behaviours (Lory et al., Reference Lory, Gregori, Huff, Lee and Rendon2024; Miltenberger et al., Reference Miltenberger, Valbuena and Sanchez2019). According to the functional approach, interfering behaviours serve a function or a purpose for the student. That is, it is assumed that the student engages in the interfering behaviours because of the consequences they generate. When the function or functions of interfering behaviours are identified, a function-based intervention can be designed and implemented to reduce these behaviours. Research has repeatedly demonstrated that function-based interventions are more likely to be effective and long-lasting when compared to interventions that did not determine the function of interfering behaviours (Jeong & Copeland, Reference Jeong and Copeland2020; Lory et al., Reference Lory, Gregori, Huff, Lee and Rendon2024).
The process for identifying the function of interfering behaviours is termed functional behaviour assessment (FBA). FBA is an umbrella term encompassing different strategies to identify the variables in the student’s environment influencing the interfering behaviour (Anderson et al., Reference Anderson, Rodriguez and Campbell2015). Different methods exist to identify the function of interfering behaviours. One method is called the antecedent-behaviour-consequence (ABC) assessment, a form of descriptive assessment involving direct observation of a student’s behaviour (B) in the classroom and the antecedents (A) and consequences (C) of the interfering behaviour (Thompson & Borrero, Reference Thompson, Borrero, Fisher, Piazza and Roane2021). A teacher should follow several steps to conduct an ABC assessment and select a function-based intervention. First, the teacher should select the primary interfering behaviour of concern and write and describe the behaviour in observable, measurable, and clear terms. Next, the teacher should define the location and time to observe the behaviour and collect information on the antecedents and consequences of the interfering behaviour. Once data involving antecedents and consequences of the interfering behaviour are collected over several days and context (e.g., two observations per day across 5 days), the teacher should analyse the data to raise a hypothesis about the function or functions of the interfering behaviour. Based on the identified function, the teacher then selects an intervention that addresses the function. Because the ABC assessment relies on the observation of naturally occurring events, the teacher must be familiar and proficient with the procedures and data collection strategies (Ellingson et al., Reference Ellingson, Miltenberger, Stricker, Galensky and Garlinghouse2000). As such, developing clear descriptions of the primary interfering behaviour, identifying antecedents and consequences surrounding the behaviour, and interpreting the results obtained are crucial for selecting and developing effective function-based interventions in classrooms.
Educator Training
Results of studies evaluating performance training strategies for staff and educators indicate that traditional lecture-based models of didactic instruction are typically ineffective in producing acceptable performance during the implementation of strategies to address interfering behaviours in the classroom (e.g., Casey & McWilliam, Reference Casey and McWilliam2011; Clark et al., Reference Clark, Cushing and Kennedy2004). Hands-on, criteria-based training (e.g., Parsons et al., Reference Parsons, Rollyson and Reid2012) is typically needed to increase staff and educators’ skills to implement these strategies. A criteria-based strategy normally used for training staff in mental health services and educators in school settings is behavioural skills training (BST).
BST involves four key components: instruction, modelling, rehearsal, and feedback (Parsons et al., Reference Parsons, Rollyson and Reid2012). During instruction, the trainer describes the targeted skill and the rationale for using it to the teacher in observable and measurable terms. During modelling, the trainer models the skill for the teacher and requires the teacher to role-play the skill. Finally, during rehearsal and feedback, the trainer observes the teacher performing the skill and provides feedback during or after rehearsal of the skill. Data are collected on the accuracy of the teacher’s performance based on the predetermined procedural fidelity checklist. The teacher rehearses the skill and receives feedback until reaching a predetermined mastery criterion.
BST has been widely used to train teachers to perform a variety of behavioural strategies in the classroom (Kirkpatrick et al., Reference Kirkpatrick, Akers and Rivera2019). Specifically, research has demonstrated the effectiveness of BST for training teachers on how to complete an FBA in the classroom (Calderhead et al., Reference Calderhead, Vargo, Mitchell, Koupaei, McKinney and Pasaoa2021; Samudre et al., Reference Samudre, Allday and Lane2022). However, research investigating training strategies is still needed. Dutt et al. (Reference Dutt, Chen and Nair2024) conducted a scoping review to identify the content and delivery methods of FBA and function-based intervention training provided to in-service teachers. Their results showed a gap in the literature regarding the use of experimental designs to evaluate the effectiveness of such training. The review identified only nine studies that employed experimentally sound methodologies. Furthermore, all but one of the identified studies were conducted in the United States, demonstrating a lack of research in international contexts. Therefore, this study seeks to evaluate the effects of BST in teaching educators from South Africa and Vietnam to conduct an ABC assessment of interfering behaviour, thereby offering a scalable solution for improving classroom management in under-resourced educational systems. This line of research has the potential to improve social justice (Pritchett et al., Reference Pritchett, Ala’i-Rosales, Cruz and Cihon2022) in research while promoting inclusive and compassionate strategies (Weiss, Reference Weiss, Sadavoy and Zube2022) for students’ behaviour in the classroom.
Research Questions
In this study, we aimed to address the following research questions:
-
1. How effective is BST in increasing the procedural fidelity of educators from South Africa and Vietnam when implementing the ABC assessment of students’ interfering behaviour?
-
2. How acceptable do educators from South Africa and Vietnam find the BST training procedures and the skills they acquired during the training?
Methods
Participants
Recruitment was conducted using purposive sampling (Campbell et al., Reference Campbell, Greenwood, Prior, Shearer, Walkem, Young, Bywaters and Walker2020), in which participants were selected based on their occupation (i.e., educator) and their availability to attend research sessions. The study in South Africa was reviewed and approved by the Human Subjects Research Institutional Review Board of the University of Missouri–St. Louis (Project # 2063404) and the Humanities and Social Science Research Ethics Committee of the University of the Western Cape (Reference # HS21/10/54). The study in Vietnam was reviewed and approved by the same committee at the University of Missouri–St. Louis (Project # 2094863). According to the Ethics Committee for Research Involving Human Subjects of the University of Education at Vietnam National University, Hanoi, the study met the criteria for exemption from ethical review.Footnote 1 All participants received an information sheet about the study (translated into Vietnamese for participants from Vietnam) and provided verbal consent prior to the start of the study.
South Africa
Three educators from Cape Town, South Africa, participated in this study. Renee was a 33-year-old ColouredFootnote 2 female (self-reported) who spoke English and Afrikaans. She graduated with a bachelor’s degree in psychology and had 9.5 years of experience as an educator. Throughout her career, she worked as an ESL teacher for K-5 and middle school students, an early childhood development teacher, a trainer and researcher, and a special education teacher. Illana was a 44-year-old Coloured female (self-reported). She spoke English as her first language and earned a master’s degree in paediatric occupational therapy. Illana had 14 years of experience as an educator in the roles of lecturer, clinical supervisor, workshop facilitator, and early childhood development researcher. Camille was a 59-year-old Coloured female (self-reported) who spoke English as her first language. She earned a master’s degree in child and family studies and had 18 years of experience as an English and special education teacher. Participants were recruited via a local university in the Cape Town metropolitan area.
Vietnam
Four Vietnamese educators living in the Hanoi region of Vietnam participated in this study. All participants worked with children who exhibited interfering behaviours and had a behaviour intervention plan. Linh was a 30-year-old female with a bachelor’s degree in nursing and 5 years of experience working with children in a paediatric ward. Mai was a 31-year-old female with a master’s degree in clinical psychology. Mai worked in a private school as a school psychologist and had 9 years of experience working with students in the school setting. She spoke fluent English and did not require a translator. Lan was a 26-year-old female with a bachelor’s degree in early childhood education and 4 years of experience as an educator. Minh was a 31-year-old male with a bachelor’s degree in social work. He had been working for 5 years with children with mental health disorders. Linh, Lan, and Minh spoke Vietnamese as their primary language and required a translator throughout the study.
Settings
Cape Town
All experimental sessions were conducted in person in Cape Town, South Africa, in February 2022. Because of the restrictions imposed by the COVID-19 pandemic, sessions were conducted outside the school setting in distraction-free rooms rented from a hotel and a community centre. Each participant worked individually with one of the experimenters. Experimenters were two behaviour analysts at the doctoral level with experience in clinical and school settings. The experimenter used a laptop to teach the different training phases and play video vignettes. Participants watched the videos on the computer across the table.
On average, training sessions lasted approximately 30 min per module, while probe sessions lasted up to 10 min. Sessions were conducted for each participant over 2 consecutive days.
Vietnam
All experimental sessions were conducted in person in Hanoi, Vietnam, in February 2023. Each participant worked individually with the first or the second author, with bilingual Vietnamese interpreters for three of the four participants. Interpreters were Vietnamese graduate-level psychology or special education students from Vietnam National University in Hanoi. Sessions were conducted in a large conference room with tables and chairs. Experimenters (the first two authors) were two behaviour analysts at the doctoral level with experience in clinical and school settings. The experimenters presented training materials via a laptop for each training phase and to play the video vignettes. Each participant was also given a printed version of the slides translated into Vietnamese.
On average, training sessions lasted approximately 40 min per module, while probe sessions lasted up to 15 min. Sessions were conducted for each participant over 2 to 3 consecutive days. Sessions were longer in Vietnam due to consecutive translation.
Materials
We developed video vignettes for probe and training sessions. Videos featured confederates playing the role of students and teachers in a classroom. Each video displayed different topographies of interfering behaviour and different maintaining functions (a table with the list of all videos developed is available in the supplementary material). A total of five series of videos were developed for probe sessions. Each series was developed to depict a different function of interfering behaviour and was composed of three 5-min videos, with a target student engaging in a specific topography of behaviour across the series. A total of 10 videos were developed for training sessions on writing operational definitions of behaviour. Each video was approximately 30-s long and displayed a confederate engaging in a different topography of interfering behaviour. Finally, 10 videos were developed for training sessions on ABC data collection. Each video was approximately 1.5 min and contained three opportunities for ABC data collection of a specific target behaviour. Videos developed for probe sessions were not used during teaching sessions and vice versa. We randomised the order in which participants would watch each video during probes and BST sessions. During all sessions, videos were paused every 20–30 s to allow for the translation of dialogues to Vietnamese participants. Translators were instructed not to answer questions unrelated to the dialogues in the videos.
Each participant received a set of printouts of the slide presentations. Participants were also provided with data sheets and blank sheets where they practised and demonstrated the skills assessed in the study. Participants from Vietnam were provided with translated versions of slide printouts and data sheets.
Response Measurements and Data Collection
The main dependent variable was the percentage of correct responses during ABC assessment probes, hereafter referred to as ‘ABC probes’. During ABC probes, participants were required to perform the steps involved in conducting an ABC assessment of students’ interfering behaviour displayed in the videos. Participants used data sheets developed for the study to complete the ABC assessment for the following steps of the process: (a) operational definition — that is, the description of the target interfering behaviour; (b) ABC data collection; and (c) conditional probability and graphing. (The table depicting each step for conducting the ABC assessment process and a list of observable responses used for scoring are available in the supplementary material.) Because participants registered their answers and steps of the ABC assessment in the data sheets provided by the experimenters, data collection was conducted via permanent product. We used a procedural fidelity checklist (see supplementary material) to score participants’ permanent product. We calculated participants’ correct responses by dividing the total number of steps performed correctly by the total number of possible steps and multiplying by 100. Scoring of participants’ permanent product was conducted immediately after each probe session. Participants were not informed of their scores during probe sessions.
Secondary dependent variables were the percentage of correct responses to knowledge tests and practice trial responses during the BST phases, and the mean score in the Training Acceptability Survey. Participants’ responses to knowledge tests were scored based on a predetermined rubric. Responses that matched the predetermined rubric were scored as correct. All other responses or blanks were scored as incorrect. We calculated participants’ correct responses by dividing the total number of questions responded correctly by the total number of questions in the test and multiplying by 100. Participants’ performances during BST were recorded using a permanent record according to the procedural fidelity checklist described previously. We scored performances during practice trials as described in ABC assessment probes, except that only the step trained during the specific BST module was scored. Scoring of participants’ permanent product was conducted immediately after each probe session. Finally, the mean score from the Training Acceptability Survey (adapted from Reimers et al., Reference Reimers, Wacker and Cooper1991) was collected after participants completed all training and probe sessions. The survey comprised 12 Likert-scale questions (see Table 1). Participants responded to each question based on a 7-point Likert scale, where a score of 1 represented not at all (e.g., clear) and a score of 7 represented very (e.g., clear). A high score demonstrated high acceptability, whereas a low score showed low acceptability.
Training Acceptability Rating Form for Participants From South Africa and Vietnam
Note. Adapted from ‘Evaluation of the Acceptability of Treatments for Children’s Behavioral Difficulties: Ratings by Parents Receiving Services in an Outpatient Clinic’ by T. M. Reimers, D. P. Wacker, and L. J. Cooper, 1991, Child & Family Behavior Therapy, 13(2), pp. 57–58 (https://doi.org/10.1300/J019v13n02_04). The survey contained 7-point Likert-scale questions, where a score of 1 represented not at all (e.g., clear), a score of 4 represented neutral, and a score of 7 represented very (e.g., clear). A high score demonstrated high acceptability, whereas a low score showed low acceptability. To calculate a score of overall acceptability, we reversed the scores for Questions 9 and 10 so that a score of 1 was transformed to 7, a score of 2 was transformed to 6, a score of 3 was transformed to 5, and vice versa. A score of 4 remained the same.
Experimental Design
We used a concurrent multiple-probe design across participants (Horner & Baer, Reference Horner and Baer1978) to evaluate the effects of BST on the levels of correct responses during probe sessions and knowledge tests. The multiple-probe design was selected to demonstrate the participant’s acquisition of the steps involved in the FBA before and after implementing the different training modules. Figure 1 depicts the sequence of conditions of the study. ABC probes were conducted before and after each training phase. Knowledge tests were conducted at the beginning and at the end of the study to evaluate the effects of didactic training. The mastery criterion for knowledge tests was one session with 80% correct responses. The mastery criterion for ABC probes was two consecutive sessions with at least 80% correct responses.
Sequence of Procedures and Phases of the Study.
Note. Sequence of conditions and training phases. ABC Probe and Knowledge Test were conducted before training started. Phase 1 was didactic training only on the functional approach to classroom behaviour management. BST started at Phase 2 with training in writing operational definitions of target behaviour, followed by Phase 3, ABC data collection, and Phase 4, conditional probability and graphing. ABC probes were conducted between Phases 2 and 3 and Phases 3 and 4. After the last training phase (Phase 4), ABC Probe and Knowledge Test were again conducted.
Participants received didactic training on the foundations of the function-based approach to interfering behaviour, and BST for the steps to conduct the ABC assessment of a student’s behaviour in the classroom. The training was conducted across four phases: Phase 1, the functional approach to understanding interfering behaviour in the classroom (didactic training); Phase 2, writing good operational definitions of behaviour; Phase 3, free-operant collection of ABC data; and Phase 4, calculating conditional probabilities, creating a graph, and identifying the potential function of the target behaviour (see Figure 1). All steps of BST (i.e., instruction, modelling, rehearsal, and feedback) were implemented in Phase 2 to Phase 4. Participants had to reach the mastery criterion of 100% correct responses for two consecutive practice trials in each training phase prior to completing ABC probes and moving on to the next phase.
Interobserver Agreement
A second observer scored responses across all phases of the study for participants from South Africa. We collected interobserver agreement (IOA) for 23% of sessions for Renee, 28% for Illana, and 35% for Camille. We scored IOA by comparing point by point the scoring of the procedural fidelity checklist between the two observers. We calculated IOA by dividing the number of point-by-point agreements by the number of point-by-point agreements plus disagreements and multiplying by 100. The mean percentages of observer agreement for probe and training sessions were 88.7% (range: 81%–91.5%) for Renee, 91% (range: 73%–99%) for Illana, and 88% (range: 75%–100%) for Camille.
IOA was collected on probe sessions only for participants from Vietnam. We collected IOA for 20% of probe sessions for Linh, 50% for Mai, 43% for Lan, and 50% for Minh. The mean percentages of observer agreement for probe and training sessions were 86% (no range) for Linh, 98% (range: 93%–100%) for Mai, 92% (range: 85%–100%) for Lan, and 98% (range: 92.5%–100%) for Minh.
Procedures
Knowledge tests
Participants completed a written knowledge test before and after the training modules. Four to five questions were developed based on the content of each training phase (except Phase 1, which had 14 questions) for a total of 25 questions. The same questions were presented as pre- and post-training knowledge tests. The test comprised multiple-choice questions, short-answer questions, and fill-in-the-blank questions. Participants completed the test in 20–30 min. Mastery criterion was one session with 80% correct responses.
Probes
Probes were conducted before and after each training phase to determine the effects of training procedures on participants’ accuracy in conducting the ABC assessment. At the beginning of each probe session, the experimenter handed data sheets to the participants, set the computer in front of the participant, started the video, pointed to the target student, and informed what the target behaviour was (e.g., ‘This is Luciana, and her interfering behaviour is aggression’).
The experimenter played all three videos from the series; participants were allowed as much time as they needed to complete the probe. Participants were expected to complete all tasks during probes: write an operational definition of the target behaviour; collect free operant ABC data of the target behaviour; and calculate the conditional probability of antecedents and consequences for the target behaviour, graph the conditional probability, and identify the potential function of the target behaviour. Data sheets were titled with the specific step of the ABC probe (e.g., ‘Operational Definition’, ‘ABC Data Collection’). There were no programmed consequences for correct or incorrect responses. The mastery criterion was two consecutive sessions with at least 80% correct responses.
Didactic training
During Phase 1, participants received a 30-min lecture on the concepts and assumptions related to the functional approach for the intervention of interfering behaviour. The content presented during the lectures included myths about the causes of interfering behaviour, how to differentiate interfering behaviour from annoying behaviour, the concept of positive and negative reinforcement, the functions of interfering behaviour, and the ABC of behaviour. The experimenter delivered the lecture during didactic training and responded to any questions. Participants were allowed to take notes in their handouts and ask questions throughout the lectures.
Behavioural skills training (BST)
We implemented BST during Phases 2 to 4, as described in the experimental design section. These phases were taught with the four key components of BST: instruction, modelling, rehearsal, and feedback. During instruction, the experimenter explained the specific steps of the ABC assessment using presentation slides. Presentations contained examples of the correct performance of each step on the ABC assessment. Each presentation lasted approximately 15 min for participants from South Africa and about 25 min (due to consecutive translation) for participants from Vietnam. After the presentation, the experimenter allowed participants to practise the specific skill for each step during practice trials. During Phase 2 and Phase 3, the experimenter used video vignettes and then asked the participants to write the operation definition of the target behaviour (Phase 2) and to collect ABC data for the target student (Phase 3). During Phase 4, participants were provided with a written set of ABC data from which to calculate the conditional probability of antecedents and consequences and draw bar graphs of the conditional probability. During Phase 4, participants were also expected to identify the potential function of the interfering behaviour based on the data displayed in the graphs. During the practice trials of Phases 2 to 4, the experimenter provided positive and corrective feedback. If necessary, the experimenter modelled again the expected completion of the target task for each phase. The mastery criterion for each training phase was 100% correct responses for two consecutive practice trials.
Data Analysis
The multiple-probe design is a single-case design that allows for the demonstration of intervention effects with individual participants if some participants respond as predicted to the intervention and others do not. We interpreted the experimental results for each participant using visual inspection of the level, trend, and variability of data points in accordance with baseline logic for multiple-probe designs (Kazdin, Reference Kazdin2021).
Results
South Africa
Figure 2 displays the performance of participants from South Africa during knowledge tests and ABC probe sessions. (Training data are available in the supplementary material.) Renee (top panel) displayed low levels of correct responses during the first probe session. Levels of correct responses for the knowledge probe were similarly low. Operational definition training did not affect Renee’s correct response levels. After ABC data collection training, Renee’s responses increased to around 70% and reached the criteria after the final training phase. The middle panel depicts Illana’s data. Illana showed low levels of correct response during pre-training probes. Levels of correct response during probe sessions increased gradually after each teaching phase and reached criteria after training in conditional probability and graphing. Camille’s data are depicted in the bottom panel. Similarly, Camille showed low levels of correct responding during pre-training probes, a gradual increase in correct responding after each training phase, and mastery criteria after conditional probability and graphing training. All participants showed low or zero levels of correct responding (M = 5%) during pre-training knowledge tests. Their correct responses increased after BST (M = 84%) but not to the mastery criteria. They all required one remedial teaching session, after which they responded with 100% accuracy.
ABC Probe and Knowledge Check Data for Participants From South Africa.
Note. ABC probe data for all participants from South Africa across all training phases. OD = operational definition; ABC = antecedent-behaviour-consequence assessment.
Vietnam
Figure 3 depicts the performance of participants from Vietnam during knowledge tests and ABC probe sessions. (Training data are available in the supplementary material.) Linh (top tier) displayed zero correct responses during the first probe session. Levels of correct responses for the knowledge test were also low at 20%. Training on operational definitions increased correct responses to 30% and maintained this level after ABC data collection training. Linh reached the criteria after the final training phase with two probes, each at 93%. The post-training knowledge test showed correct responses of 89%.
ABC Probe and Knowledge Check Data for Participants From Vietnam.
Note. ABC probe data for all participants from Vietnam across all training phases. OD = operational definition; ABC = antecedent-behaviour-consequence assessment.
Mai (second tier) showed low levels of correct response during pre-training probes (M = 26.5%). She scored 44% on the correct responses in the pre-training knowledge test. Following operational definition training, correct responses decreased to 21%. After ABC data collection training, Mai’s correct responses increased to around 69%. Mai reached the criteria after the final training phase with two probes, each at 88% and 89%. Her post-training knowledge test showed correct responses of 89%.
Similarly to Linh, Lan (third tier) showed zero levels of correct responding during pre-teaching probes. Linh responded correctly to 11% of questions in the pre-training knowledge test. Following operational definition training, correct responses decreased to 5%. After ABC data collection training, Lan’s responses increased to 50%. Lan reached mastery criteria after the final training phase. The post-training knowledge test showed correct responses of 80%.
Finally, Minh (bottom tier) showed zero levels of correct responding during pre-training probes. The pre-training knowledge test scored at 44%. Training on operational definitions increased accuracy to 18%. After ABC data collection teaching, Minh’s correct responses increased to around 76%. Minh reached the criteria after the final teaching phase with two probes, each at 85% and 87%. The post-training knowledge check showed correct responses of 89%.
Social Validity
All participants responded to the Training Acceptability Survey after completing the training (see Table 1 for specific items and mean scores). The average acceptability rate was 6.4 (range: 4–7) for participants from South Africa and 6.10 (range: 4.5–7) for participants from Vietnam. A score of 1 indicated low acceptability, and a score of 7 indicated high acceptability. In general, the outcomes of the Training Acceptability Survey showed participants were highly satisfied with the training they received.
Discussion
The purpose of this study was to assess the effectiveness of BST in training educators from South Africa and Vietnam on how to conduct an ABC assessment to identify the function of interfering behaviour within the classroom. Seven educators were taught to operationally describe the target behaviour, collect ABC data, perform conditional probability analyses for antecedents and consequences, graph their results, and identify potential functions of interfering behaviour. The results of this study replicated years of research demonstrating the effectiveness of BST in teaching an evidence-based assessment to participants without formal training in behaviour interventions (see Kirkpatrick et al., Reference Kirkpatrick, Akers and Rivera2019, for a review on the topic).
The primary objective of applied research is to develop effective interventions that enhance the quality of life for individuals across various communities and with diverse behavioural needs. As such, an essential consideration is the external validity of its findings, which refers to the extent to which the results of a specific procedure can be applied beyond the confines of a particular research setting (Kazdin, Reference Kazdin2021). To the best of our knowledge, no published studies have demonstrated the effectiveness of BST in teaching ABC assessments to educators in South Africa and Vietnam. The results of the present study contribute to the literature demonstrating that BST is an effective approach for instructing educators from different cultural backgrounds, particularly those in countries where behaviour change strategies are still emerging. The results also indicate the efficacy of BST when translation is involved.
The results of the knowledge test implemented pre-training showed that participants initially attributed the causes of interfering behaviour to factors external to the classroom. For instance, one question on the knowledge test inquired, ‘Where should we seek an explanation for current interfering behaviour?’ This question featured multiple-answer options, as follows: (a) In the student’s diagnosis/disability, (b) In the student’s early years of age, (c) In the student’s current environment, (d) In the student’s family history, and (e) All of the above. For the training provided to participants, the correct answer was option (c), in the student’s current environment. During pre-test, all participants consistently chose option (e), all of the above. These findings align with prior research, which has demonstrated a common misperception among educators concerning the causes of interfering student behaviour within the classroom. Young and Martinez (Reference Young and Martinez2016) administered an online survey with teachers enrolled in an introductory course on behaviour analysis applied to educational settings to identify their perceptions related to students’ interfering behaviour. Their results indicated that more than 75% of respondents concurred that interfering behaviour in the classroom could be attributed to factors such as inadequate parenting and past trauma. These misconceptions may lead teachers and school professionals to develop adversarial interactions with parents due to their perspective of parental responsibility for student behaviour. Additionally, these misconceptions can result in teachers developing negative attitudes towards students, especially those with disability who engage in interfering behaviour. Finally, because one cannot change the student’s diagnosis or past trauma, teachers might refrain from taking steps to implement individualised, evidence-based strategies for interfering behaviour in the classroom.
Skinner and Hales (Reference Skinner and Hales1992) demonstrated that teachers changed their explanations for the causes of interfering behaviour to a behavioural perspective after completing a course focused on behaviour strategies for classroom management. In the current study, post-test data revealed that all participants indicated that one must concentrate on the student’s immediate environment (option c) to identify the causes of interfering behaviour. The findings in the current study are like those of Skinner and Hales’s study. Participants’ responses in the knowledge post-test showed that BST effectively shifted their responses for the cause of interfering behaviour from factors beyond their control to environmental variables in their immediate context. That is not to say that educators should ignore students’ diagnoses, family situations, and past traumas. Such information is essential for supporting the development of effective and student-centred interventions. Educators should gather data related to the student’s condition in various contexts. However, the causes (i.e., function) of the interfering behaviour should be analysed from the perspectives of the antecedent and consequence conditions in the student’s immediate environment.
This shift in mindset is crucial for fostering the adoption of function-based strategies in classroom management. Providing educators with professional development opportunities to enhance their knowledge and practical skills of best classroom management practices can bolster their confidence and their application of evidence-based interventions (Dwarika, Reference Dwarika2020). Most importantly, this mindset shift could lead to a preventative approach to classroom management rather than punishment-based strategies, which have traditionally been prevalent among teachers in South Africa (Statistics South Africa, 2023). In Vietnam, individuals with disability are often viewed as a burden to their families (Truong et al., Reference Truong, Mire, Day, Ni and Keller-Margulis2023), and there is still much stigma associated with mental health and developmental disorders, particularly in low-income communities and minority groups (Poon et al., Reference Poon, Cassaniti, Karan and Ow2022). These changes represent a step towards social justice in educational practices (Weiss, Reference Weiss, Sadavoy and Zube2022), as they can promote inclusive and compassionate classroom methods, especially for children residing in regions with limited financial and educational resources.
The current study exemplifies one type of initiative that can advance social justice in research by improving representativeness, particularly for under-investigated populations (Pritchett et al., Reference Pritchett, Ala’i-Rosales, Cruz and Cihon2022). African educational systems have historically struggled with political instability and economic challenges, resulting in ill-prepared teachers, insufficient resources, limited government funding, and inadequate parental support (du Plessis & Mestry, Reference du Plessis and Mestry2019). This issue is particularly prevalent in rural South Africa, where poverty levels are more pronounced than in urban areas. The replication of the study in Vietnam, where stigma leads most individuals with disability being underserved (Truong et al., Reference Truong, Mire, Day, Ni and Keller-Margulis2023), contributes towards advancing social justice in research. By including participants from diverse ethnic backgrounds, this study makes a contribution, albeit small, to the representativeness of findings. Furthermore, this study represents an action towards the dissemination of empirically grounded practices to educators in underserved communities.
On a broad scale, both representative and localised research holds the potential to draw the attention of authoritative figures towards a crucial shift from punitive policies to proactive strategies for managing classroom behaviour in developing countries. Applying function-based interventions and strategies to build rapport between students and teachers can significantly increase the academic skills and prosocial behaviours of individual students, the classroom, and the entire school (Sugai & Simonsen, Reference Sugai and Simonsen2012). These interventions offer the potential to enhance the overall quality of life for both students and educators when implemented at a school-wide level. That is, function-based interventions can be effective at decreasing the rates of interfering behaviours among students and addressing other typical challenges encountered in educational institutions (Bradshaw et al., Reference Bradshaw, Waasdorp and Leaf2012), particularly as it relates to students with disability.
An essential aspect to consider in advancing behaviour strategies within the classroom is the feasibility and acceptability of these strategies among educators and teachers. The results from the Training Acceptability Survey (Table 1) indicated that participants expressed a high level of satisfaction. Participants from South Africa presented with an average rating of 6.4 out of 7, while participants from Vietnam presented an average rating of 6.1. State et al. (Reference State, Harrison, Kern and Lewis2017) assessed high school teachers’ perceptions regarding the acceptability and feasibility of behaviour interventions commonly employed to address the interfering behaviour of students with emotional and behavioural challenges. Their findings revealed that teachers are more inclined to implement behaviour strategies they perceive as feasible and directly relevant to their teaching practices.
The current study has a few limitations that are noteworthy. First, the videos for probe sessions and training modules featured adults in contrived scenarios rather than children engaging in more authentic behaviour. The scenarios portrayed in the videos may have been exaggerated compared to what one might encounter in a real-world setting. Additionally, due to COVID-19 restrictions in South Africa and limited access to participants’ work settings in Vietnam, we were not able to evaluate whether the skills learned during BST would transfer effectively to real classroom settings, where conditions may differ from those shown in the training videos. Future research should assess for generalisation of skills to ensure the generality of the training program. Regarding the study’s experimental design, ideally, pre-training (i.e., baseline) phases would have at least three data points to evaluate trend, level, and variability. Because of constraints related to the availability of participants and researchers at the international sites, not all participants were exposed to the minimum expected number of pre-training probe sessions. However, the replication of the study with participants from South Africa demonstrates the effectiveness of the training procedure in teaching educators to conduct an ABC assessment of interfering behaviours.
The findings of the current study replicate prior research by demonstrating the potential effectiveness of BST in teaching educators how to conduct an ABC assessment. Moreover, this study expanded on these results by including participants from South Africa and Vietnam. The outcomes of this research highlight the potential of BST to be widely employed in communities with limited or no access to behaviour analytic training. This can significantly enhance the accessibility of effective and ethically sound interventions for classroom management procedures.
Although significant, the outcomes of this study should be interpreted with caution due to the small number of participants and limited reach. Future research should emphasise the implementation of BST in training educators within larger groups to scale up the dissemination of behaviour technologies. Although BST is effective in group settings (Lloveras et al., Reference Lloveras, Tate, Vollmer, King, Jones and Peters2022), its cross-cultural applicability requires further investigation. Moreover, research should explore the effectiveness of BST when applied across different languages and involve translators to assist educators who may not share the same language as the trainer.
Supplementary material
To view supplementary material for this article, please visit https://doi.org/10.1017/jsi.2025.10002
Acknowledgements
We thank Luciana Du Plessis-Lang, John Holste, Eric Lafeber, and Allie Moehrle for their assistance with the creation of the materials and videos for this study, and Nguyen Thi Hoai Phuong, Hoang Thi Thanh Hue, and Van Tue Anh for their assistance with the translation of research material from English to Vietnamese. We thank Meral Koldas for her insightful feedback on the manuscript.
Funding
This study was supported by the University of Missouri South African Educational Program (UMSAEP) and UMSL College of Education Research Award.
Competing interests
We have no known conflicts of interest to disclose.
Open access
