The Irie Classroom Toolbox

This paper describes the development of ‘The Irie Classroom Toolbox’, a school-based violence prevention, teacher-training program for use with children aged 3-6 years. In depth interviews were conducted with Jamaican preschool teachers, who had participated in a trial of a classroom behavior management program, at post-test (n=35) and five years later (n=20). An on-going process evaluation was also conducted. Teachers’ preferred behavior management strategies and training methods were documented, and enablers and barriers to implementation identified. Teachers were most likely to adopt strategies that they liked, found easy to use, and were effective. These included paying attention to positive behaviour and explicitly teaching children the expected behaviour. Teachers preferred active, hands-on training strategies based on social-cognitive theories. Enablers to intervention implementation included positive teacher-facilitator relationships, choice, collaborative problem-solving, teachers recognizing benefits of the intervention, group support, and provision of materials. Barriers to intervention implementation were also identified. These data were integrated with behaviour change theory (the Behaviour Change Wheel and Theoretical Domains Framework) to develop an intervention grounded in common core elements of evidence-based programs while also utilising teachers’ perspectives. The resulting program is low cost, adaptable intervention that should be suitable for training preschool teachers in other low-resource settings.


Introduction
Early childhood is a particularly sensitive period as experiences in early childhood have long term effects on brain function, cognition and psychosocial functioning.A safe (free from physical and psychological harm), secure (consistent and predictable), and nurturing (sensitive and responsive) early childhood caregiving environment promotes child physical and mental health over the long term. 1 Training young children's caregivers in appropriate child behavior management strategies and in how to provide a secure and nurturing environment can reduce harsh caregiving and reduce violence against children. 2 There are a number of recent studies of violence prevention programs implemented with parents of young children in low and middle-income countries. 3,4However, less focus has been given to schoolbased violence prevention programs in LMIC, especially in the early childhood years. 57][8][9] In LMIC settings, school-based violence prevention programs have the potential to reduce the use of harsh punishment practices by teachers, including the use of corporal punishment. 5,10,11olence prevention is a leading public health problem in Jamaica and the wider Caribbean. 12Integrating violence prevention programming into the Jamaican preschool network is an important component of the primary prevention of violence for several reasons.Firstly, there is near universal enrolment of children aged 3-6 years in preschool education.
Secondly, there is a prevalence of 12% of children aged 5-6 years with externalising disorders and these children do no access appropriate services. 13Thirdly preschool classrooms have significant numbers of children with behaviour problems; in a study in 24 inner-city schools, 364/1733 children (21%) had four or more symptoms of conduct problems by teacher report. 14Furthermore, the three children in each class with highest levels of conduct problems were observed to display a median of 12 aggressive/destructive acts per hour indicating that these preschool classrooms are characterised by high levels of aggression.Finally, corporal punishment is widely used in preschools, despite the fact that it is banned by law. 15[18][19][20] Comprehensive programs, such as the Seattle Social Development Project use a combination of training programs. 21,22Although packaged violence prevention programs are available, the costs of these programs include initial purchase, on-going training, and materials and are prohibitively high for use in wide scale dissemination in LMIC.In addition, the programmes can be complex to deliver and have not been developed to match the context in early childhood classrooms in LMIC.These classrooms are often characterised by large class sizes, high child/staff ratios, few resources and paraprofessional staff.We found that making substantial adaptations to the Incredible Years Teacher Training Programme (IY-TT) led to benefits to teachers' behaviour that were of a similar magnitude to those produced using a minimally adapted version that required significantly more on-going support and supervision for teachers. 11The finding that substantial adaptations resulted in a more efficient intervention program, combined with the growing call for freely available programmes, 23 indicated a need for program developed specifically for use in low-resource contexts.
Available programs share many common core elements including core content and the use of behavioral / social learning and social cognitive theory.Hence, a more scalable and sustainable approach may be to transport evidence-based principles used in a range of effective programmes rather than evidence-based programs.McLoed et al. (2017) identified common practice elements to improve children's social-emotional competence and behavior in early childhood classrooms. 24Fourteen items were rated as essential (choices, emotion regulation, error correction, ignoring, instructive feedback, modeling, opportunities to respond, praise, premack statements, problem-solving, promoting behavioral competence, promoting teacher-child relationship, scaffolding, and social skills).Chorpita & Daleiden mapped the core components of evidence-based treatments for mental health problems in children and adolescents.The most common components of treatments for oppositional and aggressive behaviours were praise, time-out, rewards, commands and problem-solving. 25rland et al. (2008) identified common core elements of behavioral treatments for children's disruptive disorders and their parents including therapeutic content, treatment techniques, working alliance and treatment parameters (e.g.duration). 26For caregiver-mediated interventions, common treatment techniques included psychoeducation, assigning/reviewing homework, role playing/behavioral rehearsal, modeling, providing materials and reviewing goals and progress.This paper describes the development of 'The Irie Classroom Toolbox', a teacher training program with the following characteristics: i) it was developed specifically for use in early childhood classrooms in LMIC, ii) it has in-built flexibility and guidelines for adaptation, iii) it is available through a Creative Commons Licence (Attribution-NonCommercial-ShareAlike 4.0 International License), and iv) aims to facilitate widespread adoption of key child behaviour management strategies.The primary aims of 'The Irie Classroom Toolbox' are to 1) prevent violence against children by early childhood teachers and ii) prevent the early development of antisocial behavior in young children. 27'Irie' is a Jamaican word that describes feeling 'at peace and in harmony with oneself and with the world in general'.

Methods
The 'Irie Classroom Toolbox' was developed using guidance from the UK Medical Research Council framework for developing and evaluating complex interventions. 28,29The process of developing the toolbox involved integrating evidence, practice, and theory including: i) the data from the qualitative and process evaluation of an efficacy trial and ii) implementation science principles on the development of theory-informed interventions.The study progressed through three phases.In phase 1, data from the qualitative and process evaluation of an efficacy trial were used to identify enablers and barriers to intervention implementation in the Jamaican context.In phase 2, the data from phase 1 were integrated with theory to design the core intervention and core implementation components.In phase 3, the information from phase 1 and 2 were used to develop a theory of change and to design the intervention materials and structure.

Methods
We conducted a cluster-randomised trial of an adapted version of an evidence-based intervention, the Incredible Years Teacher Training Program (IY-TT) in 24 community preschools situated in inner-city areas of Kingston, Jamaica.Full details of the trial have been published previously and brief details are given here. 11,14Community preschools cater to children aged 3-6 years and are provided through community organisations (often churches) with oversight from the government.Parents pay a small fee and provide the necessary school resources (e.g.crayons, books, pencils).Preschools were selected for inclusion in the trial based on the following inclusion criteria: i) had 3-4 classes of children, ii) had at least 20 children in each class, iii) were situated in a specified geographical area and iv) all teachers consented to participate in the trial.Fifty schools in three education zones were surveyed and twenty-four of these schools met all the inclusion criteria and were recruited into the trial.
Twelve schools (with 37 teachers) were randomly assigned to receive the teacher-training programme.The training was delivered over a six-month period (November-April) by two facilitators and involved eight full-day teacher-training workshops and four 1-hour sessions of in-class support delivered once a month over four months.Teachers also received resources including educational materials (e.g.blocks, manipulatives, play doh), picture cards to assist with teaching core concepts, home-made bingo games, copies of positive notes home, stickers and stamps, behaviour planning forms and a small hand puppet.

Qualitative Interviews
Individual, in-depth, semi-structured interviews were conducted with teachers who had participated in the training programme at the end of the intervention period and again five years later.Thirty-seven intervention teachers were included in the evaluation and at the end of the intervention period, two teachers were unavailable for interview.All remaining teachers (n=35) were interviewed individually by a research assistant who was not involved in the study.Teachers had a mean age of 38 years, had been teaching for an average of 12 years, thirty-three (94%) of the teachers were female and only three were qualified teachers.Five years later, a purposive sample of 20 of these 33 teachers (1-2 teachers per school) were interviewed (eighteen females and two males).Using the post-test quantitative data from the trial, we created a proxy measure of teacher competence in the intervention which was the ratio of teacher positives to teacher negatives across one day of observation.Twenty teachers were selected: 10 who scored above the median on this measure and 10 who were scoring below the median, with at least one teacher from each of the twelve schools.Interviews were audiotaped and transcribed and transcriptions were checked for accuracy against the audiotape.Each teacher was allocated an ID number that was recorded on the audiotape prior to the interview and used on the transcript.Audiotapes were deleted once transcription and checking were complete.The data were analysed using the framework approach, which was designed for applied policy research and is particularly suitable for research with specific questions focussed on a priori issues. 30The framework approach involves a five step process: i) reading and rereading the transcripts, ii) identifying themes and subthemes and constructing an index of codes grouped into categories, iii) applying the codes to the individual transcripts, iv) rearranging the data in charts of each theme and/or subtheme with entries for each respondent under each theme, and v) examining the charts to identify the key characteristics of the data.One research assistant applied the codes to the data and prepared the thematic charts.Inter-rater reliability checks using percent agreement were conducted between the rater and the PI on fifteen transcripts from the first round of interviews and eight transcripts from the second round with high levels of agreement (>90%).Ongoing queries were discussed and resolved.

Process Evaluation
The intervention was delivered by two facilitators: one facilitator conducted all the teachertraining workshops and the second facilitator co-facilitated at the workshops and provided the in-class support.Throughout the intervention period, facilitators completed record forms after each workshop and after each in-class support session that documented: i) teachers' challenges, questions and suggestions for modifying the strategies to their context, ii) facilitator perspectives on the barriers and enablers to implementation at the level of the school, classroom, teacher and children, and iii) descriptions of how barriers were overcome.
Facilitators also kept an on-going log of their interactions with teachers and their reflections on the training.The data from the record forms and logbooks were analysed by extracting information on: i) the barriers and enablers to teachers' use of the strategies, and ii) common problems teachers face with each individual strategy and potential solutions.

Qualitative interviews at post-test
Themes from the qualitative interviews at the end of the intervention were broadly categorized as enablers and barriers to teachers' implementation of the intervention (Table 1).
Enablers included factors related to the intervention content and the intervention implementation.

Intervention Content :
Teachers reported what characteristics of the individual strategies introduced through the program led to their adoption in the classroom.Three main characteristics of preferred strategies were identified: i) the strategy was effective, ii) the teacher liked using the strategy, and iii) the strategy was easy to use."The praise, the greetings and coaching are the ones I use most.They are effective and easy to implement and they don't cost anything."(Effective and easy to use) "I like the smile on their face and they are happy, they make me happy."(Teacher and children enjoy it.) The most common preferred strategies included paying attention to positive behaviour (e.g. through praise, using rewards, narrating) and teaching children the required behaviour (e.g.teaching classroom rules).
Intervention Implementation: Teachers identified key characteristics of the training that led to change in their classroom behavior management practices.Teachers reported that the practical, hands-on aspects of the training were the most effective at helping them learn the strategies, (e.g.rehearsal and practice, group work, role plays) and these were preferred over the activities involving watching video vignettes followed by discussion.
"When you watch the videos and talk, you're kinda getting it.When it comes to activity time and you're participating, the actual activities help us a lot to come back and teach it."(Rehearsal and Practice) "The skits that you dome as a person has to do things before I can grasp it so doing that it help me to remember."(Skits/Role play) "When we don't grasp everything, they always come in to the classroom to help us to get it right."(In-class support) Teachers also reported the effect of modeling by the facilitators on their learning and motivation.
"When she do her praising, the way you look at her, you want to come and do the same thing.
When you see that, you want to come and do the same man.You're the adult and when she do it, the way you feel, just imagine the children."(Facilitators modeling strategies) The teachers also commented on other motivational aspects of the training including that the training methods used were fun, collaborative, supportive and non-critical.The barriers identified by teachers included difficulties with the intervention itself and external and internal factors that prevented their use of the strategies.

Difficulties with intervention.
Teachers reported on their difficulties with the individual strategies and it was evident that many of these difficulties were overcome with problemsolving on how to use the strategy and/or choosing a different strategy.

"What I had to do was just think and if that didn't work you had to think about why you think (that it didn't work)" (Problem-solving) "I try it and if it doesn't work, I try another one." (Choice)
Teachers also reported that some difficulties were resolved over time with practice.

"If it doesn't work then you go a step higher and it doesn't come over night and it's something that we have to practice, practice, practice." (Practice)
However, many teachers reported persistent difficulties using the strategies to manage children's misbehaviour (e.g.ignoring misbehaviour, giving consequences and time-out).

Sometimes it hard to look beyond the bad things. It hard fi nuh see the bad and see the good.
(Strategy is difficult to use)

"No care how you put him close to you and say this is your time out. He finds time to give trouble same way." (Strategy does not work)
Factors that prevent teachers' use of the strategies.Other barriers to teachers' implementation of the strategies were factors relating to the school (e.g.large class sizes), the teacher (e.g.difficult to change old habits), the children (e.g.children's socialisation at home) and government policies (e.g.demands of the curriculum and dress code) (Table 2).

"In here is clustered so I can't get to circulate to everybody and the heat and the crowded classroom, if I get a smaller amount of kids next year it would be best." (School factors)
"With me, it's a self-thing.Don't allow the self to get too upset and then you'll be able to deal with it better.Don't make it a personal issue or get upset easily when they behave bad." (Factors related to self)

Qualitative interviews at 5-year follow up
Five years after the end of the intervention, teachers reported that the most commonly used strategies were praising children, giving rewards, explicitly teaching children the expected behavior, giving clear instructions, and interactive reading (Table 2).Teachers' rationale for using these strategies was similar to that reported in the earlier round of qualitative interviews: teachers reported that these strategies were effective, they and/or the children liked the strategy, and they were easy to use.Engaging in problem-solving and having a choice of different strategies to manage children's misbehaviour continued to be strong themes.Strategies that were not used included consequences and time-out.The main reasons for not using these strategies included: i) strategy had unintended consequences (e.g. would distract other children), ii) they were difficult to use, iii) they were ineffective, iv) teacher and/or children didn't like the strategy, v) they prefer to use alternative strategies/don't need to use that particular strategy, and vi) they had forgotten about that strategy.Time-out was particularly problematic for teachers and the majority of teachers found this strategy very difficult to use due to the lack of space in the classroom and because children would find other things to do in time-out and/or would distract other children: "He was in the time-out chair, doing all sort of things to disturb the class and everybody.The children were laughing as if it was something good."There were no obvious differences in preferred strategies for teachers with different levels of competence at post-test; all teachers reported sustained use of praise and explicitly teaching the expected behaviour and the majority didn't use time-out.

Process Evaluation
Through the process evaluation, facilitators' perceptions of the enablers and barriers to teachers' implementation of the strategies were documented.

Enablers.
The key enablers identified by facilitators were positive facilitator-teacher relationships, the provision of choice and autonomy, teachers recognising the effects of their behaviour on the children and engaging teachers in collaborative problem-solving.These factors were considered to be instrumental in overcoming teachers' resistance to the intervention.Teachers were willing to to try the strategies partly because of the confidence and skills they had gained from attending workshop and partly because they liked the facilitators and didn't want to disappoint them, demonstrating the importance of positive relationships.Providing teachers with the option of choosing strategies that fit their teaching style and personal preferences also helped to overcome resistance to individual strategies.In the early stages of implementation, teachers often disagreed with individual strategies, found them difficult to use in their context and/or felt awkward and uncomfortable using them.
Provision of autonomy and choice helped to overcome teachers' reluctance to try aspects of the intervention and had the added advantage of modeling this behavior for teachers so that they could see the value of using it with the children.When teachers used the strategies, they were motivated to continue when they recognized the benefits to the children and to themselves.Teachers were supported to recognize these benefits by the facilitator during the in-class support sessions and through self-monitoring when completing homework assignments.The importance of engaging the teachers in collaborative problem-solving to ensure that they gained confidence and competence in using the strategies in a range of contexts, with different children and for different child behaviors was also a recurring theme in the process evaluation.

Barriers. Examples of barriers identified through the process evaluation included i) teachers
have inappropriate expectations of young children (for example, teachers think that once children have been taught a behavior, they should perform it consistently without further reminders), ii) norms supporting the use of harsh punishment (e.g. a belief that some children can only be managed through using corporal punishment and/or that some behaviors deserve to be punished using corporal punishment), iii) high child/staff ratios and insufficient resources, combined with teachers' inappropriate expectations and/or poor planning skills can lead to extended periods of time when children are not engaged from learning activities, iv) many teachers are not reflective about their own teaching practices and tend to be reactive rather than proactive, v) teachers need support to help them to generalise their use of the strategies to different contexts and for different behaviors across the school day, and vi) many teachers lack confidence in and commitment to using the strategies, especially when they are newly learned and the teacher has had insufficient practice in the classroom.The process evaluation also highlighted the persistent difficulties teachers faced implementing time-out for misbehavior.In some instances, time-out was over-used and teachers resorted to time-out for even minor misbehaviors.In other instances, time-out was ineffective as the teachers were unable to manage the child in time-out while continuing to monitor and teach the other children in the class.
Throughout the process evaluation, common problems faced by teachers when implementing the individual strategies were documented.These common problems were used to create a list of common problems and potential solutions list for each of the main strategies to be included in the Irie Classroom Toolbox (see supplementary online material).

Summary of Qualitative and Process Evaluation
The qualitative and process evaluation produced some comparable findings related to enablers and barriers to teacher implementation of the intervention and to teachers' preferred and less preferred strategies.Teachers and facilitators reported that the strategies related to managing misbehaviour were most difficult to implement and recognized the importance of problemsolving, choice and positive and supportive relationships in motivating teachers to start using, and encouraging teachers' sustained use of the strategies.Some unique information was also gained from each set of informants.Teachers provided information on the reasons they adopted aspects of the content of the intervention in the classroom (strategies that are easy to use, that are perceived as effective and that the teacher and children enjoy) and also stated their preferred training methods and the underlying rationale for their preferences.Facilitators identified barriers at the level of the teachers whereas the teachers primarily identified barriers related to external factors (e.g.school, government policies, children).

Intervention Content: Identifying the Core Intervention Components
The content of the Irie Classroom Toolbox was informed by the work of Embry who advocates for the use of 'evidence-based kernels' in prevention programs targeting behavior change. 31,32Evidence-based kernels are described as 'irreducible units of behavior change technology' that have been shown through experimental evaluation to affect behaviour and they are appropriate for use across contexts, that is, at school and at home. 31,33Examples include verbal praise, non-verbal transition cues, positive notes home, and time-out.7][18][19] The Irie Classroom Toolbox was developed to include evidence-based kernels that are i) used across a range of evidence-based child behavior modification interventions and ii) acceptable, feasible and effective in the Jamaican preschool setting based on the data from the in-depth interviews and process evaluation.

Intervention Process: Identifying the Core Implementation Components
The theoretical frameworks used to categorise the implementation components included the Behavior Change Wheel 34 and the Theoretical Domains Framework. 35,36The Behavior Change Wheel is based on a systematic review of existing frameworks of behavior change interventions.The core of the model is represented by the COM-B system which states that three factors: Capability, Opportunity, and Motivation are required for a behavior to occur and to be sustained (Figure 1).This model can be was used to design interventions by determining what behavior change is required and what aspect of the system needs to be targeted.Hence, once the barriers and enablers to performing the behaviors have been identified, a decision can be made as to whether the individual needs to gain skills, increase motivation, and/or be provided with opportunity to use the behavior.The Theoretical Domains Framework consists of a set of fourteen domains to describe the factors influencing behavior change: knowledge, skills, social/professional role and identity, beliefs about capabilities, optimism, beliefs about consequence, reinforcement, intentions, goals, memory, attention and decision processes, environmental context and resources, social influences, emotion, and behavioral regulation. 35This framework provides a comprehensive coverage of factors influencing behavior and provides a theory-informed approach to identifying appropriate evidence-based behavior change techniques to target specified behaviors. 36,37By combining the COM-B system of the Behavior Change Wheel with the Theoretical Domains Framework it is possible to identify what factors influencing an individual's capability, motivation, and opportunity need to be targeted to effect behavior change.Once these factors have been identified, behavior change techniques can be mapped onto these key determinants of behavior change. 36In the development of the 'Irie Classroom Toolbox', the barriers to teachers' implementation of the strategies and teachers' preferred behavior change techniques were identified through the qualitative and process evaluation.These barriers were linked through the COM-B system to the domains from the Theoretical Domains Framework.
Behavior change techniques were then chosen to address each barrier, prioritising teachers' preferred techniques where possible.

Intervention Content: Identifying the Core Intervention Components
Table 3 shows the strategies or 'evidence-based kernels' that are included in the Irie Classroom Toolbox.The strategies were divided into four modules: i) creating an emotionally supportive classroom environment, ii) preventing and managing child behavior problems, iii) teaching social and emotional skills, and 4) individual and classwide behavior planning.
Teachers are thus provided with a menu or 'toolbox' of strategies that they can use to manage children's classroom behavior and teachers choose strategies according to the needs of the children, their classroom context, and their own preferences.Each strategy can be used alone or in conjunction with other strategies.A particular emphasis is placed on the strategies that were preferred by teachers in the qualitative and process evaluation giving four key components: i) paying attention to positive behavior, ii) explicitly teaching children the required behaviors, iii) telling children what to do, not what not to do (clear instructions), and iv) interactive reading.The non-preferred strategies were the strategies to manage child misbehaviour.Given the importance of equipping teachers with skills to manage child misbehaviour, training teachers in the use of a discipline hierarchy to manage child misbehavior was incorporated as a fifth key component.The discipline hierarchy integrated preferred (e.g.paying attention to positive behaviour, clear instructions, explicitly teaching skills) with non-preferred (e.g.withdraw attention, consequences) strategies.Teachers are trained how to choose appropriate strategies when faced with a variety of child behavior problems and how to move up or down the discipline hierarchy depending on the effect of the strategy on the child's behavior.Time-out was omitted due to the significant amount of time required to train teachers to use it, its lack of acceptability and feasibility, and its perceived ineffectiveness in the Jamaican preschool context.

Intervention Process: Identifying the Core Implementation Components
Table 4 shows how the barriers to implementation identified through the qualitative and process evaluation were linked through the COM-B system to the domains from the Theoretical Domains Framework and then the behavior change techniques mapped to address each barrier.For example, one barrier was that teachers believe that harsh punishment is necessary to control child behavior and do not believe that positive disciplinary practices are effective for certain behaviors and/or certain children.This barrier affects teachers' motivation and was mapped to the theoretical domain 'Beliefs about Consequences'.
Suggested behavior change techniques include providing feedback on the effect of the teachers' behavior on the child(ren) and assigning homework for teachers to record the effect of using the strategies on individual and class-wide child behavior.Evidence, theory, and practice were integrated throughout; although the behavior techniques used were underpinned by theory, teacher preferences and practical considerations relating to feasibility were used to inform the development of the toolbox.For example: i) the intervention components were chosen to be feasible in the context, ii) some barriers are not amenable to intervention (e.g.school factors (e.g.lack of resources, class size)) and hence collaborative problem-solving on how to use appropriate behavior management strategies is a key aspect of the intervention, iii) teachers' preferred methods of training were prioritised when possible (e.g.rehearsal and practice, group work, role plays, positive feedback), and the training workshops were designed to be fun and engaging, iv) facilitators' modeling the use of the strategies in their interactions with teachers was found to be a very powerful training method and this became a core feature of the training program, v) teachers are trained in how to choose strategies that fit their own preferences, needs and teaching styles through the use of role plays, rehearsal, practice, and collaborative problem-solving, and vi) the toolbox includes games, songs, activities and short pictorial story books that require no additional resources and that teachers can use effectively with relatively large classes of young children.

Phase 3: Intervention Development
In the final phase, two further steps were conducted: i) the core intervention content and core implementation components identified through phase 2 were integrated to develop the 'Irie Classroom Toolbox' and ii) the information from phase 1 and phase 2 were used to design the theory of change.
The five key intervention components are emphasised throughout the program (see Figure 2).Although a full menu or 'toolbox' of strategies is introduced, these strategies and their underpinning rationale are reinforced the most.Each key component represents a package of related strategies; for example, 'paying attention to positive behaviour' incorporates verbal and non-verbal praise, specific feedback on children's correct response/behaviour, rewards, narrating, and positive notes home.The intervention also includes a 'menu' of other tools or strategies that teachers can use to provide an emotionally supportive classroom environment and to manage young children's behaviour This combination of choice and prescribed content ensures that teachers have the opportunity to choose their own strategies and tailor the intervention to their own context, while ensuring that all teachers use a small number of key components that have evidence of effectiveness in the Jamaican preschool setting.This is designed to maximise teacher engagement, reduce resistance, and to promote continued use of appropriate strategies and its inherent flexibility makes it suitable for wide-scale implementation.
The core implementation components include behaviour change techniques to promote teacher skills (e.g.rehearsal and practice), motivation (e.g.use of specific positive feedback), and opportunity (e.g.provision of necessary resources).The training is delivered through facilitated teacher-training workshops, individual in-class support sessions, and text messages.Teachers are given practical classroom assignments after each in-class support session and receive intervention manuals and materials (picture cards and story books) to use with the children (Table 5).Intervention facilitators are trained to deliver the training using the same behaviour change techniques that they will in turn use to train teachers.Facilitators are also trained in how to form a therapeutic alliance with teachers and manage resistance.
Two training manuals are available: a fully scripted training manual to deliver the facilitated workshops and a manual with protocols for the in-class support sessions.
The Theory of Change is shown in Figure 3. Teachers are trained in the core intervention components using the core implementation components.Through participation in the program teachers gain skills and motivation and are afforded the opportunity to use the strategies.As teachers begin to use the strategies, with support and feedback during in-class support sessions, they perceive benefits to the children and attribute these benefits to their use of the strategies.Strategies that are easy to use and strategies that are liked by teachers and children are integrated into teachers' daily practices.As teachers gain in competence they learn to generalise their use of the strategies to other situations and contexts.This leads to improved child outcomes and improved teacher well-being which in turn acts as a feedback loop to sustain teachers' use of the strategies in the classroom.Teachers' use of violence is reduced due to i) teachers use alternative strategies to manage child behavior, ii) child behavior has improved, and iii) teachers have increased well-being.

Discussion
This paper describes how evidence, theory, and practice were used to inform the development of a violence-prevention, teacher-training program for use with early childhood teachers in Jamaica.The key features of the intervention are: i) it was developed with close participation of Jamaican preschool teachers and incorporates teachers' preferred behaviour management strategies and preferred training methodologies, ii) it is theory-informed with a clear theory of change, and iii) it was designed to be feasible in low resource contexts.][26] Although the Toolbox includes a wide range of strategies, a selection of core behaviour management practices are packaged into chunks to form five key components which are repeated and reinforced throughout the intervention.This is aligned to the 'seven, plus or minus two' rule of psychology and helps prevent teachers from being overloaded with information.A similar approach has been recommended for parenting interventions to improve child development in LMIC. 38Four of these five key components are based on teachers' preferences.Research in Chile has shown that training teachers to incorporate novel practices into their existing skill repertoire is more difficult than training teachers to use skills and activities that are 'native' or common in the setting. 39Incorporating teachers preferences into the intervention is designed to increase its acceptability and effectiveness.Teachers most preferred strategies post-intervention were explicitly teaching the required behaviors and use of praise and rewards; these strategies were also used frequently at five year follow-up.These preferred strategies correspond to the key principles underpinning the Seattle Social Development Program (SSDP), a comprehensive intervention designed to prevent antisocial behaviors and promote child competencies, with evidence for long term gains in the US context. 8,40The SSDP strategy suggests that to promote healthy behaviors children need to be: i) taught the required skills, ii) provided with the opportunities to use the skills, and iii) provided with positive feedback and recognition. 41Other preferred strategies were interactive reading and clear instructions.In addition to building on teachers' preferences and strengths, the Toolbox includes core content fundamental to child behaviour management programs that was less preferred by teachers.These were strategies to manage child misbehaviour such as ignoring and use of consequences.To enhance the acceptability of these less preferred practices, a discipline hierarchy that integrates preferred and non-preferred strategies was incorporated as a fifth key component.This discipline hierarchy also incorporates teachers' use of choice and problem-solving skills that were identified as enablers to intervention implementation through the process evaluation.Problem-solving skills have been reported to be an important implementation strategy in integrated nutrition and stimulation interventions in LMIC and for effective scaling of evidence-based perinatal interventions in South Africa. 42,43e enablers to intervention implementation included factors that promoted teachers' skills, motivation, and opportunity.5][46][47] Other aspects of implementation that have been found to be important in ECD implementation research include group support from peers and provision of intervention materials and these were also recognised as enablers to intervention implementation by Jamaican preschool teachers in this study. 44,45These common enablers across several studies of implementation of ECD programs correspond to the following domains in the Theoretical Domains Framework: i) skills, ii) beliefs about consequences, iii) social influences, and iv) environmental context and resources.This suggests that these four domains may be particularly relevant for LMIC contexts.
The processes used to train teachers include all six categories of behavior change communication that have been shown to be effective in health programs in LMIC: informational techniques, performance-based techniques, problem-solving, social support, materials, and media. 48The most effective programs were found to use three or four of these categories and the authors conclude that this is because participants are engaged at the behavioral level (e.g.performance techniques), the social level (e.g.social support), the sensory level (e.g.materials), and the cognitive level (e.g.problem-solving, informational techniques).The Toolbox also includes what Han & Weiss (2005) describe as the 'essential ingredients' of a sustainable teacher-implemented mental health program in that the intervention is i) acceptable to teachers, ii) effective, iii) feasible to implement with minimal resources, and iv) flexible. 49Finally, the training is designed to be entertaining, engaging and fun which is important for maintaining teachers interest. 50The importance of interventions being acceptable, feasible, and fun was also highlighted in the evaluations of an early childhood parenting program implemented in Brazil and Zimbabwe. 44ere are several advantages to using theory-informed behavior change interventions.
Firstly, by linking the behavior change techniques through theory to the problems that need to be addressed through the intervention, the hypothesised mechanisms of change are made explicit and this allows for an exploration of the mediators of intervention effectiveness with further refinement of theory. 5146][47]53 Combining formative research with the literature on common core components of effective interventions has been used to develop a parent program to prevent child abuse in South Africa. 54The unique factor in the development of the Irie Classroom Toolbox is the integration of three sources of information: i) formative research, ii) the literature on common core components, and iii) behaviour change theory, to create an implementation intervention for use with early childhood teachers in Jamaica.This process of integrating qualitative data with evidence-based theory and practice is appropriate to develop, adapt, and refine ECD interventions in LMIC.The process would also be relevant for other public health and education programs.
One of the primary aims of the Irie Classroom Toolbox is to prevent violence against children by teachers.Corporal punishment at school is prohibited in 131 countries and permitted in 68. 55However, even in countries with a legal prohibition, corporal punishment in schools continues to be widespread.Gershoff reported the prevalence of school corporal punishment in 29 countries with a legal ban and the prevalence estimates ranged from 13% to 97%. 56Thus, while legislative reform is necessary, additional actions are required including advocacy and public education campaigns and training for teachers in positive discipline practices.Although the Irie Classroom Toolbox, was developed for the Jamaican preschool context, the content and process of training is likely to have wide applicability to other LMIC with an established preschool network.The toolbox can also be adapted for use in the early grades of primary school and we have used an adapted version with grade one teachers in Jamaican primary schools.The toolbox is being made available through a Creative Commons Licence to the international community and has potential to make a strong contribution to the global agenda on violence prevention.Example:  Demonstration of the required behavior (through role plays and demonstrations)  Rehearsal and practice to learn new skills  Supervised practice in using the strategies in the classroom  Classroom assignments to encourage teachers to practice using the skills in different contexts Teachers forget to use the strategies  Provide lesson plans and activity guides to help teachers integrate the use of the strategies into their everyday teaching and learning activities  Help teachers to make a detailed plan when, how and how often they will use the strategies  Engage teachers in collaborative goal setting.Children are often not engaged in learning activities (e.g.teacher asks them to put their heads on the desk until lunchtime).

 Teacher manuals  In-class support
Technique: Action planning, Provision of Activities Example::  Provide games, story books and songs that teachers can use during transition times (e.g.waiting for lunch)  Help teachers to make a detailed plan for how they will ensure that children are engaged in learning activities

A
semi-structured interview guide was used at both time points and interviews were conducted in a quiet location within the community preschool and lasted between 45 minutes to 1 hour.At post-intervention, the interview guide focussed on i) what aspects of the training content and process teachers liked best and were most effective, and ii) barriers and enablers affecting teachers' implementation of the strategies.At the 5-year follow-up, the main focus of the interviews was on the extent to which teachers continued to use the strategies over time and what factors affected sustainability of teacher implementation.Teachers were asked to sort cards describing the strategies into three categories: i) used a lot, ii) used sometimes, and iii) not used.Teachers were then asked to explain the reasons for their sustained use or nonuse of each individual strategy.All teachers gave signed informed consent to participate in the study at both time points and ethical approval for the research was given by the University of the West Indies Ethics Committee (approval numbers: ECP 222, 2008/2009, ECP 148, 10/11).
'evidence-based kernels' can be combined to form 'behavioral vaccines' which are daily practices to promote longer-term behavior change.Key characteristics of evidence-based kernels and behavior vaccines are that they are low cost, have easily discernible positive effects, and are easy to use.Evidence-based, behavioural interventions for use in early childhood classrooms, including school-wide, teacher-training, and child-training programs were reviewed to identify the evidence-based kernels used.Content was informed mainly through Peacebuilders school-wide program, the IY-TT program, the PATHS child training Secondly, mapping behavior change-techniques onto barriers to intervention implementation through theory provides a framework for adapting interventions to other contexts by making changes to the content, context and/or to the training of facilitators. 52For example, i) researchers can adapt the intervention based on barriers that are relevant in their context, ii) the intervention can be delivered through alternative means, for example, a different configuration of workshops and in-class support can be used and/or the intervention can be delivered through a different modality, provided the core intervention and implementation components are intact, and iii) as multiple behavior change techniques can be used to address each barrier to intervention implementation, if a shorter or less intensive intervention is required, it may be possible to select a subset of intervention components and retain effectiveness on teacher and child outcomes.Thirdly, the transparent nature of the intervention facilitates staff training.For example, mapping intervention components to the COM-B framework for behavior change helps facilitators identify what intervention components need to be prioritised to enhance teachers' skills, motivation, and/or opportunity (depending on teacher need), thus providing a concrete framework for facilitators to use.Also facilitators can use the framework to identify barriers that individual teachers are facing and prioritise the appropriate behavior change techniques to address that barrier in their interaction with the teacher.Finally, utilising a theoretical framework makes the intervention more transparent and facilitates communication with policy makers.

Figure 1 :Figure 2 :
Figure 1: The COM-B system for understanding behavior Text-messages, Provision of resources Example: Fortnightly text messages to remind teachers to use the strategies  Provision of intervention manuals, picture cards and story books and an 'I am an Irie Classroom Teacher' for affirmation for teachers to display in their classroom at the end of the intervention period.Teachers find it difficult to generalise the use of the strategies to new situations Skills  Facilitated workshop  In-class support  Assignments Technique: Demonstration, Role play.Rehearsal, Homework Example:  Demonstrations of how to use the strategies across different situations and in different contexts  Rehearsal and practice involving using the strategies in different situations and for different child behaviors to aid in generalisation of recently learned skills  Supervised practice in using the strategies in the classroom  Classroom assignments to encourage teachers to practice using the skills in different contexts MOTIVATION Feedback on behavior, Self-evaluation/self-monitoring, Homework Example:  Positive, supportive feedback from facilitator and peers in workshop after demonstrating use of the strategies  Positive, supportive feedback from coach during in-class support sessions. Prompting teacher to identify what s/he did well after performing the skills  Positive, supportive feedback on classroom assignments Teachers believe that use of harsh punishment techniques is necessary to control child behavior/do not believe that more positive Feedback on outcome of behavior, Self-monitoring of outcome of behavior, Homework, Modeling Example:  Pointing out the effects of the teachers' behavior on the children during in-class support sessions. Prompting teachers to identify the effects of their behavior on themselves. Classroom assignments: teachers record the effects of using the strategies on children. Facilitators model the strategies in their interactions with teachers so that teachers have direct experience of how the strategies lead to changes in behavior and motivation.Teachers are frustrated by child misbehavior Emotion  Facilitated workshop  In-class support Technique: Coping skills, Cognitive restructuring/Reframing Example:  Assist teachers in reframing child behaviors to change their emotions when faced with perceived misbehavior. Help teachers to develop individual strategies to manage frustration/negative emotions in the classroom. Promoting fun: activities are designed to promote teacher and child enjoyment.Teachers believe that an effective teacher is a strong disciplinarian Professional Role  Facilitated workshop  In-class support Technique: Feedback on behavior and Outcome of behavior, Cognitive restructuring/reframing, Identification of self as a role-model Example:  Positive, supportive feedback to enhance self-efficacy in their use of the strategies  Encourage teachers to view use of the strategies as integral to being an 'effective early childhood teacher'  Encourage teachers to view themselves as a role model for the children. Assist teachers in reframing child behaviors and teachers' responses to those behaviors to change their cognitions Lack of reflective practice: reactive rather than proactive Goals  Assignments  In-class support Technique: Self-monitoring, Action Planning, Goal setting, Homework Example:  Self-monitoring of use of strategies through classroom assignments  Help teachers to make a detailed plan when, how and how often they will use the strategies  Help teachers to observe children's responses to judge the effectiveness of their behavior. Engage teachers in collaborative goal setting.Strategies do not always work and/or strategies sometimes Self-monitoring, Homework, Therapeutic Alliance, Reframing, Coping skills, Problem-Solving Example:  Help teachers to observe children's responses to judge the effectiveness of their behavior. Classroom assignments: teachers record the effects of using the strategies on children. Assist teachers in reframing child behaviors to change their emotions when faced with perceived misbehavior. Help teachers to develop individual strategies to manage their own frustration/negative emotions. Engage teachers in problem-solving.Teachers have personal problems with interfere with their use of the strategies Emotion  Assignments  In-class support Technique: Positive relationships/Therapeutic alliance Example::  Develop positive, supportive relationships with each teacher. Use reflective listening/empathy OPPORTUNITY Norms reinforce use of harsh punishment practices Social influences  Facilitated workshop  In-class support Technique: Social processes of encouragement, pressure and support Example:  Group support: teachers share experiences, successes and challenges  Group activities: teachers work together in small groups to practice using the strategies and to engage in joint problem-solving. School support: all teachers within a school are trained to promote changes school norms Teachers have insufficient resources and the classroom context does not support the use of the strategies Environmental Context and Resources  Provide materials  In-class support  Teacher manuals Technique: Resource provision, Problem-Solving Example:  Provision of necessary resources: linoleum, educational materials, picture cards, puppet. Engage teachers in collaborative problem-solving to overcome barriers to implementing the strategies. Activities provided for teachers can be integrated into on-going teaching and learning activities Teachers lack time to use strategies that require teaching and rehearsing skills and playing games Environmental Context and Resources  Teacher manuals  In-class support Technique: Action planning, Goal setting, Provision of Activities Example::

Table 1 :
Interview themes with preschool teachers at the end of the intervention period

Table 2 :
Preschool teachers' reports of the extent to which they continue to use the training in the classroom five years after the end of the intervention

Table 3 :
Content of the Irie Classroom Toolbox

Table 4 :
Using COM-B and the Theoretical Domains Framework to map behavior change techniques to barriers to implementation Use role plays and skits followed by discussion to lead teachers to understand about developmentally appropriate practices for young children (e.g. to help teachers have appropriate expectations of child behavior, understand the reasons for child behavior and the theoretical principles underlying evidence based behavior management practices).

Table 5 :
The Irie Classroom Toolbox: content, structure and process of training