This study used a randomized controlled trial (RCT) design. Schools (N = 20) were matched in pairs using school size and a socioeconomic distribution key (reference?) based on a combination of the median income and level of education of the schools’ parents, and the number of cases reported to child services. The pairs were then randomly allocated to either immediately start intervention or a 1-year wait-list control by an independent third party blinded to school names/details. Schools were randomized to intervention or 1-year waitlist control conditions after all schools for each year were recruited 6-months prior to intervention delivery for the year 2023 (year 1; 10 schools) and 2024 (Year 2; 10 schools).

All schools were recruited from Oslo, which is divided into 15 districts. From the 116 elementary schools in Oslo, 14 schools were excluded because of prior exposure to TIKiS. From a final pool of 112, all schools were invited and 20 schools (18%) agreed to participate in the study. Schools were largely recruited from Kompetanseløftet for Inkluderende Praksis (The competence lift for inclusive practice). Kompetanseløftet is a meeting attended by principals and representatives from schools and kindergartens in Norway to learn of ways to increase competence in these institutions. TIKiS was presented at this meeting because it was consistent with this focus on school competence, and thus fits with an overall strategy of the schools within these districts.

After Kompetanseløftet, interested schools emailed the project team, and were sent a more detailed description about the study and invited to an online information meeting. The deadline for enrolling in the project was 1 week after this online meeting. The final sample of schools were recruited based on the school leaders’ engagement and consent to take part in the intervention.

Recruitment and retention can be difficult in school interventions (other school intervention refs) and were challenging during the pilot study (Bølstad et al., 2023). As such, in the current trial we worked closely at the municipality level and then with the school leaders so that they could support and encourage the teachers to engage in the study. Meetings were held with the school leaders after recruitment and prior to the intervention to plan the next year and to increase engagement. Some schools required extra meetings and support to plan delivery.

The initial 3-h introductory seminar for all school staff was held before the start of the school year. The seminar was given to all staff to ensure wide support and understanding of the TIKiS approach which is part of increasing consistency of responses throughout the school environment. The seminar was led by certified Tuning in to Kids facilitators, including the second author. Content included teaching about children’s emotional competence (emotion expression, understanding, empathy and emotion regulation), the role of temperament, the relationship between needs and emotions, beliefs about emotions (meta-emotion), the different ways adults respond to children’s emotions (e.g., emotion dismissing, coaching styles), the role of self-care and the emotions that occur in a school setting.

In the end of September and beginning of October 2023 and 2024, group delivery of TIKiS began with teachers in Grades 1–4. The group sessions were delivered by trained facilitators using a structured TIKiS manual. Each session included handout materials and followed a structured outline that included introductory exercises, group reflection, roleplays, teaching of concepts, exercises and home activities. Teachers attended six group sessions, delivered once every 2 weeks for 90 min during the Autumn semester in the first half of the school year (September to December). The groups consisted of 6–8 teachers in Year 1 and required a minimum of four participants to be delivered. Due to significant illness and absenteeism, group sizes were increased to 8–12 teachers in Year 2. Teachers who were not present for sessions were sent materials from the session and asked to read this prior to the next session. They could also receive a summary of the session over the phone, in person or via email from the TIKIS facilitator.

TIKiS was delivered in the teachers’ workplace during their working hours. School leaders scheduled the dates, times and rooms for intervention delivery that were suitable for the teachers. The leaders were responsible for ensuring that the teachers knew when and where to meet for each session, and a register of attendance was kept by the TIKiS facilitators.

TIKiS used roleplays in each session to draw on teacher’s experiences with students and practice how emotion coaching could be used during these interactions. Roleplays included use of scripted roleplays that contrasted emotion dismissing with emotion coaching response styles, fishbowl roleplays where the whole groups engaged in the roleplay; and pairs/threes doing unscripted roleplays using teacher’s own examples to practice the skills with support from facilitators. Additionally, teachers were encouraged to use the ideas taught in the sessions in their interactions with students during the week and were given home activities between each session to build skills. See Table 2 for details on session content.

School leaders received three 90-min group sessions to support teachers learn and apply emotion coaching skills. The sessions were led by certified TIKiS facilitators, who were part of the project team and experienced employees in pedagogical psychological services and the school health services, respectively. This was to ensure school leader engagement and long-term sustainability of the intervention, and to help the school leaders support their teachers to implement the TIKiS content.

TIKiS group facilitators were recruited from the municipal school health services (nurses in the school health services, Skolehelsetjenesten), city-wide educational-psychological services (pedagogues and psychologists from Pedagogisk psykologisk tjeneste) and municipal mental health services (Oslohjelpa), to deliver the 6-session TIKiS groups to teachers. Each TIKiS group was delivered by two facilitators, one from school health services and one from pedagogical psychological services. These employees have complementary competences and roles in schools. The municipal services had a more mental-health-oriented perspective, and mostly had familiarity with the schools they were working in. The facilitators from the educational-psychological services had expertise with group delivery with teachers, as well as understanding the teacher’s role in the classroom. Additionally, they often work with the same students without collaborating with each other, therefore the project aimed to improve the relationship between the two institutions as a wider goal for the intervention. All TIKiS facilitators had at least a master’s degree in psychology, pedagogy, special pedagogics or similar.

All TIKiS group facilitators completed a two-day TIKiS training over 16 h. Facilitators received training in March 2023 and in May 2024, a few months before intervention delivery. This training was similar to the Tuning in to Kids facilitator training but with adaptations for running groups with teachers in the primary school setting. Training included the program philosophy, research and the content for each of the six group sessions plus how to deliver all the main exercises and processes of the program.

Following training, an additional collaboration workshop was held shortly before the TIKiS groups began with the second author (Skoe). This 90-min workshop explored how to enhance the relationship between the school health services and the pedagogical psychological service employees as they worked together to deliver the intervention to the groups, how to ensure fidelity to the program manual, working with teacher’s motivation and engagement, and mastery of the group facilitator role. The workshop summarized the core elements of TIKiS and included discussion and roleplay practice to ensure facilitators were competent with using roleplay.

Supervision was provided for facilitators by the second author on three occasions during the delivery of each TIKiS group: once immediately prior to session 1; after session 2 and after session 4. Supervision addressed the content being delivered in the sessions, explored issues and challenges arising (i.e., resistance to the program content), and built the understanding of the facilitators about core program elements (such as the role of meta-emotion philosophy, how to support teacher’s self-care or building the emotion coaching skills).

This study used a modified version of the Coping with Children’s Negative Emotions Scale (CCNES; Fabes et al., 1990) to measure emotion socialization practices by exploring teacher’s reactions to student’s negative emotions. The questionnaire was originally designed for parents and translated to Norwegian by psychologists for a study of TIK with parents of kindergarten aged children (Bølstad et al., 2021). This modified measure was adapted for use with Norwegian teachers (CCNES-Teacher: Bølstad et al., 2023) and shortened to 35 items from the original 72 because teachers were not compliant with long questionnaires in the pilot study. Additionally, some of the subscales were changed or removed to be consistent with other TIK studies to improve validity of the sub-scales, and to ensure the CCNES was more culturally appropriate. Two subscales were added: Empathy/Acknowledgement of emotions, and Distraction.

Empathy/Acknowledgement is consistent with Gottman et al. (1996) concepts of the 5 steps of emotion coaching, and Distraction, measures teachers’ use of distraction as a strategy for responding to children’s emotions—also found by Gottman et al. (1996) to be an emotionally dismissive response style. In addition, the Punitive Reactions, Distress Reactions and Emotion Focused Subscales were removed because of poor validity (Bjørk et al., 2020; Bølstad et al., 2021). Subscales were combined into two dimensions measuring supportive or non-supportive emotion socialization practices. The Emotion Coaching (supportive) dimension consisted of Empathy/Acknowledgement, Expressive Encouragement and Problem Solving; the Emotion Dismissing dimension consisted of Minimization Reactions and Distraction. Using confirmatory factor analysis, the 35-item version of the CCNES was found to have acceptable reliability and validity. For more details see Koleini et al. (2026).

In the 35-item version, CCNES presents seven scenarios, with five different responses to each scenario, so each response corresponds to one of the five emotion socialization subscales. Teachers were asked to respond with the likelihood they would respond with each reaction on a scale of 1 (very unlikely) to 7 (very likely). For example, “If a student’s jacket has become damaged during recess, and they have become angry and disruptive, I would (a) help the student find out how the jacket can be repaired (Problem Solving), (b) support the student in the fact that it is terrible to have your jacket damaged (Empathy), (c) ask the student to calm down and say that the jacket can be repaired (Minimization Reactions), (d) encourage the student to express their anger in words (Expressive Encouragement) or (e) ask the student if they have any fun plans for the weekend (Distraction).

The Affect Integration Inventory 18 (AII; Solbakken and Monsen, 2017) was used to measure teacher’s affect integration, which is the ability to understand and incorporate emotional experiences into well-adjusted cognition and behavior. The measure captures emotion awareness and regulation of the adult, but rather than dampening or restricting emotions it involves the capacity to allow oneself to experience the emotion and express it appropriately. Higher levels of affect integration can protect against psychopathology (Monsen et al., 1996) and has been used as a measure of psychological function and predictor of responsiveness to interventions. The AII presents 18 statements and asks the participant to rate how well these statements describe them from 0 (does not fit at all) to 9 (fits completely). For example, “When something very sad happens, I can cry and feel very relieved afterwards.” Psychologist Bølstad shortened the AII 3.1 from 112 questions to 18 questions for this study using factor analyses conducted by one of the authors of AII (Solbakken). The AII was shortened to 18 items to prevent questionnaire fatigue.

The AII questions load to multiple constructs, and the shortened version consists of three higher order scales, global affect integration (18 items), capacity for affect experience (11 items), and capacity for affect expression (7 items), as well as six discrete affect scales with three items each; integration of joy, integration of anger, integration of tenderness, integration of sadness, integration of shame and integration of jealousy. The AII is a Norwegian scale and did not require translation. Good internal reliability, structural validity, construct validity and discriminant validity for the scale has been established for the 42-item version of AII in Norway (Solbakken and Monsen, 2021). The AII-18 factors had acceptable internal reliability and factor validity (Koleini et al., 2026).

This study used three scales to measure teacher life satisfaction, burnout and distress.

The Satisfaction with Life Scale (SWLS; Diener et al., 1985) is a 5 item questionnaire that measures life satisfaction. The measure presents five statements and participants rate how much they agree with each statement on a scale of 1 (does not fit) to 7 (fits perfectly) providing a total score. An example item includes, “I am content with my life”. SWLS shows good fit, internal reliability and convergent validity in various samples (Chinni and Hubley, 2014; Moksnes et al., 2014).

The Bergen Burnout Inventory (BII; Matthiesen, 1992; Salmela-Aro et al., 2011) is a 9-item measure of burnout, with three subscales, Exhaustion, Cynicism and Inadequacy, with three items in each. On a scale of 1 (completely disagree) to 6 (completely agree), participants were asked how much they agree with the following nine statements. For example, “I feel overwhelmed at work”. The English version of BBI was obtained from Salmela-Aro et al. (2011), translated to Norwegian and back translated to English by master’s students in psychology. BBI had showed good fit, internal reliability, and predictive validity for Estonian and Finnish managers (Salmela-Aro et al., 2011).

The Hopkins Symptoms Checklist 10 (HSCL-10; Derogatis et al., 1974; Hesbacher et al., 1980; Strand et al., 2003) is a 10-item measure of distress. On a scale of 1 (not at all) to 4 (a lot), participants are asked how much they have experienced the following symptoms in the past week. For example, “Feeling anxious”. HSCL-10 is based on HSCL-25 which was translated from English to Norwegian by Hesbacher et al. (1980) and later shortened to 10 items by Strand et al. (2003). HSCL-10 consists of two subscales, depression, and anxiety, with 6 and 4 items, respectively, that together create distress. HSCL-10 has shown good internal reliability, structural validity and external validity among Scandinavian adolescents (Haavet et al., 2011; Kleppang and Hagquist, 2016).

The Classroom Assessment Scoring System K3 (CLASS; Pianta et al., 2008) was used as an independent observation measure of the classroom emotional climate. CLASS consists of three domains, emotional support, classroom organization and instructional support, with three or four dimensions in each. Emotional support is the domain that most closely captures emotion climate and reflects the emotional connection and social interactions between teachers and students, the teacher’s awareness, and ability to effectively address their students’ emotional and academic needs, and the teacher’s interest in the student’s opinions. Teachers were scored on a scale of 1 (low range, few if any expected behaviors are seen) to 7 (high range, many expected behaviors are seen most of the time) for each dimension depending on the frequency and quantity of behaviors seen.

Three modifications were made to CLASS for the current study. First, teachers were observed in four cycles instead of six. This was due to a limited number of observers Second, teachers did not receive feedback after being observed because CLASS was used as an assessment tool and not as a method of intervention, which it is often used as, as this would make it harder to discern whether changes in behavior were due to TIKiS, or due ot the feedback. Last, if there was more than one teacher in the classroom, observers were instructed to focus on the lead teacher who was receiving the TIKiS intervention. This was to ensure that changes observed in the classroom were due to changes in the teacher who received TIKiS, and not because of other teachers.

Observers were a mix of employees from pedagogical psychological services, master and bachelor students in psychology, pedagogic psychological counseling, and special pedagogics. They all participated in 14 h of certified training by Teachstone and had to pass an exam with 80% reliability. Additionally, Interrater reliability was computed for 10% of all CLASS observations with double coding by a second coder. All coders were blind to the condition of the class (whether they were intervention or control).

Program differentiation indicated how different TIKiS was from other interventions or programs that the teachers had previously or were currently exposed to Durlak and DuPre (2008) and Humphrey et al. (2016). Program differentiation helps determine whether observed effects are masked by involvement in similar previous interventions. Program differentiation was measured using a scale developed for this study, drawing inspiration from questionnaires used for other interventions, such as the ROBUST school program (Rege et al., 2025). At baseline, participants were asked on a scale of 1 to 5, to what extent they had worked systematically with strategies or measures that focused on “verbal and nonverbal communication”, “student behavior”, “relaxation techniques”, “emotions”, “social competence”, “relationships between teachers and students”, “interacting with children when they are emotional”, or “teachers’ own emotional wellbeing” during the previous year. The measure consisted of 8 items, one for each strategy. A summed total score for program differentiation was calculated.

Reach is how many people the intervention reached, and how representative the sample was of the rest of the population. This was measured by the proportion of eligible schools and teachers that participated in the intervention, and how different they were from those who did not participate. as well as the proportion of eligible teachers in a school who participated in the intervention. Except for schools that had participated in the pilot study, all public elementary schools in Oslo were invited to participate in the intervention.

Implementation fidelity measures whether the intervention was delivered as intended. This study examined content fidelity or intervention adherence, which measures how much of the predetermined content was covered. After each TIKiS group session, facilitators completed a checklist indicating whether core components had been covered.

The dosage is a measure of how much of the intervention school leaders and teachers received. Dosage was measured by facilitators taking attendance for each session with school leaders and teacher groups, and by calculating the percentage of sessions teachers attended. Attendance at the introductory seminar was recorded separately in the baseline questionnaire. Teachers and school leaders who were present at the 3-h introduction seminar completed the baseline questionnaire prior to the seminar, providing a tally of attendance.

Adjustment or adaptation refers to changes that were made to the intervention during the implementation period. Deviations in content adherence were collected using a checklist questionnaire filled in by the facilitators after each session, and other differences from the manual were noted down qualitatively as they occurred.

Acceptability, feasibility and appropriateness were measured using self-report validated questionnaires, intervention feedback questionnaires, and focus group interviews.

The self-report questionnaire used the Acceptability of Intervention Measure (AIM), the Intervention Appropriateness Measure (IAM) and the Feasibility of Intervention Measure (FIM), which are three questionnaires used to measure acceptability, feasibility, and appropriateness (Weiner et al., 2017). Each measure presents five statements, and participants indicate how much they agree with each statement on a scale of 1 (completely disagree) to 5 (completely agree). For example, “TIKiS meets my approval”. The AIM, IAM and FIM questionnaires were administered post intervention in April.

After each TIKiS group session both teachers and facilitators answered a brief feedback questionnaire where they are presented with five statements asking to what degree they agreed to each statement on a scale of 1 (not at all) to 5 (to a large extent). The questionnaire evaluated whether the session material was useful/not-useful, easy/difficult, helped teachers respond to students’ emotions, helped teachers be better at handling their own emotional reactions at school, and whether teacher’s believed emotional competence was important at school.

Qualitative interviews were conducted after the intervention with school leaders to provide greater depth in understanding of school staff experience of the intervention.

In addition to the quantitative measures, participants were recruited to take part in focus group interviews to explore their experience of the intervention, leadership support, collaborative relationships and perceived school leader ownership of the implementation of TIKiS. The intent of the focus group interviews was to have a more exploratory, bottom-up way to uncover themes and experiences regarding the implementation process that could supplement the existing quantitative measures. There were two focus group interviews: one for the facilitators, and one for those in the school leadership. For the leadership interviews, recruitment was conducted via email with the school contact and principal. Recruitment for the facilitator interviews was done by contacting the facilitators directly. Interviews were conducted online, lasted up to 90 min and were conducted by a clinical psychologist with whom the participants had no prior contact.