Implementation process
Attendance
Attendance lists of 42 TeamUp groups over a period of six to 39 weeks included a total of n = 2183 children (ngirls = 948, nboys = 1109, nmissing = 126). More boys attended TeamUp sessions (50.8%) than girls (43.4%). A session was attended by a mean of 8.5 children (Mgirls = 4.2, Mboys = 4.1), with large differences across centres and groups (ranging from 1 to 31 children per session). Table 2 shows the attendance percentages of eight newly started TeamUp groups, considering only children whose attendance was registered for the whole period of 12 weeks. Children attended on average 31.3% of the sessions during the 12 weeks period, translating to 1.4 times per month. Overall, more boys (52.2%) than girls (40.2%) were reached by TeamUp activities, while girls joined more frequently (38.4%) than boys (28.6%). The majority of the children (75.1%) attended one to four sessions within their first 12 weeks.
Table 2
Attendance Percentages for Newly Started TeamUp Sessions
|
n
|
Percentage of sessions attended (standardised for 12-week period)
|
Overall
|
209
|
31.3%
|
Girls
|
84 (40.2%)
|
38.4%
|
Boys
|
109 (52.2%)
|
28.6%
|
Missing data
|
16 (7.7%)
|
12.0%
|
Age groups a
|
|
|
6-9 year-olds
|
62 (29.7%)
|
27.5%
|
10-14 year-olds
|
78 (37.3%)
|
40.9%
|
15-17 year-olds
|
22 (10.5%)
|
37.6%
|
Missing data
|
47 (22.5%)
|
17.2%
|
|
n
|
Percentage of children attending
|
Number of sessions a child attended in 12-week period
|
209
|
|
1-2 sessions
|
|
50.7%
|
3-4 sessions
|
|
24.4%
|
5-6 sessions
|
|
9.1%
|
7 or more sessions
|
|
15.8%
|
Note. a Individual children’s age is self-reported or based on available registers, thereby 22.5% missing values
Reasons for non-attendance and suggestions for improving attendance
Most interviewed children reported their frequent participation in TeamUp sessions and stated diverse potential reasons for non-attendance, including official appointments, school and family commitment, other concurrently offered activities, being ill or having forgotten about the session’s starting time.
“I didn’t even know that there was TeamUp today. I just happened to see a TeamUp facilitator in the hallway and they told me. We didn’t even know”, another adds that “I only found out because I came down stairs looking for my friend. And someone told me that there was TeamUp” (nine to 11 year-old Farsi/Dutch-speaking girls).
COA personnel mentioned to suspect the lack of structure within the reception centre context, ongoing stressors for families and unawareness of the offered services at the centre, to possibly explain low attendance. Facilitators mentioned the challenges of mobilising children for the session, not having sufficient time prior to the sessions as well as the requirement of positivity, leadership and perseverance.
Implementation quality
Facilitator fidelity: individual-level
Table 3 shows for each of the required session elements the percentage rated as “done” and “not done” following the session observations. Note that due to the different team compositions at each time point the changes between T1 and T2 should not be interpreted as changes over time, but as observations on two occasions. Overall, 49.2% of the observations showed the required session elements as ‘done’ at T1 and 58.2% of the observations showed the required session elements at T2.
The three intervention elements most often observed as ‘done’ at both T1 and T2 were ‘preparing with the team’ (89.0 and 89.9%), ‘following the Child Safeguarding Policy’ (88.6 and 88.0%), and ‘giving each child an equal opportunity’ to participate during the session (80.8 and 79.5 %). The three intervention elements least observed at both T1 and T2 were ‘providing settling moments’ (16.4 and 7.2%), ‘managing children’s high energy’ (24.7 and 33.7%), and ‘setting rules and limits for play’ (38.6 and 36.7%). Some of the session elements had a high number of not applicable (NA) and missing values (see Table 3) and should be interpreted with caution. The percentages were calculated by excluding the missing and NA scores, however, some of the session elements, like discussing alarming behaviour and making referrals, were probably scored as “not done” instead of NA in many observations, influencing the percentages.
Table 3
Percentage Individual-level Fidelity for Each of the Session Elements
|
T1 (n = 73 observations)
|
T2 (n = 83 observations)
|
Session elements
|
Not done
|
Done
|
Not done
|
Done
|
Mobilisation
|
54.8
|
45.2
|
58.6
|
41.4
|
Prepare w/ team
|
11.0
|
89.0
|
10.1
|
89.9
|
Prepare safe physical space
|
34.2
|
65.8
|
20.3
|
79.7
|
Interaction walk-in w/children
|
34.2
|
65.8
|
30.7
|
69.3
|
Greet children
|
60.3
|
39.7
|
41.3
|
58.7
|
Say goodbye to children
|
46.6
|
53.4
|
16.9
|
83.1
|
Give positive feedback
|
64.4
|
35.6
|
49.4
|
50.6
|
Give opportunities to participate
|
19.2
|
80.8
|
20.5
|
79.5
|
Group collaboration
|
57.5
|
42.5
|
50.6
|
49.4
|
Support settling moments
|
83.6
|
16.4
|
92.8
|
7.2
|
Manage high energy
|
75.3
|
24.7
|
66.3
|
33.7
|
Indicate boundaries of play area +
|
74.5
|
25.5
|
57.1
|
42.9
|
Set rules and limits for play
|
61.4
|
38.6
|
63.3
|
36.7
|
Support excluded children +
|
36.5
|
63.5
|
38.5
|
61.5
|
Include children w/ specific needs +
|
70.0
|
30.0
|
0.0
|
100.0
|
Address challenging behaviour +
|
39.0
|
61.0
|
33.3
|
66.7
|
Actively discuss and reflect
|
17.1
|
82.9
|
31.3
|
68.7
|
Discuss alarming behaviour +
|
64.7
|
35.3
|
61.9
|
38.1
|
Make referral to COA +
|
100.0
|
0.0
|
81.5
|
18.5
|
Follow child safeguard policy
|
11.4
|
88.6
|
12.0
|
88.0
|
Mean
|
50.8
|
49.2
|
41.8
|
58.2
|
Note: Percentages are adjusted for missing values and NA, the items with > 30% missing data and NA at T1 and T2 are indicated with +.
All of the observed facilitators (n = 81) also completed self-report fidelity checklists. For the individual-level fidelity at T1, the correlations between the observed and self-rated items ranged between Spearman’s ρ = -.15 and ρ = .54. The highest correlations were for the items “including children with specific needs” (.54) and “discussing alarming behaviour” (.49), which were both scored as “not done” by many facilitators, likely inflating the correlations. The item “mobilisation” also had a relatively high correlation between the observed and self-rated scores, ρ = .42. The remainder of the correlations were around ρ = .2. At T2 the correlations showed a similar pattern but were somewhat smaller, with most of the correlations around ρ = .15. The disagreements in all items were because the self-rating was scored as “yes, we did this” while the observer scored “no, this was not done”.
Facilitator fidelity: team-level
Table 4 shows for each of the required session elements the percentages of the teams scoring ‘not done’, ‘partly done’ and ‘very well done’ during the sessions following observations. The observed teams did not necessarily consist of the same facilitators at each time point. Overall, the percentage of teams that showed ‘partly done’ and ‘very well done’ at T1 and T2 was 72.5% and 73.0%, respectively. The three intervention elements most observed in the teams were correctly using play materials (100.0 and 100%), the provision of middle/main activities (100.0 and 95.5%), and implementing sports-based and active activities (90.0 and 100.0%). The three least observed intervention elements were ‘offering creative movement and dance activities’ (22.7 and 31.8%), ‘offering body-awareness activities’ (36.4 and 27.3%), and ‘working explicitly on a specific psychosocial support theme’ (27.3 and 40.9%).
Table 4
Team-level Fidelity for each of the Session Elements
|
T1 (n = 22 teams)*
|
T2 (n = 22 teams)*
|
Session elements
|
not done
|
partly done
|
very well done
|
not done
|
partly done
|
very well done
|
Opening: check-in
|
18.2
|
54.5
|
27.3
|
36.4
|
36.4
|
27.3
|
Opening: body warm-up
|
22.7
|
40.9
|
36.4
|
22.7
|
18.2
|
59.1
|
Middle/main act.
|
0.0
|
50.0
|
50.0
|
4.5
|
27.3
|
68.2
|
Closing: cooling-down
|
40.9
|
40.9
|
18.2
|
36.4
|
22.7
|
40.9
|
Closing: check-out
|
22.7
|
54.5
|
22.7
|
31.8
|
36.4
|
31.8
|
Sportive and active activity.
|
9.1
|
36.4
|
54.5
|
0.0
|
27.3
|
72.7
|
Creative and dance activity.
|
72.7
|
27.3
|
0.0
|
72.7
|
22.7
|
4.5
|
Body awareness act.
|
63.6
|
27.3
|
9.1
|
72.7
|
18.2
|
9.1
|
Correct use of materials
|
0.0
|
45.5
|
54.5
|
0.0
|
31.8
|
68.2
|
Use routine
|
22.7
|
50.0
|
27.3
|
27.3
|
40.9
|
31.8
|
Demonstration
|
27.3
|
59.1
|
13.6
|
22.7
|
54.5
|
22.7
|
Group organisation
|
9.1
|
36.4
|
54.5
|
4.5
|
27.3
|
68.2
|
Build-up of act.
|
27.3
|
50.0
|
22.7
|
13.6
|
54.5
|
31.8
|
Session flow
|
9.1
|
63.6
|
27.3
|
18.2
|
45.5
|
36.4
|
Adapt to age, needs etc.
|
22.7
|
36.4
|
40.9
|
9.1
|
50.0
|
40.9
|
Work on focus theme
|
72.7
|
27.3
|
0.0
|
59.1
|
31.8
|
9.1
|
Mean
|
27.6
|
43.8
|
28.7
|
27.0
|
34.1
|
38.9
|
Note: * Teams consist of different team members at T1 and T2.
All of the observed facilitators (n = 81) also completed self-report fidelity checklists for their team (2 to 6 facilitators per team). Unfortunately, the inconsistencies between facilitators of the same team were so large that the results of the self-report per team could not be interpreted well and are therefore not reported.
Facilitator Competencies
Table 5 shows the percentages of the observations scored as ‘harmful’, ‘absence of the competency’, ‘competency partly present’ and ‘mastery’ at both time points. Overall, facilitators demonstrated higher ‘adequacy’ (defined as the sum of ‘partly’ and ‘mastery’) than ‘inadequacy’ scores (defined as the sum of ‘harmful’ and ‘absence of competency’). On average, 82.9% of the observations demonstrated adequate competency at T1 and 88.4% at T2. Particular strengths were observed on empathy (98.6 and 91.6% adequacy at T1 and T2, respectively) and team collaboration (98.6 and 97.5% adequacy at T1 and T2, respectively). The main observed weaknesses were the behaviour management of children (36.4% and 10.9% inadequacy at T1 and T2, respectively) and giving feedback (27.8% and 18.1% inadequacy at T1 and T2, respectively).
When comparing facilitators who did a session at both T1 and T2 based on their experience, the competencies of facilitators significantly improved. Those with less than 4 months of experience, t(24) = 5.66, p < .001, (Mdifference = 3.08, 95% CI = 1.96 to 4.2), showed a large effect (Cohen’s d = 1.2) and those with at least 4 months of experience, t(20) = 1.76, p = .043), improved with a small-medium effect (d = 0.48). Given the small sample size, some facilitators having been observed twice and their varying levels of experience as facilitator (1 month to 5 years), these results need to be interpreted carefully.
Table 5
Percentages of Observations Showing Each Level of Competency
|
T1 (n = 73 observations)
|
T2 (n = 83 observations)
|
Items
|
harmful
|
absent
|
partly
|
mastery
|
harmful
|
absent
|
partly
|
mastery
|
Empathy
|
0.0
|
1.4
|
55.5
|
43.1
|
0.0
|
8.4
|
45.8
|
45.8
|
Connection
|
0.0
|
8.3
|
68.1
|
23.6
|
0.0
|
12.0
|
53.0
|
35.0
|
Non-verbal
|
0.0
|
12.5
|
62.5
|
25.0
|
0.0
|
15.7
|
53.0
|
31.3
|
Adaptable+
|
0.0
|
23.6
|
63.9
|
12.5
|
0.0
|
13.4
|
63.4
|
23.2
|
Feedback
|
1.4
|
26.4
|
61.1
|
11.1
|
0.0
|
18.1
|
59.0
|
22.9
|
Inclusive
|
0.0
|
23.6
|
55.6
|
20.8
|
0.0
|
6.0
|
60.3
|
33.7
|
Behaviour management+
|
0.0
|
36.4
|
56.1
|
7.5
|
1.2
|
9.7
|
61.4
|
27.7
|
Group management+
|
0.0
|
18.1
|
63.8
|
18.1
|
0.0
|
17.7
|
44.3
|
38.0
|
Team collaboration
|
0.0
|
1.4
|
40.3
|
58.3
|
0.0
|
2.5
|
37.3
|
60.2
|
Mean
|
0.2
|
16.9
|
58.5
|
24.4
|
0.1
|
11.5
|
53.1
|
35.3
|
Note: Percentages are adjusted for missing values in the items indicated with a +.
Self-efficacy pre-post training
On average, more respondents rated their ability to handle specific aspects of the TeamUp intervention as high or very high after the training (79.5%), compared to before (62.1%). The paired-samples t-test showed that the total score increased significantly from pre- to post-training (Mpre = 44.5, SD = 6.7 and Mpost = 49.4, SD = 6.7, t(39) = 5.42, p < .001). On most elements, participants rated their perceived self-efficacy higher after the training compared to before. Participants continued to perceive a few elements as challenging, such as dealing with children’s strong emotions and challenging behaviours, as well as implementing movement-based and creative activities, and energy-release activities (see Table A in supplementary materials for more details).
Perceptions on implementation and outcomes
Facilitator perceptions
Based on the reporting of the surveys amongst the facilitators (N=99), we learned that they enjoyed their role (84.8%), the interaction with the children (89.9%) and their team (75.8%), see Table 6. In line with the observed fidelity items, survey respondents felt more comfortable providing sports and active games (85.9%), compared to (creative) body-movement activities (59.6%). Many facilitators noted feeling uncomfortable with being the centre of attention, a few were positive about gaining confidence over time. A third to one fourth of respondents stated to experience a little difficulty or stress (27.5-41.3%) in their role, fewer reporting a lot and very much difficulty or stress (3.3-11.6%). Particularly, dealing with the children’s behaviour, hearing children’s experiences, noticing emotions or behaviours were regarded as difficult or stressful.
Table 6
Perceptions of Facilitators in Percentages (n = 99)
Motivational and satisfaction factors
|
Not at all
|
A little
|
Neutral
|
A lot
|
Very much
|
Liking facilitator role
|
0.0
|
2.2
|
5.5
|
37.4
|
54.9
|
Liking interaction with children
|
0.0
|
0.0
|
1.1
|
28.9
|
70.0
|
Liking working in their team
|
1.1
|
2.2
|
14.3
|
30.8
|
51.6
|
Feeling appreciated and recognized by TeamUp
|
1.1
|
6.8
|
21.6
|
31.8
|
38.6
|
TeamUp to benefit them personally
|
2.3
|
3.4
|
12.6
|
48.3
|
33.3
|
TeamUp to benefit them professionally
|
5.7
|
8.0
|
22.7
|
43.2
|
20.5
|
Self-efficacy
|
|
|
|
|
|
Comfortable with sports/games
|
0.0
|
2.2
|
4.4
|
36.3
|
57.1
|
Comfortable with body-movement
|
1.1
|
8.7
|
26.1
|
40.2
|
23.9
|
Comfortable with body-awareness
|
0.0
|
3.3
|
18.5
|
43.5
|
34.8
|
Comfortable with role/responsibilities
|
0.0
|
2.2
|
7.7
|
46.2
|
44.0
|
Burden (difficulty or stress experienced)
|
|
|
|
|
|
With children’s stories/experiences
|
19.8
|
37.2
|
31.4
|
10.5
|
1.2
|
With children’s behaviours
|
19.6
|
41.4
|
28.3
|
10.9
|
0.0
|
With running session
|
38.5
|
38.5
|
19.8
|
3.3
|
0.0
|
With team members
|
49.5
|
27.5
|
17.6
|
1.1
|
0.0
|
With remembering TeamUp session/content
|
34.1
|
40.7
|
17.6
|
7.7
|
0.0
|
Perceived support
|
|
|
|
|
|
Usefulness of team/intervision meetings
|
0.0
|
3.4
|
13.8
|
39.1
|
43.7
|
Support from Volunteer Coordinators
|
0.0
|
3.4
|
12.6
|
29.9
|
54.0
|
Support from Senior Trainers
|
0.0
|
4.6
|
13.8
|
35.6
|
46.0
|
Usefulness of information/ communication from TeamUp team
|
2.3
|
4.6
|
28.7
|
42.5
|
21.8
|
Perceived impact
|
Very negative
|
Negative
|
Neutral
|
Positive
|
Very positive
|
On children’s emotional wellbeing
|
0.0
|
0.0
|
5.6
|
67.8
|
26.7
|
On children’s behaviours
|
0.0
|
2.2
|
8.9
|
75.6
|
13.3
|
On children’s social abilities or relations
|
1.1
|
0.0
|
6.7
|
63.3
|
28.9
|
On children’s emotional regulation
|
0.0
|
0.0
|
24.7
|
70.6
|
4.7
|
Children feeling at ease
|
0.0
|
2.2
|
18.7
|
61.5
|
17.6
|
Note: Percentages are adjusted for missing data and “don’t know/don’t want to say” replies, max percentages of these is 14.1%.
|
While managing children’s behaviours and being confronted by children’s experiences was difficult, this in turn also motivated facilitators to continue with TeamUp. Noticing children’s enjoyment during the sessions appeared to be a primary motivator. A few volunteers stated to perceive a strong benefit of TeamUp for the participating children – TeamUp “transforms difficulties into positivity for the things we do for them.” and another expressed that “through these activities we help [the children] to forget for a little [ongoing worries] while making contact and playing with the child.” Overall, facilitators perceived TeamUp to have a positive to very positive impact on children’s emotional wellbeing (85.9%), behaviours (80.8%) and social abilities or relations with other children (83.8%). One survey participant stated “we [facilitators] are making a difference in their [the kids’] lives, however small it may be”. Moreover, participating facilitators were positive about the TeamUp intervention, its methodology, capacity-building and mentoring support structures.
Evaluation of facilitators
Most children spoke very positively about the facilitators, describing them as ‘nice’, ‘good’ or having ‘passion’. Some mentioned facilitators’ irritability, lack of fairness, inconsistency or leniency with the enforcement of session rules and discipline. COA personnel strongly appreciated the enthusiasm, dedication and self-reliance of facilitators. A few staff desired more depth in the intervention, training, or increased facilitator self-confidence, calm or a more suitable dress code for facilitating movement-based activities. All stakeholders were critical of facilitator turnover, indicating its effect on quality and therefore impacting children’s sense of trust, relationship-building between children and facilitators as well as within facilitator teams and the collaboration with COA.
“they [children] need a safe environment. Once this is not safe, they will show other behaviour, do other things, they will walk away […] If they [children] come back to tell their story, [it is] important that there is trust. […] then part of the team are leaving again. This happens... In general a lot collapses and you have to rebuild this. […] It costs a lot of time and energy.” (COA staff, centre in Gilze)
Respondents from all stakeholder groups, expressed the added value of having at least one male and/or someone with a migrant background within the facilitator team, allowing children to have a diversity of role models. Trust, bonding and showing genuine interest and care was said to be vital to the facilitator role and overall methodology. COA personnel expressed valuing facilitators’ independent and neutral role, offering children attention, interaction and connection with adults who are external to the asylum seeker centre.
Perceived implementation challenges
Managing children’s diverse levels of energy and children displaying strong emotions or challenging behaviour during TeamUp sessions appeared to be one of the primary difficulties for facilitators. The open group nature of the intervention results in varying group sizes and composition of children participating in each session, requiring facilitators’ continuous flexibility and adaptation to children’s individual and collective needs.
“[When] we feel the energy [is getting] really high, [then we are] sitting down, smell the flowers [breathing exercise where children are guided to pretend to smell a flower, thus breathing in deeply]. Sometimes they are just hyper-energetic. And then we are like ‘oh we need something to stop this’. First we sit there and they roll over the groups [of children] (laughing), then after the ritual [e.g. clapping routine] they chill, they are quiet, so you can explain the next game or do something, really” (facilitators, centre in Oisterwijk)
Other challenges included managing children’s expectations, their diverse requests for different activities, finding suitable activities to engage teenagers and to prevent below six year-olds from joining the sessions (given the target age range of the TeamUp intervention). Survey respondents also mentioned language barriers and request for e.g. native Arabic or Farsi-speakers within the team. In line with the observations, facilitators explained that session implementation often depended on team stability, dynamics, communication and collaboration. In turn, high turnover of facilitators usually impeded this team work. Usually, session implementation quality and rapport between facilitator and children improved over time (also see section on competencies). Despite the experienced difficulties, most facilitators – who participated in the survey and/or FGDs – voiced appreciating and growing within their role. They perceived challenges and learnings as rewarding and motivating.
Suggestions for implementation improvements
Children’s feedback differed and provided concrete ideas for improvements. Generally, children desired sessions to be longer and more frequent. Many also requested fewer activities, each lasting for a longer time. In their opinion, changing activities frequently within a session seemed to break the momentum and resulted in frustration or reduced satisfaction. Several respondents suggested the repetition of activities to offer children predictability and structure. A few children felt that explanations and instructions of activities took too long, despite TeamUp’s facilitation methodology of using non-verbal demonstration and flow to present new games and rules. Children alluded to their wish to be fully immersed in the activities, to feel a sense of achievement and recognition for their efforts. All stakeholders indicated a need for TeamUp to be adapted to meaningfully engage adolescents.
Perceived outcomes
All stakeholders perceived the TeamUp intervention positively. Activities allowed children to experience positivity and normalcy, release high energy or strong emotions (e.g. expressing anger), and build peer-relationships in a socially and emotionally safe space with trustworthy adults. Children would strengthen social-emotional abilities through “a playful way of learning”. COA staff particularly appreciated that TeamUp offers an additional referral platform, supporting their (social) work in the centres.
Children’s accounts of TeamUp sessions, alluded to them experiencing a time and space of (emotional) safety and normalisation, an opportunity to play and move, be seen, heard and taken into account, as well as to interact and connect with peers.
“I like these games, it reminds me [of] when I was little.”, “I feel like I am playing happily […] it reminds me of my country.” (7-10 year-old Arabic speaking children). “You can’t meet people if you spend your time at home, so coming here and meeting up with friends gives you a good feeling.” “it’s fun and it’s good for your health” (15-17 year-old Tigrinya-speaking AMVs).
As children often described the activities in great detail, requiring focus and coordination, alludes to them experiencing a sense of being in the present moment. This might also indicate that children were fostering their social, cognitive and physical abilities (e.g. movement and playing resources) during TeamUp sessions. Several interviewed children expressed their need for more fairness and consistent rules.
“I used to fight when I was at the former [asylum seeker centre] but I don’t fight anymore”; “[when someone annoys me], I feel like I don’t want to play with them”; “I don’t fight [when someone else pushes me], I just tell him to leave me alone”. (6-17 year-old Tigrinya-speaking children).
Facilitators described various individual children showing improved self-regulation and behaviour over a period of time and when attending sessions regularly. For example, children showed a reduction of displayed irritability, anger, aggression or frustration when losing a game. Some also increasingly listened to the instructions, were able to choose to take a “time out” or to apologise to others after conflict. Other perceived changes in children were increased participation, interaction, collaboration, and trust, e.g. holding hands or playing with peers of another gender and/or ethic group. Facilitators and COA staff observed children exhibiting more self-confidence, and appearing more comfortable and relaxed within the session environment, usually after weeks of shyness and reluctance to join in. A few children seemed to increasingly “feel freer to be themselves” or getting out of their “comfort zones”. Hence, they explored boundaries with facilitators, or expressed their needs, e.g. suggesting new or adaptation of activities. This was perceived as an increase in sense of assertiveness and agency.
“once it just happened that there was just girls [in the session] and they asked ‘can we dance, but can you close the blinds? So people can’t look!?’ and […] we didn’t even think about that - that [this] would [could] have been a barrier. So it’s just one of those things, as you go along. And there was a point where the girls would say ‘now I want to teach everyone this move’ and they were leading [them]” (facilitators, implementing sessions in various centres)
Despite giving various examples, facilitators were reluctant to make general statements about improvements observed in children’s behaviour. For example, children continued to manifest impulsive, aggressive and bullying behaviours throughout the sessions. Even the interviewed children expressed their frustration and disappointment about the aggressive behavior of peers, whilst also describing their own readiness to respond physically when feeling angry, upset or annoyed. Some conflicts were challenging to mitigate, especially due to communication (e.g. language barriers), the attendance of different children every week and frequently rotating facilitators. In spite of these challenges, all facilitators perceived TeamUp to support and contribute to children’s socio-emotional learning, peer interaction and psychosocial wellbeing. While COA staff were usually not present during TeamUp sessions, they argued that TeamUp positively contributed to children’s socio-emotional development and psychosocial wellbeing.