The Circle of Friends intervention: a research synthesis

Figures & data

Figure 1. PRISMA search and selection diagram.

Table 1. General information on intervention.

Nr.	Article	Setting	Country carried out	Duration	Delivered by	‘Focus child(ren)’	Variant of intervention
1	Barrett and Randall (2004)	Mainstream primary school	UK (Scotland)	6 weeks/weekly à 30-min	Headteacher and Educational Psychologist (EP) in Training	1 student (boy) rejected by peers	Adapted version: ‘Focus child’ always present, circle for all circle members. 6 students in the circle – 4 boys, 2 girls.
2		Primary school	N/A	6 weeks/weekly meetings	Teacher, deputy class teacher and community education worker	3 students either rejected or neglected by their peers	Adapted version: 3 circles with 3 ‘focus children’, whole class involved. Circle for all circle members. ‘Focus child’ was not nominated officially and didn’t know that circles had been set up around them.
3	Frederickson and Turner (2003)	Mainstream primary schools	UK	6 weeks/weekly meetings	Intervention Group (IG) (Phase 1): EP student and teacherControl group (CG) (Phase 2):EP and teacher for the initial meeting and a staff member for the weekly meetings	20 students (19 boys, 1 girl, aged 6–12) with SEN (EBD and 30% with significant LD) in different schools and classrooms	Original version:2–5 peers in the circle, nominated by teachers.
4	Frederickson, Warren, and Turner (2005)	Mainstream primary schools	UK	6–10 weeks/weekly meetings	Assistant EP; Most of the staff and teachers participated as observers	Phase 1: 14 students (11 boys, 3 girls, aged 6.8–11.3) with SEN (7 with MLD, 6 with EBD and 1 with ASD) Phase 2: 7 students out of the 14 students from phase 1 (4 with MLD, 2 with EBD and 1 with ASD)	Original version: 6–8 students in the circle.
5	Kalyva and Avramidis (2005)	Preschool	UK	3 month/weekly à 30 min	Teacher	5 children (5 boys, aged 3.10–4.7) with ASD: 3 in the IG, 2 in the waiting CG	Original version: 25 peers/5 in each of the circles (15 girls and 10 boys).
6	Miller et al. (2003)	Mainstream elementary school	USA	Duration n/a/weekly à 30 min	N/A	3 students (boys, aged 11) with mild disabilities (EBD, developmental disability, academic difficulties) who were isolated/ignored	Original version: 3 circles; 4 peers (2 girls, 2 boys, aged 10–11) for each circle.
7	O’Connor (2016)	Mainstream	N/A	3 month/weekly à 30–40 min, 12 sessions	Teacher (usually)	Student (boy, aged 10) with Asperger Syndrome and learning needs	Adapted version: ‘Focus child’ always present
8	Shotton (1998) Educational	N/A	N/A	6 weeks/weekly à 30 min	EP (author of study)	Student (girl, aged 13) displaying anxiety and becoming school refuser	Adapted version: ‘Focus child’ always present, circle for 2 additional students (girls not having friends).4 peers (girls) in the circle
9	Smith and Cooke (2000)	Reception class	N/A	Approx. 7 weeks	EP and teacher	Student (boy, aged 5) at risk of exclusion	Adapted version: Whole class intervention combined with behavioural management framework and involvement of parents at planning meeting.
10	Whitaker et al. (1998)	Primary and secondary schools (Mainstream schools (6) and school for students with moderate LD (1))	UK	3–17 meetings at the time of evaluation/weekly à 20–30 min	Staff of the Autism outreach team and teacher	7 students with ASD	Original version: Peers for the circle nominated by teachers, other students with EBD and/or students, who would also profit included in the circle. 6–8 peers in the circle.

Table 2. Study information.

Nr.	Article	Sample	Methods	Measurement points	Outcomes for ‘focus child(ren)’/circle members/other class members/others (e.g. teacher)
1	Barrett and Randall (2004)	N n/a for whole class N = 6 (‘focus child’: N = 1; circle members: N = 5)	QuantitativeStudents Sociometric questionnaire (peer acceptance: how much children like to play with each member of the class; enjoyment rating of circle members)Staff and parents Evaluation questionnaires (unclear if open questions included)QualitativeStudents Whole class discussion	T1: pre-intervention T2: post-intervention	For ‘focus child’ QuantitativeNo effect on peer acceptance by the whole class.Increase in peer acceptance by the circle members. Enjoyment rating for the CoF: 5/10 QualitativeWould only remain friends with boys in the group. Staff reported limited effect on peer relationships. Parentsreported that children liked the CoF. Rating for the group 10/10 regarding help for their children. For circle members Quantitative Peer acceptance (no values available) increased for 2 circle members, no effect for 2 members and decreased for 1 circle member; enjoyment rating: 10/10 QualitativeWould stay friends with all of the circle members. Class members: Reported changes in the friendliness of the circle members For class members Qualitative Most of the non-circle members reported interest in taking part in a circle in the future.
2		N n/a for whole class ‘focus child’: N = 3	QuantitativeStudentsMy Class Inventory for actual and preferred class cohesion scores.Social skills questionnaireQualitativeStudentsClass discussion Interviews TeachersInterviews	T1: pre-intervention T2: mid-point of intervention T3: post-intervention	For ‘focus child’ Self-report: Increase in friendships for 2/3 children. More playtime with other circle members for 2/3 children. More awareness with whom from their circle playing is possible for all 3 children. Increase of 51% in scores on self-reported social skills questionnaire for one child. For circle members N/A – whole class involved For class members QuantitativeHigh increase in scores on self-report social skills. Enhanced class cohesion scores (MCI) decrease of the gap between mean, actual and preferred scores at T3. Enhanced communication and relationships between adults and students stated. Teachers reported a positive change in the class.
3	Frederickson and Turner (2003)	N/a for whole class and circle members ‘Focus child’: N = 20	IG (Phase 1) and wait-list CG (Phase 2)QuantitativeStudentsSociometric Rating Scale Self-Perception Profile for ChildrenMy Class Inventory (MCI-SF) TeachersTeacher’s Rating Scale of Child’s Actual BehaviourMy Class Inventory (MCI-SF)	IG (Phase 1) T1: pre-intervention T2: post-interventionCG (Phase 2) within-subject designT1–T2–intervention–T3	For ‘focus child’ IG (Phase 1 – between-group analyses)Positive effects on social acceptance by class (η^2 = .35) and circle members (η^2 = .31) from T1 to T2. No significant effects (students’ Self-Perception Profile and teachers’ ratings) for Child’s Actual Behaviour and My Class Inventory. CG (Phase 2 – within group analyses)Measured at three times: positive effects on social acceptance by class members (η^2 = .49) Significant increase of student self-perception on the Global Self-Worth scores (η^2 = .42). Significant increase in social acceptance scores (η^2 = .48) rated by the teacher. No changes in MCI-SF scores. No effect on the perception of social acceptance or behavioural conduct. For circle members N/A For class members CG (Phase 2) Decrease in competitiveness and experienced difficulty.
4	Frederickson, Warren, and Turner (2005)	N n/a for whole class and circle members ‘Focus child’: N = 14/N = 7 for follow-up (T4) drop out due to several reasons	QuantitativePhase 1Sociometric ratings: LITOP Questionnaire from the Social Inclusion Survey (SIS)Guess Who peer assessment–adapted version	Phase1T1: pre–interventionT2: 3-5 days post whole class meetingT3: post-interventionPhase 2T4: follow up – approx. 18 weeks post T3	For ‘focus child’Phase 1Whole class acceptance increased (η^2 = .30) and rejection decreased (η^2 = .40) after the whole class meeting from T1 to T2. The significant decrease (η^2 = .36) in rejection of the ‘focus child’ was only maintained at T3 for circle members Phase 2 Rejecting ratings decreases for circle members at T4 (η^2 = .29) and acceptance ratings return to baseline. No follow-up-effects for whole class, ratings return to baseline. For circle members N/A For class members N/A
5	Kalyva and Avramidis (2005)	N = 30 (‘Focus child’: N = 5, out of them, 2 in the waiting CG; circle members N = 25)	IG = 3/CG = 2 Quantitative Observations: 3 h for each child/1 h at T1, T2 and T3. Nr. of responses ‘focus child’ to peers’ contact initiatives.Nr. of initiation attempts of ‘focus child’.	T1: pre-intervention T2: post-intervention T3: follow-up (two months after T2)	For ‘focus child’ Significant increase in the nr. of successful responses and initiations and a significant decrease in the nr. of unsuccessful responses and initiations of ‘focus child’ in IG compared to ‘focus child’ in the CG – maintained at T3. No significant difference in responses and initiations rates between IG and CG at T1. For circle members N/A For class members N/A
6	Miller et al. (2003)	N = 30 (‘Focus child’: N = 3; circle members: N = 12/4 per circle; CG: N = 15)	Quantitative Observations: approx. 9 min during circle and recess time on % and form of interaction CoF Sociogram	Multiple measurements during: T1: pre-intervention T2: intervention T3: maintenance	For ‘focus child’Increase in appropriate and decrease in inappropriate interactions as well as no-interactions at lunch for all 3 students. Increases in friendly play at recess for 2 students For circle members Circle 1 peers T2 and T3: increase in appropriate interactions and decrease in no-interactions. Inappropriate interactions increased in T2 and decreased in T3. Circle 2 peers T2 and T3: increase in appropriate interactions and decrease in inappropriate and no interactions. Circle 3 peers T2: increase in appropriate interactions and decrease in inappropriate and no interactions. T3: decrease in appropriate interactions and remain of inappropriate and increase in no interactions For class members N/A
7	O’Connor (2016)	N/A for circle members ‘Focus child’: N = 1	Quantitative Belonging scales–adapted version of Psychological Sense of School Membership ScaleSocio-metric measure: Social Inclusion Survey Qualitative Observations–STAR approach	T1: baseline (over three months) T2: post-intervention	For ‘focus child’ Experiences higher global self-esteem scores and feeling socially accepted.Growth of peers’ willingness to make social initiations and relationships with ‘focus child’.Increase of likelihood to accept ‘focus child’s’ behaviour (empathy). Increase in the frequency and quality of social contacts between ‘focus child’s’ peer groups. For circle members N/A For class members N/A
8	Shotton (1998)	N n/a for the whole class N = 5 (‘Focus child’: N = 1 circle members N = 4)	QuantitativeStudentsSociometric questionnaire Qualitative Students Talks, discussions (not clearly described) Teachers, parents Talks, discussions (not clear).	T1: pre-intervention T2: post-intervention	For ‘focus child’ Happier at school (teacher and mother’s feedback, but not on the ‘focus child’s perception). More talkative and more confident (teacher feedback); Gained peers with whom she would talk to. For circle members Felt less isolated and much happier at school. 2 other students having no friends had more friends (self-report). For class members N/A
9	Smith and Cooke (2000)	N n/a for whole class ‘Focus child’: N = 1	Quantitative Observations Qualitative Classroom staff, mother Reports	T1: pre-intervention T2: post-intervention	For ‘focus child’ Improved social skills and less antisocial behaviour at T2. For circle members N/A–whole class involved For class members Motivated to improve their skills and focus on positive aspects.
10	Whitaker et al. (1998)	N n/a for whole class N = 59 (Circle members N = 52 and ‘focus child’ N = 7) N = 7 (circle facilitators)	Qualitative Interviews with: school staff involved in the circle,the focus, his/her parents, and thecircle members. QuantitativeB/G Steem for the circle members (N = 3 classes, out of 6). For other school staff no information on the instrument.	no clear pre/post design	For ‘focus child’TeachersReported enhanced quality of contacts and higher nr. of new friends (circle and non-circle members).Less adult support needed.Decrease of anxiety (4/7 teachers).Improved behaviour due to problem-solving focus. Negative: ‘increased egocentricity’. ParentsReported that children were more extroverted.More playtime with same-aged students.For circle membersTeachersReported development of empathy (all teachers).Greater self-esteem (4/7 teachers).Enhanced skills in joining discussions (3/7).Distress with ‘focus child’s’ behaviour (2/7).Self-reportAbility to offer help (32/52).Growth of self-expression and empathy.Would continue with the circle (48/52).No reciprocal relationships between them and ‘focus child’ B/G Steem: higher self-esteem for 70% compared to less than 50% of their class members (non-circle). For class membersN/A

Barrett, W., and L. Randall. 2004. “Investigating the Circle of Friends Approach: Adaptations and Implications for Practice.” Educational Psychology in Practice 20 (4): 353–368. doi:https://doi.org/10.1080/0266736042000314286.

 Google Scholar

Frederickson, N., and J. Turner. 2003. “Utilizing the Classroom Peer Group to Address Children’s Social Needs: An Evaluation of the Circle of Friends Intervention Approach.” The Journal of Special Education 36 (4): 234–245. doi:https://doi.org/10.1177/002246690303600404.

 Web of Science ®Google Scholar

Frederickson, N., L. Warren, and J. Turner. 2005. ““Circle of Friends”—An Exploration of Impact over Time.” Educational Psychology in Practice 21 (3): 197–217. doi:https://doi.org/10.1080/02667360500205883.

 Google Scholar

Kalyva, E., and E. Avramidis. 2005. “Improving Communication between Children with Autism and Their Peers through the ‘Circle of Friends’: A Small‐scale Intervention Study.” Journal of Applied Research in Intellectual Disabilities 18 (3): 253–261. doi:https://doi.org/10.1111/j.1468-3148.2005.00232.x.

 Web of Science ®Google Scholar

Miller, M. C., N. L. Cooke, D. W. Test, and R. White. 2003. “Effects of Friendship Circles on the Social Interactions of Elementary Age Students with Mild Disabilities.” Journal of Behavioral Education 12 (3): 167–184. doi:https://doi.org/10.1023/1025556226951.

 Google Scholar

O’Connor, E. 2016. “The Use of ‘Circle of Friends’ Strategy to Improve Social Interactions and Social Acceptance: A Case Study of a Child with Asperger’s Syndrome and Other Associated Needs.” Support for Learning 31 (2): 138–147. doi:https://doi.org/10.1111/1467-9604.12122.

 Google Scholar

Shotton, G. 1998. “A Circles of Friends Approach with Socially Neglected Children.” Educational Psychology in Practice 14 (1): 22–25. doi:https://doi.org/10.1080/0266736980140103.

 Google Scholar

Smith, C., and T. Cooke. 2000. “Collaboratively Managing the Behaviour of a Reception Class Pupil.” Educational Psychology in Practice 16 (2): 235–242. doi:https://doi.org/10.1080/713666056.

 Google Scholar

Whitaker, P., P. Barratt, H. Joy, M. Potter, and G. Thomas. 1998. “Children with Autism and Peer Group Support: Using ‘Circles of Friends’.” British Journal of Special Education 25 (2): 60–64. doi:https://doi.org/10.1111/1467-8527.t01-1-00058.

 Google Scholar