A randomised controlled trial of the FRIENDS for Life ...€¦ · The FRIENDS for Life programme is a cognitive behavioural based programme designed to reduce childhood ... Further,

THIS PAPER briefly outlines the researchon childhood anxiety, resilience andbelonging in school. Each of these

factors is important in understandingchildren’s mental health and wellbeing. Thepositive capacity and accessibility of schoolsto promote resilience and wellbeing throughearly intervention and prevention pro-grammes, including those based on cogni-tive behavioural approaches is discussed.Specifically, evidence for the cognitivebehavioural FRIENDS for Life programme toimprove emotional health is highlighted.overall, the importance of training teachersas lead facilitators in the universal delivery ofthe programme in the ‘secure base’ ofschool is supported.

Anxiety disorders are the most commonform of psychological distress in childhoodand youth (Cartwright-Hatton et al., 2004),with prevalence reported as high as 21 percent (Kashani & orvaschel, 1990) with moststudies estimating around 10 per cent (Carr,2006). Anxiety can have negative conse-quences in many areas including educa-tional attainment and social functioning(Pine, 1997). Research has linked highanxiety with low cognitive performance asexcessive anxiety impairs concentration onacademic tasks due to biased attention tonegative cues (Wood, 2006). School atten-dance can be affected and there is anincreased risk of premature withdrawal fromschool (Van Ameringen et al., 2003). There

Educational & Child Psychology Vol. 33 No. 2 69© The British Psychological Society, 2016

A randomised controlled trial of theFRIENDS for Life emotional resilienceprogramme delivered by teachers in Irish primary schoolsRichard Ruttledge, Eileen Devitt, Gabrielle Greene, Mary Mullany, Elizabeth Charles, Joanne Frehill & Maura Moriarty

The FRIENDS for Life programme is a cognitive behavioural based programme designed to reduce childhoodanxiety and promote emotional resilience. Teachers are in a unique position to monitor children who are atrisk and to intervene early with preventive social and emotional learning programmes. This study wasdesigned to replicate very positive international evaluations of the FRIENDS for Life programme for anxietyreduction and extend the evidence base by investigating effects on strengths based qualities such as self-concept, coping and school connectedness. Further, for the first time in an Irish context primary school teacherswere the lead facilitators of the programme, with 709 children aged 9 to 13 years in a representative sampleof 27 primary schools from across Ireland. Schools were allocated to an intervention group or a wait-listedcontrol group. Teachers were trained and supported to deliver the programme by educational psychologists.Quantitative and qualitative data including measures of anxiety, self-concept, coping, school connectednessand social validity indicated that the FRIENDS for Life programme was very positively received by children,parents and teachers. The programme was implemented successfully by teachers and resulted in positiveoutcomes for students including improved emotional wellbeing, greater coping skills and an enhanced senseof connectedness with school.Keywords: anxiety; emotional resilience; children; school-based intervention; universal.

is evidence to suggest that childhood anxietyproblems, left untreated, significantlyincrease the risk of mental health difficultiesin adulthood (Bittner et al., 2007).

Resilience has been defined as thecapacity of a person to prevent, minimise orovercome the damaging effects of adversity(Grotberg, 1997). Adversity can include lifeevents such as maternal depression, maritaldiscord, experience of abuse, bereavement,divorce or separation from a significantperson in a child’s life (McCrory &Cameron, 2009). Rutter (2006) describes aresilient person as having high self-conceptand confidence, possessing good socialproblem-solving skills and being instilledwith a sense of self-efficacy. Studies havefound that the following characteristics werepositively associated with resilience: usingeffective coping strategies (Werner & Smith,1992), effectively controlling negative feel-ings (Eisenberg et al., 2004), and using avail-able social supports (Jonzon & Lindblad,2005).

The need to belong is a powerful motiva-tion and has multiple and substantial effectson emotional and cognitive processes (Fred-erickson & Baxter, 2009). Maslow (1962)detailed a hierarchy of human needs inwhich a need to belong was central to theacquisition of knowledge. Having a sense ofbelonging to a school community is likely tohave a positive effect on crucial factorsincluding engagement with learning, mentalhealth and happiness. Grotberg (1997)states that for children in middle childhood(aged 5 to 12) school may play an even moreimportant role than the family unit, since itexposes them to the powerful influence ofteacher support and peer networks. Notbelonging may lead to disaffection, disen-gagement from learning, anxiety anddepression (Frederickson & Baxter, 2009).Stewart et al. (2004) found that schoolswhere students reported more feelings ofconnectedness to adults and peers werestrongly associated with higher pupil-ratingsof resilience.

Interventions based on cognitive behav-ioural approaches are recommended as afirst line intervention for anxiety (NationalInstitute for Health and Clinical Excellence,2008). The core principle of cognitive behav-ioural theory is that people are not disturbedby things, but by the views they take of them(Greig, 2007). Children with anxiety disor-ders have been found to misperceiveambiguous events as threatening. Cognitivebehavioural interventions aim to increase achild’s awareness of unhelpful or irrationalcognitions and understanding of the effectsthese have on behaviour and emotions(Crosbie et al., 2011).

It can be difficult for children withanxiety symptoms to access appropriate andtimely therapeutic intervention (Barrett &Pahl, 2006). Children needing support areoften not reached by front line mentalhealth services (Essau, 2005). In a UK studyFord et al. (2003) found that 53 per cent ofchildren with significant emotional disordershad no contact with any mental healthservice over an 18-month period. Childhoodanxiety is frequently overlooked as thesechildren are likely to be shy, co-operative andcompliant (Essau et al., 2012). Socio-economic disadvantage is a risk factor forchildhood anxiety (Sawyer at al., 2001).Children from disadvantaged communitiesare not only less likely to receive intervention(Misfud & Rapee, 2005), but also more likelyto terminate the intervention prematurely(Kazdin et al., 1997). Therefore, it is increas-ingly important to consider how effectiveinterventions can be made accessible tochildren with anxiety symptoms or thosewho are at risk of developing anxiety disor-ders.

Wood (2006) found that using a cogni-tive behavioural based intervention toreduce anxiety resulted in improved schoolperformance and social functioning. A meta-analysis conducted by Durlak et al. (2011) of213 school-based, universal social andemotional learning programmes found thatcompared to controls, participants demon-strated significantly improved social and

70 Educational & Child Psychology Vol. 33 No. 2

Richard Ruttledge et al.

emotional skills, attitudes, behaviour, andacademic performance that reflected an 11 percentile point gain in achievement. An important finding was that regular schoolstaff were well placed to deliver theseprogrammes.

Schools have been identified as having akey role in the provision of prevention andearly intervention programmes for child-hood anxiety (Neil & Christensen, 2009).This can serve to reduce many of thecommon barriers to intervention in theclinic setting, such as time, location, stigma,transport and cost (Barrett & Pahl, 2006). Inschools, prevention programmes may beuniversal, selected or indicated (Mrazek &Haggerty, 1994). Universal programmes aredelivered to all students and are aimed atenhancing general mental health (Neil &Christensen, 2009). Selective programmesare targeted at students who have been iden-tified as being at risk of developing disor-ders, for example, those having an anxiousparent (Spence & Dadds, 1996). The finalmethod is an indicated or tertiary approachdelivered to students with early or mildsymptoms of a disorder to prevent moresevere problems emerging. Universalprogrammes to reduce anxiety are advanta-geous for schools as they target a largenumber of students regardless of risk status,help to reduce difficulties in screening forinclusion in targeted intervention groupsand have the potential to reduce the inci-dence of anxiety disorders through earlyintervention (Essau et al., 2012).

Gilligan (1998) details some of the‘potential power of school experiences,arguing that school life offers vulnerablepupils a wide range of opportunities to boostresilience, by acting as a complementarysecure base, providing many opportunitiesfor developing self-esteem and self-efficacy,and opportunities for constructive contactwith peers and adults’ (McCrory &Cameron, 2009, p.8).

In support of building capacity withinschools Macklem (2011) has argued thatgroup cognitive behavioural interventions

delivered by school staff work better thanthose facilitated by researchers or cliniciansfrom outside the school. A systematic reviewby Neil and Christensen (2009) found that ahigher percentage of trials involving teacherprogramme leaders were successful in signif-icantly reducing the symptoms of anxietythan trials involving mental health profes-sionals, researchers or graduate students.However, the authors highlight that theeffect sizes in the studies where teachersacted as programme leaders were slightlysmaller than those trials employing otherprogramme leaders. Effectiveness trialsinvolving teacher programme leaders mayhave produced smaller effects as teachers areless experienced than mental health profes-sionals in the delivery of psychological inter-ventions. However, it has been argued thatalthough the effects for social and emotionalprogrammes in schools may be ‘small tomoderate in statistical terms; they representeffects in the real world that are importantand relatively large’ (Weare & Nind, 2011,p.64).

The FRIENDS for Life programmeThe FRIENDS for Life programme is a cogni-tive behavioural based early intervention andprevention programme, delivered tochildren in group or universal formats, anddesigned to help cope with feelings ofanxiety, fear and worry by developing self-esteem, teaching coping skills andpromoting resilience in a simple, structuredway (Barrett, 2012). The programme hasbeen supported by the World Health organ-isation (2004) as an anxiety intervention that‘appears to be efficacious across the entirespectrum, as a universal preventionprogramme, as a targeted preventionprogramme and as a treatment’ (p.43). It isalso listed on the National Registry ofEvidence-Based Programs and Practices(2013) maintained by the United StatesDepartment of Health and Human Services.

A series of studies, including randomisedcontrol trials, have reported positiveoutcomes when the programme is delivered

Educational & Child Psychology Vol. 33 No. 2 71

A randomised controlled trial of the FRIENDS for Life emotional resilience programme…

universally to whole classes of children (Stal-lard, 2010). The first study to evaluate theeffectiveness of FRIENDS for Life involved 489children (aged 10 to 12) and showed a signif-icant reduction in anxiety symptoms (Barrett& Turner, 2001). These findings were repli-cated by Lowry-Webster et al. (2003) in astudy involving 594 young people (aged 10to 13), which found that gains were main-tained a year after completing the FRIENDSfor Life programme. Long-term positiveeffects of the programme were demon-strated by Barrett et al. (2006) who foundthat a sample involving 692 children andyoung people (aged 9 to 16) showed signifi-cant reductions in anxiety three years laterwhen compared with control groups. Thisstudy also found that the impact of theprogramme was stronger at four-monthfollow-up than immediately after completingthe programme. Stallard et al. (2008) evalu-ated FRIENDS for Life with 106 children(aged 9 to 10) and found that significantimprovements in emotional health, namelyanxiety and self-esteem, were maintained atthree-month and 12-month follow-up respec-tively. Essau et al. (2012) evaluated FRIENDSfor Life with 638 children (aged 9 to 12) andfound that participants in the programmeexhibited significantly fewer anxiety anddepressive symptoms and lower perfec-tionism scores than children in the controlgroup at 12-month follow-up.

overall, whilst international evidence onthe universal delivery of the programme hasbeen positive and has the potential to reachmore children, studies have used a variety offacilitators that are not available universally,such as psychologists, school nurses andresearch students. Teachers in schools areavailable universally and have been facilitatorsof the programme previously (Barrett et al.,2006; Lowry-Webster et al., 2003). Therefore,the accessibility of the programme couldpotentially increase further following specifictraining of teachers as lead facilitators.

Research in Ireland and the UK hasfound that school nurses (Stallard et al.,2008) and psychologists (Crosbie et al.,

2011) are effective in delivering the FRIENDSfor Life programme. Mixed results haveemerged in terms of teacher delivery. Stal-lard et al. (2014) found differences between‘health-led’ and ‘teacher-led’ intervention infavour of the former. In contrast, otherstudies report that teachers can deliver theprogramme equally as well as psychologists(Barret & Turner, 2001). This suggestsfurther research on supporting teacherdelivery of the programme is warranted,particularly as teachers are viewed as expertsin providing education to children and inunderstanding their specific school context(Shute, 2012).

Study aimsThis study was undertaken to developfurther the evidence base for using theFRIENDS for Life programme in Irish primaryschools. Teacher-led delivery of theprogramme has not previously been evalu-ated in an Irish context. Specifically it soughtto investigate the following hypotheses:1. In comparison with children in the

control group, children who participatein the FRIENDS for Life programme, deliv-ered by teachers, will report a signifi-cantly lower level of anxiety symptoms atthe end of the intervention.

2. In comparison with children in thecontrol group, children who participatein the FRIENDS for Life programme, deliv-ered by teachers, will report significantlyhigher self-concept at the end of theintervention.

3. In comparison with children in thecontrol group, children who participatein the FRIENDS for Life programme, deliv-ered by teachers, will report a signifi-cantly higher level of coping self-efficacyat the end of the intervention.

4. In comparison with children in thecontrol group, children who participatein the FRIENDS for Life programme, delivered by teachers, will report a signif-icantly higher sense of school connected-ness at the end of the intervention.



5. Teachers trained and supported byeducational psychologists are able toeffectively deliver the FRIENDS for Lifeprogramme.

MethodDesignThis study used a randomised controlleddesign whereby schools were used as the unitof random assignment to either an interven-tion or a to a wait-list control group. Blockrandomisation by school was employed tofacilitate matching on potentially importantcharacteristics within the schools grouped(mixed sex, single sex, urban, rural, desig-nated socioeconomic disadvantage [DEIS],non-DEIS, emotional behavioural difficulties[EBD] special school). Within each group ofschools, names were drawn at random to jointhe intervention group, with the othersbeing assigned to the control group. Therewere two independent variables: betweensubjects (intervention or control group) andwithin subjects (time). Children, parentsand teachers completed measures at threedifferent time points: Time 1 (early January2013), Time 2 (early April 2013) and Time 3(late June 2013). The dependent variablesincluded measures of anxiety, self-concept,behaviour, coping, school connectednessand social validity. This enabled comparisonbetween children in the intervention groupwho received the FRIENDS for Lifeprogramme, delivered by their teachers, andthose in the control group who receivedSocial, Personal and Health Education(SPHE) lessons as part of the normal Irishprimary school curriculum between Time 1and Time 2. It also allowed for a 12-weekfollow-up of programme effects for the inter-vention group at Time 3.

SampleA priori power analysis was carried out usingG*Power (Faul et al., 2009) to find out thenumber of participants that would berequired for the study. G*Power is a stand-alone power analysis programme for statis-tical tests commonly used in the social and

behavioural sciences. The effect size formixed between-within subjects (2x2) analysisof variance (ANoVA) is calculated by thepartial eta squared statistic (small effect: .01;moderate effect: .06; large effect: .14; Cohen,1988). Selecting a partial eta squared effectsize of .14 and an alpha level of .05 inG*Power led to a sample size calculation of70 participants. This sample allowed for apower of .80 to be achieved, which is consid-ered by Ellis (2010) to be acceptable in orderaccurately to reject, or accept, the nullhypothesis. A total of 709 children from 27primary schools across Ireland participated.The large sample size ensured more thanadequate power to detect statistically signifi-cant changes over the duration of the studyfor the overall group.

Schools from across Ireland wererecruited to participate in the study byeducational psychologists from the NationalEducational Psychological Service. To ensurethe sample was representative a variety ofschools were included from both urban andrural areas. Single sex schools, mixed sexschools and schools assigned designatedsocioeconomic disadvantaged (DEIS) statuswere part of the sample. A number of specialschools for children with significantemotional and behavioural difficulties(EBD) were also included.

of the 27 schools, 13 were drawn atrandom and assigned to the interventiongroup leaving 14 schools in the wait-listcontrol group. The assignment of schools tothe intervention or control group wasmanaged using block randomisation byschool type (mixed sex, single sex, urban,rural, DEIS, non-DEIS, EBD special school).This ensured that each group had the samecomposition of school types. Table 1 detailshow schools in the intervention and controlgroups were matched to reduce betweengroup variability.

The age of participants at Time 1(January 2013) ranged from 9 to 13 years(Mean age=10.83 years; SD=.70). out of the709 participants 346 were male and 363 werefemale. The number, gender, mean age and



school setting of participants by group areshown in Table 1.

Intervention GroupThirteen schools were assigned to the inter-vention group, according to their schooltype, leading to the inclusion of 333children. Information packs for parents weredistributed by participating schools. Thechildren and their parents were informedabout the FRIENDS for Life programme, thepurpose of the study and that they would becontacted at three time points over the2012/2013 academic year to complete anumber of questionnaires. Children whodeclined to be part of the research studywere permitted to participate in theprogramme if they so wished or were facili-tated to engage in other activities offered bythe school such as Art or Physical Education.The average age was 10.88 years (SD=.70).

Information sessions for participatingschools were provided by educational psychol-ogists where teachers were appraised of thepurpose and protocols of the study, trainingrequirements and follow-up support. Subse-

quently, 34 teachers from the interventiongroup schools attended the compulsoryaccredited FRIENDS for Life two-day trainingworkshop in November 2012, delivered byeducational psychologists from the NationalEducational Psychological Service. Twoteachers per classroom then delivered the 10-session programme on a weekly basis to awhole class in their school, as part of theSPHE curriculum, between January and April2013. Each child was given a workbook whichwas used throughout the duration of theprogramme. Educational psychologistsaccredited as trainers of the FRIENDS for Lifeprogramme were available for consultation ifthe teachers required support throughout theprogramme. They also checked in directlywith teachers at the beginning, middle andend of the intervention.

Parents of participating children wereinvited to attend two parent psycho-educa-tional workshops jointly facilitated by theteachers delivering the programme and theschool’s allocated educational psychologist.one was held at the beginning and the otherhalf-way through the programme.



Table 1: Participant data by group.

Participant data Intervention group Control group (N=333) (N=376)

Gender and Mean Age: N Mean Age N Mean AgeMale 173 10.98 (SD=.70) 173 10.89 (SD=.79)Female 160 10.76 (SD=.69) 203 10.71 (SD=.61)Total 333 10.88 (SD=.70) 376 10.79 (SD=.70)

School Location:Urban 169 228Rural 164 148

School Type:All Male 97 83All Female 105 120Mixed Sex 107 161EBD Special School 24 12

School Status:Non-DEIS 226 224DEIS 107 152

Control GroupFourteen schools were assigned to thecontrol group, leading to the inclusion of376 children. Similar to the interventiongroup the children and their parents wereinformed about the FRIENDS for Lifeprogramme, the purpose of the study andthat they would be contacted at three timepoints over the 2012/2013 academic year tocomplete a number of questionnaires. Inorder to meet ethical guidelines thesechildren were then invited to participate inthe programme three months later than theintervention group (between April and June2013). This ensured that all children in boththe intervention and control groupsreceived the intervention, albeit at differenttime points. The average age was 10.79 years(SD=.70). Prior to beginning the programme33 teachers from the 14 control groupschools attended the compulsory accreditedFRIENDS for Life two-day training workshopin March 2013. The 10-session programmewas then delivered in these schools to awhole class on a weekly basis. Similar to theintervention group, two parent psycho-

educational workshops were also facilitatedat the beginning and half-way through theprogramme.

FRIENDS for Life programme contentAs shown in Table 2 the FRIENDS for Lifeprogramme consists of 10 structured sessionsusing behavioural, physiological and cogni-tive strategies to teach children how to iden-tify feelings associated with various kinds ofemotional distress; how to relax; how toidentify unhelpful thoughts and to changethese to more helpful thoughts; how to over-come everyday problems and build onsuccess (Stallard et al., 2008). The sessionsinvolve a mixture of group work, role plays,workbook exercises, games, and interactiveactivities. Some tasks are completed at homewith the participant’s family in order to prac-tise new skills learned.

FRIENDS for Life group facilitator training for teachersPrimary school teachers from participatingschools (N=27 schools, minimum of twoteachers per school) were trained to deliver



Table 2: The FRIENDS for Life 10 session emotional resilience programme.

Session Content

Session 1 Introduction to FRIENDS for Life; understanding and accepting difference

Session 2 Introduction to feelings

Session 3 Introduction to body clues and relaxation

Session 4 Self-Talk: Helpful (Green) and Unhelpful (Red) thoughts

Session 5 Challenging Unhelpful (Red) into Helpful (Green) thoughts

Session 6 Introduction to Coping Step Plans

Session 7 Learning for our role models and building support teams

Session 8 Using a Problem Solving Plan

Session 9 Using the FRIENDS for Life skills to help ourselves and others

Session 10 Review, generalising skills and planning for the future

FRIENDS for Life by their school educationalpsychologist and three other educationalpsychologists, all accredited as trainers bythe programme licensor Pathways. Teachersassigned to the intervention group receivedthe training in November 2012 and thosefrom the control group received the trainingin late March 2013. The training was deliv-ered at different time points to ensure nocontamination within the control groupduring the waiting period between Januaryand April 2013. An intensive two-day trainingprogramme was delivered in order to famil-iarise teachers with the programme and toensure fidelity of implementation. The work-shops also covered the underlying theorybehind cognitive behavioural approaches,self-concept, attachment theory, anxiety,depression and ethical issues when runningthe programme. The training involvedreviewing each session of the programmeusing dialogue, role plays and exercises.Each teacher received a leader’s manual(Barrett, 2012) providing a detailed plan ofeach of the 10 sessions.

Fidelity of implementationA checklist was used to ensure teachers deliv-ered the programme with fidelity. After eachsession teachers ticked the checklist toconfirm that each part of the programmehad been completed. This checklist was thenreturned to the researchers at the end of theprogramme. In addition, a timetable wasgiven to each teacher detailing the time scaleand completion dates to be adhered to.

Ethical issuesThis study was guided by the PsychologicalSociety of Ireland’s Code of Professional Ethics(2010). Prior to taking part in the studyinformed written consent was obtained fromthe participants and their parents. Thechildren and their parents were made awareof their right to withdraw from the study atany time. Descriptions of each measure wereprovided for parents and steps were taken toensure that completed questionnairesremained confidential. However, it was

explained to participants that their parentswould be informed if they reported signifi-cant concerns or if their scores on the stan-dardised test measures were in the elevatedrange. After collecting the data a standardprotocol was followed whereby parents ofparticipants who rated themselves in theelevated range were contacted by educa-tional psychologists from the research teamand advised as to appropriate next steps fortheir child. In addition, educational psycho-logists were available for consultation withschool staff to discuss any concerns theymight have had around programme imple-mentation or other matters arising duringthe course of the sessions.

MeasuresSpence Children’s Anxiety Scales (SCAS)Child and parent versions of the SCAS(Spence, 1998) were used. The question-naires assess anxiety in the areas of socialphobia, separation anxiety, panic attack/agrophobia, physical injury fears, obsessivecompulsive disorder and generalised anxietydisorder. Internal consistency and test-retestreliability of the SCAS have been reported assatisfactory, with Cronbach Alpha coeffi-cients well above .70 and a test-retest correla-tion coefficient of .60 (Spence, 1998).DeVellis (2003) recommends that the Cron-bach alpha coefficient should ideally beabove .70. In the present study the Cronbachalpha coefficients for the child and parentversion of the SCAS were both .89.

Beck Self-Concept Inventory for Youth (BSC-Y)The BSC-Y (Beck et al., 2005) is a self-reportscale about thoughts, feelings and behav-iours relating to self-concept. Childrendescribe how frequently each statement istrue for them. Internal consistency and test-retest reliability of the BSC-Y have beenreported as satisfactory, with Cronbach alphacoefficients above .80 and test-retest correla-tion coefficients above .74 (Beck et al.,2005). In the present study the internalconsistency of the BSC-Y was high with aCronbach Alpha coefficient of .89.



Coping Efficacy Scale (CES)The CES (Sandler et al., 2000) is a question-naire developed for children to assess howsatisfied they are with their handling of theirproblems in the past and their level of confi-dence about handling future problems.Internal consistency and test-retest reliabilityof the CES have been reported as satisfac-tory, with Cronbach alpha coefficients gener-ally above .82 and a test-retest correlationcoefficient of .75 (Sandler et al., 2000). In the present study, the Cronbach alphacoefficient for the CES was .83.

School Connectedness Scale (SCS)The SCS (Resnick et al., 1997) is a question-naire which measures self-reported happi-ness, belonging, safety and closeness toothers at school, as well as treatment byteachers. The internal consistency of the SCShas been reported by Sieving et al. (2001) assatisfactory with Cronbach alpha coefficientsranging from .75 to .82. In the present studythe Cronbach alpha coefficient was .80.

FRIENDS Social Validity Measures (SVMs)The SVMs (Barrett et al., 1998), designed tomeasure child, parent and teacher satisfac-tion with the programme, were completed atthe end of the programme. These scalesgathered information on how much partici-pants had learned about feelings and how tocope with them, and how often they usedskills taught in the programme. Respondentsanswered questions by ratings on a four-point Likert scale as follows: 1=Very useful;2=Somewhat useful; 3=A little useful; and 4=Not at all useful. other questions enquiredabout aspects of the programme thatchildren enjoyed, parent views on the useful-ness of the programme and teacher views onthe opportunities and challenges in usingthe programme. This measure has been usedin a number of studies; however no psycho-metric data have been reported. In thepresent study the Cronbach alpha coeffi-cients for the child, parent and teachermeasures were .89, .88 and .68 respectively,indicating acceptable internal consistency.

ResultsThis section presents findings on the effecton participants’ psychological wellbeingafter receiving the FRIENDS for Lifeprogramme; data on social validity; and datarelating to programme implementation byteachers in Irish primary schools. A series of2 (Time: Time 1, Time 2) x 2 (Group: Inter-vention, Control) mixed between withinsubjects ANoVA were conducted using theStatistical Package for Social Sciences (SPSS)– Version 20.0 (IBM Corporation, 2011) inorder to determine if there were significantprogramme effects for the interventiongroup when compared with the controlgroup. An alpha level of α=.05 was used forall tests, with the Bonferroni correctionbeing applied as appropriate. Data arepresented at Time 3 to provide evidence thatgains were maintained by the interventiongroup at 12-week follow-up. Children in theintervention and control groups did notdiffer significantly on any of the measurescompleted at Time 1 (January 2013). A fulldata set of child self-report measures wasavailable for 301 participants in the controlgroup and 338 participants in the interven-tion group, indicating a response rate of 90 per cent and reducing the effect ofmissing data. All teachers returned thefidelity checklist confirming that they haddelivered all 10 sessions of the programme insequence and covered the key components.

AnxietyThe impact of the programme on children’sanxiety was determined by analysing theseven scores on both the self-report (SCAS-Child) and parent (SCAS-Parent) measures.Bonferroni adjusted alpha levels of α=.007per test (.05/7) were used to account formultiple comparisons and to reduce thechance of Type 1 errors.

The mean Total Anxiety score at Time 1(see Table 3 below) on the SCAS-Child forparticipants in the intervention group was24.13 (SD=16.16) and the mean score forparticipants in the control group was 23.09(SD=16.08). There was no significant differ-





Table 3: Means and Standard Deviations of self-report measures for children in theIntervention and Control Groups at Time 1, Time 2 and Time 3.

Variables Time 1 Time 2 Time 3(January 2013) (April 2013) (June 2013)

Mean (SD) Mean (SD) Mean (SD)

SCAS-ChildTotal AnxietyIntervention 24.13 (16.16) 19.48 (15.16) 15.33 (12.63)Control 23.09 (16.08) 19.73 (14.38) 16.66 (12.11)

Separation AnxietyIntervention 3.48 (3.06) 2.64 (2.78) 1.96 (2.22)Control 3.22 (3.01) 2.68 (2.72) 2.14 (2.40)

Social PhobiaIntervention 4.60 (3.69) 3.75 (3.51) 3.15 (3.07)Control 4.24 (3.43) 4.03 (3.54) 3.55 (3.02)

Obsessive CompulsiveIntervention 4.59 (3.77) 3.53 (3.48) 2.43 (2.97)Control 4.36 (3.70) 3.52 (3.43) 2.62 (2.88)

Panic/AgrophobiaIntervention 2.70 (3.62) 2.10 (3.05) 1.54 (2.66)Control 3.02 (3.67) 2.14 (2.84) 1.83 (2.55)

Physical Injury FearsIntervention 3.53 (2.93) 3.02 (2.74) 2.68 (2.49)Control 3.34 (2.80) 3.14 (2.60) 2.90 (2.50)

Generalised AnxietyIntervention 5.29 (3.36) 4.42 (2.99) 3.57 (2.75)Control 4.91 (3.31) 4.22 (3.03) 3.63 (2.63)

BSC-YSelf-ConceptIntervention 41.68 (9.29) 44.16 (9.24) 44.86 (9.29)Control 42.04 (9.23) 41.69 (9.11) 43.11 (9.26)

CESCopingIntervention 21.21 (4.01) 22.18 (3.84) 22.03 (4.08)Control 21.35 (3.99) 21.44 (3.97) 21.93 (3.94)

SCSSchool ConnectednessIntervention 24.36 (3.97) 24.54 (4.14) 24.72 (4.44)Control 24.49 (3.86) 23.91 (4.17) 24.04 (4.28)

ence between the groups’ mean TotalAnxiety scores at Time 1. After receiving theFRIENDS for Life programme there was adecrease in the intervention group’s meanTotal Anxiety score at Time 2 to 19.48(SD=15.16). over the same time period thecontrol group’s mean Total Anxiety scoredecreased to 19.73 (SD=14.38). A 2 (Time:Time 1, Time 2) x 2 (Group: Intervention,Control) mixed between within subjectsANoVA was conducted, and found thatthere was no significant interaction betweentime and group, indicating that the reduc-tion in the intervention group’s mean TotalAnxiety scores between Time 1 and Time 2did not reach significance when comparedto the control group. From Table 3 it can beseen that the intervention group’s meanTotal Anxiety score dropped further to 15.33(SD=12.63) at Time 3 giving some indicationthat the reduction between Time 1 and Time 2 was maintained and improved upon12 weeks after completing the programme,but without availability of a similar compar-ison for controls.

Exploration of the SCAS-Child meansubscale scores found that there was nosignificant differences between the interven-tion and the control groups at Time 1 forSeparation Anxiety, Social Phobia, obsessiveCompulsive, Panic/Agrophobia, PhysicalInjury Fears and Generalised Anxiety.Analysis of the mean subscale scores betweenTime 1 and Time 2 using 2x2 mixed betweenwithin subjects ANoVA indicated that whilstmean scores for Separation Anxiety, obses-sive Compulsive, Panic/Agrophobia, PhysicalInjury Fears and Generalised Anxietyreduced for the intervention group, thereductions were not significant (p>0.007)when compared with the control group.

A significant interaction was foundbetween group (Intervention, Control) andtime (Time 1, Time 2) for Social Phobia,Wilks’ Lambda=.97, F(1,637)=8.929, p=.003,ηρ²=.014, showing that there was a signifi-cant decrease in the intervention group’smean score on this subscale, aftercompleting the FRIENDS for Life programme

when compared with the control group. Thisreduction was maintained by the interven-tion group at Time 3.

Data from the SCAS-Parent for both Time1 and Time 2 were available for 304 partici-pants in the intervention group and 339controls, representing a response rate of 91per cent. There was no significant differencebetween the intervention and control groups’Time 1 SCAS-Parent mean Total Anxiety orsubscale scores. From Table 4 it can be seenthat the mean score on the SCAS-Parent atTime 1 for the intervention group was 15.84(SD=12.89) and the mean score for thecontrol group was 14.72 (SD=9.19). Afterreceiving the FRIENDS for Life programmethere was a decrease in the interventiongroup’s mean Total Anxiety score at Time 2 to13.84 (SD=11.94). over the same period thecontrol group’s mean Total Anxiety scoredecreased to 13.95 (SD=10.64). A 2 (Time:Time 1, Time 2) x 2 (Group: Intervention,Control) mixed between within subjectsANoVA was conducted, and found that therewas no significant interaction between timeand group indicating that the reduction inthe intervention group’s mean Total Anxietyscores between Time 1 and Time 2 did notreach significance when compared with thecontrol group. Data were available at Time 3for 254 participants in the intervention groupand indicated a mean Total Anxiety score of13.00 (SD=12.81), giving some indication thatthe reduction between Time 1 and Time 2was maintained and improved upon 12 weeksafter completing the programme, but withoutavailability of a similar comparison forcontrols.

A significant interaction was foundbetween group (Intervention, Control) and time (Time 1, Time 2) for Separa-tion Anxiety, Wilks’ Lambda=.98,F(1,656)=10.691, p=.001, ηρ²=.016, showingthat there was a significant decrease in theintervention group’s mean score on thissubscale, after completing the FRIENDS forLife programme, when compared with thecontrol group. This reduction was main-tained by the intervention group at Time 3.





Table 4: Means and Standard Deviations of SCAS-Parent for Children in the Intervention and Control Groups at Time 1 and Time 2.

Variable Time 1 Time 2(January 2013) (April 2013)

Mean (SD) Mean (SD)

SCAS-ParentTotal AnxietyIntervention 15.84 (12.89) 13.84 (11.94)Control 14.72 (9.19) 13.95 (10.64)

Separation AnxietyIntervention 3.16 (3.20) 2.51 (2.79)Control 2.56 (2.52) 2.44 (2.61)

Social PhobiaIntervention 4.01(3.50) 3.64 (3.19)Control 3.83 (2.74) 3.86 (2.84)

Obsessive CompulsiveIntervention 1.47 (2.31) 1.19 (2.15)Control 1.48 (2.08) 1.19 (2.14)

Panic/AgrophobiaIntervention 1.19 (2.34) 1.14 (2.34)Control 1.14 (2.34) 0.93 (1.77)

Physical Injury FearsIntervention 2.90 (2.47) 2.59 (2.40)Control 2.59 (2.40) 2.82 (2.42)

Generalised AnxietyIntervention 3.10 (2.66) 2.77 (2.34)Control 2.77 (2.34) 2.71 (2.12)

Self-conceptThe mean BSC-Y score at Time 1 for partici-pants in the intervention group was 41.68(SD=9.29) and the mean score for partici-pants in the control group was 42.04(SD=9.23). There was no significant differ-ence between the intervention and controlgroups’ Time 1 mean BSC-Y scores. A 2 (Time: Time, Time 2) x 2 (Group: Inter-vention, Control) mixed between withinsubjects ANoVA was conducted in order todetermine if there was a significant effect forthe FRIENDS for Life programme on partici-pants’ self-concept scores when comparedwith the control group. A significant interac-tion between group and time was found,Wilks’ Lambda=.96, F(1,638)=24.612, p=.000,ηp²=.037, indicating that there was a signifi-

cant increase in the intervention group’sself-concept when compared with thecontrol group. From Table 3 and Figure 1 itcan be seen that the intervention groupmaintained the gains made between Time 1and Time 2 at 12-week follow-up (Time 3)with a mean score of 44.86 (SD=9.29).

CopingThe impact of the programme on children’scapacity to cope with problems was deter-mined by analysing scores between Time 1and Time 2 (see Table 3). There was nosignificant difference between the interven-tion and control groups’ mean CES scores atTime 1. A 2 (Time: Time 1, Time 2) x 2(Group: Intervention, Control) mixed

between within subjects ANoVA wasconducted in order to determine if the therewas a significant effect for the FRIENDS forLife programme on participants’ mean CESscores when compared with the controlgroup. A significant interaction betweengroup and time was found, Wilks’Lambda=.99, F(1,638)=7.738, p=.006,ηp2=.012, indicating that there was a signifi-cant increase in the intervention group’scoping efficacy when compared with thecontrol group.

School connectednessThe impact of the programme on partici-pants’ happiness, belonging, safety andcloseness to others at school, as well as treat-

ment by teachers was determined byanalysing scores on the SCS. There was nosignificant difference between the interven-tion and control groups’ mean SCS scores atTime 1. A 2 (Time: Time 1, Time 2) x 2(Group: Intervention, Control) mixedbetween within subjects ANoVA wasconducted in order to determine if the therewas a significant effect for the FRIENDS forLife programme on participants’ mean SCSscores when compared with the controlgroup. A significant interaction betweengroup and time was found, Wilks’Lambda=.99, F(1,638)=7.389, p=.007,ηp2=.011, indicating that there was a signifi-cant increase in the intervention group’sSCS mean scores when compared with the



Figure 1: Intervention and Control Groups’ mean self-concept scores at Time 1, Time 2 and Time 3.

40

41

42

43

44

45

46

1 2 3

Mea

n BS

C-Y

Scor

es

Time

Intervention Control

control group. This increase was maintainedby the intervention group at Time 3.

Social validityThe majority of the children in the study(N=640) completed a Social Validity Measure(SVM) and found the FRIENDS for Lifeprogramme ‘very useful’ or ‘somewhatuseful’ (N=490, 68 per cent). Some high-lighted benefits of the programme includingincreased self-esteem: ‘made me feel good’;reduced anxiety: ‘less worries’; increasedlearning: in terms of ‘coping’ strategies; andbetter connectedness to school. A specificaspect which was ranked highly as being‘helpful’ was the ‘relaxation’ element of theprogramme (N=474, 74 per cent); otheruseful aspects included using ‘green thoughts’(N=181, 28 per cent) and the ‘step plan’(N=111, 17 per cent). In terms of theprogramme’s generalisation and usefulnessto other areas many students stated that it‘helped me with a problem’. Specific examples of current and future situations included: ‘it helped me control my anger’; ‘to become a betterfriend’; ‘if it wasn’t for FRIENDS for Life, I wouldn’t have boxed in my boxing match’; and‘it helped me join a new club’. Some ideas forimprovement included having ‘more fun, moregames and more activities’ and having ‘moretime’.

Parental qualitative data from the SVM(N=460, 65 per cent) complemented datagathered from quantitative measures wherechildren were reported as having ‘a lot’ or‘some’ coping skills (53 per cent) followingthe programme to deal with feelings, forexample, ‘my child commented one day thatFRIENDS for Life makes my stomach feel better…he was aware of his tension subsiding.’ In termsof generalisation to other situations, someparents noticed increased empathy forothers, for example: ‘my son is getting better atchoosing his behaviour and trying to put himselfin the position to think how his behaviour affectsothers.’ Moreover, parents commented onhow the programme also helped their family.Instead of increased school connectednessas experienced by the children, parents

reported increased family connectedness.The home activities in the FRIENDS for Lifeprogramme encouraged increased familyinvolvement and positive thinking which wascommented on by parents. For example:‘each day we discuss the day at mealtime as afamily unit. Myself as a parent I also benefit fromchanging negative to positive thoughts.’ Finally,some parents remarked that the two parents’information sessions were difficult to attendas only 30 per cent of parents whocompleted the SVM reported attending thetwo sessions. A suggestion made to involveparents included ‘making the informationsessions available on DVD’.

Teachers from each of the 27 schoolscompleted the SVM and rated theprogramme as ‘very useful’ in buildingresilience (68 per cent) and felt thatchildren learned ‘a lot’ of new skills (52 percent). In terms of benefits, teacher responseswere in agreement with the quantitative find-ings in terms of reduced anxiety, increasedcoping, increased self-esteem and increasedschool connectedness: ‘I really enjoyed theprogramme and think the philosophy and skillswere excellent in helping children deal with diffi-cult and anxious situations’; ‘it provides excellentcoping strategies for dealing with difficult situa-tions’; and ‘the programme was a very positiveexercise, the children enjoyed the lessons and I dofeel their outlook in the classroom has improved’.Girls, children with emotional difficultiesand those that were more anxious were seenas particular beneficiaries: ‘I think that those inmy class who are ‘worriers’ benefited a lot from thecourse. All children enjoyed it and learned some-thing useful from it’; ‘I think it particularly helpedthe girls that were timid, reserved and shy before thebeginning of the programme. I feel they interactmore with the class now in class and group discus-sions’. The programme was felt to have bene-fited the general class ethos. Teachers statedhow the language in the class has changed:‘the children now have a common language toexpress how they are feeling’ and ‘I find that theFRIENDS language is used naturally throughoutthe day’. Some barriers for teachers includedthe time taken to deliver the programme



(N=9, 33 per cent), the workbook (N=6, 22per cent) and the age group targeted by theprogramme (N=3, 11 per cent).

DiscussionThe quantitative and qualitative results indi-cate that the FRIENDS for Life programmewas positively evaluated by children, parentsand teachers. The programme was imple-mented successfully by teachers in Irishprimary schools of all types and resulted inpositive outcomes for students includingimproved emotional wellbeing, greatercoping skills and an enhanced sense ofconnectedness with school. The findings willnow be considered in relation to the initialguiding hypotheses.

AnxietyThe first aim of the study was to investigatethe effect of the FRIENDS for Life programmedelivered universally on children’s feelings ofanxiety. The hypothesis that the interventiongroup would report significantly lower levelsof anxiety than the control group betweenTime 1 and Time 2 was not supported asreductions in Total Anxiety on the SCAS-Child and SCAS-Parent scales did not reachsignificance. other studies have found reduc-tions in anxiety levels immediately after theintervention (Barrett & Turner, 2001; Liddle& Macmillan, 2010); however, some studieshave noted effects were not fully evident untila longer time had elapsed than the currentresearch study (Barrett et al., 2006; Lowry-Webster et al., 2003). The FRIENDS for Lifeprogramme is skills-based and children needtime to practise the skills learned. It has beenreported that younger children (aged 9 to10) show treatment gains immediately afterthe intervention, whereas older children(aged 11 to 12) show anxiety reduction at six-and 12-month follow-up (Essau et al., 2012).Therefore, the current design may have beentoo short for this age cohort to show signifi-cant results. Participants in the interventiongroup reported significant reductions atTime 2, when compared with the controlgroup on the Social Phobia subscale. In addi-

tion, scores were significantly lower at Time 2for the intervention group on the SCAS-Parent Separation Anxiety subscale. Thedecrease in anxiety symptoms was alsosupported by the reduction over time in thenumber of self-reported worries on the SCAS-Child. However, on balance there is insuffi-cient evidence to accept the hypothesis.

Self-conceptThe second aim of the study was to investi-gate the effect of the FRIENDS for Lifeprogramme on children’s self-concept. Thehypothesis that the intervention groupwould report significantly greater improve-ments in self-concept than the control groupbetween Time 1 and Time 2 was supportedby ratings on the BSC-Y. In addition, thesegains were maintained, and continued toimprove, at 12-week follow-up (Time 3).Previous research has also reported improve-ments and maintenance, at long-term follow-up, of self-concept gains followingparticipation in the FRIENDS for Lifeprogramme (Liddle & Macmillan, 2010; Stallard et al., 2008). Data from the SVMindicated that children reported that they‘feel good’, parents reported that theirchildren appeared ‘more confident’, andteachers noted that the programme ‘openedup self-esteem within the children’.

CopingThe third aim of the study was to examinethe effects of a universal delivery of theFRIENDS for Life programme on children’scoping self-efficacy beliefs. Coping self-effi-cacy beliefs refer to individuals’ beliefs abouttheir ability to cope with external stressors(Pisanti, 2012). People with higher levels ofcoping self-efficacy beliefs tend to approachchallenging situations in an active andpersistent way, whereas those with lowerlevels tend to direct greater energy tomanaging increasing emotional distress(Bandura, 1997). The hypothesis that parti-cipants in the intervention group wouldreport significantly higher levels of copingself-efficacy than those in the control group



between Time 1 and Time 2 was supported byratings on the CES. This finding wasevidenced further by qualitative data gath-ered from the SVM, where participantsreported that they learned skills aboutcoping when feeling worried. Parent reportsindicated that they found the programmeenhanced their child’s coping skills andteachers commented on the ‘excellent copingstrategies for dealing with difficult situations’. Theprogramme devotes considerable time tousing and developing ‘Coping Step Plans’which mirror the cognitive behaviouralstrategy of gradual desensitisation to anxioussituations. Furthermore, participantsreported on current situations where theywere using the skills learned in theprogramme to cope better with anger, friend-ships and sports. In anticipation of future lifestressors many children described how theycould apply the skills learned during theFRIENDS for Life programme to events such astransferring to secondary school.

School connectednessThe fourth aim of the study was to investi-gate the impact of the FRIENDS for Lifeprogramme on participants’ sense of schoolconnectedness. The hypothesis that theintervention group’s sense of happiness,belonging, safety and closeness to others inschool would improve relative to the controlgroup between Time 1 and Time 2 wassupported. These gains were maintained at12-week follow-up (Time 3). The benefits ofthe programme in terms of school connect-edness were evidenced further by data fromthe SVM. Children reported ‘liking schoolbetter’ and their teachers noted a more ‘positive outlook’ in school and that the‘FRIENDS for Life language is used naturallythroughout the day’. This finding emphasisesthe importance of social and emotionalliteracy becoming part of routine educa-tional practice as advocated by numerousempirical studies (Durlak et al., 2011).Furthermore, parents described how theskills learned during the programme led toan improved sense of family connectedness.

Teacher implementation of the FRIENDS for Life programmeThe fifth and final aim of the study was toexamine if teachers, trained and supportedby educational psychologists were able todeliver the FRIENDS for Life programmeeffectively. The hypothesis was supported byquantitative data showing positive outcomesfor children, as assessed by participantsthemselves, their parents and teachers. Thisfinding was further triangulated by qualita-tive data which were overwhelmingly positivein attesting to teachers’ capacity to deliverthe programme. Whilst there was no inde-pendent evaluation of the programmeimplementation, through objective observa-tion, fidelity checklists completed by the twoteachers in each classroom and consultationwith their school’s educational psychologistindicated high adherence to theprogramme. The value of teacher-led socialand emotional interventions has beendemonstrated to be equally effective to thatof outside professionals (Barrett & Turner,2001) or indeed more effective (Durlak etal., 2011). It should be noted that in thecurrent study the teachers were deliveringthe FRIENDS for Life programme for the firsttime. Therefore, it could be posited that theimpact of the programme on participantswould be greater as teachers become morepractised at delivering it.

Study strengthsThe major strength of the study was its largesample size and the large number of schoolsrepresenting a variety of settings and socio-economic backgrounds. other notablestrengths included the random assignmentof schools to intervention or control condi-tions, the presence of a control group, use ofa standardised evidence-based interventionprotocol to ensure fidelity of implementa-tion, and the use of reliable and valid assess-ment measures. In addition, there was a highcompletion rate from participants, parentsand teachers in returning standardised testmeasures.



Study limitationsIn considering the findings of the presentstudy it is important to acknowledge its limi-tations. Sandler (1999) suggests that theeffects of prevention programmes should bejudged by how well they change targetedoutcomes over time, rather than in terms ofimmediate effects. The skills-based andcognitive behavioural nature of the FRIENDSfor Life programme indicates that longer-term outcomes are particularly importantfor assessing the true effect of the interven-tion. In the present study there was less timefor changes to take effect and impact ondaily life; the reported results may, therefore,underestimate the true clinical impact of theprogramme (In-Albon & Schneider, 2007).International research has found largereffect sizes for the programme whenchildren are followed up at six and 12 months (Essau et al., 2012).

Whilst the assignment of schools to theintervention or control group was managedto ensure that each group had a similarcomposition of school type and setting thedata gathered are nested within differentschools and the programme was delivered bydifferent teachers. Therefore, it could bepostulated that the positive effects reportedin this study were not solely due to thechildren’s participation in the FRIENDS forLife programme. For example, some schoolsmay have been more motivated towardsprevention than others and individualteachers may have been more effective intheir programme delivery. Future researchcould consider using advanced modellingtechniques such as hierarchical linearmodelling to account for factors such asschool effects specifically. Using measuressuch as the My Class Inventory (Fisher &Fraser, 1981) would also explore school andclassroom ethos.

As highlighted in the SVM this was thefirst time that the teachers had delivered theFRIENDS for Life programme. Anecdotalevidence suggests that teachers stuck rigidlyto the children’s workbook, which was notalways culturally or age appropriate.

Teachers were confined to a tight timelinewithin which to deliver the programme. Thismay have led them to work too prescriptivelythrough the manual and acted as a barrier toworking more creatively. It is likely thatteachers would be more effective afterrepeated delivery, as they would becomemore familiar with programme content andstructure. Stallard (2010) highlights thateffective delivery of the FRIENDS for Lifeprogramme can be influenced by a range ofvariables including leader commitment andconfidence in discussing a range of socialand emotional health issues as part of theschool curriculum.

The use of behavioural self-report meas-ures for children raises the issue of internalvalidity as participants are often prone togiving socially desirable responses (Sellitz etal., 1961). It would have been beneficial tohave used a teacher measure of child self-concept and anxiety to detect change in thisarea. A measure of childhood depressivesymptoms would also have been useful giventhe co-morbidity of anxiety and depressivesymptoms in children and young people.Furthermore, the timeline of the study didnot facilitate further qualitative explorationsuch as focus groups or follow-up interviewswith participants, parents and teachers. Like-wise, additional investigation of the impactof the programme with focus groups of‘targeted’ populations, such as those withhigh anxiety, depression, lower self-esteemor learning difficulties could form the basisof future research.

International research has found thatprogrammes which actively encourage theinvolvement of parents, local communitygroups and agencies are more likely to havea positive impact on pupil behaviour,resilience, mental health and learning(Brooks, 2006). Although two parent psycho-educational sessions were delivered in eachschool as part of the programme, no atten-dance records were taken by the researchteam. This would have enabled theresearchers to ascertain if children whoseparents attended the sessions gained most



because of their parents’ increased involve-ment and heightened awareness of theFRIENDS for Life programme.

Future directionsContemporary research has provided clearevidence about the positive correlationbetween social and emotional learning andacademic success and the links betweenemotional distress and poor academic attain-ments (Payton et al., 2008). An area of futureinvestigation would be to ascertain ifchildren’s attainment scores rose aftercompleting the FRIENDS for Life programme.

The transition from primary to secondaryschool has been found to be particularlystressful as children adjust to their neweducational and social environment(Greene & ollendick, 1993). It is a time ofheightened anxiety for both children(Bloyce & Frederickson, 2012) and theirparents (o’Brien, 2003). Further longitu-dinal research could assess programmeimpact over a longer period of time (Maggin& Johnson, 2014) and allow examination ofassociated helpful effects in relation to tran-sition from primary to secondary school.

Jennings and Greenberg (2009) high-light the importance of teachers’ social andemotional competence and wellbeing in thedevelopment of supportive teacher-pupilrelationships, effective classroom manage-ment, and successful social and emotionallearning programme implementation. AnAdult Resilience programme has been devel-oped by Professor Paula Barrett to comple-ment the suite of FRIENDS programmes.Future research could consider the effect ofcompleting the Adult Resilience programmeon teacher delivery of the FRIENDS for Lifeprogramme with children.

The content of the programme workbookwas raised as a barrier by a number ofteachers, children and parents. Like all socialand emotional learning programmesFRIENDS for Life needs to be adapted to the

needs of the group. Given that theprogramme was developed in Australia, manyof the activities were culturally unsuitableand need to be modified to suit an Irishcontext. Another challenge emerging fromteacher SVM responses was the time taken todeliver the programme. Teachers made anumber of suggestions as to how to overcomethis difficulty such as teaching each sessionover two lessons or spreading the programmeover the course of the academic year ratherthan the suggested 10-week format. Educa-tional psychologists have the requisitepsychological consultation skills and theoret-ical knowledge to support teachers in theadaptation and evaluation of the programmewith fidelity in school contexts for universaland targeted populations.

ConclusionIn summary, it has been shown that teachersare in an optimal position to promoteresilience in children through delivery of theFRIENDS for Life programme. Schoolsprovide opportunities to target largenumbers of children and make evidenced-based intervention and preventionprogrammes more accessible at a universallevel. The challenge is to continue deliveringevidenced-based programmes with fidelitythrough ongoing support and coaching.Educational psychologists have the specialistexpertise, knowledge and skills to train andsupport school staff in delivering, evaluatingand adapting such programmes. Thissupport will enable the expansion of suchprogrammes in Ireland nationally to reachmore children, potentially leading toimproved emotional wellbeing, greatercoping skills and an enhanced sense ofconnectedness with school. The currentstudy has added to the body of internationalevidence for the efficacy of the FRIENDS forLife programme and shown that educationalpsychologists are ideally placed to promotemental health and wellbeing in schools.



AcknowledgementsThe authors would like to express their grat-itude to the teachers, parents and childrenwho participated in the study; and AifricMcArdle who volunteered as a research assis-tant. A word of thanks also to ProfessorNorah Frederickson (Emeritus Professor ofEducational Psychology, University CollegeLondon) and Dr Andy Fugard (UniversityCollege London) for their advice and guidance.

Address for correspondenceDr Richard RuttledgeEducational & Child Psychologist,National Educational Psychological Service,Department of Education & Skills,Elm House,Cavan, Co. Cavan,Ireland.Email: [email protected]



Bandura, A. (1997). Self-efficacy: The exercise of control.New York: W.H. Freeman.

Barrett, P.M. (2012). FRIENDS for Life: Group leaders’manual for children (6th ed.). Brisbane: BarrettResearch Resources Pty Ltd.

Barrett, P.M. & Pahl, K.M. (2006). School-based inter-vention: Examining a universal approach toanxiety management. Australian Journal of Guidance and Counselling, 16(1), 55–75.

Barrett, P.M. & Turner, C. (2001). Prevention ofanxiety symptoms in primary school children:Preliminary results from a universal school-basedtrial. British Journal of Clinical Psychology, 40(4),399–410.

Barrett, P.M., Farrell, L.J., ollendick, T.H. & Dadds,M. (2006). Long-term outcomes of an Australianuniversal prevention trial of anxiety and depres-sion symptoms in children and youth: An evalua-tion of the FRIENDS program. Journal of ClinicalChild and Adolescent Psychology, 35(3), 403–411.

Barrett, P.M., Lowry-Webster, H.M., Turner, C. &Johnson, S. (1998). The FRIENDS program integritychecklists and social validity measures. Unpublishedmanuscript. School of Applied Psychology, Griffith University, Queensland, Australia.

Beck, J., Beck, A., Jolly, J. & Steer, R. (2005). Beck Youth Inventories (2nd ed.). Texas: HarcourtAssessment.

Bittner, A., Egger, H.L., Erkanli, A., Costello, E.J.,Foley, D.L. & Angold, A. (2007). What do child-hood anxiety disorders predict? Journal of ChildPsychology and Psychiatry, 48(12), 1174–1183.

Bloyce, J. & Frederickson, N. (2012). Intervening toimprove the transfer to secondary school. Educational Psychology in Practice, 28(1), 1–18.

Brooks, J.E. (2006). Strengthening resilience inchildren and youth: Maximising opportunitiesthrough the schools. Children and Schools, 28(2),69–76.

Carr, A. (2006). The handbook of child and adolescentclinical psychology: A contextual approach (2nd ed.).London: Routledge.

Cartwright-Hatton, S. Roberts, C., Chitsabesan, P.,Fothergill, C. & Harrington, R. (2004). System-atic review of the efficacy of cognitive behaviourtherapies for childhood and adolescent anxietydisorders. British Journal of Clinical Psychology,43(4), 421–436.

Cohen, J.W. (1988). Statistical power analysis for thebehavioural sciences (2nd ed.). Hillsdale, NJ:Lawrence Erlbaum Associates.

Crosbie, G., Charles, E., Peel, R., Moran, J., Cullen,A., Carroll, D. & Buckley, C. (2011). An evalua-tion of the FRIENDS for Life programme in Irishschools. The Irish Psychologist, 38(1), 14–20.

DeVellis, R.F. (2003). Scale development: Theory andapplications (2nd ed.). Thousand oaks, CA: Sage.

Durlak, J.A., Weissberg, R.P., Dymnicki, A.B., Taylor,R.D. & Schellinger, K.B. (2011). The impact ofenhancong students’ social and emotionallearning: A meta-analyis of school-baseduniversal interventions. Child Development, 82(1),405–432.

Eisenberg, N., Spinrad, T.L., Fabes, R.A., Reiser, M.,Cumberland, A., Shepard, S.A., Valiente, C..Losoya, S.H.. Guthrie, I.K. & Thompson, M.(2004). The relations of effortful control andimpulsivity to children’s resiliency and adjust-ment. Child Development, 75(1), 25–46.

Ellis, Paul D. (2010). The essential guide to effect sizes: An introduction to statistical power, meta-analysis andthe interpretation of research results. Cambridge:Cambridge University Press.

Essau, C.A. (2005). Use of mental health servicesamong adolescents with anxiety and depressivedisorders. Depression and Anxiety, 22(3), 130–137.

Essau, C.A., Conradt, J., Sasagawa, S. & ollendick,T.H. (2012). Prevention of anxiety symptoms inchildren: Results from a universal school-basedtrial. Behavior Therapy, 43(2), 450–464.

Faul, F., Erdfelder, E., Buchner, A. & Lang, A. (2009).Statistical power analyses using G*Power 3.1:Tests for correlation and regression analyses.Behavior Research Methods, 41(4), 1149–1160.

References

Fisher, D.L. & Fraser, B. J. (1981). Validity and use ofthe My Class Inventory. Science Education, 65(2),145–156.

Ford, T., Goodman, R. & Meltzer, M. (2003). Serviceuse over 18 months among a nationally represen-tative sample of British children with psychiatricdisorder. Clinical Child Psychology and Psychiatry,8(1), 37–51.

Frederickson, N. & Baxter, J. (2009). Belonging. In N. Frederickson & S. Dunsmuir (Eds.), Measures of children’s mental health and psychologicalwellbeing: A portfolio for education and health profes-sionals. London: GL Assessment Limited.

Gilligan, R. (1998). The importance of schools andteachers in child welfare. Child and Family SocialWork, 3(1), 13–26.

Greene, R.W. & ollendick, T.H. (1993). Evaluation ofa multimodal program for sixth graders in transi-tion from elementary to middle school. Journal ofCommunity Psychology, 21(2), 162–176.

Greig, A. (2007). A framework for the delivery ofcognitive behaviour therapy in the educationalpsychology context. Educational & ChildPsychology, 24(1), 19–35.

Grotberg, E.H. (1997). The International ResilienceProject: Findings from the research and theeffectiveness of interventions. In B. Bain et al.(Eds.), Psychology and education in the 21st Century:Proceedings of the 54th Annual Convention of theInternational Council of Psychologists. Edmonton:IC Press.

IBM Corporation. (2011). Statistical Package for theSocial Sciences – Version 20.0. Armonk, NY: IBMCorp.

In-Albon, T. & Schneider, S. (2007). Psychotherapy ofchildhood anxiety disorders: A meta-analysis.Psychotherapy and Psychosomatics, 76(1), 15–24.

Jennings, P.A. & Greenberg, M.T. (2009). The pro-social classroom: Teacher social and emotionalcompetence in relation to student classroomoutcomes. Review of Educational Research, 79(1),491–525.

Jonzon, E., & Lindblad, F. (2005). Adult femalevictims of child sexual abuse. Journal of Inter-personal Violence, 20(6), 651–666.

Kashani, J. & orvaschel, H. (1990). A communitystudy of anxiety in children and adolescents.American Journal of Psychiatry, 147(3), 313–318.

Kazdin, A., Holland, L. & Crowley, M. (1997). Familyexperience of barriers to treatment and prema-ture termination from child therapy. Journal ofConsulting and Clinical Psychology, 65(3), 453–463.

Liddle, I. & Macmillan, S. (2010). Evaluating theFRIENDS programme in a Scottish setting. Educational Psychology in Practice, 26(1), 53–67.

Lowry-Webster, H., Barrett, P. & Lock, S. (2003). A universal prevention trial of anxiety symptoma-tology during childhood: Results at one-yearfollow-up. Behaviour Change, 20(1), 25–43.

Macklem, G.L. (2011). Evidence-based school mentalhealth services: Affect education, emotional regulationtraining and cognitive behavioral therapy. New York:Springer.

Maggin, D.M. & Johnson, A.H. (2014). A meta-analytic evaluation of the FRIENDS program forpreventing anxiety in student populations. Educa-tion and Treatment of Children, 37(2), 277–306.

Maslow, A.H. (1962). Towards a psychology of being.Princeton: D. Van Nostrand Company.

McCrory, E. & Cameron, S. (2009). Resilience. In N. Frederickson & S. Dunsmuir (Eds.), Measures of children’s mental health and psychologicalwellbeing: A portfolio for education and health profes-sionals. London: GL Assessment Limited.

Misfud, C. & Rapee, R. (2005). Early intervention forchildhood anxiety in a school setting: outcomesfor an economically disadvantaged population.Journal of the American Academy of Child and Adoles-cent Psychiatry, 44(10), 995–1004.

Mrazek, P. & Haggerty, R. (1994). Reducing risks formental disorders: Frontiers for preventive interventionresearch. Washington, DC: National Academy Press.

National Institute for Health & Clinical Excellence.(NICE) (2008). Public Health Guidance 12:Promoting children’s social and emotional wellbeing inprimary education. Accessed 27 July 2012, from:http://www.nice.org.uk/nicemedia/pdf/PH012Guidance.pdf

National Registry of Evidence-Based Programs andPractices (2013). Intervention summary: Friendsprogram. Accessed 27 June 2014, from:http://nrepp.samhsa.gov/ViewIntervention.aspx?id=334

Neil, A.L. & Christensen, H. (2009). Efficacy andeffectiveness of school-based prevention andearly intervention programs for anxiety. ClinicalPsychology Review, 29(3), 208–215.

o’Brien, M. (2003). Making the move: Students’,teachers’ and parents’ perspectives of transfer from firstto second-level schooling. Dublin: Marino Instituteof Education.

Payton, J., Weissberg, R.P., Durlak, J.A., Dymniccki,A.B., Taylor, R.D., Schellinger, K.B. & Pachan, M.(2008). The positive impact of social and emotionallearning for kindergarten to eighth-grade students:Findings from three scientific reviews. Chicago:Collaborative for Academic, Social andEmotional Learning.

Pine, D. (1997). Childhood anxiety disorders. CurrentOpinion in Pediatrics, 9, 329–339.

Pisanti, R. (2012). Coping self-efficacy and psycho-logical distress: Results from an Italian study onnurses. The European Health Psychologist, 14(1),11–14.

Psychological Society of Ireland. (2010). Code of profes-sional ethics. Accessed 26 July 2012, from:http://www.psihq.ie/2010%20Code%20of%20Ethics.doc



Resnick, M.D., Bearman, P.S., Blum, R.W., Bauman,K.E., Harris, K.M., Jones, J., Tabor, J., Beuhring,T., Sieving, R.E., Shew, M., Ireland, M.,Bearinger, L.H. & Udry, J.R. (1997). Protectingadolescents from harm: Findings from thenational longitudinal study on adolescent health.Journal of the American Medical Association,278(10), 823–832.

Rutter, M. (2006). Implications of resilience conceptsfor scientific understanding. Annals of the New York Academy of Sciences, 1094, 1–12.

Sandler, I.N. (1999). Progress in developing strate-gies and theory for the prevention and treatmentof anxiety and depression. Prevention and Treat-ment, 2, Article 9.

Sandler, I.N., Tein, J.Y., Mehta, P., Wolchik, S.A. &Ayers, T. (2000). Coping efficacy and psycho-logical problems of children of divorce. ChildDevelopment, 71(4), 1097–1118.

Sawyer, M.G., Arney, F.M., Baghurst, P.A., Clark, J.J.,Graetz, B.W., Kosky, R.J., Nurcombe, B., Patton,G.C., Prior, M.R., Raphael, B., Rey, J.M., Whaites,L.C. & Zubrick, S.R. (2001). The mental healthof young people in Australia: Key findings fromthe child and adolescent component of thenational survey of mental health and wellbeing.Australian and New Zealand Journal of Psychiatry,35(6), 806–814.

Selltiz, C., Jahoda, M., Deutsch, M. & Cook, S. (1961)Research methods in social relations. New York: Holt,Rinehart & Winston, Inc.

Shute, R.H. (2012). Promoting mental healththrough schools. The Psychologist, 25(10),752–755.

Sieving, R.E., Beuhring, T., Resnick, M., Bearinger,L.H., Shew, M., Ireland, M. & Blum, R.W. (2001).Development of adolescent self-report measuresfrom the National Longitudinal Study of Adoles-cent Health. Journal of Adolescent Health, 28(1),73–81.

Spence, S.H. (1998). A measure of anxiety symptomsamong children. Behaviour Research and Therapy,36(5), 545–566.

Spence, S.H. & Dadds, M.R. (1996). Preventing child-hood anxiety disorders. Behaviour Change, 13(4),241–249.

Stallard, P. (2010). Mental health prevention in UKclassrooms: The FRIENDS anxiety preventionprogramme. Emotional and Behavioural Difficulties,15(1), 23–35.

Stallard, P., Simpson, N., Anderson, S. & Goddard, M.(2008). The FRIENDS emotional health preven-tion programme: 12-month follow-up of auniversal school-based trial. European Child andAdolescent Psychiatry, 17(5), 283–289.

Stallard, P., Skryabina, E., Taylor, G., Phillips, R.,Daniels, H., Anderson, R. & Simpson, N. (2014).Classroom-based cognitive behaviour therapy(FRIENDS): A cluster randomised controlledtrial to prevent anxiety in children througheducation in schools (PACES). The Lancet Psychi-atry, 1(3), 185–192.

Stewart, D., Sun, J., Patterson, C., Lemerle, K. &Hardie, M. (2004). Promoting and buildingresilience in primary school communities:Evidence from a comprehensive ‘healthpromoting school’ approach. International Journalof Mental Health Promotion, 6(3), 26–33.

Van Ameringen, M., Mancini, C. & Farvolden, P.(2003). The impact of anxiety disorders oneducational achievement. Journal of Anxiety Disor-ders, 17(5), 561–571.

Weare, K. & Nind, M. (2011). Mental health promo-tion and problem prevention in schools: Whatdoes the evidence say? Health Promotion Interna-tional, 26(1), 29–69.

Werner, E. & Smith, R. (1992). Overcoming the odds:High-risk children from birth to adulthood. New York:Cornell University Press.

Wood, J. (2006). Effects of anxiety reduction onchildren’s school performance and social adjust-ment. Developmental Psychology, 42(2), 345–349.

World Health organisation (WHo) (2004). Preven-tion of mental disorders: Effective interventions andpolicy options. Geneva: WHo.



Copyright of Educational & Child Psychology is the property of British Psychological Societyand its content may not be copied or emailed to multiple sites or posted to a listserv withoutthe copyright holder's express written permission. However, users may print, download, oremail articles for individual use.

Documents

A randomised controlled trial of the FRIENDS for Life ...€¦ · The FRIENDS for Life programme is a cognitive behavioural based programme designed to reduce childhood ... Further,