An observational study of recess quality and physical activity in urban primary schools

Abstract

Background
To date, there is scant literature that examines the recess context concurrent with, but separate from, levels of physical activity. The primary purpose of the current study was to examine how recess quality impacted physical activity levels, and how this was moderated by gender. A secondary purpose was to examine if differences in children’s engagement in activities occurred between recess sessions scored as low- or high- quality.
Methods
This was an observational study of children at 13 urban elementary schools in the U.S. Across the 13 schools, data were collected at 55 recess sessions, with 3,419 child-level observations ( n= 1,696 boys; n= 1,723 girls). Physical activity data were collected using Fitbit accelerometers, recess quality data were collected using the Great Recess Framework – Observational Tool (GRF-OT), recess engagement data were collected using the Observation of Playground Play (OPP), and basic psychological need satisfaction (BPNS) data were collected using a modified version of the BPNS for recess physical activity survey. Primary analyses were conducted using Hierarchical Linear Modeling (HLM) with children nested within recess sessions.
Results
Gender moderated the relationship between adult engagement and moderate-to-vigorous physical activity (MVPA) (b= .012; 95% CI .001, .024), student behavior and MVPA (b= -.014; 95% CI -.021, -.007), and student behaviors and light physical activity (b= .009, 95% CI .003, .015). Both boys and girls engaged in more play during recess sessions scored as high quality on the GRF-OT. Children reported higher levels of basic psychological need satisfaction at recesses sessions scored as high quality on the GRF-OT.
Conclusions
Results of the current study showed that the quality of the recess environment, and the interactions of both adults and students in that environment, need to be taken into consideration in future school-based recess studies.

Full text

1
Preprint:Pleasenotethatthisarticlehasnotcompletedpeerreview.
Anobservationalstudyofrecessqualityandphysical
activityinurbanprimaryschools
CURRENTSTATUS:UNDERREVIEW
WilliamVincentMassey
OregonStateUniversity
william.massey@oregonstate.eduCorrespondingAuthor
ORCiD:https://orcid.org/0000-0002-4002-3720
MeganBStellino
UniversityofNorthernColorado
JohnGeldhof
OregonStateUniversity
DOI:
10.21203/rs.2.24385/v1
SUBJECTAREAS
HealthPolicy HealthEconomics&OutcomesResearch
KEYWORDS
Schoolhealth,Adultengagement,Play,Obesity

2
Abstract
Background
Todate,thereisscantliteraturethatexaminestherecesscontextconcurrentwith,butseparate
from,levelsofphysicalactivity.Theprimarypurposeofthecurrentstudywastoexaminehowrecess
qualityimpactedphysicalactivitylevels,andhowthiswasmoderatedbygender.Asecondary
purposewastoexamineifdifferencesinchildren’sengagementinactivitiesoccurredbetweenrecess
sessionsscoredaslow-orhigh-quality.
Methods
Thiswasanobservationalstudyofchildrenat13urbanelementaryschoolsintheU.S.Acrossthe13
schools,datawerecollectedat55recesssessions,with3,419child-levelobservations(n=1,696
boys;n=1,723girls).PhysicalactivitydatawerecollectedusingFitbitaccelerometers,recessquality
datawerecollectedusingtheGreatRecessFramework–ObservationalTool(GRF-OT),recess
engagementdatawerecollectedusingtheObservationofPlaygroundPlay(OPP),andbasic
psychologicalneedsatisfaction(BPNS)datawerecollectedusingamodifiedversionoftheBPNSfor
recessphysicalactivitysurvey.PrimaryanalyseswereconductedusingHierarchicalLinearModeling
(HLM)withchildrennestedwithinrecesssessions.
Results
Gendermoderatedtherelationshipbetweenadultengagementandmoderate-to-vigorousphysical
activity(MVPA)(b=.012;95%CI.001,.024),studentbehaviorandMVPA(b=-.014;95%CI-.021,
-.007),andstudentbehaviorsandlightphysicalactivity(b=.009,95%CI.003,.015).Bothboysand
girlsengagedinmoreplayduringrecesssessionsscoredashighqualityontheGRF-OT.Children
reportedhigherlevelsofbasicpsychologicalneedsatisfactionatrecessessessionsscoredashigh
qualityontheGRF-OT.
Conclusions
Resultsofthecurrentstudyshowedthatthequalityoftherecessenvironment,andtheinteractions
ofbothadultsandstudentsinthatenvironment,needtobetakenintoconsiderationinfutureschool-
basedrecessstudies.

3
Background
Lowlevelsofphysicalactivity(PA)remainaproblemthatcontributetothehighobesityratesseenin
children.Increasingly,schoolshavebecomeafocusforphysicalactivityinterventionsamongst
researchers(1)particularlygiventheamountoftimechildrenspendinthisenvironment.
Unfortunately,childrenintheUnitedStates(U.S.)continuetolackmeaningfulopportunitiesfor
physicalactivityatschool,despiteresearchthatshowstimespentengaginginphysicalactivity
makespositivecontributionstoacademics(2).AsidefromPE,school-basedrecessisalsoaprime
opportunityforphysicalactivitywithintheschoolday.Specifically,recesshasbeenshowntoaccount
for42%ofchildren’sopportunitiestobephysicallyactiveinschool(3),andupto44%ofstepcounts
duringtheschoolday(4).Despitethis,datafromthe2012–2013academicyearintheUnitedStates
(U.S.)suggestthat60%ofschooldistrictshavenoformalpolicyregardingdailyrecess.Ofschool-
districtsthathavearecesspolicy,only22%requiredailyrecessforelementaryschoolstudents,with
lessthanhalfoftheserequiringatleast20minutesofrecessperday(5).Whilerecentlegislative
effortshavebeenmadetopromoterecessatthestatelevel,onlysevenoutof50U.S.statesrequire
dailyrecessforchildrenduringtheschoolday(6).
Perhapsmoreconcerningisthetrendsdescribedabovedisproportionallyaffectchildrenfrom
disadvantagedbackgrounds.Datashowthatchildrenwhogotolarge,urbanschools;schoolswitha
highminoritypopulation;andschoolswithlow-incomelevelsaretheleastlikelytogetaccessto
recess,andoftenreporttheshortestamountoftimededicatedtorecess(7,8).Itisplausiblethat
environmentalfactorsaffectchildren’saccesstophysicalactivityopportunitiesinurbanandlow-
incomeschoolsystems.Notably,bullyingandaggressivebehaviorhavebeenreportedonthe
playgroundaturbanelementaryschools(9,10),whichcouldresultinfeweropportunitiesforstudents.
Furthermore,accesstospaceandequipmentarethoughttobecentraltorecessfacilitation(11),
whichcouldbelackingatlow-income,and/orurbanschools.Forexample,asystematicreview
conductedbyErwinetal.(4)suggeststhataddingmoreplaygroundequipmentandprovidinga
structuredrecessyieldsthelargesteffectonphysicalactivityduringrecess.Yet,budgetaryrestraints
couldlimitthepurchaseofequipmentinlow-incomeschooldistricts.Thus,thereisaneedtoconsider

4
bothaccesstorecess,andthequalityoftherecessenvironmentforpromotingphysicalactivity
throughouttheschoolday,particularlyinlow-incomeandurbanenvironments.
Asidefromaddressingdisparitiestophysicalactivityopportunitiessuchasrecess,aneedalsoexists
toexaminebarriersto,andfacilitatorsfor,physicalactivitywhenchildrendohaveaccessto
discretionarytimeduringtheschoolday.Onecommonareaoffocushasbeenondifferencesbetween
activitylevelsforboysandgirls,giventhatdatahasconsistentlyshownthatgirlsarelessactive
duringrecessperiods(12–14).Thesedatasuggestthatsocialdeterminantsmightplayarolein
children’sbehaviorontheplayground.Inexaminingbarrierstophysicalactivity,Pawlowskiand
colleagues(15,16)reportedseveralgenderedissuesthatmaylimitphysicalactivityduringrecess.
Forexample,elementaryschoolgirlsreportedwantingtoplaysports(i.e.,ballgames),butthose
wereusuallydominatedbytheboysontheplayground(15).Moreover,activitiesatrecesscanbe
labeledalonggenderlines,withgirlsbeingexpectedtoconformtomoresedentaryactivities(16).
Whilegirlsseemmoreinterestedincrossinggenderbordersatrecess(17),boysinthePawlowskiet
al.study(16)dominatedthelargerplaygroundspaces(i.e.,footballpitch)oftenexcludinggirlsduring
thegame.Thus,itappearsthatthereisaneedforbothvarietyofgamesandplayspaces,aswellas
inclusivebehaviorontheplayground,inanefforttomoreeffectivelypromotephysicalactivityat
recess.Indeed,researchershavereportedgirlsengageinsimilarlevelsofmoderate-to-vigorous
physicalactivity(MVPA)asboyswhenplayingteamsports(14)andthatprovidinganactivityofthe
weekinterventioncanyieldgainsinphysicalactivity(18).
Anotherimportantconsiderationforexaminingbehavioratrecessishowperceptionsofphysicaland
emotionalsafetymightimpactphysicalactivitybehaviorduringrecess.Previousresearchwithchild
participantshassuggestedthatconflictisaregularpartoftheplaygroundexperience(19,20).
Similarly,behavioralobservationshavesuggestedthatbullyingregularlytakesplaceduringrecess
periods(10).Intermsofphysicalactivityduringrecess,childrenhavespecificallydiscussedconflict
asabarriertoplayforboysandgirlsalike.Moreover,childrenhavereportedthatastrongerpresence
byadultsatrecessmighthelptofacilitatehigherlevelsofplay,astherewouldbeamoreneutral
mediatortoweighinongames(16).Thisnotionwassupportedbyarecentstudythatshowedadult

5
engagementandsupervisionasasignificantpredictorofplayforboysandgirlsduringrecess(21).
Additionalworkhasshownthatthepresenceofrecessinterventionschangesplaypatterns,
specificallyreducingnon-engagementandincreasingengagementinmoretraditionalgamessuchas
four-square,hopscotch,anduseoflooseequipment(22)
Inconsideringthepotentialbarrierstophysicalactivityatrecess,multipleinteractingfactorsareat
play.Accesstoresources,schoolpolicies,thephysicalenvironment,varioussocialdeterminants,and
studentbehaviorsallplayvariousrolesinfacilitating,orimpeding,physicalactivityatrecess.
Understandingoptimalstandardsforfacilitatingphysicalactivityduringrecessiscurrentlyneeded,
especiallyforthoseinurbanand/orlow-incomeschoolswhomayhavelimitedopportunities.Todate,
thereisscantliteraturethatexaminestherecesscontextconcurrentwith,butseparatefrom,levels
ofphysicalactivity.Recently,Masseyandcolleagues(23)developedanobservationaltooltomeasure
environmentalandsocialdeterminantsofelementaryschoolrecess.Specifically,theGreatRecess
Framework–ObservationalTool(GRF-OT)measuresthesafetyandstructureoftheplayground,adult
engagementandsupervision,aswellasstudentbehaviors.GiventheoverlapbetweentheGRF-OT
andresearchingexaminingpotentialbarriersandfacilitatorstorecessphysicalactivity,thereisa
needtoexaminehowtheabove-mentionedcontextualfactors,asmeasuredbytheGRF-OT,impact
physicalactivitylevels.Assuch,theprimarypurposeofthecurrentstudywastoexaminehowrecess
qualityimpactedphysicalactivitylevels,andhowthiswasmoderatedbygender.Asecondary
purposewastoexamineifdifferencesinchildren’sengagementinactivitiesoccurredbetweenrecess
sessionsscoredaslow-orhigh-quality.
Methods
InstitutionalReviewBoardapprovalwasobtainedpriortothestartofanystudyprocedures
(ConcordiaUniversityWisconsinID:932380-3;926512-1).Additionally,allprotocolsandprocedures
wereapprovedbytheresearchboardattheschooldistrictlevel,aswellastheprincipalateach
school.Inanefforttoensurewecouldincludeamaximumnumberofchildrenontheplayground,a
passiveconsentprotocolwasfollowed.Consentformsweresenthometoeachstudentandparents
andstudentsweregiventheopportunitytooptoutofthecurrentstudy.Inaccordancewithschool

6
districtpolicyonpassiveconsentprotocols,personallyidentifyinginformationwasnotcollectedon
participants.
Participants
Participantsincludedchildrenat13urbanelementaryschoolsintheU.S.Acrossthe13schools,
accelerometerdatawerecollectedat55recesssessions,with3,419child-levelobservations(n =
1,696boys;n = 1,723girls).Observationsofengagementinrecessactivitieswerecollectedat61
recesssessionswith4,528child-levelobservations(n = 2243boys;n = 2285girls).Thenumberof
childrenwithineachrecesssessionrangedfrom12to117withanaverageof62.16(SD = 26.34)
childrenperrecess.Allschoolsenrolledinthisstudyservedchildreninlow-incomeareas,with
publiclyavailabledatashowingthat78.8%ofthestudentpopulationiseconomicallydisadvantaged
(12-out-of-13schools > 50%ofeconomicallydisadvantagedstudents;range = 22.5%− 99.6%).Ofthe
13schools,10wereinthepublic-schoolsystem,withthreeofthese10schoolsbeinglanguage
immersionschools.Twelveofthe13schoolswereexclusivelyelementaryschools(i.e.,grades1–5),
andoneschoolalsoservedmiddleschoolstudents(i.e.,grades4–8).Enrollmentateachschool
rangedfrom253studentsto690students,withanaverageof436studentsperschool.
Measures
Physicalactivity.TheFitbitFlex™isawristworntriaxialaccelerometerthatusesproprietary
algorithmstoestimatestepscountsandtimespentinvariousactivitylevels.TheFitbitFlexprovides
themostsimplisticuserdisplayofallFitbitproducts,withonlyLEDlightstorepresentprogress
towardsdailygoals(thedefaultsettingis10,000steps;2000stepsperdotshown).Thiswasthought
tobeadvantageous,asparticipantswouldnotbeabletodirectlymonitortheirstepcountsofphysical
activitylevelsduringrecess.TheFitbitFlexcanbysynchedwirelesslytoasmartphoneortablet,and
providesinformationonstepscountsandtimespendinvariousactivitylevels(i.e.,sedentary,light,
moderate,vigorous).Forthepurposesofthecurrentstudy,theresearchteamcreatedanonymous
accountsforeachdevicethatcouldonlybeaccessedbytheresearchteam.Eachaccountwas
assignedtoeitheramaleorfemaleuser,withnationalaveragesforheightandweightbeingusedfor
userdemographicinformation.Fitbitswereplacedonstudentsintheirclassroom,orinthelunch

7
room,approximately30minutespriortothestartofrecess.Dataassessorsrecordedtheexactstart
andstoptimesofrecesssothatdatacouldbeextractedtomatchthetimestamp.Datawerehoused
byathird-partyvendor(FitabaseLLC,SanDiego,California)andprocessedusing60secondepochs
withinthenotedtimestamp,themostsensitivesettingavailableforthisdevice.Inchild-based
studies,bothwaist-worn(24)andwristworn(25)Fitbitdevices(FitbitOneandFitbitCharge,
respectively)havebeenshowntohaveconsistentlevelsofstepcountswithActigraph
accelerometers,yetmayover-estimateabsolutenumberofsteps,aswellastimespentinMVPA.
Additionalresearchinyoungadultpopulationshasshownmoderatevaliditybetweenthewrist-worn
FitbitFlexandthewrist-wornActigraphGT3X + infree-livingconditions(26),yettheFitbitflex
showedhigherlevelsofvariability,andwasmorelikelytounder-estimatedactivitylevels.
Recessquality.TheGreatRecessFramework–ObservationalTool(GRF-OT)wasusedasameasureof
recessqualityinthecurrentstudy.TheGRF-OTrepresentsfourdomainsofrecessthatincludesafety
andstructureoftheplayground,adultsupervisionandengagement,studentbehaviors,and
transitionstoandfromtherecessspace(23).Inthecurrentstudydatawascollectedonthreeofthe
foursub-scales(transitionswereexcluded,astheyaccountforthetimesimmediatelybeforeand
afterrecessandthefocusofthecurrentstudywasPAduringrecess).Itemsarescoredona4-point
scalebyaliveobserverwhowaspresentatrecess(4 = highestquality;1 = lowestquality).Thesafety
andstructuresub-scaleexaminesthephysicalenvironmentandaccesstoequipment;theadult
engagementandsupervisionsub-scaleexaminesthenumberofadultspresent,theirproximityto
students,andwhetherornottheyengagewithstudentsontheplayground;andthestudent
behaviorssub-scalesexaminesstudentengagement,initiationofplay,conflict,andconflict
resolution.EachitemanditsassociatedscoringprocedurecanbefoundinMasseyetal.2018(20).A
completescoringmanualwithdetailedinstructionsisavailableatwww.greatrecessframework.org.
Datainthecurrentstudysuggestacceptablelevelsofinternalconsistentforthesafetyandstructure
sub-scale(α = .806),adultengagementandsupervisionsub-scale(α = .736),andstudentbehavior
sub-scale(α = .788).Previousresearchhasshownsupportfortheinter-raterreliabilityandfactorial
validityoftheGRF-OT(23).

8
Engagementinrecessactivities.Thedifferenttypesofactivitieschildrenengagedinduringrecess
weremeasuredusingtheObservationsofPlaygroundPlay(OPP)(22).TheOPPallowsobservesto
codeengagementin32commonrecessactivitiesacrosseightdifferentplaydomains.Observersare
alsoabletowriteinobservedbehaviorswithineachoftheeightdomainstoensureallrecess
activitiesarecaptured.Previousresearchhasbeenreportedonthereliabilityofthisassessmenttool
(22).
Basicpsychologicalneedssatisfaction(BPNS).Asub-sampleoffourthandfifthgradestudents(n =
820)completedamodifiedversionofthebasicpsychologicalneedsatisfactionscale(27).Theoriginal
21-itemquestionnairedesignedtoassessindividualperceptionsofautonomy(7items),competence
(6items),andrelatedness(8items)needsatisfactionatworkwasmodifiedtospecificallyexamine
children’sneedsatisfactionduringrecessandreducedtotwo-itemsperscaletoensure
comprehensionforayoungerpopulation.Additionally,twoitemsrelatedtophysicalandemotional
safetyatrecesswereaddedtothescale.Allresponsescorrespondedtoa5-pointLikertscale(5 =
highneedsatisfaction)onitemssuchas“IfeellikeIcansaymyideasaboutwhatIwanttodoat
recess”(autonomy),“KidstellmeIamgoodatthingsIdoatrecess”(competence),“Ireallylikethe
kidsIplaywithatrecess”(relatedness),and“IamsafewhenIamplayingatrecess”(safety).
Procedures
Withtheexceptionofoneschoolthatconductedconcurrentindoorandoutdoorrecessperiods,all
recessdatacollectionperiodswereconductedoutside.DatacollectiontookplacebetweenFebruary
andMayinalargecityintheMidwesternregionoftheUnitedStates.Recessperiodsrangedfrom12
minutesto40minutesinlength(M = 21.12minutes;SD = 5.83minutes)andprimarilyincluded
traditionallunchrecessperiods.Schoolsmaintainedvariableschedules,withsomeschoolssending
groupsofstudentsoutsideallatonce,whileothersrotatedthesessionswithdifferentchildrenand
differentsupervisors(e.g.,onlyfirstthroughthirdgradersatrecessone,followedbyonlyfourthand
fifthgradersatrecesstwo).
Outcomeassessorsarrivedtotheschoolapproximately60minutesbeforethescheduledrecess
sessiontoensurestudentswereproperlyfittedwithactivitymonitoringdevices.Eachdatacollection

9
periodcontainedfourstudyteammembers.Twomembersoftheteamwereassignedtoensure
complianceintermsofproperlywearingtheactivitymonitoringdevices.Theothertwoteam
memberscollectedobservationaldatathroughouttheentirerecesssession.DatausingtheGRF-OT
werecollectedbythePIoratrainedgraduatestudent.Inallcases,therecessenvironmentwas
completelyvisibletotheoutcomeassessor,andtheoutcomeassessormovedthroughoutthe
playgroundinadiscreetmannerinanefforttoobservepatternsofinteractionandbehavior.Final
scoringofeachitemwascompletedimmediatelyaftertherecesssessionandtookintoaccountthe
aggregatepatternsofbehaviorthroughoutthedurationoftherecesssession.DatausingtheOPP
werecollectedatfive-minuteintervalsacrosseachrecessperiod.OPPdatawerethenaveragedto
createacompositelevelofstudentengagementindifferentactivitiesforeachrecesssession.
DataAnalysis
Priortodataanalysis,FitbitFlexdatawerescreenedanddevicesregistering0stepcountsinthe
recesstimerecordingwereeliminatedfromthedataset.Timerecordingsofthebeginningandending
ofeachrecesssessionwerekepttoallowforspecificityindataextractionwhenexaminingrecess-
basedphysicalactivity.Furthermore,Fitbitnumberswereloggedandtrackedforeachrecesssession
toensurewhichdeviceswereinuseforeachsession,andwhichdeviceswerereturnedattheendof
eachrecesssession.Giventhevaryingtimesandnumberofstudentsacrossrecesssessions,we
convertedphysicalactivitydatatothepercentoftimespentinMVPAorlightphysicalactivity(LPA)
duringrecessandusedthesepercentagesasthedependentvariableinprimaryanalyses.
PrimaryanalyseswereconductedusingHierarchicalLinearModeling(HLM)withchildrennested
withinrecesssessions.Interceptsfreelyvariedacrossrecesssessions,whileallprimarypredictors
wereenteredasfixedeffects.Anunconditionalnestedmodelwasfirsttestedtoexaminepossible
recess-leveleffectsforalldependentvariables(i.e.,physicalactivitylevels).Next,modelswerefitted
inwhichrecessqualityscoreindicators(i.e.,adultengagementandsupervision,studentbehaviors,
safetyandstructure)wereenteredaspredictorsofphysicalactivitylevelswhilecontrollingforschool
asafixedeffect.Moderationwasalsotestedbyexaminingtheinteractionbetweengender(alevel
onepredictor)andrecessqualityindicators(aleveltwopredictor)onlevelsofphysicalactivityduring

10
recess.Significantinteractionswereprobedtoexaminethesimpleslopesandinterceptsasafunction
ofgenderusingtheformuladepictedbelowandasdescribedbyPreacherandcolleagues(28),where
ý00representstheintercept,ý01,ý10andý11aretheregressioncoefficients,xrepresentsthefocal
predictor,andwrepresentsthemoderatorvariable.
yij=ý00+ý10x1+ý01w1+ý11w1x1
Finally,trendsacrossrecessqualitywereexaminedrelativetohigh-andlow-qualityrecesssessions
(i.e.,oneSDaboveandoneSDbelowthesamplemean).Ingeneral,high-qualityrecesssessionswere
characterizedbysafephysicalenvironments(e.g.,lackofhazardousmaterials),abroadrangeof
equipmentandactivitiestoengageinplay,prosocialstudentbehaviors(e.g.,initiatinggames,
positivecommunication,lackofphysicalviolence)andpresentandengagedadults.Incontrast,low-
qualityrecesssessionswereoftencharacterizedasunsafeenvironments(e.g.,glassandhazardous
debris),limitedornoequipmenttouseforgameplay,verbalandphysicalconflicts,anddisengaged
adults.AggregateprofilesofrecesssessionsintheupperandlowerquartileforGRF-OTscoreswere
createdtoexaminedifferentialpatternsinthegamesandactivitiesinwhichchildrenparticipatein,
andthepsychologicalneedsatisfactionofchildrenduringrecess.
Results
DescriptivestatisticswerecalculatedforallvariablesunderstudyandcanbefoundinTable1.
ResultsofthenullmodelscanbefoundinTable2(MVPA)andTable3(LPA).

11
Table1
Descriptivestatistics.
Variable Mean SD Range(possiblerange)
TotalRecessQualityScore 41.15 7.44 19–54(14–56)
SafetyandStructureof
Environment 14.96 3.49 6–20(5–20)
AdultEngagementand
Supervision 10.61 2.49 5–16(4–16)
StudentBehaviors 15.58 3.37 5–20(5–20)
   
PhysicalActivity 
Percentoftimespentin
MVPA 50.54% 35.96% 0-100%
Percentoftimespentin
LPA 36.42% 29.20% 0-100%
   
PsychologicalNeed
Satisfaction
34.34 4.74 12–40(10–40)
Autonomy 7.94 1.91 2–10(2–10)
Competence 6.77 1.84 2–10(2–10)
Relatedness 8.23 1.68 2–10(2–10)
Safety 8.41 1.67 2–10(2–10)
EngagementinRecess
Activities(percentof
students)

Equipment 0% 1% 0–8%
Organizedactivities 30% 24% 0–92%
Anti-socialbehavior 1% 3% 0–22%
Non-engagedinplay 30% 22% 0–84%
Nature 2% 6% 0–34%
Running/chasinggames 17% 15% 0–76%
Traditionalplayground
games 16% 14% 0–68%
Roughandtumbleplay 2% 7% 0–52%
   
Table2
UnconditionalnestedmodelforpercentoftimespentinMVPAduringrecess
Estimationofcovarianceparameters 
Parameter Estimate s.e.WaldZ p-value 95%CIof
the
estimate
ICC
Residual .122 .003 40.99 < .001 0.116,
0.128 
InterceptVariance
(RecessSession) .009 .002 3.97 < .001 .005,.014 0.066
 
Table3
UnconditionalnestedmodelforpercentoftimespentinLPAduringrecess
Estimationofcovarianceparameters 
Parameter Estimate s.e.WaldZ p-value 95%CIof
the
estimate
ICC
Residual .082 .002 41.00 < .001 .078,.086 
InterceptVariance
(RecessSession) .003 .001 3.59 < .001 .002,.006 0.041
 
AnexaminationofpredictorsofthepercentoftimespentinMVPAatrecessshowedgenderasthe
onlysignificantpredictorinthecurrentstudy(p = .001).However,moderationanalysesrevealedthat
gendermoderatedtherelationshipbetweenadultengagementandMVPA(p = .046),andstudent

12
behaviorandMVPA(p = < .001).ResultscanbefoundinTable4.Simpleslopesanalysesindicated
thatgenderwasnotasignificantpredictorofpercentoftimespentinMVPAatlow(M-1SD;b =
− .116;p = .107),moderate(M;b = − .085;p = .287)orhigh(M + 1SD;b = − .055,p = .542)levelsof
adultengagementandsupervision.However,ascanbeseeninFig.1,thedifferencebetweenboys’
andgirls’percentoftimeinMVPAwasminimizedashigherlevelsofadultengagementand
supervisionwereobserved.Inexaminingstudentbehaviors,simpleslopesanalysesrevealedthat
genderwasasignificantpredictorofpercenttimeinMVPAatlow(M-1SD;b = − .381;p = < .001),
moderate(M;b = − .429;p = < .001)andhigh(M + 1SD;b = − .477,p = < .001)levelsofprosocial
studentbehavior.AscanbeseeninFig.2,boys’percentoftimeinMVPAwashigheratrecess
sessionsinwhichmoreprosocialstudentbehaviorswereobserved;whereasgirls’percentoftimein
MVPAwasloweratrecesssessionsinwhichmoreprosocialstudentbehaviorswereobserved.
Table4
EstimatesofeffectsonpercentofMVPAduringrecess
Estimationofcovarianceparameters
Parameter Estimate s.e.WaldZ p-value 95%CIoftheestimate
Residual .110 .003 40.99 < .001 .105,.115
Intercept
Variance
(recess)
.009 .002 4.12 < .001 .006,.014
Estimationoffixedeffects
Parameter Estimate s.e.Ttest
statistic
p-value 95%CIoftheestimate
Intercept .641 .093 6.87 < .001 .454,.827
Gender − .215 .065 -3.32 .001 − .341,− .088
School − .002 .005 − .365 .717 − .012,.008
Safetyand
Structureof
Environment
− .002 .006 − .380 .705 − .014,.010
Adult
Engagement
and
Supervision
− .003 .008 − .354 .724 − .018,.013
Student
behaviors .003 .006 .519 .605 − .009,.015
Gender*
Safetyand
Structureof
Environment
.006 .004 1.32 .186 − .003,.014
Gender*
Adult
Engagement
and
Supervision
.012 .006 1.99 .046 .001,.024
Gender*
Student
behaviors
− .014 .004 -3.725 < .001 − .021,− .007
 
InexaminingLPA,genderwasonceagaintheonlysignificantpredictorofLPA(p = .004).Genderalso
moderatedtherelationshipbetweenstudentbehaviorsandLPA(p = .005).Resultscanbefoundin

13
Table5.Simpleslopesanalysesrevealedthatgenderwasasignificantpredictorofpercenttimein
LPAatlow(M-1SD;b = .256;p = < .001),moderate(M;b = .286;p = < .001)andhigh(M + 1SD;b
= .317,p = < .001)levelsofprosocialstudentbehavior.Specifically,girlsrecordedhigherlevelsof
LPAduringrecesssessionswithhighlevelsofprosocialstudentbehavior;whereasboysrecorded
lowerlevelsofLPAduringrecesssessionswithhighlevelsofprosocialstudentbehavior(seeFig.3).
Table5
EstimatesofeffectsonpercentofLPAduringrecess
Estimationofcovarianceparameters
Parameter Estimate s.e.WaldZ p-value 95%CIoftheestimate
Residual .074 .002 41.00 < .001 .071,.078
Intercept
Variance
(recess)
.003 .001 3.60 < .001 .002,.006
Estimationoffixedeffects
Parameter Estimate s.e.Ttest
statistic
p-value 95%CIoftheestimate
Intercept .201 .063 3.21 .002 .076,.326
Gender .152 .053 2.86 .004 .048,.257
School .004 .003 1.18 .241 − .003,.010
Safetyand
Structureof
Environment
.004 .005 .885 .379 − .004,.012
Adult
Engagement
and
Supervision
.008 .005 1.52 .133 − .003,.019
Student
behaviors − .006 .004 -1.41 .163 − .014,.002
Gender*
Safetyand
Structureof
Environment
− .002 004 − .679 .497 − .009,.005
Gender*
Adult
Engagement
and
Supervision
− .008 .005 -1.50 .133 − .017,.002
Gender*
Student
behaviors
.009 .003 2.83 .005 .003,.015
 
EngagementInRecessActivities
Followinganalysesofphysicalactivityatrecess,patternsofplaywerecomparedforrecesssessions
atleastonestandarddeviationabovethemeanrecessqualityscoreandatleastonestandard
deviationbelowthemeanrecessqualityscore(n = 952boys;n = 952girls).AscanbeseeninFig.4,
thelargestdifferenceswereseeninnon-engagementinplay(e.g.,talkingwithfriends,watching
others),with61percentofgirlsatlow-qualityrecesssessionsnon-engagedinplayascomparedto22
percentofgirlsathighqualityrecesssessions.Girlsathighqualityrecesssessionsalsoparticipated
inmoreorganizedgamesthangirlsatlow-qualityrecesssessions(e.g.,dance,kickball,soccer;23%

14
vs.9%),andmoretraditionalplaygroundactivities(e.g.,four-square,jumpropes;21%vs.9%).As
seeninFig.5,asimilarpatternwasobservedforboysasitrelatedtonon-engagement(10%athigh
qualityrecesssessionsvs.36%atlowqualityrecesses),participationinorganizedactivities(52%at
highqualityrecesssessionsvs.37%atlowqualityrecesses),andparticipationintraditional
playgroundactivities(15%athighqualityrecesssessionsvs.5%atlowqualityrecesses).
Finally,basicpsychologicalneedsatisfactionscoreswerecomparedforthoseattendinghigh-quality
recesssessions(i.e.,atleastoneSDabovethemean)andthoseattendinglow-qualityrecesssessions
(i.e.,atleastoneSDbelowthemean).Resultsindicatethatchildrenatahigh-qualityrecesssession
reporthigherlevelsofbasicpsychologicalneedsatisfactionatrecessthanchildrenatalow-quality
recesssession(t(324) = 2.65;p = .004).Anexaminationoftheindividualsub-scalesrevealedno
groupdifferencesincompetenceorrelatedness,howeverthoseatahigh-qualityrecessreported
higherlevelsofperceivedautonomyatrecesst(333) = 1.79;p = .037)andhigherlevelsofperceived
safetyatrecess(t(306) = 4.52;p < .001).
Discussion
Theprimarypurposeofthecurrentstudywastoexaminehowcontextualfeaturesoftheplayground
impactedphysicalactivitylevels,andhowthiswasmoderatedbygender.Asecondarypurposewas
toexaminepatternsofengagementatrecesstobetterunderstandhowthequalityofthe
environmentshapesphysicallyactivebehaviorduringthisdiscretionarytimeperiod.Resultsofthe
currentstudyshowedthatthequalityoftherecessenvironment,andtheinteractionsofbothadults
andstudentsinthatenvironment,needtobetakenintoconsiderationinfutureschool-basedrecess
studies.InexaminingpatternsofMVPA,resultsshowedthathigherlevelsofadultengagementand
supervisionreducedthedisparitybetweenboys’andgirls’physicalactivitylevelsatrecess.This
findingissupportedbypreviousresearchthatshowshigherlevelsofadultengagementpredicthigher
levelsofstudentengagementatrecess(21),thatchildrenreportteacherengagementasapositive
influenceonrecess(29),andthatteachersupportisafacilitatorofactivityatrecess(30).Moreover,
givenpreviousreportsofsocialbarriersgirlsfaceinbeingphysicallyactiveatrecess(e.g.,boys
dominatingequipmentandspace;15,16),itislikelythatengagedadultsontheplaygroundhelpto

15
facilitateequityinaccesstoplaygroundresourcesduringrecess.
Anexaminationofphysicalactivitylevelsatvariouslevelsofprosocialstudentbehaviorsshowedthat
boysparticipatedinhigherlevelsofMVPAduringrecesssessionswithhighpro-socialbehaviors.For
boys,atahigh-qualityrecesssession,higherlevelsoforganizedgames(i.e.,soccer,basketball,
football)likelyaccountedforhigherlevelsofMVPA.Notably,whenprosocialbehaviorswerehigh(i.e.,
limitedfighting,limitedarguments,highlevelsofgameinitiation,highlevelsofconflictresolution)
gameswentonuninterrupted,allowingboystofurtherengageinMVPAduringrecess.Incontrast,
girlsparticipatedinhigherlevelsofMVPAduringrecesssessionswithlowpro-socialbehaviors.While
contrarytoexpectations,thefindingsrelatedtogirlsincreasedratesofMVPAatrecesssessionswith
lowpro-socialbehaviorscanpartiallybeexplainedbyactivitypatternsatrecess.Popularactivitiesfor
girlsathighqualityrecesssessionsincludedtraditionalplaygroundgames(e.g.,4-square,hopscotch)
aswellasorganizedactivities(e.g.,dance),whicharegenerallyalignedwithLPA,asopposedto
MVPA.AsshowninFig.3,LPAwashigherforgirlsatrecesssessionswithhighlevelsofpro-social
behavior.Thus,whenprosocialbehaviorswerehigh,gamesandactivitiesinwhichgirlswere
participatinginhadfewerinterruptions.However,atlow-qualityrecesssessions,orrecesssessions
characterizedashavinglowlevelsofprosocialbehavior(i.e.,higherratesoffightingandarguing,low
levelsofconflictresolution),itispossiblethatotherwiseunengagedgirlsengagedinintermittent
boutsofMVPA.Thesedataareconsistentwithpreviousdataofboysandgirlsplaypatternsatrecess
(15,16,21,22),yetalsounderscoretheimportanceofexaminingthecontextinwhichphysicalactivity
occurs.Notably,whileincreaselevelsofPAremainsagoalofmanyresearchers,datafromthe
currentstudyunderscorestheneedtounderstandthequalityofthePAenvironmentasitrelatesto
children’sholisticdevelopment.
Previousresearchhasalsoexaminedpsychologicalneedsatisfactionandmotivationforphysical
activityatrecess.Notably,StellinoandSinclair(27)reportedthatpsychologicalneedsatisfactionwas
predictiveofphysicalactivitymotivationatrecess,andthatautonomywaspredictiveofphysical
activityduringrecess.Resultsofthecurrentstudysupportthesedata,andfurthersuggestthata
high-qualityrecessenvironmentcanhelpfacilitatebasicpsychologicalneedsatisfaction.Notably,

16
childrenathigh-qualityrecesssessionsreportedhigheroverallpsychologicalneedsatisfaction,higher
autonomyforrecessphysicalactivity,andhigherlevelsofperceivedsafetythanthoseatlow-quality
recesssessions.Thesefindingssuggestthatstructuringtherecessenvironmenttoincludeavariety
ofplayopportunitiesandtrainingadultstobeactivelyengagedinrecessareautonomysupportive,as
opposedtocontrollingstrategies.Whilerecessremainsoneoffewdiscretionarytimeperiodsduring
theschoolday,datasuggeststhatimplementingevidence-basedpracticesduringrecesscanhelpto
enhancethisdiscretionarytimeandsatisfystudent’sbasicpsychologicalneedsofautonomy,
competence,andrelatedness.
StrengthsAndLimitations
Takentogether,themajorstrengthsofthisstudyincludeamultifacetedexaminationofrecesswithin
anunder-researchedpopulation–urbanelementaryschoolstudentsinlowincomeschools.Further,
theresultsofthisstudyshedimportantlightonfuturerecessresearch.Notably,theequivocal
findingsofvariousinterventionsonphysicalactivitypromotionandsocialbehaviors(31–33)warranta
morenuancedunderstandingofrecess,particularlyinunder-resourcedcommunities.Thecurrent
studysuggeststheneedformulti-facetedinterventionsthatconcurrentlyfocusonincreasedaccess
torecess,increasedaccesstoequipmentandplayspaces,positiveandencouragingadultbehavior,
andpro-socialstudentbehavior.Indeed,perhapstheshort-comingsinpreviousinterventionstudies
relytosingularordual-purposeinterventions,ratherthanafocusonmultipleinterventionstaking
placesimultaneously.
Themajorlimitationsinthecurrentstudyarethereducedsamplesizeduetoanalysistakingplaceat
thegrouplevel,thecross-sectionalnatureofthedatacollection,andlimitationswiththeuseofFitbits
tocaptureobjectivephysicalactivityinchildpopulations.Becausechildrenengageinmore
intermittentboutsofPA,particularlyduringrecess,shortermeasurementintervalsarethoughtto
providemoreaccurateestimatesoftimespentinvariousPAintensities.Inastudyexamining
differencesinMVPAforchildrenduringphysicaleducationclasses,McClainandcolleagues(34)
compareddirectobservationandaccelerometryatvariousepochlengths.Resultsshowedthat
estimatesofMVPAwerelowerwhenusingaccelerometry,ascomparedtodirectobservation

17
regardlessofepochlength.However,longerepochlengthswereassociatedwithlowerMVPAcounts.
Similarly,morerecentresearchhasshownlowerestimatesofMVPAinchildrenwhenusinga60
secondepochasopposedtoa15-or5-secondmeasurementperiod(35).Moreover,Bandaand
colleagues(35)reportedthatincreasedepochlengthmayover-estimateLPAinchildrenasincreased
epochtimewasassociatewithsedentarybehaviorbeingreclassifiedasLPA.Giventhis,itispossible
thatMVPAwasunder-estimatedbasedonthedurationandintensityofvariousactivitiesengagedin
duringrecess;whereasLPAmayhavebeenover-estimated.However,concurrentobservationaldata
ofengagementinrecessactivitiessupportstheoverallpatternofresults;specifically,higherlevelsof
engagementinactivitiesassociatedwithincreasedPAduringhigh-asopposedtolow-qualityrecess
sessions.
Conclusions
Recessremainsacriticalopportunityforchildrentobephysicalactiveduringtheschoolday.Results
ofthecurrentstudysuggestthatincreasingadultengagementandfacilitatinghigherlevelsofpro-
socialbehaviorareimportanttonotonlyphysicalactivitypromotionatrecess,butalsochildren’s
psychologicalneedsatisfaction.
Abbreviations
PA
PhysicalActivity
US
UnitedStates
PE
PhysicalEducation
MVPA
Moderate-to-VigorousPhysicalActivity
GRF-OT
GreatRecessFramework–ObservationalTool
LPA
LightPhysicalActivity
OPP
ObservationsofPlaygroundPlay
BPNS

18
BasicPsychologicalNeedSatisfaction
HLM
HierarchicalLinearModeling
SD
StandardDeviation
Declarations
EthicsApproval.EthicsapprovalwasprovidedbytheinstitutionalreviewboardatConcordia
UniversityWisconsin(ID:932380-3;926512-1).Theneedforconsentwaswaivedasnoindividual
leveldatawerecollected.ThePI(WVM)waspreviouslyfacultyatConcordiaUniversityWisconsin.
ConsentforPublication.Notapplicable.
AvailabilityofDataandMaterials.Thedatasetsusedand/oranalysedduringthecurrentstudy
areavailablefromthecorrespondingauthoronreasonablerequest.
CompetingInterests.Theauthorsdeclarethattheyhavenocompetinginterests
Funding.FundingforthisstudywasprovidedbyPlayworksEducationEnergized.Thefundingagency
wasnotinvolvedindatacollection,analysis,orinterpretation.
AuthorContributions.WVMdesignedthestudy,trainedalldataassessors,assistedwithdata
collectionanalyzedthedata,interpretedthedata,anddraftedthemanuscript.MBSassistedwith
studydesign,interpretation,andwritingofthemanuscript.JGassistedindataanalysis,data
interpretation,andeditingofthemanuscript.Allauthorshavegiventheirfinalapprovalforthework
tobepublishedandhaveagreedtotakeaccountabilityforallaspectsofthework.
Acknowledgments.NotApplicable.
References
1.  D o bb i n s M ,  H us s o n H ,  D eC o r by  K ,  La R o cc a  R L.  S c ho o l - b as e d  ph y s i c al  a c t iv i t y
programsforpromo t i n g  p h y s ic a l  a c ti v i t y  a n d f i t n e ss  i n  ch i l d r e n a n d a d o le s c e n ts
aged6to18.Cochra n e  D a t a b as e  Sy s t  R ev .  2 01 3 F e b  28 ; ( 2 ): C D 0 07 6 51 .
2.  R a sb e r ry  C N , L e e  SM ,  R ob i n  L,  L a r is  B A , R u s s el l  L A ,  C oy l e  K K,  e t  a l.  T h e a s s o ci a t i o n
betweenschool-based  p h y s i c a l  ac t i v i t y , i n c l u d in g  p h ys i c a l e d u c at i o n ,  an d  a ca d e mi c
performance:asyste m a t i c  r e v ie w  o f t h e  li t e r a t u re .  P r e v M e d.  2 01 1  Ju n ; 5 2 S u p pl

19
1:S10-20.
3.  R o be r t  Wo o d J o h n s on  F o un d a ti o n .  R ec e s s r u l e s : W h y t h e  u nd e r va l u e d p l a y ti m e  m ay
beAmerica’sbestinve s t m e n t  f o r h e a lt h y  k id s  a n d h e a lt h y  s c ho o l s r e p o rt  [ I n t e rn e t ] .
2007.Availablefrom:
http://www.rwjf.org/co n t e n t / d a m /f a r m/ r e po r t s / re p o r ts / 2 0 07 / r w jf 1 8 06 0
4.  E r wi n  H , A b e l M ,  B ei g h l e A ,  N o la n d  M P,  W or l e y B ,  R i g gs  R . T h e  Co n t r i b ut i o n  of  R e ce s s
toChildren’sSchool-Day  P h y s i c a l  A ct i v i t y .  J P h y s  Ac t  H ea l t h .  20 1 2 M a r 1 ; 9 (3 ) : 4 4 2– 8 .
5.  U S C e n te r s  f o r D i s e as e  C on t r o l.  S t r a te g i e s f o r  s u pp o r ti n g  r e ce s s  in  e l e m en t a ry
schools,updateforthe  2 0 1 2 – 13  s c ho o l  y ea r  [ I n te r n e t] .  A v a il a b l e  fr o m :
https://www.cdc.gov/h e a l t h y s c h oo l s / n pa o / p df / L W P_ R ec e s s _B r i e f_ 2 0 12 _ 1 3. p d f.
6.  S h af e r  M,  W hi t e h o us e  E . S t a t e p o l i c i e s o n  ph y s i c al  a c t iv i t y  i n  s ch o o l s [ I n t e r ne t ] .
CouncilofStateGovernm e n t s ;  2 01 8 [ c i t e d  20 2 0 J a n  20 ] .  A va i l a b le  f r o m:
http://knowledgecenter . c s g . o r g /k c / s y st e m /f i l e s / C R _a c t iv i t y _ s c ho o l . p df
7.  B a rr o s  RM ,  Si l v e r  E J,  S t e i n R E .  Sc h o o l r e c e ss  a n d g r o up  c l a ss r o o m b eh a v io r .
Pediatrics[Internet].200 9  [ c i t e d  2 01 9 J u n  2 5] ; 1 2 3( 2 ) :4 3 1 –6 .  A va i l a b le  f r o m :
http://pediatrics.aappub l i c a t i o n s . o rg / c g i / do i / 1 0 .1 5 4 2/ p e ds . 2 00 7 - 28 2 5
8.  M on n a t S M , L o u n s be r y M A F , M c K en z i e  TL ,  C h an d l e r R F .  As s o c ia t i o n s b e t w ee n
demographiccharac t e r i s t i c s  a n d p h y si c a l  a ct i v i t y  p ra c t i c es  i n  N ev a d a s c ho o l s .  Pr e v
Med[Internet].2017F e b  [ c i t e d 2 0 2 0 J a n 2 0 ] ; 9 5: S 4 –9 .  A va i l a b le  f r o m:
https://linkinghub.elsevier . c o m / r e t r ie v e / p ii / S 0 0 91 7 43 5 1 63 0 23 6 5
9.  A n de r s o n- B u t ch e r  D , N e ws o m e W S,  N a y S .  S oc i a l  Sk i l l s  I n t er v e n ti o n  d ur i n g
ElementarySchoolRe c e s s :  A  V i s ua l  A n al y s i s . C h i l d  S ch .  2 00 3 ;2 5 ( 3 ): 1 3 5– 4 6 .
10.  M a s se y  W V,  S t e ll i n o  M B,  H ol l i d a y  M,  G o db e r se n  T , R o d ia  R ,  K uc h e r G ,  e t a l .  T h e
impactofamulti-comp o n e n t  p h ys i c a l a c t i v i t y p r o g ra m me  i n l o w - i n co m e e l e m en t a ry
schools.HealthEducJ.2 0 1 7 ; 7 6 (5 ) : 5 17 – 53 0 .

20
11.  H u b e rt y  J L , B e e t s M W , B e i g hl e  A , W e l k G .  E n vi r o n m en t a l m o di f i c a t i o ns  t o  in c r e a se
physicalactivityduringre c e s s :  p r e li m i n a ry  f i n d in g s  f ro m r e a d y f o r  r ec e s s.  J  P h y s A c t
Health.2011Sep;8S u p p l  2 : S 24 9 -2 5 6 .
12.  V i c i a n a J ,  M ay o r g a- V e g a D , M a r t ín e z - B a en a  A.  M o de r a te - t o - V ig o r o u s P h y si c a l  A ct i v i t y
LevelsinPhysicalEduca t i o n ,  S c h o ol  R e c es s ,  a nd  A ft e r - S c ho o l  T im e :  In f l u e n ce  o f
Gender,Age,andWe i g h t  S t a t u s.  J  P h y s A c t  He a l t h . 2 0 1 6; 1 3( 1 0 ): 1 1 1 7– 2 3 .
13.  S a i n t - M au r ic e  P F , W e l k G J ,  S il v a  P ,  S ia h p u sh  M , H u b er t y  J . A s s e ss i n g  c hi l d r e n’ s
physicalactivitybehavior s  a t  r e c e ss :  a  mu l t i - m et h o d a p p ro a c h.  P e d ia t r  E xe r c  Sc i .
2011Nov;23(4):585–99.
14.  B a q u et  G , R i d g e rs  N D , B l a e s A ,  A uc o u tu r i e r  J , V a n  Pr a a g h E ,  Be r t h oi n  S .  Ob j e c ti v e l y
assessedrecessphys i c a l  a c t i v i t y  in  g i r l s  an d  b oy s  f ro m  hi g h  a nd  l o w s o c io e c o n om i c
backgrounds.BMCP u b l i c  H e a lt h .  2 0 14  F eb  2 1; 1 4 :1 9 2 .
15.  P a w lo w s k i C S ,  T jø r n h ø j- T h o m se n  T,  S c h ip p e r ij n  J ,  T r o el s e n J .  B a r ri e r s  f or  r e c es s
physicalactivity:agende r  s p e c i f i c q u a l i t a ti v e  f o cu s  g ro u p e x p l o ra t i o n . B M C P u b li c
Health.2014Jun23;1 4 ( 1 ) : 6 3 9.
16.  P a w lo w s k i C S ,  E rg l e r  C,  T j ø r nh ø j - T h om s en  T ,  Sc h i p pe r i j n  J , T r o e ls e n  J .  “ Li k e  a S o c c er
CampforBoys”:AQu a l i t a t i v e  E x pl o r a ti o n  o f G e n d er e d A c t i v i t y P a t t e rn s  i n  Ch i l d r e n’ s
Self-OrganizedPlaydur i n g  S c h o o l R e c e ss .  E ur  P h y s E d uc  R e v.  2 0 15  A ug ; 2 1 (3 ) : 2 75 –
91.
17.  B o y l e D E ,  M ar s ha l l  N L , R o b es o n  WW . G e n d er  a t P l a y :  F ou r t h -G r a d e G i r l s a n d  Bo y s  on
thePlayground.AmB e h a v  S c i . 2 0 0 3 J u n  1; 4 6 (1 0 ) : 1 32 6 –4 5 .
18.  S t e l l i n o M B , S i n c l a ir  C D ,  Pa r t r i d ge  J A , K i n g  KM .  D if f e r e nc e s  in  c h i l d r en ’ s  r ec e s s
physicalactivity:recessa c t i v i t y  o f  t h e w e ek  i n t e rv e n ti o n .  J  S ch  H ea l t h .  20 1 0
Sep;80(9):436–44.
19.  R i d g e rs  N D , S t r a tt o n  G , M c Ke n z i e T L .  R el i a b i l i t y  an d  v al i d i t y  of  t h e S y s t e m f o r

21
ObservingChildren’sAc t i v i t y  a n d  R el a t i o ns h i p s d u r i n g P l a y ( S O C AR P ) . J  P h y s A c t
Health.2010Jan;7(1) : 1 7 – 2 5 .
20.  P a w lo w s k i C S ,  S ch i p p er i j n  J ,  T jø r n h ø j- T h o m se n T ,  T r o el s e n J .  G i v i n g c h i l d re n  a v o i c e :
Exploringqualitativeper s p e c t i v e s  on  f a ct o r s  i n fl u e n ci n g  r e ce s s p h y s ic a l  a c ti v i t y .  E u r
PhysEducRev.2018  F e b  1 ; 2 4( 1 ) :3 9 – 5 5.
21.  M a s se y  W V,  S t e ll i n o  M B,  F r as e r  M.  I n d iv i d u a l a n d  en v i r o nm e nt a l  c o rr e l a t es  o f  sc h o o l-
basedrecessengage m e n t .  P re v  M ed  R ep .  2 01 8 S e p  1; 1 1 :2 4 7 –5 3 .
22.  M a s se y  W V,  K u B ,  S t el l i n o  M B.  O bs e r v at i o n s o f  p l a yg r o un d  p la y  d u ri n g  e le m e nt a r y
schoolrecess.BMCRe s  N o t e s  [ In t e r n e t] .  2 0 18  O ct  2 3 [ c i t e d 2 0 1 9 A u g 2 0 ]; 1 1 .
Availablefrom:https://w w w . n c b i. n l m . ni h . g o v/ p m c/ a r t i c le s / P M C6 1 99 6 97 /
23.  M a s se y  W V,  S t e ll i n o  M B,  M ul l e n  S P,  C l a as s e n J ,  W i l k is o n  M . D e v el o p m en t  of  t h e g r e a t
recessframework– o b s e r v a ti o n a l  t o ol  t o  m ea s ur e  c on t e x tu a l  a n d b e ha v i o ra l
componentsofeleme n t a r y  s c h oo l  r e ce s s . B M C P u b l i c H e a lt h .  2 0 18  M ar  2 2; 1 8 (1 ) : 3 94 .
24.  H a m ar i  L , K u l l b e rg  T ,  Ru o h on e n  J,  H e i n on e n  OJ ,  D í a z- R o d rí g u e z N ,  L i l i u s J ,  e t  a l.
Physicalactivityamong c h i l d r e n :  o bj e c t i v e m e as u r em e n ts  u s i n g F i t b it  O n e ® a n d
ActiGraph.BMCResN o t e s  [ I n t e r ne t ] .  2 01 7 A p r  2 0 [ c i t ed  2 0 19  A ug  2 0] ; 1 0 . A v a il a b l e
from:https://www.ncb i . n l m . n i h . go v / p mc / a rt i c l e s / PM C 5 39 7 82 8 /
25.  V o s s  C,  G a rd n e r R F ,  D ea n P H ,  H ar r i s K C .  V a li d i t y  o f C o m me rc i a l  A c ti v i t y  T r ac k e rs  i n
ChildrenWithCongenital H e a r t  D i s ea s e . C a n  J C a r d io l .  2 0 17 ; 3 3( 6 ) : 79 9 – 80 5 .
26.  S u s h am e s A ,  E dw a r ds  A , T h o m ps o n F ,  M cD e r mo t t  R,  G e be l  K .  Va l i d i t y a n d  Re l i a b il i t y
ofFitbitFlexforStepCou n t ,  M o d e ra t e  to  V i g o ro u s  Ph y s i c al  A c t i v it y  a n d A c t i v it y
EnergyExpenditure.Plo S  O n e .  2 01 6 ;1 1 ( 9) : e 0 16 1 2 24 .
27.  S t e l l i n o M B , S i n c l a ir  C D .  Ps y c h ol o g i c a l P r e d ic t o r s o f  C h il d r e n ’ s R e c es s  P hy s i c a l
ActivityMotivationandBe h a v i o r .  R es  Q E x e r c S p o rt .  2 0 13  J u n 1 ; 8 4( 2 ) : 16 7 – 76 .
28.  P r e a ch e r  K J,  C u r r an  P J , B a u e r D J .  C om p ut a t i o na l  T o ol s  f o r P r o b in g  I n t e ra c t i o ns  i n

22
MultipleLinearRegress io n ,  M u l t i l e v el  M o de l i n g , a n d L a t e n t C u r v e A n a ly s i s .  J E d u c
BehavStat.2006De c  1 ; 3 1 ( 4 ): 4 3 7– 4 8 .
29.  P a r r i s h A - M , Y e a tm a n H ,  I v e rs o n D ,  R u ss e l l  K . U s i n g i n t e r vi e w s a n d  pe e r  pa i r s  to
betterunderstandhow  s c h o o l  e nv i r o n me n ts  a f f ec t  y o un g  c hi l d r e n ’s  p l a yg r o u nd
physicalactivitylevels:aq u a l i t a t i v e  s tu d y . H e a lt h  E d u c R e s . 2 0 1 2 A p r; 2 7 (2 ) : 2 6 9– 8 0.
30.  S t a n l e y R M , B o s ho f f  K , D o l l m an  J.  V o i c e s i n  t h e p l a y gr o u n d:  A  q ua l i t a ti v e  e x pl o r a ti o n
ofthebarriersandfac i l i t a t o r s  o f  l u nc h t i m e p l a y.  J  S ci  M e d S p o rt .  2 0 12  J a n
1;15(1):44–51.
31.  B u n d y A ,  E ng e l e n L ,  W yv e r S ,  T r an t e r P ,  R a ge n  J , B a u m an  A,  e t  a l.  S y d ne y  P la y g r o un d
Project:ACluster-Rando m i z e d  T r i a l t o  I n c re a s e P h y s ic a l  A c ti v i t y ,  P la y ,  a nd  S o ci a l
Skills.JSchHealth.2017  O c t  1 ; 8 7( 1 0 ): 7 5 1– 9 .
32.  M a y fi e l d  C A , C h i l d S ,  W ea v e r R G , Z a r r e tt  N ,  B ee t s M W , M o o re  J B .  Ef f e c ti v e n e ss  o f  a
PlaygroundIntervention f o r  A n t i s o ci a l ,  P r os o c i a l,  a n d P h y s ic a l  A c ti v i t y  B e ha v i o rs .  J
SchHealth.2017Ma y  1 ; 8 7 ( 5 ): 3 3 8– 4 5 .
33.  F a r m er  V L,  W i l l i a ms  S M,  M an n  JI ,  S c h o fi e l d  G , M c P he e  J C , T a y lo r  R W . T h e e f f e ct  o f
increasingriskandcha l l e n g e  i n  t h e s c h oo l  p l a yg r o un d  o n p h y si c a l  a ct i v i t y  a nd
weightinchildren:aclus t e r  r a n d o mi s e d c o n tr o l l e d t r i a l  ( P L AY ) .  I n t J  O b e s 2 0 05 .
2017;41(5):793–800.
34.  M c c l a in  J J ,  A b ra h a m T L , B r u ss e a u T A ,  T ud o r - Lo c k e C .  E p oc h  L en g t h a n d
AccelerometerOutputs  i n  C h i l d r e n : C o m pa r is o n  t o D i r e c t O b s er v a t i o n.  M ed  S c i S p o r ts
Exerc[Internet].2008 D e c  [ c i t e d 2 0 2 0 J a n 2 1 ] ; 40 ( 1 2) : 2 0 80 – 7.  A v a il a b l e  fr o m :
https://insights.ovid.com/ c r o s s r e f ? an = 00 0 0 57 6 8- 2 0 08 1 20 0 0- 0 0 0 10
35.  B a n d a J A ,  Ha y d el  K F , D a v i l a T ,  D e sa i  M , B r y so n  S , H a s ke l l  W L,  e t  a l.  E f f e ct s  o f
VaryingEpochLengths ,  W e a r  T im e  A lg o r i t h ms ,  a nd  A ct i v i t y  C ut - P o i n ts  o n  E st i m a te s
ofChildSedentaryBeh a v i o r  a n d P h y s ic a l  A c ti v i t y  f r o m A cc e l e r om e te r  D a ta .

23
PappalardoF,editor.P L O S  O N E [ I n t e r ne t ] .  2 01 6 M a r 3  [ c i t e d 2 0 20  J a n
21];11(3):e015053 4 .  A v a i l a b le  f r o m:
https://dx.plos.org/10.1 3 7 1 / j o u rn a l . p on e . 0 15 0 53 4
Figures
Figure1
GenderxAdultengagementandsupervisioninteractionforpercentageoftimeinMVPA
duringrecess.

24
Figure2
GenderxStudentbehaviorinteractionforpercentageoftimeinMVPAduringrecess.
Figure3
GenderxStudentbehaviorinteractionforpercentageoftimeinlightphysicalactivityduring
recess.

25
Figure4
Differencesingirls'engagementinplayathigh-versuslow-qualityrecesssessions
Figure5
Differencesinboys’engagementinplayathigh-versuslow-qualityrecesssessions
SupplementaryFiles
Thisisalistofsupplementaryfilesassociatedwiththispreprint.Clicktodownload.

26
PsychNeedSatisfactionQuestionnaire_Short.docx