Development and validation of a model of motor competence in children and adolescents

(1)

ContentslistsavailableatScienceDirect

Journal

of

Science

and

Medicine

in

Sport

j o u r n a l ho me p ag e :w w w . e l s e v i e r . c o m / l o c a t e / j s a m s

Original

research

Development

and

validation

of

a

model

of

motor

competence

in

children

and

adolescents

Carlos

Luz

a,b,∗

_,

_Luis

_P.

_Rodrigues

d,e

_,

_Gabriela

_Almeida

c

_,

_Rita

_Cordovil

f a_Laboratory_of_Motor_Behavior,_Faculdade_de_Motricidade_Humana,_Universidade_de_Lisboa,_Portugal

b_Escola_Superior_de_Educac¸_ão_de_Lisboa,_Instituto_Politécnico_de_Lisboa,_Portugal c_Faculdade_de_Ciências_da_Saúde,_Universidade_Fernando_Pessoa,_Portugal

d_Escola_Superior_de_Desporto_e_Lazer,_Instituto_Politécnico_de_Viana_do_Castelo,_Portugal e_CIDESD,_Portugal

f_Laboratory_of_Motor_Behavior,_CIPER,_Faculdade_de_Motricidade_Humana,_Universidade_de_Lisboa,_Portugal

a

r

t

i

c

l

e

i

n

f

o

Articlehistory:

Received23December2014 Receivedinrevisedform15June2015 Accepted2July2015

Availableonline10July2015 Keywords: Motorcompetence Physicaleducation Quantitativeinstruments Locomotor Manipulative Stability

a

b

s

t

r

a

c

t

Objectives:Thisstudywasaimedatdevelopingaquantitativemodeltoevaluatemotorcompetence(MC) inchildrenandadolescents,tobeapplicableinresearch,education,andclinicalcontexts.

Design:Cross-sectional.

Methods:Atotalof584children(boysn=300)withagesbetween6and14yearswereassessedusing ninewellknownquantitativemotortasks,dividedintothreemajorcomponents(stability,locomotor andmanipulative).StructuralequationmodellingthroughEQS6.1wasusedtofindthebestmodelfor representingthestructuralandmeasurementvalidityofMC.

Results:ThefinalMCmodelwascomposedbythreelatentfactorscloselyrelatedwitheachother.Each factorwasbestrepresentedbytwooftheinitialthreemotortaskschosen.Themodelwasshowntogive averygoodoverallfit(!2₌_12.04,_p₌_.061;_NFI₌_.982;_CFI₌_.991;_RMSEA₌_.059).

Conclusions:MCcanbeparsimoniouslyrepresentedbysixquantitativemotortasks,groupedintothree interrelatedfactors.Thedevelopedmodelwasshowntoberobustwhenappliedtodifferentsamples, demonstratingagoodstructuralandmeasurementreliability.Theuseofaquantitativeprotocolwith few,simpletoadministerandwellknown,motortasks,isanimportantadvantageofthismodel,since itcanbeusedinseveralcontextswithdifferentobjectives.Wefinditespeciallybeneficialforphysical educationsteacherswhohavetoregularlyassesstheirstudents.

1. Introduction

Inthelasttwodecadesagrowingbody ofevidencesuggests

thatearlyMotorCompetence(MC)isofparamountimportancefor

developinganactiveandhealthylifestyle.1_MC_is_used_as_a_global

termtodescribeaperson’sabilitytobeproficientonawiderange

ofmotoractsorskills.2_This_ability_has_been_described_in_the

lit-eraturealsoasmotorcoordination,motorperformance,ormotor

proficiency.Intheinitialphasesofmotordevelopment,children’s

MCinvolvesthemasteryoffundamentalmotorskillsthatarethe

foundationsforthemasteryofspecializedmotorskills.Ithasbeen

reportedthatphysicalactivity,3,4_{cardiorespiratory}_fitness,5,6

phys-icalfitness,7_and_perceived_physical_competence,8,9_have_positive

effectsandassociationswithMC,aswellasaninverseassociation

∗ Correspondingauthor.

E-mailaddress:carlosmiguelluz@gmail.com(C.Luz).

withweightstatus10_in_children_and_adolescents._This_has_provided

emergingevidencetothetheoreticalmodelproposedbyStodden

andcolleagues.1

MCasatheoreticalconstructis consideredtobesubdivided

intolocomotor(e.g.,leaping,gallopingorverticaljump),stability

(e.g.,dynamicand staticbalance)and manipulative(e.g.,

catch-ing,throwingandkicking)11_proficiency._However,_this_structure

isnotalwaysreflectedinresearchand/orclinicalsettingswhere

MCconstitutesthesubjectofinterest.Severalstandardizedtests

(e.g.,TGMD, KTK),and a number of different non-standardized

protocols,12,13_found_in_the_literature,_are_deemed_to_evaluate_MC

butdonotfollowthetheoreticalMCconstruct.Forexample,the

TGMDdoesnotevaluatestabilityandtheKTKdoesnotevaluate

manipulativeproficiency.Furthermore,mostinstrumentsand

pro-tocolsarerestrictedtoaspecificage,ornarrowage-range,either

duetothedevelopmentalrestrictedagewindowofthemotortasks,

ortothenatureoftheusedscoringprocedures(quantitativeor

qualitative).

http://dx.doi.org/10.1016/j.jsams.2015.07.005

(2)

C.Luzetal./JournalofScienceandMedicineinSport19(2016)568–572

Thisgreatdiscrepancybetweentheacceptedtheoretical

con-structofMCanditsapplicationinresearchand/orclinicalsettings

showsthelackofarobustconceptualandworkingmodelofMC

thatcouldbesuccessfullyusedindifferentsettingsand

develop-mentalages.Toourknowledge,nostudieshavebeenpresented

thatvalidatethetheoreticalMCmodelstructureusingthe

origi-nalthreecategories.Somestudieshaveusedstructuralequation

modellingorconfirmatoryfactoranalysistechniquestolookfor

thestructuralvalidityofinstruments(e.g.,M-ABC,14_TGMD15₎_but

theinstrumentsthemselveswerenotinfullagreementwiththe

theoreticalMCmodel.

Themain objectiveof thisstudy wastoestablish aworking

developmentalmodel ofMC,based onthree domains

(locomo-tor,stability,andmanipulative)ofthetheoreticalconstructofMC.

Wehypothesizedthateachofthesethreecategoriesisrepresented

byageindependentsignificantmotortasksthatcanbeobjectively

measured(product).

Toachievethispurposewehaveassessedchildrenusingseveral

motortasksrepresentativeofeachMCcategoryintheliterature,

andworkedwiththedatatofindarepresentativeandparsimonious

modelofMCusingspecificStructuralEquationsModelling(SEM)

techniques.

2. Methods

Atotalof584children(300males),aged6to14years(M=10.60,

SD=2.40), participated in this study. Children were randomly

selectedfrompublicPortugueseschoolsandhadnoknown

learn-ing disabilitiesor pre-existing motor limitations. A localethics

committeeapprovalwasobtainedandparentsprovidedwritten

informedconsent.TwoPhysicalEducation(PE)teacherswith10

years of experiencewere trained to collectthedata in regular

scheduledclasses(eachteacheralwaysassessedthesamegroup

oftasks).

Three tests for each MC category (stability, locomotor, and

manipulative)11_were_selected_from_the_most_used_protocols_and

instrumentsinthemotordevelopmentliterature.Inclusioncriteria

werebeingquantitative(product-oriented)motortestswithouta

markeddevelopmental(age)ceilingeffect,andoffeasible

execu-tion.

Stabilitytestswere:(a)balancebeams16_—walking_backwards

onthreebalancebeamswith3minlengthbutofdecreasingwidths

(6,4.5and3cm).Eachparticipanthadthreeattemptsperbeam

andeachattempthadthemaximumscoreof8points.Thetotal

scorewasgivenbythesumofpointsinthethreebalancebeams

(72totalpossiblepoints);(b)Shiftingplatforms16_—moving

side-waysfor20susingtwowoodenplatforms(25cm×25cm×2cm).

Eachsuccessfultransferfromoneplatformtotheotherwasscored

withtwopoints(onepointforeachstep).Participantsweregiven

two trials and only the best score was considered; (c)

jump-inglaterally16_—jumping_sideways_with_two_feet_together_over_a

woodenbeamasfastaspossiblefor15s.Eachcorrectjumpscored

1pointandthebestresultovertwotrialswasconsidered.

Locomo-tortestswere:(a)hoppingononelegoveranobstacle16_—jumping

overastackoffoamblocks5cmhighwithonefoot,reachingthe

floorwith thesame foot. Aftera successful attemptwith each

foot,the heightwas increasedby adding one foam block.

Par-ticipantsreceivedthree,twoor onepoint(s)foreachsuccessful

performanceonthefirst,secondorthirdtrial,respectively.

There-fore,eachchild hadthree attempts ateachheightand oneach

foot.Thetestingwasstoppedwhenaheighttrialwasnot

success-fullycompletedwithbothfeet.Thetotalscorewasgivenbythe

sumofpointsatalltheheights;(b)shuttlerun(SHR)17_—running

atmaximalspeedtoalineplaced10mapart,pickingupablock

ofwood,runningbackand placingit onor beyondthestarting

line.Thenrunningbacktoretrievethesecondblockandcarryit

backacrossthefinishline. Thefinalscorewasthebesttimeof

thetwotrials;(c)standinglongjump(SLJ)18_—jumping_with_both

feetsimultaneouslyasfaraspossible.Thefinalscore(thebetter

of2attempts)wasthedistance(inm)betweenthestartingline

andthebackoftheheelatlandingclosesttotheline.

Manipula-tivetestswere:(a)walltosstest19_—throwing_a_tennis_ball_with

anoverarmactionagainstthewall(2mdistance),attemptingto

catchitwithbothhandsover30s.Thefinalscorewasgivenby

thebetterresult(numberofcatches)in2attempts;(b)throwing

velocity20_—throwing_a_baseball_{(circumference:}_22.86_cm;_weight:

142g) at a maximum speed against a wall using an overarm

action.Threetrialswereperformed,andthefinalscorewasgiven

by the best result; and (c) kickingvelocity20_—kicking _a _soccer

ballno.4 (circumference:64cm,weight: 350g) at amaximum

speedagainstawallusingakickingaction.Threetrialswere

per-formed,andthefinalscorewasgivenbythebestresult.Ballpeak

velocity wasmeasuredwitha Pro IIStalkerRadarGunin both

tests.

Participantscompletedageneralwarm-upbeforethebeginning

ofthetests.Then,groupsoffivestudentswereevaluatedinthe

sametaskorder.Participantsobservedademonstrationofthe

pro-ficienttechniqueandhadtheopportunitytoexperimentwitheach

taskonetimebeforetheirperformance.Motivationalfeedbackwas

given;howevernoverbalfeedbackonskillperformancewas

pro-vided.Inthethrowing/kickingtasks,childrenwereinstructedto

throw/kicktheballasfastastheycould.

Inordertoassesstheplausibilityandvalidityofourtheoretically

drivenMCmodel,weusedaspecialmulti-groupconfirmatory

fac-toranalysis,knownasstackmodel.21_In_the_first_two_steps_of_this

procedure, thefullsample(584 subjects)wasrandomlysplitin

halfmaintainingthesexproportionality.Thefirsthalf(292

sub-jects)wasusedasacalibrationsample(tosettheinitialbestmodel

toentailMCaccordingtothetheoreticalframework),andthe

sec-ondasavalidationsample,usedtoassurethatthepreviouschosen

model (factors andloadingitems)wasable toreproduceevery

otherdata.

Onthethirdphase(crossvalidity)weformallytestedfor

mea-sureandstructuralinvariancebetweenthetwosplithalves.Totest

formeasureinvariance,theformalstructurefromthecalibration

samplewasimposedonthevalidationsamplewhileallparameters

wereleftfree.Usingamorerestrictedapproach(tightcross

vali-dation),structuralinvariancewasalsoimposedtothevalidation

sample,withallparametersconstrainedtothecalibrationmodel

values.

Theabsolutefitofthemodels(individualandmulti-group

anal-ysis)wasevaluatedusingtheSatorraandBentlerscaledchi-square

(!2₎₍₁₉₉₄₎_with_correction_for_{non-normality,}_while_the_relative_fit

wasassessedusingthecomparativefitindex(CFI),thenormedfit

index(NFI),andthegoodnessoffitindex(GFI).Fortheseindices,

valuesover.95andupto1.0aredeemedindicativeofagoodfit.

The root mean square error of approximation (RMSEA) and

respective confidence intervals (CI) were used for evaluating

howwellthemodel-impliedreproducedthevariance-covariance

matrixofthedata,keepinginmindthatRMSEAvaluesaslowas

.06representagoodfittothemodel.21–23_EQS_Lagrange_Multiplier

Tests(LMT)foraddinganddeletingparameterswereinterpreted

withinthetheoreticalframeworkforeachmodeltestedinthe

cal-ibrationphase,andalterationsmadeaccordingly.Variableswere

considered for deletion when LMT suggested that such

proce-dureresultedinasignificantimprovementofthemodelfit.Each

consecutivemodelwascomparedwiththeprevioususingthe

chi-squareanddegreesoffreedomchange,andwasonlyretainedwhen

thiscomparisonshowedstatisticalsignificance.Allanalyseswere

conductedusingtheEQS6.1computerprogram(Multivariate

(3)

C.Luzetal./JournalofScienceandMedicineinSport19(2016)568–572 Table1

Indicesforevaluatinggoodnessoffitofmodelsindifferentphases.

!2 _p _NFI _CFI _RMSEA

Phase1—Calibrationsample(n=292)

Model1(9variables) 159.39(24df) .000 .881 .896 .139

Phase2—Validationsample(n=292)

Phase3—Multi-groupanalysis(n=584)

Measurementinvariance 24.20(12df) .019 .984 .992 .042

Structuralinvariance 30.44(15df) .010 .980 .990 .042

3. Results

Inthecalibrationphase(phase1)theinitialformulationofthe

MCmodelwassetaccordingtothetheoreticalformulationwith

threefactors(stability,locomotor,and manipulative),and three

possibleitems (motorskill tasks)accounting (loading)oneach

factor.Departingfromthis theoreticalmodel,weexaminedthe

solutionaccordingtothesignificanceoftheloadingcoefficients,

R2_values_of_the_variables_equations,_and_indices_of_overall_and

rel-ativefit(!2_,_CFI,_NFI_and_RMSEA)._EQS_Lagrange_Multiplier_tests_for

addinganddeletingparameterswereusedtoimprovethemodel

fit,accordingtotheoreticalinterpretation.Asaresultofthis

proce-dure,threevariables(hoppingononelegoveranobstacle,walking

onabalancebeam,andtossingandcatchingaball)were

consecu-tivelydroppedfromtheoriginalmodel,resultinginafinalmodelof

MCwithsixmotortasks(jumpinglaterally,shiftingplatforms,SHR,

SLJ,throwingvelocity,kickingvelocity)andthreecorrelated

fac-torsshowingaverygoodoverallfit(!2₌_12.04,_p₌_.061;_NFI₌_.982;

CFI=.991;RMSEA=.059;CI(RMSEA)=(000–.106).Thisfinal

stan-dardizedsolutionisshowninFig.1,andthefitindexesvaluesfor

thefourconsecutivemodelstestedcanbeseeninTable1.

Inthesecondstep(validationphase),datafromthesecondhalf

ofthesamplewastestedusingthefinalspecifiedmodelfromthe

calibrationsample.Overallindicesshowedaverygoodadjustment

ofthismodeltothedata(seeTable1),similartotheonefoundin

thevalidationsample.

Inthethirdstep(crossvalidationphase),inordertotestfor

thecrossvalidityofthemodelweformallytestedformeasureand

structuralinvariancebetweenthetwosplithalves.Totestfor

mea-sureinvariance,theformalstructurefromthecalibrationmodel4

samplewasimposedonthevalidationsamplewhileallparameters

wereleftfree.Indices(!2₌_24.20,_p₌_.019;_NFI₌_.984;_CFI₌_.993;

RMSEA=.059)fortheoverallfitofthis multi-groupmodelwere

good(seeTable1).Usingamorerestrictedapproach(tightcross

validation),structural invariance wasalsoimposed to the

vali-dationsample,withallparametersconstrainedtothecalibration

model4loadingvalues.Finalresultscontinuedtoshowagood

over-allfit(!2₌_30.44,_p₌_.010;_NFI₌_.980;_CFI₌_.990;_RMSEA₌_.042),_and

theformaltestingfordifferencesbetweentheimposed

parame-ter’svaluesshowednosignificantvalues.So,thesolutionfoundfor

thecalibrationsample(model4)showedaverygoodadjustment

totheotherhalfofthedata,provingitsvalidityforinterpretingthe

MCmodel.

4. Discussion

TheaimofthisstudywastoestablishamodelofMC,based

onatheoreticallystructuredividedintolocomotor,stability,and

manipulative domains. In this endeavor, quantitative

(product-oriented) motor test protocols without a developmental (age)

ceilingeffect,andoffeasibleexecution,wereused.Ourpurpose

when selecting only product-oriented tests was to ensure an

objectiveevaluationandagoodsensitivitytodiscriminateamong

competencelevelsacrossages.20

TheuseofSEMfortestingthisspecificmodelisofgreatutility

sinceitallowstoworkfromthedataforreachingafinalsolutionfor

aMCstructurethatrepresentswellthecommunality(represented

bythecovariance)andtheuniquecharacteristics(non-explained

covariance)oftests(items) andcategories (factors).Theoverall

adjustmentindices,alongwiththeindividualcoefficientsforthe

pathsinvolved(factor-item;factor–factor)providesarationalefor

includingorexcludingeachitem(test),orfactor(category),orpath

(representativenessoftheteststomarka category),toa better

representationofthefullmodel.

Inthevalidationphaseourresultsconfirmedtheexistenceof

three latentfactorsrepresentingthe stability(shifting platform

andjumpinglaterally),locomotor(SHRandSLJ),and

manipula-tive(moving platformand jumping laterally)categories of MC,

eachonebestrepresentedbytwooftheinitialthreemotortasks

chosen.Thesethreefactorsshowtoreproduceverywellthree

dis-tinctiveaspectsofMC,asprovedbytheinexistenceofanychange

JUMP.

Stabil

ity

SHIFT. SLJ

Locomotor

SHR KICK

Man

ipu

lative

THROW .86 .84 .65 .88 .91 .86 .83 .94 .64

Fig.1.Pathdiagramofthemodelformotorcompetencewithcompletely standard-izedvaluesforcoefficientsandcovariances.

(4)

C.Luzetal./JournalofScienceandMedicineinSport19(2016)568–572

suggestedinthefactor-itemstructurebythemodificationindices.

In addition,and in accordance withthetheoretical framework,

thesethreecategories(factors)provedtobecloselyrelatedwith

eachother.Overall,this modelpresenteda verygoodfit tothe

data,21–23_suggesting_that_it_can_be_used_to_represent_(and_assess)

MC.

Inthesecondphase,thereplicationoftheinitiallyfoundmodel

structureresultedalsoinaverygoodfittotheotherhalfsample

data(calibrationsample),indicatingagoodreliabilityofthemodel

toreproduceMCdata.Inthelaststepweformallytestedfor

mea-sureandstructuralinvariancebetweenthetwohalf-splitsamples,

inordertocross-validateformeasurementandstructural

invari-ance.Inbothcases,thetestedmodelshowedaveryadequatefit,

concludingforitsoverallvalidityforinterpretingMCinchildren

andadolescents.

Therefore,ourresultspostulatethatMCcanbeadvantageously

representedbylocomotor(SHRandSLJ),stability(movingplatform

andjumpinglaterally),andmanipulative(kickandthrowvelocity)

categoriesofmovementskills,andthatthelatentessenceofeachof

thesecategoriescanbeobjectivelymeasuredbytwoquantitative

motortasks.Thismodelpresentsseveraladvantagesforresearch,

education,andclinicalsettings.

Thefirstadvantageistheuseofasetofmotortaskswidelyused

inpastresearchsettingsasrepresentativeofMCcategories.11,16,20

Thesecondadvantageistheparsimonyofthemodel.Unlikeother

models’protocolsthatuseseveralmotortasks,suchastheTGMD24

orM-ABC25_or_even_some_{non-standardized}_protocols,13_our_final

modelisonlycomprisedbysixfeasibletests.Thethirdadvantage

isthatthismodelusesobjective(quantitative)measures.

Quali-tativemethodsarefocusedontheprocess,providinginsightinto

theformorcharacteristicsofthemovement;therefore,requiring

agreater knowledgeofthemovementcomponentsand usually

requirealotoftimetoanalyzethedata.Quantitativeapproaches

arefocusedonlyonthefinalproductandenableafaster

assess-mentoftheperformanceoutcomewithahighlevelofreliability

over time.26 _These_methods_are _sensitive_{discriminators}_among

competencelevelsacrosschildhoodandearlyadolescence,20_and

arecorrelatedwithqualitativeprocess-orientedassessmentsofthe

skills.27–29_Moreover,_quantitative_methods_also_do_not_require_a

highlevelofexpertiseandtrainingoftheevaluators,asusually

recommendedinqualitativemethods,13_since_the_lack_of

subjec-tivityinherenttothequantitativeapproachespermitsthateven

lessexperiencedobserverscanapplyit.Theentireprotocoltakes

about10minperparticipant;howeverchildrencanbegroupedin

smallgroups,reducing theaveragetimeneededforassessment.

Furthermore,the resultsinformation can beimmediatelyused,

makingitahugeadvantagefortheuseinPEclasses,andsports’

environments.Thefourthadvantageisthatthemotortasksused

donothaveaceilingeffectoverdevelopmentalyears,andsothe

samemodel and protocolcan beusedfrom childhoodtoadult

years. Thefifth advantageis thatthe model,giving the

magni-tudeofthecorrelationsbetweenfactors,suggeststhepossibilityto

obtainaglobalcompositescoreofMC,inadditiontothecategories’

scores.

This model is representative of MC and can be used by

researchers,PEteachers, andhealth and sports training

profes-sionals,inordertoobjectivelymonitormotordevelopment.This

MCmodelseemspromising,butfurtherresearchiswarrantedto

replicatethecurrentresults.

This study has some limitations. The results confirmed the

agreementofsixofthemwiththetestedmodel,nevertheless,and

inordertoachieveamoreaccuraterepresentationofMC,abroader

rangeofmotortestscouldbewarrantedinnextstudies.In

addi-tion,inordertobeconsistentwiththenumberoftrialsforeach

skill,theuseofthreetrialscouldbemoreappropriatetoselectthe

bestscore.

Itisalsoimportanttonotethatoursamplehadchildrenfrom6

to14yearsoldandtheresultsmightevenhaveabetteradjustment

forseparategroupsofageandgender.Futureinvestigationsshould

takeintoconsiderationageandgender.

5. Conclusion

OurresultssupporttheideathatMCcanbeevaluatedthrougha

protocolwithsixmotortasksthatrepresentthethreemajorlatent

variables ofMC(i.e.,stability,locomotorandmanipulative). We

suggestthattheuseofa quantitativeapproachwithfewmotor

taskswithoutacellingeffect,whicharerepresentativeofthemajor

MCcomponents,isagoodalternativetotheexistingtesting

proto-cols.Becausethetestedmotortasksareeasytoassess,PEteachers

oreventrainedclassroomteacherscanusethismodelregularlyin

theirpracticesandevaluations.

Practicalimplications

• Briefandeasytoadministerevaluationmodelrepresentativeof

MC,whichcanbeusedbyseveralprofessionalstoobjectively

monitorMCinseveralcontexts.

• TheteachingofmotorskillsshouldbeintegratedintothePE

cur-riculumactivities,andteacherscouldusethismodelprotocolto

assesschildren’sMC.

• Thisregularassessment canhelpteacherstodevelopthebest

approachandexercisestoimprovetheirstudent’sMC.

Acknowledgments

Wewouldliketothanktheschools,childrenandparentsfor

their participation in this study. We also thank everyone who

helpedwiththedatacollection.

References

1.Stodden David F, Goodway Jacqueline D, Langendorfer Stephen J et al. A developmental perspective on the role of motor skill competence in physical activity: an emergent relationship. Quest 2008; 60(2):290–306.

http://dx.doi.org/10.1080/00336297.2008.10483582.

2.Fransen Job, D’Hondt Eva, Bourgois Jan et al. Motor competence assess-mentinchildren:convergentanddiscriminantvaliditybetweentheBOT-2 ShortFormandKTKtestingbatteries.ResDevDisabil2014;35(6):1375–1383.

http://dx.doi.org/10.1016/j.ridd.2014.03.011.

3.LopesVP,RodriguesLP,MaiaJARetal.Motorcoordinationaspredictorof physicalactivityinchildhood.ScandJMedSciSports2011;21(5):663–669.

http://dx.doi.org/10.1111/j.1600-0838.2009.01027.x.

4.Holfelder Benjamin, Schott Nadja. Relationship of fundamental movement skills and physical activity in children and adoles-cents: a systematic review. Psychol Sport Exerc 2014; 15:382–391.

http://dx.doi.org/10.1016/j.psychsport.2014.03.005.

5.HagaMonika.Physicalfitnessinchildrenwithhighmotorcompetenceis dif-ferentfromthatinchildrenwithlowmotorcompetence.PhysTher2009; 89(10):1089–1097.http://dx.doi.org/10.2522/ptj.20090052.

6.VandendriesscheJoricB, VandorpeBarbara,Coelho-e-SilvaManuelJetal. Multivariateassociationamongmorphology,fitness,andmotorcoordination characteristicsinboysage7to11.PediatrExercSci2011;23(4):504–520.

7.Hands B, Larkin D, Parker H et al. The relationship among physical activity, motor competence and health-related fitness in 14-year-old adolescents. Scand J Med Sci Sports 2009; 19(5):655–663.

http://dx.doi.org/10.1111/j.1600-0838.2008.00847.x.

8.BarnettLisaM,MorganPhilipJ,BeurdenEricetal.Perceivedsportscompetence mediatestherelationshipbetweenchildhoodmotorskillproficiencyand ado-lescentphysicalactivityandfitness:alongitudinalassessment.IntJBehavNutr PhysAct2008;5:40.http://dx.doi.org/10.1186/1479-5868-5-40.

9.BarnettLisaM,MorganPhilipJ,VanBeurdenEricetal.Areversepathway? Actualandperceivedskillproficiencyandphysicalactivity.MedSciSportsExerc 2011;43:898–904.http://dx.doi.org/10.1249/MSS.0b013e3181fdfadd. 10.LopesVítorP,StoddenDavidF,BianchiMafaldaMetal.Correlationbetween

BMIandmotorcoordinationinchildren.JSciMedSport2012;15(1):38–43.

(5)

C.Luzetal./JournalofScienceandMedicineinSport19(2016)568–572 11.GallahueDavid,OzmunJohn,GoodwayJackie.Understandingmotor

develop-ment:infants,children,adolescents,adults,7thedNewYork,NY,McGraw-Hill, 2012.

12.Hume Clare, Okely Anthony, Bagley Sarah et al. Does weight sta-tus influence associations between children’s fundamental movement skills and physical activity? Res Q Exerc Sport 2008; 79(2):158–165.

http://dx.doi.org/10.1080/02701367.2008.10599479.

13.OkelyAD,BoothML.Masteryoffundamentalmovementskillsamongchildren inNewSouthWales:prevalenceandsociodemographicdistribution.JSciMed Sport2004:358–372.

14.Schulz Joerg, Henderson Sheila E, Sugden David A et al. Structural validity of the Movement ABC-2 test: factor structure comparisons across three age groups. Res Dev Disabil 2011; 32(4):1361–1369.

http://dx.doi.org/10.1016/j.ridd.2011.01.032.

15.KimSeonjin,KimMinJoo,ValentiniNadiaCetal.Validityandreliabilityof theTGMD-2forSouthKoreanchildren.JMotBehav2014;46(5):351–356.

http://dx.doi.org/10.1080/00222895.2014.914886.

16.ErnstKiphard,SchillingF.KörperkoordinationstestfürKinder:KTKManual, Göt-tingen,BeltzTest,2007.

17.Vicente-Rodríguez Germán, Rey-López Juan P, Ruíz Jonathan R et al. Interrater reliability and time measurement validity of speed-agility field tests in adolescents. J Strength Cond Res 2011; 25(7):2059–2063.

http://dx.doi.org/10.1519/JSC.0b013e3181e742fe.

18.AraIgnacio,MorenoLuis,LeivaMariaTetal.Adiposity,physicalactivity,and physicalfitnessamongchildrenfromAragón,Spain.Obesity(SilverSpring)2007; 15(8):1918–1924.http://dx.doi.org/10.1038/oby.2007.228.

19.BeashelPaul.AQASportExamined,Cheltenham,NelsonThornes,2004.

20.StoddenDavidF,GaoZ,LangendorferSJetal.Dynamicrelationshipsbetween motorskillcompetenceandhealth-relatedfitnessinyouth.PediatrExercSci 2014;26(3):231–241.http://dx.doi.org/10.1123/pes.2013-0027.

21.ByrneBarbara.StructuralEquationModelingwithEQS:BasicConcepts, Applica-tions,andProgramming,Mahwah,NJ,LawrenceErlbaumAssociates,2006.

22.BentlerPM.EQS6StructuralEquationsProgramManual,Encino,CA,Multivariate Software,Inc,2006.

23.RaykovTenko,MarcoulidesGeorgeA.AFirstCourseinStructuralEquation Mod-eling,Mahwah,NJ,L.ErlbaumAssociates,2006.

24.UlrichDA.TestofGrossMotorDevelopment,Austin,TX,Pro-ED.Inc,1985.

25.HendersonSE,SugdenDA.MovementAssessmentBatteryforChildrenManual, London,ThePsychologicalCorporationLtd,1992.

26.SprayJudithA.Recentdevelopmentsinmeasurementandpossible applica-tionstothemeasurementofpsychomotorbehavior.ResQExercSport1987; 58(3):203–209.http://dx.doi.org/10.1080/02701367.1987.10605449. 27.MallyKristiK,BattistaRebeccaA,RobertonMaryAnn.Distanceasa

con-trol parameter for place kicking. J Hum Sport Exerc 2011;6(1):122–134.

http://dx.doi.org/10.4100/jhse.2011.61.14.

28.Stodden David F, Langendorfer Stephen J, Fleisig Glenn S et al. Kine-matic constraints associated with the acquisition of overarm throwing, part II:Upper extremity actions.Res QExerc Sport 2006; 77(4):428–436.

http://dx.doi.org/10.5641/027013606X13080770015166.

29.Stodden David F, Langendorfer Stephen J, Fleisig GS et al. Kinematic constraints associated with the acquisition of overarm throwing, part I: Upper extremity actions. Res Q Exerc Sport 2006; 77(4):417–427.