Effects of home environment and center-based child care quality on children’s language, communication, and literacy outcomes

(1)

ContentslistsavailableatSciVerseScienceDirect

Early

Childhood

Research

Quarterly

Effects

of

home

environment

and

center-based

child

care

quality

on

children’s

language,

communication,

and

literacy

outcomes

夽

Ana

Isabel

Pinto

a,∗

_,

_Manuela

_Pessanha

b

_,

_Cecília

_Aguiar

c,1

a_University_of_Porto,_Department_of_Psychology,_Portugal b_Polytechnic_Institute_of_Porto,_College_of_Education,_Portugal c_UIPCDE,_ISPA_{– University}_Institute,_Portugal

a

r

t

i

c

l

e

i

n

f

o

Articlehistory: Received2August2011

Receivedinrevisedform17June2012 Accepted1July2012

Keywords: Childcarequality Homeenvironmentquality

Languageandcommunicationdevelopment Earlyliteracy

a

b

s

t

r

a

c

t

Thisstudyexaminedthejointeffectsofhomeenvironmentandcenter-basedchildcarequalityon children’slanguage,communication,andearlyliteracydevelopment,whilealsoconsideringprior devel-opmentallevel.Participantswere95children(46boys),assessedastoddlers(meanage=26.33months; Time1)andpreschoolers(meanage=68.71months;Time2)andtheirfamilies.Atbothtimes,children attendedcenter-basedchildcareclassroomsinthemetropolitanareaofPorto,Portugal.Resultsfrom hierarchicallinearmodelsindicatedthathomeenvironmentandpreschoolquality,butnotcenter-based toddlerchildcarequality,wereassociatedwithchildren’slanguageandliteracyoutcomesatTime2. Moreover,thequalityofpreschoolclassroomsmoderatedtheassociationbetweenhomeenvironment qualityandchildren’slanguageandearlyliteracyskills–butnotcommunicationskills–atTime2, sug-gestingthepositivecumulativeeffectsofhomeenvironmentandpreschoolquality.Findingsfurther supporttheexistenceofadetrimentaleffectoflowpreschoolqualityonchildren’slanguageandearly literacyoutcomes:positiveassociationsamonghomeenvironmentqualityandchildren’sdevelopmental outcomeswerefoundtoreducesubstantiallywhenchildrenattendedlow-qualitypreschoolclassrooms. © 2012 Elsevier Inc. All rights reserved.

1. Effectsofhomeenvironmentandcenter-basedchild carequalityonchildren’slanguage,communication,and literacyoutcomes

The demand for early childhood education and care (ECEC) outside the family context in Portugal has grown in response totheincrease in theproportion ofworking mothers.Portugal hasthehighest rateof mothersworkingfull time inthe Euro-pean Union. In 2005, 69.1% of mothers with children under 2years and 71.8% of mothers with children between 3 and 5 were employed (Organization for Economic Co-operation and Development[OECD],2007).

Overthepasttwodecades,substantialeffortshavebeenmade bythePortuguese governmenttoincrease thecoveragerateof ECEC services.Sociopolitical changes that occurredin Portugal, mainlyfollowingthe1974revolution,increasedwomen’saccessto

夽 ThisworkwasfundedbyFCT–Fundac¸ãoparaaCiênciaeaTecnologia,through researchgrants“POCTI/PSI/35207/1999”and“POCI/PSI/58712/2004”.

∗ Correspondingauthorat:FaculdadedePsicologiaedeCiênciasdaEducac¸ãoda UniversidadedoPorto,RuaAlfredoAllen,4200-135Porto,Portugal.

Tel.:+351226079748.

E-mailaddress:ana@fpce.up.pt(A.I.Pinto).

1 _Now_at_ISCTE_–_University_Institute_of_Lisbon.

laborandrestartedthedebateonECEC.In1975,areportfromthe UnitedNationsEducational,Scientific,andCulturalOrganization (UNESCO),highlightedtheprecariousstateofpreschoolservices inPortugalandstatedtheneedtocreate12,000newclassrooms forECEC(UNESCO,1982).Thisreportinfluencedtheadoptionofa publiclawrecommendingtheimplementationofanetworkof pub-licpreschoolservices(Bairrão,Leal,Abreu-Lima,&Morgado,1997; Vasconcelos,Orey,Homem,&Cabral,2002).Until1981,officialdata oninfant-toddlerchildcareandpreschoolservicesinPortugalwere scarceandunreliable.Inthelasttwodecades,thecoverageratefor childrenbetween4monthsand3yearsofagehasincreased drasti-callyfrom5.8%in1984(Vasconcelosetal.,2002)to30.2%in2008 (GabinetedeEstratégiaePlaneamento,2009).Inthemainland,the coveragerateofeducationalservicesforchildrenbetween3and 6yearshasincreasedfrom32%in1984(Vasconcelosetal.,2002) to78.8%in2008(MinistériodaEducação,2010).However, state-fundedearlyeducationalservicesarestillinsufficient(Conselho NacionaldeEducação,2011).

Aspreviouslymentionedin aPortuguesestudybyPessanha, Aguiar,and Bairrão (2007), ECECservices for children between 4months and 3years of ageare all dependentonthe Ministry of Solidarity and Social Security and are set up either by the Ministryitselforbyotherprivateentities(e.g.,individuals, cooper-atives,ornon-proﬁtorganizations)(Bairrão,Barbosa,Borges,Cruz,

(2)

&Macedo-Pinto,1990;Vasconcelosetal.,2002).RegardingECEC servicesforchildrenfrom3to6yearsofage,publicandprivate preschoolservicesformanationalnetworkunderthe responsibil-ityoftheMinistryofEducation,whoseaimistoextendpreschool educationprovisiontoallchildrenofthisagegroup(Bairrãoetal., 1990;MinistériodaEducac¸ão,2000).

In recent years, Portugal has made notable progress in the formulation andimplementation ofECEC policywithincreased enrollmentrates,theexpansionofpubliclyfundedpreschool pro-grams, and recognition of the critical role of early educational experiences.AccordingtotheOECD,almost90%ofchildrenaged 5–6yearswereenrolledinECECin2006.Morerecently,the cen-tralgovernmenthasmandateduniversalpreschoolfor5-year-olds (MinistériodaEducac¸ão,2009).

Consideringtheresearchﬁndingsonthecompensatoryeffects of high-quality ECEC for disadvantaged children (see Peisner-Feinberg & Yazejian, 2010), it can be argued that the quality ofpreschool contextsisof crucial relevancein Portugal, where the risk of poverty is around 17.9% for the overall population andover20.6% forfamilieswithchildren(InstitutoNacionalde Estatística,2010).Moreover,maternaleducationislowwhen com-paredtootherEuropeanUnioncountries, with70%of mothers havingcompleted less than upper secondary education (OECD, 2008).

Overthepastdecades,theincreaseinthenumberofinfants, toddlers,andpreschoolersreceivingECEChasgeneratedan abun-danceofresearchontheeffects ofearlychildcare experiences onchildren’s developmental outcomes. A considerable number of studies have demonstrated the short and long-term effects ofdevelopmentallyappropriate,high-qualityECEConchildren’s cognitive, language, and social outcomes as well as on later schoolachievement(Burchinal,Roberts,Nabors,&Bryant,1996; Burchinal,Peisner-Feinberg,Pianta,&Howes,2002;Howesetal., 2008; Mashburn et al., 2008; Pessanha, 2008; Vandell, Belsky, Burchinal,Steinberg,&Vandergrift,2010).Forexample,the large-scaleNationalInstituteonChildHealthandHumanDevelopment [NICHD]childcarestudy(1997,1998a,1998b,2000,2006) inves-tigatedtheinteractions betweenchild, homeenvironment, and ECECcharacteristicstohelpexplainhowchildrendevelopedover time.Theresultsweregenerallyconsistentwithﬁndingsfromother studiesanddemonstratethatthequalityofboththehomeandECEC environmentsmatter.Whencareenvironmentsweremore stim-ulatingandwellorganized,childrenhadbettervocabulary,more advancedattentionandmemoryskills,andpeerrelationships. Fur-thermore,inthepreschoolyears,childrenwhospentmorehours incenter-basedchildcaredisplayedmoreadvancedlanguageand cognitiveskills.

Given previousfindings on thepositive effects of preschool experienceonprimaryschoolperformance(seeSammonsetal., 2002,2003),thefirstmajorEuropeanlongitudinalstudy– Effec-tive Provision of Pre-School Education (EPPE; Sylva, Melhuish, Sammons, Siraj-Blatchford, & Taggart, 2008) – was conducted to analyze the long-term influence of preschool on children’s performance in school. EPPE measuredpreschool quality using theEarlyChildhoodEnvironmentRatingScale-Revised(ECERS-R; Harms, Clifford, & Cryer, 1998) and the Early Childhood Envi-ronmentRatingScale-Extension(ECERS-E;Sylva,Siraj-Blatchford, & Taggart, 2003) with a sample of 3–7year old children. The studyidentifiedarelationshipbetweenhigh-qualitypreschooland higherintellectualandsocial/behavioraldevelopmentinchildren, withsubsequentpositiveoutcomesinmathematicsandreading. However, despite earlier findings suggesting that all preschool experiencewasbeneficial,EPPEparticipantswhoattended low-qualitypreschooldidnotshowcognitivebenefitsinprimaryschool, andtheiroutcomesdidnotsignificantlydifferfromthechildren whodidnotattendpreschool(Sylvaetal.,2008).

AccordingtoAhnertandLamb(2004),cognitivetheoristshave maintainedamorepositiveviewofECEC,emphasizingthepositive impactofwell-plannedlearningenvironmentsonthecognitiveand communicativedevelopmentofchildren.However,theseauthors underlinethat,ratherthanconsideringchildcareeffectsas univer-sal,theinﬂuenceofseveralfactors(e.g.,cultural,family,andchild characteristics)shouldbetakenintoaccountwhenanalyzingthe associationbetweenthequalityofearlychildhoodenvironments andchildren’sdevelopment.

AlthoughthereisagreatdealofevidencesuggestingthatECEC canhave a long-lastinginfluenceonchild outcomes,the home environmentiswidelyrecognizedasthemostimportantinfluence onchildren’sadjustment(e.g.,NICHD,1998b),withfamily char-acteristicssuchasmaternaleducationconsistently emergingas strongpredictorsofchildren’sacademicandcognitiveoutcomes (e.g.,Abreu-Lima,Leal,Cadima,&Gamelas,2012).However,the qualityofthehomelearningenvironmenthasbeenfoundtobe moreimportantforchildren’sintellectualandsocialdevelopment thanparentaloccupation,education,orincome(Sylvaetal.,2008). Whilesinglefactorshavebeenlookedatinanumberofstudies, evidenceregardingtheinteractioneffectsofhomeandECEC qual-ityonchildren’soutcomesisscarceandtheinconsistencyofwhatis availablesuggeststheneedforadditionalresearch.Forexample,in theNICHDchildcarestudy(1997,1998a,1998b,2000,2006), sig-nificantinteractioneffectsbetweenqualityofhomeenvironment andECECqualitywerefound, indicatingthatchildrenwereless likelytobesecurewhenlowmaternalsensitivity/responsiveness wascombinedwithpoor-qualitychildcare.Further,Watamura, Phillips,Morrissey,McCartney,andBub(2011)foundcross-context influencesonchildren’ssocial-emotionaloutcomes,reportingboth detrimentaleffectsofdoublejeopardy(i.e.,low-qualityhomeand ECECenvironments)andcompensatoryeffectsofattending high-qualityECECforchildrenfromlower-qualityhomeenvironments. Bradley,McKelvey,andWhiteside-Mansell(2011)reporteda pre-dominanceofadditiveeffectsofthehomeenvironmentandECEC. However, theauthors foundsmall moderating effects of home environment features ontheimpactof ECEC,withECEC some-times compensating for low levels of stimulation and warmth inthehomeenvironment,but rarelypotentiatingtheeffects of high-qualityhomeenvironments.Withafocusonearlyliteracy skills, Anders et al. (2011) found that children from medium-qualityhomeenvironmentsbenefitedfromattendinghigh-quality preschoolwhereaschildrenfromlow-qualityhomeenvironments didnot.Theauthorsacknowledgeadiversepatternofresults, pos-siblyspecifictoGermany.However,thisfurtherstheargumentfor continuedresearchontheinterplayofhomeandECECexperiences acrossdiversesamples.

Thisstudyaimstohelpﬁllthisgapbyanalyzingboththedirect andjointeffectsofhomeenvironmentandECECqualityina Por-tuguesesample.ConsideringthelowqualityofPortuguesetoddler center-basedchild care, reportedin previous studies(Barros & Aguiar,2010;Pessanhaetal.,2007),aswellasthespeciﬁccultural backgroundofthepresentstudy’sparticipants,resultsmayprovide additivevaluetounderstandingthecomplicatedandcontextual natureofyoungchildren’searlycareandeducationexperiences.

This paper is based on a data set from the first two data pointsof alongitudinalstudyonyoungchildren’sengagement. In thefirst phase of the study,predictors of child engagement anddevelopmentwereidentifiedin1–3year-olds,namelysetting characteristics(i.e.,homeenvironmentandcenter-basedtoddler childcare),adult’sinteractivebehaviors,andchildcharacteristics. Inthesecondphaseofthestudy,afollow-upofthesechildrenat ages4–6yearswascarriedoutinordertoanalyzetheinfluence ofengagement,asobservedintoddlerchildcareclassrooms,on children’slateradaptation.Bothdatapointsincludeda consider-ablenumber of measures,withsimilarinstrumentsutilized for

(3)

measuringprimary variables. For example,at both datapoints, childengagementwascodedusingthePortuguesetranslationof the Engagement Quality Observation System (McWilliam & de Kruif,1998;Pinto,Aguiar,Barros,&Cruz,2004);childdevelopment wasmeasuredusing the Portugueseadaptation of theGriffiths Mental Development Scales (Griffiths, 1984/1996, 1986/1996); childadaptivebehaviorwasmeasuredusingthePortuguese trans-lationandadaptationoftheage-appropriatesurveyformofthe AdaptiveBehaviorScales(Sparrow,Balla,&Cicchetti,1984/n.d.); homeenvironment quality wasmeasured withthe Portuguese translationoftheage-appropriateversionoftheHomeObservation forMeasurementoftheEnvironment(HOME;Caldwell&Bradley, 1984/1997);and center-basedclassroom qualitywas measured usingthePortugueseversionoftheInfantToddlerEnvironment RatingScale(ITERS;Harms,Cryer,&Clifford,1990/1996), inthe firstdatapoint,andthePortugueseversionoftheECERS-R(Harms etal.,1998/2008)inthesecond.Similarpsychometricproperties werefoundforthesemeasuresatbothtimes(seethemethods sec-tionforamoredetaileddescription).Resultsfromthefirstphaseof thisstudyshowedthathomogeneouslylowtoddlerchildcare class-roomqualitywasassociatedwithtimechildrenspentnonengaged aswellaswithlowerpercentagesoftoddlersengagedin activi-ties(Pinto,2006).Inaddition,highertoddlerchildcareclassroom qualitywasassociatedwithchildren’scommunicationcompetence andhomequalitywasmoderatelyassociatedwithdifferent devel-opmentalareas(i.e.,cognitive,social,andcommunicationskills) (Pessanha,2008).

Buildingontheliteraturementionedabove,weexamined asso-ciationsbetweenchildren’slanguage,communication,andliteracy skills during preschool and (a) the overall quality of center-basedtoddlerchildcare,(b)theoverallqualityofthepreschool environment,and(c) theoverall homeenvironment quality.All associationswereexaminedwhilealsoconsideringchildren’sprior developmentallevelaswellasmaternaleducation.Ourhypotheses werethefollowing:

1. Theoverallqualityofcenter-basedtoddlerchildcareispositively associatedwithchildren’slanguage,communication,andearly literacyskillsduringpreschool.

2.Theoverallhomeenvironmentqualityispositivelyassociated with children’s language, communication, and early literacy skillsduringpreschool.

3. Theoverallqualityofthepreschoolenvironmentispositively associatedwithchildren’slanguage,communication,andearly literacyskills during preschool,when alsoconsidering home environmentquality,maternaleducation,andchildren’s previ-ousdevelopmentallevel.

4.Theoverallqualityofthepreschoolenvironmenthasa mod-eratingeffectontherelationshipbetweenhomeenvironment qualityandchildren’slanguage,communication,andearly lit-eracyskillsduringpreschool,whenalsoconsideringmaternal education and children’s previous developmental level. It is expectedthathigherpreschoolqualitypotentiatesthepositive effectsofhomequality.

Resultsmayprovideevidencefortheneedtoimprovethe qual-ityofECECexperiences.ThisisespeciallyimportantinPortugal, wherepubliclysupportedECECisafastgrowingsector.

2. Methods

2.1. Participants

The dataused in this study was drawn froma longitudinal studyonchildengagement.Participantsincluded95children(46

boys)andtheirfamilieslivinginthemetropolitanareaofPorto. Toprovidesomecontext,Portoisthesecondlargestmetropolitan areainPortugal,withapproximately1,500,000inhabitants(Área MetropolitanadoPorto,n.d.).Fromtheinitialsampleof120 typ-icallydevelopingchildren(60boys),weretainedparticipantsfor whomtwodatapointswereavailable:onedatapointduringthe toddleryears(Time 1)and onedatapointduringthepreschool period(Time2).Proceduresfortheinitialsamplewerepreviously describedbyPessanhaetal.(2007):15privateinfant-toddlerchild carecenterswererandomlyselected,followedbyrandom selec-tionof4children(2boysand2girls)ineachofthe30participating toddlerchildcareclassrooms.

AtTime1,childrenwere14–40months(M=26.33,SD=6.51). Mothers’ageaveraged31.21years(SD=4.77),84%weremarried, and91%workedoutsidetheirhome.Maternaleducationranged from3to18years (M=10.64,SD=4.47),with29% ofthe moth-ershavingcompletedacollegedegree.Familyincomeaveraged 1,388.63Eurospermonth(SD=735.20).Childrenattended15 class-roomsfor1-year-oldsand15classroomsfor2-year-olds(from11 privatenon-proﬁtcentersand4privatefor-proﬁtcenters).Inthese toddlerclassroomsmeangroupsizewas16.00(SD=5.07)andmean adult-childratiowas1:8(SD=1:3).

AtTime2,children’sageaveraged68.71months(SD=5.91)and familyincome averaged1574.29Eurospermonth(SD=801.67). Childrenattended41preschoolclassroomsfrom29centers(13 pri-vatenon-proﬁt,6privatefor-proﬁt,5public,and5withnocenter typeinformationavailable)inthesamegeographicarea.Inthese preschoolclassroomsmeangroupsizewas22.50(SD=4.23)and meanadult-childratiowas1:13(SD=1:4).

Inthisstudy,atTime1,thenumberoftarget-childrenineachof thetoddlerclassroomsrangedbetween2and4(M=3.17,SD=0.79). AtTime2,thenumberoftarget-childrenperpreschoolclassroom rangedbetween1and7(M=2.32,SD=1.47).

2.2. Measuresandprocedures 2.2.1. Childoutcomes

Tomeasurechildren’sglobaldevelopment,bothatTimes1and 2,weusedaPortugueseadaptationoftheGriffithsMental Develop-mentScales(Griffiths,1984/1996,1986/1996).Thismeasurecovers birthto8yearsofageandincludesatotalof468itemsacrosssix subscalesordevelopmentaldomains:locomotor(i.e.,grossmotor skills),personal–social(i.e.,independenceandinteraction),hearing andspeech,eyeandhandcoordination(i.e.,finemotorskills), per-formance,andpracticalreasoning.Rawscoresmaybeadjustedfor children’schronologicalageandconvertedintoadevelopmental quotient.ThePortugueseversionusedinthisstudyhasbeenused inpreviousresearch(Cordazzo,Almeida,&Vieira,2011;Pereira etal.,2010).Inthis study,Cronbach’salphafortheglobalscale was.98atTime1and.97atTime2.Wealsousedthe develop-mentalquotientoftheHearingandSpeechsubscale,ascollectedat Time2(˛=.84),asameasureofchildren’sreceptiveandexpressive languageoutcomes.TheGriffithsscaleswereadministeredateach child’scenterbyresearchassistantswithaPsychologydegree.The Portuguesetranslationandadaptationofthesurveyformofthe VinelandAdaptiveBehaviorScales(Sparrowetal.,1984/n.d.)was usedtomeasurechildren’sdailyfunctioning.Althoughdatawere collectedonthecommunication,dailylivingskills,and socializa-tiondomainsatbothTimes1and2,asrequiredbyourhypotheses, onlydataoncommunicationskillsforTime2arepresentedhere. Thecommunication scaleusedin this studyincluded10 items, ratedona3-pointscale(0,neverperformsbehavior;1,sometimes orpartiallyperformsbehavior;2,usuallyperformsbehavior).The questionnairewasfilledoutbypreschoolteachersandresulting scoresonchildren’scommunicationskillswereinternally consis-tent(˛=.83).

(4)

Finally,aPortugueseversionoftheConceptsAboutPrinttask (Clay,2000/2003)wasused,atTime2,for measuringchildren’s knowledgeabouttheconventionsandcharacteristicsofwritten languageandearlyliteracyskills.Thismeasureincludes23 dichoto-mousitems,scoredas1(correct) or0(incorrect)(Alves,Aguiar, Castro,&Bairrão,2004).Whilereadingaspeciallydesigned chil-dren’sbookwithchildparticipants,researchersareabletoassess children’sknowledgeofconceptssuchastitle,frontandbackcover, beginninganddirectionoftext,letter,word,etc.Previousstudies haveusedthePortugueseversionofthismeasure,reporting appro-priatescorereliabilityandvalidity(e.g.,Cadima,Leal,&Burchinal, 2010).Forthisstudy,aglobalscorereﬂectingthenumberof con-cepts known tochildren was used (˛=.85). This measure was administeredinaquietroomateachchild’scenter.

2.2.2. Classroomquality

Tomeasuretheoverallqualityoftoddlerchildcareclassroomsat Time1,weusedaPortuguesetranslationoftheITERS(Harmsetal., 1990/1996).TheITERSconsistsof35items,organizedinseven sub-scalescorrespondingtoareasofclassroomquality:furnishingsand displayforchildren,personalcareroutines,listeningandtalking, learningactivities,interaction,programstructure,andadultneeds. Eachitemisratedasa7-pointscale,withdescriptorsfor1 (inade-quate),3(minimal),5(good),and7(excellent).Translation,training, anddatacollectionproceduresforthismeasureareavailablefrom Pessanhaetal.(2007).Inthisstudy,Cronbach’salphafortheITERS overallscore,computedasthemeanscoreacross33items(2items wereremovedduetolackofvariance:item31–provisions for exceptionalchildren–anditem23),was.80.

The Portuguese translation of the ECERS-R (Harms et al., 1998/2008)wasusedto measurepreschool classrooms’ overall qualityat Time2. TheECERS-Rconsistsof 43 items,organized insevensubscalesthatmeasureclassroomspaceandfurnishing, personalcareroutines,languageandreasoning,activities, interac-tion,programstructure,andparentsandstaff.AsinITERS,each itemispresentedasa7-pointscale,withdescriptorsfor1 (inade-quate),3(minimal),5(good),and7(excellent).Datapresentedhere arebasedoninformationcollectedovertwomornings,which cor-respondsto6hofobservationonaverage. Mean percentage of within-oneagreementbetweenraterswas87.7%(SD=9.91)and meanweightedKappacoefﬁcientwas.70(SD=.20).Inthisstudy, Cronbach’salphafortheECERS-Roverallscore,computedasthe meanscoreacrossallitems,was.95.

2.2.3. Homeenvironmentquality

Tomeasurethequalityofchildren’shomeenvironmentatTime 1,weusedthePortuguesetranslation oftheHome Observation forMeasurementoftheEnvironment–InventoryforFamiliesof InfantsandToddlers(HOME;Caldwell&Bradley,1984/1997).This measureiscomposedof45dichotomousitems(1,yes;0,no), orga-nizedinsixsubscalesthatmeasureparentalemotionalandverbal responsiveness,acceptanceofthechild’sbehavior,organizationof thephysicalandtemporalenvironment,provisionofappropriate playmaterials,parentalinvolvementwiththechild,and opportu-nitiesforvarietyindailystimulation.Researchassistantswitha degreeinPsychologyvisitedchildren’sfamiliesintheirhomesand conductedasemi-structuredinterviewwithmothers.Inthisstudy, weusedanoverallscorecomputedasthesumofscoresacrossall items.Cronbach’salphaforthisoverallscorewas.90.

2.2.4. Demographicandstructureinformation

Tocollectdataonparentalandfamilycharacteristicsatboth timepoints,weusedtheFamilyQuestionnairedevelopedbythe EuropeanChildCareandEducation(ECCE)StudyGroup(1997a). Thisquestionnairewasadministeredduringthehomevisits previ-ouslydescribed.Inaddition,atbothTimes1and2,thepreschool

anddaycarestructuralcharacteristicsquestionnaire –interview withtheclassroomteacher(ECCEStudyGroup,1997b)wasused tocollectdataonteacherandclassroomcharacteristics.Maternal educationwasconsideredanindicatorofthechild’ssocioeconomic status(SES).Thedecisiontoincludethisvariableintheanalysiswas basedonresultsofpreviousstudieswhichsuggestthatmaternal educationprovidesamorereliableindexofSESthanfamilyincome oroccupationprestige(Kagan,Lapidus,&Moore,1978;Roberts, Bornstein,Slater,&Barrett,1999).

2.3. Dataanalyses

Initialdataanalysisincludedexaminationofthereliabilityof scores,computationofdescriptivestatisticsforallvariables,and computationofcorrelationcoefﬁcientsamongvariablesofinterest. Inordertoaddressaggregationbiasandtoaccountforchildren’s mobility across ECEC settings, models for cross-classiﬁed ran-domeffects weretested, usinghierarchicallinear model(HLM; Raudenbush&Bryk,2002),toestimatecomponentsofvariancein outcomesthatliebetweentoddlerchildcareclassrooms,preschool classrooms,andchildren.

Asspeciﬁedin ourhypotheses,criterion variableswere chil-dren’slanguage,children’s communicationskills,and children’s knowledgeofearlyliteracyconcepts.Resultsofthethree-levelHLM emptymodelsforeachcriterionvariableshowedthattwo-level HLMmodelswereappropriateforourstudypurposes.TheHLM modelsincludedthreevariablesatthechildlevel(developmental levelatTime1,maternaleducation,andhomeenvironment qual-ity)andonevariable,preschoolquality(ECERS-Rglobalscoreat Time2),atlevel2.Allvariablesintheanalyseswerecontinuous andcenteredatthegrandmean.Theeffectsizeforthemaineffect ofbothlevel-1andlevel-2predictorswascomputedasBxSDx/SDy.

Accordingtotheequation,theeffectsizewasobtainedby multi-plyingtheunstandardizedregressioncoefﬁcientbythestandard deviationforthepredictorandthendividingbythestandard devi-ationfortheoutcome(Morrison,Gutman,Sameroff,&Cole,2003; NICHDEarlyChildCareResearchNetwork&Duncan,2003).

Moderationanalysestestedwhetherthedirectionorstrengthof therelationshipbetweenhomeenvironmentqualityandchild out-comesvarieddependingonpreschoolclassroomquality.According toBaronandKenny(1986),themoderatorhypothesisissupported iftheinteractionbetweenthepredictorvariableandthemoderator isstatisticallysignificant.Aftertestingforstatisticalsignificance, theeffectsizesfortheseinteractionswerecomputedbyestimating thecoefficientforhomeenvironmentqualitywhenpreschool class-roomqualitywas1SDaboveand1SDbelowthemean(Morrison etal.,2003;NICHDEarlyChildCareResearchNetwork&Duncan, 2003).Effectsizesindicatethedegreetowhichchangesinstandard deviationunitsinthepredictorcorrespondtolinearchangesofone standarddeviationintheoutcome,suchthat–1indicatesastrong negativeassociation,0indicatesnoassociationand1indicatesa strongpositiveassociation.Sinceallpredictors werecontinuous variablesandthestudywasnonexperimental,.10wasconsidered asmalleffectsize,.30amoderateeffectsize,and.50alargeeffect size(Cohen,1988;NICHD,2006).Thedatabaseusedinthisstudy didnotincludemissingvalues.

3. Results

Table1presentsdescriptivestatisticsonthechild,home,and classroom variables involved in the analyses. Global quality of center-basedtoddlerchild care classroomsas measuredbythe ITERSwashomogeneouslylow–aspreviouslyreportedbyAguiar andMcWilliam(2012)andPessanhaetal.(2007)– andaverage preschoolqualityasmeasuredbytheECERS-Rwasslightlybelow

(5)

Table1

Descriptivestatistics.

Scale N M SD Range

Toddlerchildcarequality 1–7 30 2.55 0.42 1.76–3.47

Preschoolquality 1–7 41 3.32 0.89 2.01–6.09

Maternaleducation(years) 95 10.64 4.47 3–18

Time1homeenvironmentquality 0–45 95 35.48 7.22 17–45

Time1developmentallevel 95 103.83 9.95 84.21–132.35

Time2language:hearingandspeech 95 93.04 12.62 63.80–131.40

Time2communication 0–30 95 15.32 4.21 1–25

Time2earlyliteracy:conceptsaboutprint 0–23 95 7.44 4.30 1–22

Table2

Bivariatecorrelationsbetweenchildren’sdevelopmentaloutcomesandhomeandchildcarequality.

1 2 3 4 5 6 7

Childdevelopmentaloutcomes

1.Time1developmentallevel –

2.Time2language:hearingandspeech .06 –

3.Time2communication .38* _.43* _–

4.Time2earlyliteracy:conceptsaboutprint .21* _.48* _.36* _– Homeandchildcarequality

5.Toddlerchildcarequality .08 .08 .01 .06 –

6.Preschoolquality −.34* _.20* _.07 _.13 _.21* _–

7.Time1homeenvironmentquality .40* _.17 _.18 _.37* _−.03 _−.07

8.Maternaleducation(years) .39* _.18 _.17 _.23* _−.05 _−.14 _.61*

*_p_<_.05.

thescale’smidpoint.Noneofthetoddlerchildcareclassroomsand only8%ofpreschoolclassroomswereinthegoodtoexcellentrange. Althoughthemeanhomeenvironmentqualitywasrelativelyhigh, itshowedalargevariabilityamongparticipants.

Table 2 presents a matrix displaying bivariate correlations betweenchildren’sdevelopmentallevel,children’slanguage, com-munication,andearlyliteracyskills,homeenvironmentquality, maternaleducation,overallqualityofcenter-basedtoddlerchild careclassrooms,andpreschoolquality.Somecorrelationswere sig-niﬁcantlydifferentfromzero.Supportingourhypothesis,positive statisticallysigniﬁcantcorrelationswerefoundbetweenpreschool qualityandchildren’slanguageaswellasbetweenhomequality andchildren’searlyliteracyskills.

Three-levelHLMemptymodelsexaminedcomponentsof vari-ance at the child level (level 1), between toddler child care classrooms(level2),andbetweenpreschoolclassrooms(level3). Examinationof resultsof thesethree-levelHLM emptymodels foreachcriterion variableshowedthatthere wasnosigniﬁcant variabilitytobeexplainedamongtoddlerchildcareclassroomsat Time1,whereastherewassigniﬁcantbetween-groupvariancein

preschoolclassroomsatTime2.Basedontheseresults,two-level HLMmodelswereusedtoanalyzetheeffectsofpreschool qual-ityand homequalityonchildoutcomes,whenalsoconsidering childdevelopmentallevelandmaternaleducationatTime1.The between-grouppredictoratTime1,toddlerchildcareclassroom quality,asmeasuredbytheITERS,wasnotconsideredforfurther analyses.

Results forthe two-levelHLMmodels(see Table3)showed that,whenalsoconsideringchildren’spriordevelopmentallevel andmaternaleducation,homeenvironmentqualityatTime1,was associatedwithchildren’slanguageandearlyliteracyskillsatTime 2.Further,whenaccountingforboth childandfamilyvariables, preschoolqualitywasassociatedwithchildren’sconcurrentearly literacyskills.Nodirecteffectsofmaternaleducationwerefound.In addition,therewasamoderatingeffectofpreschoolqualityinthe associationbetweenhomeenvironmentqualityandtwochild out-comes:languageandearlyliteracyskills.Theinterpretationofthis moderatingeffectindicatesasynergeticpositiveeffectinwhich theinﬂuenceofhomequalityisstrongerwhenpreschoolquality is higher. Speciﬁcally, when preschool quality is one standard Table3

Mainandmoderatedﬁxedeffectsofhomeandpreschoolqualityonchildren’sdevelopmentaloutcomesatTime2.

Language:hearingandspeech Communication Earlyliteracy:conceptsaboutprint

B(se) Effectsize B(se) Effectsize B(se) Effectsize

Intercept 92.87*_(1.57) _15.05*_(0.49) _7.66*_(0.47) Maineffects Level1 Time1developmental level 0.16(0.11) .13 0.21*_(0.04) _.50 _0.09*_(0.03) _.21 Time1homeenvironment

quality

0.34*_(0.15) _.19 _0.04_(0.05) _.07 _0.17*_(0.08) _.29 Maternaleducation(years) 0.47(0.27) .17 0.10(0.10) .11 0.04(0.11) 0.04 Level2

Preschoolquality 2.52(2.40) .18 0.84(0.67) .18 1.40*_(0.61) _.29 Moderationeffectsofpreschool

quality

2 2 2

16.75* _25.12 _34.49*

PreschoolqualityXTime1home environmentquality 0.44*_(0.16) _.42₍₊₁_SD) −.03(−1SD) 0.05(0.05) 0.17*_(0.06) _.82₍₊₁_SD) .32(−1SD) *_p_<_.05.

(6)

deviationabovethemean,homequalityhasapositivemoderate effectonlanguageand astrongpositive effectonearlyliteracy skills.Whenpreschoolclassroomqualityisone standard devia-tionbelowthemean,homequalitybecomesatrivialpredictorof children’slanguageandamoderatepredictorofearlyliteracyskills.

4. Discussion

Inthisstudy,weinvestigatedassociationsbetweenthe qual-ityof familyenvironment and ECECprogramsastheyrelateto preschoolers’ language,communication, and early literacy out-comes.Intersectionsbetweenthehomeenvironment and ECEC programswereourprimaryinterest.

First,ourinitialhypothesiswasnotconfirmedaswedidnot findassociationsbetweencenter-basedtoddlerchildcarequality andchildoutcomesinthepreschoolyears.Researchonthreshold effectsintheclassroomquality-childoutcomeassociationhas sug-gestedlargerbenefitsforchildren’sdevelopmentwhenclassroom qualityisinthegoodtohighrange(Burchinaletal.,2009)and thathigh-qualityclassrooms maybenecessarytoimprove chil-dren’sdevelopmentaloutcomes(Burchinal,Vandergrift,Pianta,& Mashburn,2010).Thus,itseemsreasonabletohypothesizethatour resultscouldbeduetothehomogeneouslowqualityfoundamong participatingclassroomsinthisstudy.Ifempiricallyconfirmed,this hypothesiswouldhaveimportantimplicationsforPortuguese pol-icymakers,professionals,andparentsascenter-basedtoddlerchild careinPortugalisconsideredtobeofpredominantlylow-quality (Barros&Aguiar,2010;Pessanhaetal.,2007).

Second,thequality ofthehomeenvironmentwaspositively associatedwithchildren’slanguageandliteracyoutcomesinthe preschoolyears.Theseresultspartiallyconfirmoursecond hypoth-esisandarecongruentwithpriorresearchfindings(e.g.,Bradley etal.,2011).Theyalsoexpandonotherdevelopmentaloutcomes findingsfromaEuropeanstudywhichreportedsimilarresultsfor intellectualandsocialdevelopment(Sylvaetal.,2008).Inaddition, thehomelearningenvironmentwasfoundtobemorepredictive ofchildren’sdevelopmentaloutcomesinthepreschoolyearsthan maternaleducation,whichSylvaetal.(2008)alsosuggested.

Third,thedirectlinksofpreschoolqualitytoconcurrentchild outcomesshoweda positive impactonchildren’searlyliteracy skills, over and above the effect of home quality, as hypothe-sized.Theseresultsareconsistentwithpreviousﬁndings(Burchinal et al., 2002; Howes et al., 2008; Mashburn, 2008), and com-plementresearchthathasemphasizedtheimplicationsofECEC quality onchildren’s cognitive and social outcomes(Burchinal, Campbell,Bryant,Wasik,&Ramey,1997;NICHDEarlyChildCare ResearchNetwork, 2000; Pierrehumbert,Ramstein,Karmaniola, le Miljkovitch, & Halfon, 2002). Previous studies in Portuguese settingssupport suchevidence, reportingassociations between preschoolglobalqualityandchildren’searlyliteracyskills,namely vocabulary, concepts about print, and phonological awareness (Abreu-Limaetal.,2012).

Oneofthemostimportant studiestodate,theNICHDEarly ChildCareResearchNetworkstudy(1997,1998a,1998b,2006), whichcomparedtherelevanceofboththehomeenvironmentand ECEC,reportedconsistentresultsforallthemeasuredvariablesand indicatedthatECECqualityexplainslessvariancethandofamily variables.However,similartorecentﬁndings(e.g.,Pierrehumbert etal.,2002;Sylvaetal.,2008),wefoundtheeffectsofpreschool qualitywereasstrongasthoseofthehomeenvironmentquality.

Fourth,ourﬁndingssuggestthatpreschoolqualitymaybean importantmoderatorontherelationshipbetweenhome environ-mentqualityandchildren’slatterlanguageandearlyliteracyskills, even when also considering children’s previous developmental levelandmaternaleducation.Ourresultsshowasynergetic posi-tiveeffectofhigherpreschoolqualityontheassociationsbetween

homeenvironmentqualityandchildren’slanguageandearly liter-acyoutcomes.Thesefindingsarecomplementarytopriorstudies thatsuggestedamediatingeffectofhomeenvironmentqualityon theassociations betweenECECqualityand children’soutcomes (Lamb, 1998), and to previous results that suggested cumula-tiveeffectsofhomeandpreschoolquality(NICHD,1997,1998a, 1998b,2000,2006).Findingsreportedheresupportthe integra-tivemodelproposedbyHowes(1999,ascitedinPierrehumbert etal.,2002)sincetheysuggestcumulativeeffectsofpositive inde-pendentvariablesfromtwodifferenteducationalcontexts.Such cumulativepositive effects onchildren’soutcomessuggest that higherpreschoolqualitypotentiatesthepositiveimpactofhome environmentqualityinpredictingchildren’slanguageandliteracy outcomes.Thisevidenceisstrongerforearlyliteracyskills, high-lightingtherelevanceofpromotinghighqualityinECECsettings,as theseskillsareessentialfordevelopingproficientreadingand writ-ingskillsaswellaslong-termacademicsuccess(Skibbe,Connor, Morrison,&Jewkes,2011).Likewise,severalauthors(e.g.,Pianta, 2007)haveidentifiedearlyprintknowledgeasarelevantfeature inhigh-qualitypreschoolclassrooms.

However,themoderatingeffectsreportedhereaddtoprevious knowledgeontheintersectionsbetweenhomeenvironmentand ECECquality.Ourstudyfurtherdocumentsadetrimentaleffectof poorpreschoolquality,andsuggeststhatlow-qualityECECmay hinderorevennullifythepositiveeffectofhomeenvironment qual-ityonchildren’slanguageandearlyliteracyskills.Infact,Watamura etal.(2011)designatednichescharacterizedbyhigh-qualityhome environments and low-qualityECEC aslost resources. Our ﬁnd-ingssuggest that more can be at stake than low-quality ECEC programs’failuretopotentiatetheeffectsof high-qualityhome environments.

Thisstudyhaslimitationsthatshouldbeacknowledged.First, bothourresearchdesignandstatisticalanalysesdonotallowus tomake cause-effectinferences. Second,three factors limitthe externalvalidity ofourfindings:thelimitednumber oftoddler andpreschoolchildcareclassroomsincluded,therestricted geo-graphicalareafromwhichtheseclassroomswererecruited,and thespecificqualityfeaturesoftheseclassrooms(whichinthecase oftoddlerchildcare,limitourfindingstolow-qualityclassrooms). Despitetheseconstraints,findingscontributetoabetter under-standing of the associations between children’s developmental outcomesandcontextqualityintwoofthechild’smain microsys-tems.Inaddition,thisstudyprovidesanopportunitytoexamine thecross-culturalconsistencyoffindingsfromqualitystudies con-ductedbothintheU.S.andinEurope.

Keepinginmindthereferredlimitations,wecautiouslytrust thattheeffectsfoundinthisstudymayhighlightpossible commu-naleffects.CarneiroandHeckmanhavearguedthatECECservices areoneofthefeweffectivemeansofincreasingeconomic oppor-tunitiesfordisadvantagedindividuals and,therefore,forsociety asa whole(2002, as cited inNICHD EarlyChild CareResearch Network, 2006). Presently, in Portugal, a largenumber of chil-drenspendextendedperiodsoftimeinavarietyofECECservices –public,private,andclandestine–betweenbirthandthetime theyenterschool.Fortheﬁrstthreeyearsofchildren’slives,the qualityoftheseservicesmayrangefrominadequatetominimally adequateintermsofhealth,safety,anddevelopmental appropri-ateness(Barros&Aguiar,2010;Pessanhaetal.,2007).Although researchresultshave shown that parental practicesare a rele-vantpredictorofchildcognitiveandsocialdevelopmentoutcomes duetogeneticandenvironmentalinﬂuences,highlightingthe cru-cialimportanceofthefamilyinchildren’slives(Collins,Maccoby, Steinberg,Hetherington,&Bornstein,2000),parentsarenot act-ingalone.WhenfamiliesdecidetouseECECservices,theynotonly shouldfeelthattheyarenotharmingtheirchildren,buttheyalso have therighttoknowthe characteristicsofthecontexts they

(7)

chooseand thepossibleeffects oftheseenvironments onchild development.

TheneedformoreresearchonECECinPortugalisclear. Cost-beneﬁtanalysesonthelong-termeffectsofECECexperiencesare especially important, considering the high risk of poverty and thenumberof children involvedin ECEC(InstitutoNacionalde Estatística,2010;MinistériodaEducac¸ão, 2010).However,even well-establishedconclusionsaboutECECeffectsizesmaynotlead to policy changes. As Shonkoff (2000) claims, social scientists oftenemphasizedatafromthiskindofresearch,butthisisrarely considered in policy development. Assuming our share of co-responsibilityinthepreventionandsolutionofsocialproblems, wehopethattheknowledgegainedfromthisstudywillbeuseful forparents,teachers,researchers,andpoliticiansandcontributeto bridgingthegapbetweenresearchandpractice.

Acknowledgements

Thisstudyispartoftworesearchprojectstitledrespectively. Thequalityofchildinteractionsinfamilyanddaycaresettingsand itsinﬂuenceonchildren’ssocio-cognitivedevelopmentand Longi-tudinalStudyofChildren’sEngagementandAdaptationdeveloped bytheDepartmentofPsychologyoftheUniversityofPortoand were supportedby grants fromthe Foundation of Science and Technology(POCTI/PSI/35207/2000,POCI/PSI/58712/2004).These projectswerecoordinatedbyProfessorJoaquimBairrãotowhom weaddressourappreciationandrespect.Theteamisgratefulfor sharinghiswisdomandknowledge.

Theauthorsthanktheteachersanddirectorsofthecentersthat participatedintheprojectalongwiththeresearchteammembers, Ana José,Ana SusanaAlmeida, Carla Cid,Carla PeixotoCristina Nunes,FernandoSantos,HelenaRosário,LeenPoppe,Margarida Guimarães,MariadaPazMascarenhas,MónicaSilva,OrlandaCruz, PedroLopesdosSantos,andSílviaBarros.Thevaluablecomments andsupportofMargaretBurchinal,R.A.McWilliam,JoanaCadima, andAnneKirbyarealsoappreciated.

References

Abreu-Lima,I.,Leal,T.,Cadima,J.,&Gamelas,A.M.(2012).Predictingchildoutcomes frompreschoolqualityinPortugal.EuropeanJournalofPsychologyofEducation,

http://dx.doi.org/10.1007/s10212-012-0120-y

Aguiar, C., & McWilliam, R. A. (2012). Consistency of toddler engage-ment across two settings. Early Childhood Research Quarterly,.

http://dx.doi.org/10.1016/j.ecresq.2012.04.003

Ahnert, L., & Lamb, M. (2004). Child care and its impact on young chil-dren (2–5). In R. E. Tremblay, R. G. Barr, & R. De V Peters (Eds.), Encyclopedia on early childhooddevelopment (pp. 1–6). Montreal, Quebec: Centre of Excellence for Early Childhood Development. Retrieved from:

http://www.child-encyclopedia.com/pages/pdf/ahnert-lambangxp.pdf

Alves,R.A.,Aguiar,C.,Castro,S.L.,&Bairrão,J.(July2004).Assessmentofconcepts aboutprintusinganecologicallyvalidtaskwithPortuguesechildren.InPoster paperpresentedattheEleventhAnnualMeetingoftheSocietyfortheScientiﬁcStudy ofReadingAmsterdam,Netherlands.

Anders,Y.,Rossbach,H.-G.,Weinert,S.,Ebert,S.,Kuger,S.,Lehrl,S.,etal.(2011). Homeandpreschoollearningenvironmentsandtheirrelationstothe devel-opmentofearlyliteracyskills.EarlyChildhoodResearchQuarterly,27,231–244.

Área Metropolitana do Porto. (n.d.). Prioridades de desenvolvimento para 2007–2013(Developmentprioritiesfor2007–2013).Retrievedfrom:http://www. amp.pt/fotos/gca/amp prioridades2007a201321190370053.pdf. Bairrão,J.,Barbosa,M.,Borges,I.,Cruz,O.,&Macedo-Pinto,I.(1990).Perﬁlnacional

doscuidadosprestadosàscriançascomidadeinferioraseisanos.(National pro-fileofcaretochildrenundersixyearsold).Lisbon,Portugal:FundaçãoCalouste Gulbenkian.

Bairrão,J.,Leal,T.,Abreu-Lima,I.,&Morgado,R.(1997).Educaçãopré-escolar [Preschooleducation].InMinistériodaEducação(Ed.),EstudosTemáticos.Vol. II:AEvoluçãodoSistemaEducativoeoPRODEP(pp.17–110).Lisbon,Portugal: DepartamentodeAvaliação,ProspectivaePlaneamento.

Baron,R.M.,&Kenny,D.A.(1986).Themoderator–mediatorvariabledistinction insocialpsychologicalresearch:Conceptual,strategic,andstatistical consider-ations.JournalofPersonalityandSocialPsychology,51,1173–1182.

Barros,S.,&Aguiar, C.(2010).Assessingthequality ofPortuguesechildcare programs for toddlers. Early Childhood Research Quarterly, 25, 527–535.

Bradley,R.H.,McKelvey,L.M.,&Whiteside-Mansell,L.(2011).Doesthequalityof stimulationandsupportinthehomeenvironmentmoderatetheeffectofearly educationprograms?ChildDevelopment,82,2110–2122.

Burchinal,M.R.,Campbell,F.A.,Bryant,D.M.,Wasik,B.H.,&Ramey,C.T.(1997). Earlyinterventionandmediatingprocessesincognitiveperformanceofchildren oflow-incomeAfricanAmericanfamilies.ChildDevelopment,68,935–954. Burchinal,M.R.,Kainz,K.,Cai,K.,Tout,K.,Zaslow,M.,Martinez-Beck,I.,etal.(2009).

Earlycareandeducationqualityandchildoutcomes.Washington,DC:Ofﬁceof Planning,ResearchandEvaluation,AdministrationforChildrenandFamilies,US DHHS,andChildTrends.

Burchinal,M.R.,Peisner-Feinberg,E.S.,Pianta,R.,&Howes,C.(2002).Development ofacademicskillsfrompreschoolthroughsecondgrade:Familyand class-roompredictorsofdevelopmentaltrajectories.JournalofSchoolPsychology,40, 415–436.

Burchinal,M.R.,Roberts,J.E.,Nabors,L.A.,&Bryant,D.M.(1996).Qualityofcenter childcareandinfantcognitiveandlanguagedevelopment.ChildDevelopment, 67,606–620.

Burchinal,M.,Vandergrift,N.,Pianta,R.,&Mashburn,A.(2010).Thresholdanalysisof associationbetweenchildcarequalityandchildoutcomesforlow-income chil-dreninpre-kindergartenprograms.EarlyChildhoodResearchQuarterly,25(2), 166–176.http://dx.doi.org/10.1016/j.ecresq.2009.10.004

Cadima, J., Leal, T., & Burchinal, M. (2010). The quality of teacher-student interactions:Associationswithﬁrst graders’academic andbehavioral out-comes.JournalofSchoolPsychology,48,457–482.http://dx.doi.org/10.1016/j.jsp. 2010.09.001

Caldwell,B.M.,&Bradley,R.H.(1997).EscaladeAvaliac¸ãodoAmbienteFamiliar: Versãoportuguesa(Homeobservationforthemeasurementoftheenvironment: Portugueseversion)(UniversidadedoPorto,CentrodePsicologia,Trans.).Centro dePsicologia,UniversidadedoPorto,Portugal(Originalworkpublished1984). Clay,M.M.(2003).Segue-me,lua:Adaptac¸ãoportuguesa(Followme,moon:Portuguese version)(R.Alves&C.Aguiar,Trans.).CentrodePsicologia,Universidadedo Porto,Portugal(Originalworkpublished2000).

Cohen,J.(1988).Statisticalpoweranalysisforthebehavioralsciences(2nded.). Hills-dale,NJ:LawrenceErlbaumAssociates.

Collins,W.A.,Maccoby,E.,Steinberg,L.,Hetherington,E.M.,&Bornstein,M.(2000). Contemporaryresearchonparenting:Thecaseofnatureandnurture.American Psychologist,55,218–232.

ConselhoNacionaldeEducação.(2011).EstadodaEducação2011.AQualificação dosPortugueses[Thestateofeducation2011.ThequalificationsofthePortuguese population].Lisbon,Portugal:Author.

Cordazo,S.T.,Almeida,A.M.,&Vieira,M.L.(2011).Theinﬂuenceofplayinmotor, cognitiveandsocialperformanceinschool-agechildren.EducationalResearch, 2(9),1472–1480.

EuropeanChildCareandEducationStudyGroup.(1997a).Questionáriodafamília [FamilyQuestionnaire]CentrodePsicologia,UniversidadedoPorto,Portugal. EuropeanChildCareandEducationStudyGroup.(1997b).Questionáriodeavaliac¸ão

dascaracterísticasestruturaisdojardim-de-infânciaedacreche[Preschooland day-carestructuralcharacteristicsquestionnaire–interviewwiththeclassroomteacher] CentrodePsicologia,UniversidadedoPorto,Portugal.

GabinetedeEstratégiaePlaneamento.(2009).CartaSocial– Folhainformativan.◦₁

[Socialletter.informativepaperno.1].Lisbon,Portugal:MinistériodoTrabalhoe daSolidariedadeSocial.

Griffiths,R.(1996).AsEscalasGriffithsadaptadasaoPortuguês:Versãoparcialpara investigação(TheGriffithsScalesadaptedtoPortuguese:Partialversionforresearch purposes)(S.L.Castro,&I.Gomes,Trans.).LaboratóriodeFala,Faculdadede PsicologiaedeCiênciasdaEducaçãodaUniversidadedoPorto,Portugal(Original workpublished1984,1986).

Harms,T.,Clifford,R.,&Cryer,D.(1998).Earlychildhoodenvironmentratingscale– revisededition(ECERS-R).NewYork,NY:TeachersCollegePress.

Harms,T.,Clifford,R.,&Cryer,D.(2008).EscaladeAvaliaçãodoAmbienteemEducação deInfância– EdiçãoRevista(Earlychildhoodenvironmentratingscale–revised edition)(I.Abreu-Lima,C.Aguiar,A.M.Gamelas,T.Leal,&A.I.Pinto,Trans.). Porto,Portugal:LivPsic/LegisEditora(Originalworkpublished1998). Harms,T.,Cryer,D.,&Clifford,R.M.(1996).EscaladeAvaliaçãodoAmbientedeCreche

(Infant/toddlerenvironmentratingscale–ITERS)(A.I.Pinto&T.Grego,Trans.). CentrodePsicologia.UniversidadedoPorto,Portugal(Originalworkpublished 1990).

Howes,C.,Burchinal,M.,Pianta,R.,Bryant,D.,Early,D.,Clifford,R.M.,etal.(2008). Readytolearn?Children’spre-academicachievementinpre-kindergarten pro-grams.EarlyChildhoodResearchQuarterly,23,27–50.

Instituto NacionaldeEstatística. (2010).Sobrea pobreza,as desigualdadesea privac¸ãomaterialemPortugal[Onpoverty,inequalityandmaterialdeprivation in Portugal].Lisbon, Portugal: Author. Retrieved from: http://www.ine.pt/ xportal/xmain?xpid=INE&xpgid=ine publicacoes&PUBLICACOESpub boui= 100334100&PUBLICACOEStema=55466&PUBLICACOESmodo=2.

Kagan,J.,Lapidus,D.R.,&Moore,M.(1978).Infantantecedentsofcognitive func-tioning:Alongitudinalstudy.ChildDevelopment,49,1005–1023.

Lamb,M.(1998).Nonparentalchildcare:Context,quality,correlates,and conse-quences.InW.Damon(SeriesEd.)&I.E.Sigel,&K.A.Renninger(Vol.Eds.), Handbookofchildpsychology:Vol.IV.Childpsychologyinpractice(pp.73–133). NewYork,NY:Wiley.

Mashburn, A. J. (2008). Quality of social and physical environments in preschools and children’s development of academic, language,

(8)

and literacy skills. Applied Developmental Science, 12(3), 113–127.

http://dx.doi.org/10.1080/10888690802199392

Mashburn,A.J.,Pianta,R.C.,Hamre,B.K.,Downer,J.T.,Barbarin,O.A.,Bryant,D., etal.(2008).Measuresofclassroomqualityinprekindergartenandchildren’ developmentofacademic,language,andsocialskills.ChildDevelopment,79(3), 732–749.

McWilliam,R.A.,&deKruif,R.E.L.(1998).Engagementqualityobservationsystem III(E-QualIII).ChapelHill,NC:FrankPorterGrahamChildDevelopmentCenter, UniversityofNorthCarolinaatChapelHill.

MinistériodaEducação.(2000).Aeducaçãopré-escolareoscuidadosparaainfância emPortugal[PreschooleducationandearlychildhoodcareinPortugal]. Departa-mentodeEducaçãoBásica.Lisbon,Portugal:Author.

MinistériodaEducac¸ão.(2009).Lein.◦_85/2009_de₂₇_de_Agosto_[Law_No._85/2009_of

Agosto27].DiáriodaRepública,1.asérie,N.◦_._166._Ministério_da_Educac¸_ão.

Ministério da Educação. (2010). Perfil do aluno 2007/2008 [Student’s profile 2007/2008]. Lisbon, Portugal: Gabinete de Estatística e Planeamento da Educação(GEPE).Retrievedfrom:http://www.gepe.min-edu.pt/np4/?newsId= 411&fileName=GEPEDSEPerfilAluno200708.pdf

Morrison,F.J.,Gutman,L.M.,Sameroff,A.J.,&Cole,R.(2003).Academicgrowth curvetrajectoriesfrom1stgradeto12thgrade:Effectsofmultiplesocialrisk factorsandpreschoolchildfactors.DevelopmentalPsychology,39,777–790. NationalInstituteonChildHealthand HumanDevelopmentEarlyChild Care

ResearchNetwork.(1997).Theeffectsofinfantchildcareoninfant–mother attachmentsecurity:ResultsoftheNICHDstudyofearlychildcare.Child Devel-opment,68,860–879.

NationalInstituteonChildHealthand HumanDevelopmentEarlyChild Care ResearchNetwork.(1998a).Earlychildcareandselfcontrol,complianceand problembehaviorat24and36months.ChildDevelopment,69,1145–1170. NationalInstituteonChildHealthand HumanDevelopmentEarlyChild Care

ResearchNetwork.(1998b).Relationsbetweenfamilypredictorsandchild out-comes:Aretheyweakerforchildreninchildcare?DevelopmentalPsychology, 34,1119–1128.

NationalInstituteonChildHealthand HumanDevelopmentEarlyChild Care ResearchNetwork.(2000).Therelationofchildcaretocognitiveandlanguage development.ChildDevelopment,71,960–980.

NationalInstituteonChildHealthand HumanDevelopmentEarlyChild Care ResearchNetwork.(2006).ChildcareeffectsizesfortheNICHDstudyofearly childcareandyouthdevelopment.AmericanPsychologist,2,99–116. NationalInstituteonChildHealthand HumanDevelopmentEarlyChild Care

ResearchNetwork,&Duncan,G.J.(2003).Modelingtheimpactsofchildcare qualityonchildren’spreschoolcognitivedevelopment.ChildDevelopment,74, 1451–1475.

OrganizationforEconomicCo-OperationandDevelopment. (2007).Babiesand bosses:Reconcilingworkandfamilylife:AsynthesisofﬁndingsforOECDcountries. Paris,France:Author.http://dx.doi.org/10.1787/9789264032477-en

Organization for EconomicCo-Operation and Development. (2008).Education ata glance 2009: OECDindicators. Paris,France: Author. Retrieved from:

http://www.oecd.org/publishing/corrigenda

Peisner-Feinberg,E.S.,&Yazejian,N.(2010).Researchonprogramquality:The evidencebase.InP.W.Wesley,&V.Buysse(Eds.),Thequestforquality: Promis-inginnovationsforearlychildhoodprograms.Baltimore,MD:PaulH.Brookes Publishing.

Pereira, M., Soares, I., Dias, P., Silva, J., Marques, S., & Baptista, J. (2010). Desenvolvimento,psicopatologiaeapego:Estudoexploratóriocomcrianc¸as institucionalizadase suascuidadoras [Development, psychopathology and attachment:Anexploratorystudywithinstitutionalizedchildrenandtheir care-givers].Psicologia:ReﬂexãoeCrítica,23(2),222–231.

Pessanha,M.(2008).Vulnerabilidadeeresiliêncianodesenvolvimentodos indiví-duos:Influênciadaqualidadedoscontextosdesocializaçãonodesenvolvimento dascrianças[Vulnerabilityandresilienceinindividuals’sdevelopment:Influenceof thequalityofsocializationcontextsonchildren’sdevelopment].Lisbon,Portugal: FundaçãoCalousteGulbenkian,FundaçãoparaaCiênciaeTecnologia. Pessanha,M.,Aguiar,C.,&Bairrão,J.(2007).Influenceofstructuralfeatureson

Portuguesetoddlerchildcarequality.EarlyChildhoodResearchQuarterly,22(2), 204–214.http://dx.doi.org/10.1016/j.ecresq.2007.02.003

Pianta,R.C.(2007).Preschoolisschool,sometimes:Makingearlychildhood educa-tionmatter.EducationNext,7,44–49.

Pierrehumbert,B.,Ramstein,T.,Karmaniola,A.,Miljkovitch,R.,&Halfon,O.(2002). Qualityofchildcareinthepreschoolyears:Acomparisonoftheinﬂuenceof homecareanddaycarecharacteristicsonchildoutcome.InternationalJournal ofBehavioralDevelopment,26(5),385–396.

Pinto,A.I.(2006).Oenvolvimentodacriançaemcontextodecreche:Efeitosde carac-terísticasdacriança,daqualidadedocontextoedasinteracçõeseducativas.[Child engagementinchildcare:Effectsofchildcharacteristics,contextquality,andteacher interactionsonchildren’sobservedengagement](Unpublisheddoctoral disserta-tion).FaculdadedePsicologiaedeCiênciasdaEducação,UniversidadedoPorto, Portugal.

Pinto,A.I.,Aguiar,C.,Barros,S.,&Cruz,O.(2004).OSistemadeAvaliaçãoda Qual-idadedoEnvolvimentoIII:Umprocedimentodeavaliaçãodoenvolvimentoda criançaemcontextodecreche[TheengagementqualityobservationsystemIII: Aprocedureforassessingchildengagementinchildcare].InC.Machado,L. S.,Almeida,M.Gonçalves,&V.Ramalho(Orgs.),AvaliaçãoPsicológica:Formase Contextos(Vol.X,pp.441–448).Braga,Portugal:PsiquilíbriosEdições. Raudenbush,S.W.,&Bryk,A.S.(2002).Hierarchicallinearmodels:Applicationsand

dataanalysismethods(2nded.).London,UK:SagePublications.

Roberts,E.,Bornstein,M.H.,Slater,A.M.,&Barrett,J.(1999).Earlycognitive devel-opmentandparentaleducation.InfantandChildDevelopment,8,49–62. Sammons,P.,Sylva,K.,Melhuish,E.C.,Siraj-Blatchford,I.,Taggart,B.,&Elliot,K.

(2002).Theeffectiveprovisionofpre-schooleducation(EPPE)project:Technical paper8a–measuringtheimpactofpre-schoolonchildren’scognitiveprogress overthepre-schoolperiod.London,UK:DfES/InstituteofEducation,University ofLondon.

Sammons,P.,Sylva,K.,Melhuish,E.C.,Siraj-Blatchford,I.,Taggart,B.,&Elliot,K. (2003).Theeffectiveprovisionofpre-schooleducation(EPPE)project:Technical paper8b–measuringtheimpactofpre-schoolonchildren’ssocial/behavioural developmentoverthepre-schoolperiod.London,UK:DfES/InstituteofEducation, UniversityofLondon.

Shonkoff,J.P.(2000).Science,policy,andpractice:Threeculturesinsearchofa sharedmission.ChildDevelopment,71,181–187.

Skibbe,L.E.,Connor,C.M.,Morrison,F.J.,&Jewkes,A.M.(2011).Schoolingeffects onpreschoolers’self-regulation,earlyliteracy,andlanguagegrowth.Early Child-hoodResearchQuarterly,26,42–49.

Sparrow,S.S.,Balla,D.A.,&Cicchetti,D.V.2000.(n.d.).EscaladeComportamento AdaptativodeVineland(VinelandAdaptativeBehaviorScales:Portugueseversion forresearchpurposes).(UniversidadedoPorto,CentrodePsicologia,Trans.). Portugal:CentrodePsicologia,UniversidadedoPorto(Originalworkpublished 1984).

Sylva,K.,Melhuish,E.,Sammons,P.,Siraj-Blatchford,I.,&Taggart,B.(2008).Final reportfromtheprimaryphase:Pre-school,schoolandfamilyinﬂuencesonchildren’s developmentduringkeystage2(Age7–11).London,UK:DfES/Instituteof Educa-tion,UniversityofLondon.Retrievedfrom:http://www.dcsf.gov.uk/research/

Sylva,K.,Siraj-Blatchford,I.,&Taggart,B.(2003).Earlychildhoodenvironmentrating scale– extension(ECERS-E).StokeonTrent,UK:TrenthamBooks.

UnitedNationsEducational,Scientiﬁc,andCulturalOrganization.(1982).Parauma políticadaeducac¸ãoemPortugal[ForaneducationpolicyinPortugal].Lisbon: Portugal:LivrosHorizonte.

Vandell, D. L., Belsky, J., Burchinal, M., Steinberg, L., Vandergrift, N., & NICHD Early Child Care Research Network. (2010). Do effects of early child care extend to age 15 years? Results from the NICHD study of earlychildcare and youthdevelopment.Child Development, 81, 737–756.

http://dx.doi.org/10.1111/j.1467-8624.2010.01431.x

Vasconcelos,T.,Orey,I.,Homem,L.F.,&Cabral,M.(2002).Educaçãodeinfância emPortugal:Situaçãoecontextosnumaperspectivadepromoçãodeequidadee combateàexclusão[ChildhoodeducationinPortugal:Stateoftheartandcontexts withinanequitypromotingapproachandfightagainstexclusion].Lisbon,Portugal: ConselhoNacionaldeEducação.

Watamura,S.E.,Phillips,D.A.,Morrissey,T.W.,McCartney,K.,&Bub,K.(2011). Doublejeopardy:Poorersocial–emotionaloutcomesforchildrenintheNICHD SECCYDexperiencinghomeandchild-careenvironmentsthatconferrisk.Child Development,82,48–65.http://dx.doi.org/10.1111/j.1467-8624.2010.01540.x