Impactofparentalmigrationonthepsychosocialwellbeingof
left-behindchildrenintwoChineseprovinces:individual
experiences,familycharacteristics,andcommunitycontexts
ChenyueZhao
InstituteofGlobalHealth,UniversityCollegeLondon
Supervisor:Prof.ThereseHesketh
AthesissubmittedforthedegreeofDoctorofPhilosophy
February2017
2
Declaration
I,ChenyueZhao,declarethattheworkpresentedinthisthesisismyown.Where
information has been derived from other sources, I confirm that this has been
indicatedinthethesis.
3
Abstract
Background:Inmanycountries,largenumbersofleft-behindchildren(LBC)
growupexperiencingprolongedseparationfromtheirmigrantworkerparents.
InChina,thewellbeingof60millionLBChasbecomeasignificantchallengefor
families,communities,andthesociety.Thisthesisaimedtoinvestigatetheimpact
mechanismsofparentalmigrationonthepsychosocialwellbeingofLBC.
Methods:Childrenwererecruitedfrommigrant-sendingruralareasinZhejiang
andGuizhouprovinces.Aqualitativestudyinvestigatedtheexperiencesof
childrenandperceptionsoffamilymembersinrelationtochildpsychosocial
wellbeingundertheimpactofparentalmigration,aswellastheinterrelated
factorsinthefamilyandsocialenvironmentsandthemechanismsthroughwhich
theyaffectLBC.Aquantitativestudymeasuredchildren’spsychosocialstrengths
anddifficultiesandthefamilyandsocialfactors,withaself-administered
questionnaire,andexaminedtheeffectsofbothcurrentandpreviousparental
migration,andpotentialcausalmechanismsinvolvingdifferentcovariatesin
multiplelinearregressionmodels.
Findings:Qualitativefindingsshowedthatlengthyseparationposesconsiderable
difficultiesonmanychildren’spsychosocialwelling,especiallyemotionaldistress,
primarilythroughdisruptedattachmentrelationships.Theseeffectsmaynotbe
easilyrestoredevenifmigrantparentspermanentlyreturnedhome.The
4
psychosocialsupport(andlackof)fromtheco-residentfamily,aswellasfactors
regardingschoolperformanceandcommunitycohesion,alsomodifyorintensify
thechildwellbeingoutcomesundertheimpactofparents’absence.Quantitative
resultsindicatedthatbothcurrentandpreviousparentalmigrationwas
independentlyassociatedwithadverseoutcomes,especiallytheemotionaland
socialdimensionsofchildpsychosocialwellbeing.Relationshipbetweennuclear
familymembers,carearrangements,andavailabilityofsupportinfamilyand
socialenvironmentsseemtobeimportantcovariatesinthecausalmechanisms.
Parentaldivorceandlackofsupportmayparticularlyexacerbatechildren’s
psychosocialdifficulties.
Discussion:Parentalmigrationhasalong-lastingadverseeffectonchildren’s
psychosocialwellbeing,especiallyemotionalfunctioning.Positiveparent-child
relationshipbondsandsupportivefamilyandsocialenvironmentsarecrucial
determinantsofthewellbeingofLBC,ratherthansocioeconomicstatus.
Communitybasedprogramsmaybedevelopedtoprovideadditionalcareand
support.
5
Acknowledgements
IwouldfirstandforemostliketothankmyprimarysupervisorProf.Therese
HeskethwhoprovidedessentialguidanceandsupportthroughoutmyPhD
project.IamalsoverygratefultomysecondarysupervisorDrLeahLiforthe
insightfulhelpwiththequantitativeanalysischapter.
TherearemanypeopleinChinawithoutwhomthisresearchwouldhavebeen
impossible.ThankstotheprojectteamatZhejiangUniversityforenablingmeto
conductthisstudy,andtoKaihuaWomen’sFederationforthelocalsupportfrom
manyofitsstaffmembers.ThanksalsotothestudyparticipantsinZhejiangand
Guizhouforsharingtheirtimeandoftenpersonalexperiences.Specialthanksto
KaihuaWomen’sFederationforthecruciallocalsupportfrommanyofitsstaff
members.
Also,verydeepthankstothefriendswhohavemotivatedandinspiredme
throughoutmyPhD,inparticularJinXuandHenryLi,becauseofwhomInever
reallyfeltaloneonthisbittersweetjourney.
Finally,thankstomyparentsandmygirlfriend,whosupportedme
unconditionallyalongtheway.Andthankstomygrandfather,whoIwasunable
tospendmoretimewithduringhislastyears,whotaughtmesomuchin
becomingapersonIam,andwouldhavebeensoproudofme.
6
TableofContents
Chapter1Introduction.........................................................................................................................13
1.1ScaleofmigrationandLBCpopulationsglobally.....................................................13
1.2Contextsofmigrationandparent-childseparation................................................16
1.3Careenvironmentofleft-behindchildren...................................................................18
1.4PoliciestosafeguardthewellbeingofLBC.................................................................21
1.5ProgramstoimprovethewellbeingofLBC................................................................24
1.6Researchquestionandstudyaims.................................................................................26
1.7Thesisstructure......................................................................................................................26
Chapter2ReviewofTheoreticalPerspectives...........................................................................28
2.1Introduction..............................................................................................................................28
2.1Migrationtheories.................................................................................................................28
2.1.1Thepush-pullmodels................................................................................................28
2.1.2Neoclassicaleconomicstheory..............................................................................30
2.1.3Householdeconomicstheory.................................................................................31
2.1.4Segmentedlabor-marketandworldsystemstheory....................................32
2.2Theoriesinfamilystudies..................................................................................................33
2.2.1FamilyStressManagement....................................................................................33
2.2.2FamilyContext............................................................................................................35
2.2.3Socio-ecologicalapproachandfamilysystemstheory................................37
2.3AttachmentTheory................................................................................................................39
2.3.1Theearliestattachment..........................................................................................39
2.3.2Developmentanddisruptionofattachmentrelationship..........................41
2.3.3Migratoryseparationandattachment..............................................................42
2.4Summary....................................................................................................................................43
Chapter3ReviewofEmpiricalLiterature....................................................................................45
3.1Introductionandscopeofthereview...........................................................................45
3.1Methods......................................................................................................................................46
3.2Quantitativestudies..............................................................................................................62
7
3.2.1Characteristicsofthestudysamples..................................................................63
3.2.2ComparisonbetweenEnglish-languageandChinese-languagestudies67
3.2.3Happinessandqualityoflifemeasures.............................................................68
3.2.4Mentalhealthstatus.................................................................................................70
3.2.5Resilienceandpositiveyouthdevelopment......................................................72
3.2.6Psychosocialstrengthsanddifficulties..............................................................73
3.2.7Emotionalsymptoms................................................................................................77
3.2.8Behavioralissues........................................................................................................82
3.2.9Self-conceptandself-esteem..................................................................................84
3.2.10Factorsassociatedwithwellbeinginleft-behindchildren......................86
3.3Qualitativestudies..............................................................................................................110
3.3.1Migrationdecision..................................................................................................110
3.3.2Gendernormsandsocio-culturalcontext......................................................113
3.3.3Emotionalandbehavioralimpact....................................................................114
3.3.4Communication........................................................................................................117
3.3.5Education...................................................................................................................120
3.3.6Careenvironmentintheleft-behindfamilies...............................................123
3.4Summary.................................................................................................................................125
3.4.1Overviewandgapsintheliterature................................................................125
3.4.2Contributionofthereviewtothecurrentqualitativestudy...................127
3.4.3Contributionofthereviewtothecurrentquantitativestudy................128
3.4.4Puttingempiricalstudiesintocontexts..........................................................130
Chapter4Studycontexts..................................................................................................................131
4.1Introduction...........................................................................................................................131
4.2Studyareasandpopulations..........................................................................................131
4.2.1SocioeconomicdevelopmentandmigrationinZhejiangProvince......133
4.2.3IntroductiontoKaihua,themainprojectsite..............................................135
4.2.2SocioeconomicdevelopmentandmigrationinGuizhouProvince.......137
4.3Policyandadministrationcontexts.............................................................................138
4.3.1FamilyplanningpolicyinChina........................................................................139
4.3.2Ruralgovernmentadministrationsystem.....................................................140
8
4.3.3Women’sFederation(WF)system....................................................................142
4.3.4 Roles of government agencies on child protection and intervention
programs...............................................................................................................................143
4.4Summary.................................................................................................................................144
Chapter 5 Experiences of children, and relationships, care and support in the
familieslivingapart.............................................................................................................................145
5.1Introductionanduseofextantliterature.................................................................145
5.2Methods...................................................................................................................................147
5.2.1Studyarea..................................................................................................................148
5.2.2Sampling.....................................................................................................................149
5.2.3Interviewpreparations.........................................................................................152
5.2.3Interviewprocedure...............................................................................................154
5.2.4Rolesofthelocalcollaborator...........................................................................156
5.2.5Analysis.......................................................................................................................158
5.3Results......................................................................................................................................161
5.3.1Overviewofmigrantfamilies.............................................................................161
5.3.2Householdsocioeconomicrationalesofmigration....................................162
5.3.3Childwellbeinginabsenceofmigrantparents...........................................168
5.3.4 Care arrangements and psychosocial support in the left-behind
family......................................................................................................................................179
5.3.5Careanddiscipliningconcernsandthereturnofmigrantmother.....185
5.3.6Challengesineducationattainmentandcommunityenvironment....192
5.3.7Psychosocialsupportfromthecommunityenvironment........................197
5.4Discussion...............................................................................................................................201
5.4.1Summaryofkeyfindings......................................................................................201
5.4.2Comparisonofstudycontextswithextantinternationalliterature....204
5.4.3Attachmentrelationshipandemotionaldistress........................................206
5.4.4 Availability of support from family and socio-ecological
environments.......................................................................................................................211
5.4.5Theprocessesandeffectsofreturnmigration.............................................214
5.4.6Limitations:studysample,age,andgender.................................................217
9
5.4.7Studyimplicationsandconnectionswiththequantitativestudy........218
Chapter 6 Impact of parental migration on children’s psychosocial wellbeing
measures:aquantitativesurvey....................................................................................................222
6.1Introduction...........................................................................................................................222
6.2Methods...................................................................................................................................223
6.2.1Studypopulations...................................................................................................223
6.2.2Datacollection.........................................................................................................224
6.2.3Measures.....................................................................................................................225
6.2.4Dataanalysis.............................................................................................................236
6.3Results......................................................................................................................................237
6.3.1Sampledescriptivestatistics...............................................................................237
6.3.2Parentalmigrationstatusandpsychosocialdifficulties..........................240
6.3.3Effectsofhouseholdcharacteristics.................................................................243
6.3.4Effectsofsocialsupportandschoolperformance......................................246
6.3.5Differentdimensionsofpsychosocialwellbeing..........................................246
6.3.6Resultsincurrentlyleft-behindchildren........................................................248
6.2.7Interactioneffects...................................................................................................250
6.4Discussion...............................................................................................................................251
6.4.1Summaryofkeyfindingsandcontributionstotheliterature...............251
6.4.2Independenteffectsofparentalmigrationonpsychosocialdifficulties254
6.4.3Familycareandsocialsupport..........................................................................258
6.4.4Familywealthstatusandchildwellbeing.....................................................265
6.4.5Householdeconomicsoflabormigration......................................................268
6.4.6Schoolperformance...............................................................................................270
6.4.7Ageandgenderdisparities..................................................................................271
6.5Strengthsandlimitations................................................................................................274
Chapter7Discussionandconclusion..........................................................................................276
7.1Overallsummaryofstudyresults................................................................................276
7.2Keyfindingsandcontributions.....................................................................................277
7.2.1 The relationship-emotion mechanisms showing the impact of
migration..............................................................................................................................277
10
7.2.2Non-emotionalaspectsofchildwellbeing.....................................................281
7.2.3Atimedimensionoftheimpactfrommigration.........................................283
7.2.4Carearrangementsandpsychosocialsupport.............................................286
7.2.5Householdeconomicresources..........................................................................290
7.2.6Socioeconomicandcommunitycontexts.......................................................292
7.3Conclusion..............................................................................................................................298
7.3.1SystemssafeguardingthewellbeingofLBC..................................................298
7.3.2Stableandprotectivefamilyenvironment....................................................299
7.3.3Implicationsforpolicyandinterventions......................................................300
7.3.4Solutionstokeepfamiliestogether..................................................................302
References...............................................................................................................................................304
Annex:Quantitativesurveyquestionnaire................................................................................324
ListofTables
Table3.1:SummaryofreviewedquantitativestudiesconductedinChinaand
publishedinEnglish............................................................................................................................50
Table3.2:SummaryofreviewedquantitativestudiesconductedinChinaand
publishedinChinese...........................................................................................................................53
Table3.3:Summaryofreviewedquantitativestudiesthatwerenotconductedin
China..........................................................................................................................................................59
Table3.4:Summaryofreviewedqualitativestudies.........................................................108
Table5.1:Demographics,familycomposition,andmigrationpatternof
interviewedmigrantfamilies.................................................................错误! 未定义书签。
Table6.1:FrequencydistributionofresponsesineachSDQitem..............................227
Table6.2:Overallsamplecharacteristics...............................................................................234
Table6.3:Samplecharacteristicsbyparentalmigrationstatus...................................238
Table6.4:Psychosocialoutcomesbyparentalmigrationstatus..................................242
Table6.5:Regressioncoefficient(SE)fortotaldifficultiesscoreonparental
migrationstatuswithandwithoutadjustmentforhouseholdcharacteristics,
socialsupport,andschoolperformance.................................................................................244
Table6.6:Regressioncoefficientsforemotionalsymptoms,peerrelationships,
andhyperactiviessubscalesonparentalmigrationstatus,household
characteristics,socialsupport,andschoolperformance.................................................247
Table6.7.Regressioncoefficients(SE)fortotaldifficultiesscoreonparental
migrationstatus,householdcharacteristics,socialsupport,andschool
performanceforcurrentlyleft-behindchildrenonly........................................................249
Table6.8:Between-subgroupcomparisonsofadjustedmeantotaldifficulties
scorebykeycovariatesandparentalmigrationstatus....................................................251
12
ListofFigures
Figure2.1:TheABC-XContextualModeloffamilystress(Boss,2002).......................37
Figure2.2:Bronfenbrenner's(1979)ecologicalframeworkforhuman
development...........................................................................................................................................38
Figure3.1:Flowchartofarticleselectionprocedure...........................................................49
Figure3.2:Summaryframeworkofthequantitativeliterature......................................66
Figure4.1:Locationofthestudyprovinces...........................................................................133
Figure5.1:Conceptualframeworkofkeyfactorsintheimpactmechanismsof
parentalmigrationonchildpsychosocialwellbeing(AdaptedfromBoss,2002)202
Figure6.1:FrequencydistributionofSDQtotaldifficultiesscore...............................231
Figure6.2.FrequencydistributionofSDQsubscalescores............................................232
Figure6.3:Adjustedtotaldifficultiesscorebyparentalmigrationstatusand
covariates..............................................................................................................................................250
13
Chapter1 Introduction
1.1ScaleofmigrationandLBCpopulationsglobally
Overthepastfewdecades,unprecedentedmigrationflowsinmanypartsofthe
worldhavehadprofoundimpactsonfamilystructuresandthehomecommunities
ofmigrants(Lametal.,2013).In2013,232millionpeople–3.2%oftheworld's
population–livedoutsidetheircountryoforigin(UnitedNations,2013).Aneven
highernumber-anestimated763millionpeople-arelivingwithintheirown
countrybutoutsidetheirregionofbirth(UnitedNations,2013).Despitethat
familiesmaybeabletomigratetogether,thepopulationmovementsoftenleadto
separationbetweenhouseholdmembers,withavarietyofimplicationsonboth
sidesoffamilieslivingapart:themigrantsawayfromhome(abroadorinanother
partofthecountry),andtheleft-behindfamilymembersintheirhometown.In
particular,suchseparationisperhapsmostconcerningwhenitinvolveschildren
leftbyoneorbothmigrantparents,inthecareofothercaregivers(Asis,2006).
Inmostdevelopedcountries,theterm“leftbehind-children”isusedtodescribe
childrenwhofallbehindinwellbeing,mainlyrelatingtomaterialwellbeing,
educationandhealth.Toclarify,inthisthesis,“leftbehind-children”(LBC)refers
tochildrenwhoexperienceprolongedseparationfromtheirmigrantworker
parent(s).ThenumberofLBCishighinmanylow-andmiddle-incomecountries,
14
whereanincreasingnumberofparentshavejoinedtheglobalmovementof
workersrespondingtolaborshortagesinwealthierregions(Graham&Jordan,
2011).
Forinstance,overseasworking,includingalargenumberoffemalemigrant
workers,isparticularlycommoninsomeSoutheastAsiancountries.Areviewby
Bryant(2005)focusingonthreeSoutheastAsiancountriesestimatedthat,there
werethreetosixmillionchildreninthePhilippinesleftbehindbyparentsworking
overseas,approximatelyonemillionsuchchildreninIndonesia,andhalfamillion
inThailand.
EnormousimmigrationflowsfromLatinAmericatotheUnitedStatesalsoresulted
ingreatchallengesoftheleft-behindfamilies.Nobles(2013)reportedthatmore
than20%ofchildreninMexicoexperienceafather’smigrationbytheageof15.A
UNICEF(2007)reportsuggestedthatinEcuador,between1990and2005,the
numberofLBChasrisenfrom17,000to218,000.
EasternEuropeisanotherhighout-migrationarea,bothfromwithinandoutside
oftheEuropeanUnion,totheWesternhigh-incomecountries.InRomania,astudy
(Botezat&Pfeiffer,2014)reportedanestimated350,000children(over8%ofthe
childpopulation)werelivinginamigrantfamily,andnearly400,000childrenhad
atsomepointoneorbothparentsworkingabroad.InMoldova,itwasestimated
that22%ofallchildrenaged0-14hadoneorbothparentsabroad(UNICEF,2011).
15
DespitealackofreportsfromAfricancountriesonthenumberofLBC,thescaleof
migrationtowesterncountriesandwithinthecontinentorthecountryhasalso
beensignificant.TheWorldBank’s(2016)reportindicatedthatinternational
migrantsfromSub-SaharanAfricatotaled23.2million,ofwhom26%wereliving
inOECDcountriesand65.6%werelivingwithintheregion,whichinevitablylead
toalargenumberofchildrenleftbehindathome.
Whileoverseasmigrationisnowacommonphenomenoninmanypartsofthe
world,itispossiblethatevenmorepeoplearemigratingwithintheirown
countries,primarilyfromruraltourbanareas.Morethanhalfoftheworld’s
populationnowlivesinurbanareas(UnitedNations,2014),notleastduetothe
rapidurbanizationinAsiaandAfrica.
InChina,unevensocioeconomicdevelopmentisencouragingmigrationflowsto
thecitiesforbetterjobopportunities;themassiverural-urbanmigrationhas
driventhenumberofchildrenleftbehindinruralChinaupto61million,
accountingfor38%ofchildreninruralareasand22%ofallchildreninthe
country(All-ChinaWomen’sFederation,2013).Anestimated29millionLBClive
withneitherparent,andover2millionlivealone(All-ChinaWomen’sFederation,
2013).
16
1.2Contextsofmigrationandparent-childseparation
Globalizationinthepastfewdecadeshasresultedinanoverwhelmingtrend
towardtechnologicalandsocialintegration(Prempehetal.,2004),whichhas
affectedglobalmigrationpatterns.Thegenerallyrestrictivestancetowards
immigrationindevelopedcountriesnowcoexistswithastrongdemandforcertain
categoriesofmigrantworker,forinstanceindomesticserviceandagriculture.
Meanwhile,inthedevelopingworld,therecenturbanizationintheemerging
countrieslikeChinahasattractedtremendousmigrationflowsintothecities(Wu,
2007).Sincethereformandopeninguppolicyinthelate1970s,Chinahas
witnessedrapideconomicgrowthandmassiveurbandevelopment.Theprocessof
industrializationhasacceleratedruralreformsinChina,increasedagricultural
productivityandfreedagriculturallabor.Asaconsequence,largenumbersof
formerrurallaborershaveheadedintomoreurbanizedareas,lookingforbetter
employmentopportunitiesandsourcesofincome.Therefore,sincemid-1980s,
Chinahaswitnessedwhatsomescholarsdescribedasthelargestpeacetime
populationmovementinworldhistory(Roberts,2002).
Peoplemigrateacrosscountries,orwithinacountryfromruraltourbanareas,
mostlydrivenbythepotentialforimprovedeconomiccircumstancesandbetter
opportunitiesforhumandevelopment(UNICEF,2006).Althoughsuchincentives
usedtobeviewedonindividualbasis,nowitisnearlyimpossiblenottoconsider
thefundamentalroleplayedbythefamily.Hence,householdmigrationistypically
17
consideredalivelihoodstrategythatistakenupduetoeconomicdeficienciesof
thefamily(Laudy&Stark,1988;Stark&Lucas,1988)
However,duetostringententrypolicies,financialconstraints,andlimitedaccess
topublicgoodsinthedestination(Valtolina&Colombo,2012),inmanycases
familymembersareunabletomigratetogether.Consequently,hundredsof
millionsofchildrenaroundtheworldarebeingleftbehindbytheirmigrant
parents,manyofwhomarelower-skilledworkersfromdevelopingcountries
(Yeoh&Lam,2006).Inparticular,migrantworkersatthelowerendofthe
socio-economicspectrumoftenhavelittleornopossibilityoffamilyreunification
indestinationcountries,resultingintheirlong-termseparationfromleft-behind
familymembers.
LabormigrantsfromSoutheastAsiatypicallytakeupshort-termcontractsinthe
wealthiercountiesoftheregionandbeyond,betheywomenworkingasdomestic
servantsormenemployedatseaorinconstruction.Theirlimitedrightsof
residenceinhostcountriesmakeitlegallyorpracticallyimpossibletobringa
family.MillionsofLatinAmericansintheUSandinEuropeareunabletobring
theirchildrenbecausetheparentsareundocumentedimmigrants;evenlegally
residentmigrantsoftenfacebureaucraticandfinancialobstaclestofamily
reunification,whichcanresultinyearsofunwantedseparation(Carlingetal.,
2012).China’surbanlocalpublicresourcedistributionandmanagementsystem
posesformidableobstaclesforrural-to-urbanmigrantsaccessingpublicgoodslike
18
primaryeducationandmedicalcare.Also,rurallaborershavelowerpay,more
crowdedlivingconditions,andmorephysicallydemandingjobs.Thishasresulted
inthemajorityofmigrantworkershavingnochoicebuttoleavetheirchildren
livinginruralhometowns.Leavingchildrenbehindhasbecomeacommon
phenomenonnotonlyforlow-incomefamiliesbutalsoacrossthesocio-economic
spectruminruralChina(Heetal.,2012)
1.3Careenvironmentofleft-behindchildren
CareisdefinedbyStanding(2001)astheactofseeingtoaperson’s“physical,
psychological,emotionalanddevelopmentalneeds”.Aschildren’swellbeing
comprisesavarietyofelements,theimpactfromparentalabsenceandeffectsof
differentcharacteristicsofthecareenvironmentarecomplex.Ingeneral,
standardsoflivingshouldriseinhouseholdsthatreceiveremittancesfromfamily
migrantlaborers(Sofranko&Idris,1999).However,evidencehasindicatedthat
parentalmigrationaffectschildwellbeingthroughatrade-offbetweenincreased
familyincomeanddisruptedparentalcare(Valtolina&Colombo,2012).Thequality
ofsubstitutecarearrangementsisanimportantfactordeterminingthedegreeto
whichparentalabsenceimpactsthecare,stimulationandsupervisionofchildren
(Parreñas,2008).
Forthefamilieswhereoneorbothmigrantparentsliveapartfromthefamily,
childrenleft-behindareusuallyleftinthehometowntoberaisedbyacaregiver,
19
whomaybetheotherparent,anextendedfamilymember,orinsomecasesa
non-kinrelatedperson.Whileinmanycountriesitwastraditionallythefatherwho
wouldleaveandbecomethebreadwinnerofthefamily,nowitisincreasingly
frequentthatthemotheralsomigrates,separatelyortogetherwiththefather.
Regardlessofthemigrationpatternofthefamily,infulfillingchildcare
responsibilities,caregiversarefacedwithnumerouschallenges,including
inadequatesupport,socialstigma,isolation,disruptedleisureandretirementplans,
age-relatedadversities,angertowardthecustodialchildrenandfinancial
difficulties(Butler&Zakari,2005;Daly&Glenwick,2000;Whitleyetal.,2001).
Asiancareregimesremainstrongly‘familistic’,thatis,familieshavetheprincipal
responsibilityfortheirmembers’welfare,beitintermsofsharingincomesor
providingcaretothoseinneed(Esping-Andersen,2000).Unlikewesterncountries,
organizationofcareinmanypartsofAsiadoesnotheavilyreplyonthestate,
marketorevencommunity.TheconceptoffamilycareinmanyAsiancultures,
fromhigh-incometolow-incomesettings,involvesactiverolesofextendedfamily
members,especiallygrandparents(Ochiai,2009).InSoutheastAsia,theextended
familyadultshasbeenfoundtohavemajorinfluenceoverallaspectsofmigration
includingparticipatinginthemigrationdecision,fundingthetripandtakingcare
ofleft-behindchildren,therebymitigatingthesocialcostsofmigration(Bryant,
2005).MeanwhileinChinawherethefamilycarecultureislargelysimilar,
researchalsosuggestedthatintensivegrandchildrencare,albeitacommon
20
practice,canbebeyondthedefinedobligationsofgrandparentsandconsidered
moreafavorthanaformalobligation(Cong&Silverstein,2011).
China’s“OneChildPolicy”createdauniqueimpactonfamilystructures.The
enforcementofOneChildPolicy,despitethemorerelaxedvariationsofitincertain
areasofthecountry,hasbeenfoundtobeariskfactorforaseriesofemotional
andbehavioralproblems,basedondatafromeconomicsexperimentsconducted
withgroupsofindividualsbornjustbeforeandjustafterthepolicy’sintroduction
in1979(Cameronetal.,2013).Ahighproportionofsinglechildfamiliesin
relativelyunderdevelopedareasinruralChinashouldbeconsideredanimportant
characteristicofLBC’scareenvironment.Asrapideconomicgrowthcreatedmore
humandevelopmentopportunitiesinChina,thosewhoaretheonlychildmaybear
higherexpectationsfromthefamily,butreceivelesssupportandbemore
vulnerabletopsychosocialchallengesoncesetbackstakeplaceintheirlives(Wang
&Guo,2010).
Whatchildrenreceivefromtheirmigrantparentscanbebroadlycategorizedinto
twogroups:economicresourcesandsocio-culturalorientations(McLanahan&
Sandefur,1994).Additionalfinancialresourcesfromremittancessentbymigrant
parentscanbeusedtofacilitatechilddevelopment,includinginvestmentin
educationandthelivingenvironment.Assetssuchasgoodhousingcondition,
adequatefoodandnutrition,highqualityschoolandneighborhood,andadditional
opportunitiesforskillbuilding,arecrucialinpromotingchilddevelopment(Wen,
21
2008).Remittanceshavebeenshowntohaveapositiveeffectonnutritionand
generalhealth(Constant&Zimmermann,2013).Ontheotherhand,LBCcan
benefitfromnewinformationandideasinamoredevelopedurbansetting,that
aretransmittedbytheirparentsandwhichhavebeentermed“social
remittances”(Levitt,1998).Migrantparentsmayrepresentanadvancedmodern
lifestylethatthechildcouldliveupto,andprovideaspirationsofpursuingsuccess
inlife(Pribilsky2001).
1.4PoliciestosafeguardthewellbeingofLBC
Whetherlabormigrationisaneffectivelong-termpovertyalleviationstrategy,at
bothhouseholdandnationallevels,isstilladebatedissue(Adams&Page,2005).
Althoughmigrationwasdrivenbyeconomicincentives,theincreaseindisposable
incomeandmaterialresourcesarenotguaranteedintheleft-behindfamily
members.Particularly,childrenarevulnerabletotheeconomicshocksand
instabilityinthemigrantreceivingareas(Janson,2014).Forexample,inChina,
partlybecauseofthepreferenceforinvestinginnewhousingratherthanin
privatesavingschemes,veryfewmigrantfamilieshaveliquidassets(Janson,
2014).Consequently,theuncertaintyineconomicbenefitsfromremittancesis
rathersignificant.
Thislackofeconomicsecurityrequirestargetedpolicyandinterventionfromlocal
governmentsaswellasnongovernmentalorganizations.Corporatesocial
22
responsibilityfromtheenterpriseswheretheparentsareworkingshouldalsobe
strengthened,inadditiontoprovidingbasicemployeebenefits.Decenthousing,
accessiblepreschoolsandschoolsareamongtheessentialconditionstoenable
familiestobringtheirchildrenwiththem.
Whilepolicymakersandacademicshavebecomeincreasinglyawareofthe
economicroleofmigration,lessattentionhasbeenpaidtothepsychosocial
ramificationsforleftbehindfamilymembers(dalaGarza,2010).Yetanumberof
studieshaveindicatedthedisadvantagesinavarietyofmeasuresofpsychosocial
wellbeing,includingdepression,anxiety,loneliness,andqualityoflife(seethe
literaturereviewchapter).
InChina,theincreasingevidenceaboutthesocialanddevelopmentalcostsof
migration,ledseveralcentralgovernmentagenciestojointlycallonlocal
authoritiestoprioritizeimprovementsintheeducationandcareofLBC
(GovernmentofChina,2013)inordertomitigatetheadverseeffectsofparental
absence.MostrecentlyinFebruary2016,theStateCouncilofChinaissuedasetof
guidelinescallingfortheestablishmentofanenhancedsystemthatengages
families,governments,schoolsandsocialgroupsinprovidingbettercareand
protectionto“left-behindchildren”(StateCouncil,2016).Arescuemechanism
thatincludesinterventionmeasuressuchasmandatoryreporting,crisis
managementandassistanceevaluationwillpreventinfringementsontherightsof
thesechildren.Theguidelinesencouragegovernmentstocontractcharitiesand
23
voluntarybodiestoprovideprofessionalservices,includingmechanismsof
compulsoryreporting,intervention,assessmentandhelp.Theseorganizations
mustreportcasesthatinvolveinjury,abuseorotherharmtotheappropriate
authorityandinterveneassoonaspossible.
Theguidelinesalsolayoutmeasurestograduallydecreasethenumberof
left-behindchildren.Thegovernmentwillprovidemoreassistancesuchas
grantingfamiliesofmigrantworkersurbancitizenshiporsubsidiesinhousingor
education.Ruralmigrantworkersarealsoencouragedtoreturntotheir
hometownsandstarttheirownbusinesses.Itisexpectedthat,by2020,Chinawill
improvelawsandregulationsregardingchildprotection,andcreateasafer
environmentforchildrentogrowupespeciallyinruralareas.
Insomeothercountries,thegovernmenthasalsotakenmeasurestosafeguardthe
wellbeingofLBC.Forexample,RomaniaissuedaGovernmentalOrdinancein2006,
solicitingparentstoinformauthoritiesoftheirleavingintentions,andregistera
guardian,inordertomonitortheLBCespeciallythose“atrisk”,raiseawareness
andprovidepertinentservices.HowevertheimplementationoftheOrdinance
soonbecameineffective.Thenanewregulationwasissuedin2009,asalaw
obligingmigrantparentstonominateaguardianmeetingseveralcriteria(a
relativewithatleasttheminimumincomeguaranteedandatleast18yearsolder
thanthechildtobecaredfor,fullphysicalandpsychologicalcapacity,nodisabling
conditionsorpreviousparentingconvictions,nomorethanthreeotherchildrenin
24
care).Therearesubstantialfinesforparentsbreakingthelaw.Monthlymonitoring
visitsbythesocialworkers,withtheultimateresponsibilityforchildreplacement
instatecare,arestipulated(Pantea,2012).
EventhoughRomaniahasabroadschemeofsocialassistanceprovisionsand
services(Cerami,2006)especiallyforchildren,thehighpressurefromparental
migrationhasfoundthesystemunprepared.Somedeprivedmigrant-sending
communitiesintheeasthavethepoorestresourcesforassistanceand
interventions.Moreover,localauthoritiesmayhavetheassumptionthatmigrant
familiesdeveloptheirowninternalarrangementsandthus,relievethestateofits
role(Luca,2007).
1.5ProgramstoimprovethewellbeingofLBC
Despitethetremendouschallengesfacedbytheleft-behindfamiliesandthe
migrantsendingcommunitiesglobally,onlyfewspecificpoliciesandprograms
thataddressthewellbeingofLBChavebeenreported.
InChina,despitethegovernment’sadvocacyandconstantefforts,veryfew
systematiccareandsupportprogramsforruralLBCarebeingimplementedinthe
field,partlybecausethenon-profitsectorinChinaisundermanypoliticaland
administrativeconstraints.UNICEFChinahasreportedonpreliminary
achievementsofa“ChildWelfareDirectors”programinafewprovinces,which
25
recruitsgrassrootssocialworkerstobridgeservicegapsforvulnerablechildren,
especiallyLBC(UNICEF,2014a).Yetthisapproachdoesnotprovideessential,
concreteservicestochildrenotherthanconnectingthemwithavailableresources
andinstitutions.Neitherdoesithaveaphysicalspaceasacommunitylevel
platform.
VeryfewprogramstargetingLBCwerereportedinothercountriesaswell.One
program(Givaudan&Pick,2013)wascarriedoutinMexico,bytheMexican
InstituteofFamilyandPopulationResearch(IMIFAP).IMIFAPhasestablisheda
numberofCommunityChildDevelopmentCenters(CCDCs)thatfocuson
mitigatingthenegativeeffectsofparentalmigration,intheruralareasoftwo
Mexicanstates.BasedattheseCenters,abroadlifeskillstrainingprogram(e.g.,
cognitive-problemsolving,expressionandmanagementofemotions,
social-conflictresolution)isconducted,targetingLBCandvolunteercommunity
promoters/caregiversofchildrenunder12yearsold.
Theprogramhasbeensuccessfulintrainingpromoterswhointurnhavebecome
changeagentsintheirlocalities.Migrantparents,aswellascaregiverssuchas
grandmothers,relatives,andneighbors,acceptedtheprogramandreported
changesintheperceptionoftheirrolesinthecaringanddevelopmentofthe
babies.Positivebehavioralchangeswerealsofoundastheyattachmore
importancetohealthandnutritionissues.
26
1.6Researchquestionandstudyaims
InlightofthebackgroundsoftheLBCissues,withaspecificfocusonfamilies
affectedbyrural-to-urbanmigration,anextensiveliteraturereviewwasconducted
whichidentifiedafewmajorgaps.First,longer-termimpactofparentalmigration
onchildrenwasseldomexplored,asonlyveryfewstudiesindicatedthelasting
effectsaftermigrantparentsreturned.Second,therelationshipqualityand
dynamicsremainedunclearinLBC’sfamily,especiallybetweenmigrantparents
andtheirchildren.Third,howtheleft-behindfamilyandtheirsocialenvironments
providecareandsupporttochildrenhasnotbeencomprehensivelyassessed.
Thisthesisaimedtoanswerthisoverarchingresearchquestion:whatarethe
impactmechanismsofparentalmigrationonchildren’spsychosocialwellbeing?In
thecurrentqualitativeandquantitativestudies,thiswastobeaddressedby
exploringhowparentalmigrationmayaffectpsychosocialwellbeingofchildren
overalengthyperiodoftime,andwhataretherolesofparent-childrelationship,
familycareandsocialenvironment,inthemechanismsorpotentialcausal
pathwaysthatconnectparentalmigrationwithchildpsychosocialwellbeing
status.
1.7Thesisstructure
27
Chapter2providesanoverviewoftheoreticalperspectivesthatarerelevanttothe
mechanismsofparentalmigrationanditsimpactonchildren.Chapter3isa
comprehensivereviewofbothquantitativeandqualitativestudiesonthe
wellbeingofLBC.Chapter4introducesthesocioeconomicdevelopment,andpolicy
andadministrationcontextsofthestudyareas.Chapter5isaqualitativestudy
examiningexperiencesofchildren,andrelationships,careandsupportinthe
familieslivingapart,throughin-depthinterviewscarriedoutinonestudycounty.
Chapter6isaquantitativecross-sectionalsurveyinvestigatingpsychosocial
wellbeingoutcomesandchildandfamilycharacteristics,basedonasampleoffive
countiesintwoChineseprovinces.Chapter7isthefinaldiscussionandconclusion
basedonfindingsacrossbothqualitativeandquantitativestudiesinthisthesis.
28
Chapter2 ReviewofTheoreticalPerspectives
2.1Introduction
Thischapterreviewstheoriesthatarerelatedtochildwellbeinginmigrant
familiesfrommultipledisciplines.Accordingtotheoverallresearchquestionof
thisthesis,migrationinthesociety,familycharacteristics,andchildwellbeingare
thethreelogicallevelsordimensionsthatareinvolvedinthecausalpathways
andmechanismswhenexaminingthepsychosocialwellbeingofLBC.Connections
andinteractionsbetweenthesedimensionsarealsosuggestedinthe
interdisciplinarytheoreticalapproaches.Thereviewedtheorieswillbe
summarizedattheendofthisChapter,withreferencetotheaimsofthisthesis.
Linksbetweenthetheoreticalreviewandthefollowingchapters,includingthe
empiricalreviewaswellasthecurrentstudies,willalsobehighlighted.
2.1Migrationtheories
2.1.1Thepush-pullmodels
Theearliestmigrationtheoristsconcludedthatmigrationwasgovernedbya
"push-pull"process(Dorigo&Tobler,1983).Thissimplemodelconceivesof
migrationasdrivenbyasetofpushfactorsoperatingfromtheregionorcountryof
origin(poverty,unemployment,landlessness,rapidpopulationgrowth,political
29
repression,lowsocialstatus,poormarriageprospectsetc.),andpullfactors
operatingfromtheplaceorcountryofdestination(betterincomeandjob
prospects,bettereducationandwelfaresystems,landtosettleandfarm,good
environmentalandlivingconditions,politicalfreedometc.).
Ravenstein's1889"LawsofMigration"statedthattheprimarycauseformigration
wasbetterexternaleconomicopportunities;thevolumeofmigrationdecreasesas
distanceincreases;migrationoccursinstagesinsteadofonelongmove;
populationmovementsarebilateral;andmigrationdifferentials(e.g.,gender,
socialclass,age)influenceaperson'smobility.
ManytheoristsdevelopedvariationsofRavenstein'sconclusionsinthelater
migrationtheories.EverettLee(1966)reformulatedRavenstein'stheorytogive
moreemphasistointernal(orpush)factors.Leeintroducedasetof“intervening
obstacles”whichhavetobeovercome;examplesincludephysicaldistance,costof
makingthejourney,culturalbarrierssuchaslanguageanddifferentwaysoflife,
andpoliticalobstaclessuchasinternationalbordersandimmigrationrestrictions.
PersonalfactorsalsoplayaroleinLee’stheorizationofmigration:differentpeople
willreactdifferentlytovariouscombinationsofpushesandpulls,accordingto
theireconomicstatus,life-stageandpersonality.Forexample,asingle,
unemployedyoungadultwillrespondmoredirectlytojobandincomefactorsand
belessconcernedabouttheeducationsystemofadestination,ascomparedto
someonewhoismigratingwiththeirchildren.
30
2.1.2Neoclassicaleconomicstheory
Thepush-pullmodelsdominatedmuchmigrationthinkingduringthe
mid-twentiethcentury,untilthe1960sifnotlater.Manyofthemigration
paradigmsarethenreflectedintheneoclassicaleconomicstheory,whichisbased
onprinciplesofutilitymaximisation,rationalchoice,factor-pricedifferentials
betweenregionsandcountries,andlabormobility.Migrationcanbeassumedasa
rationalchoicewithinthebroadersocialstructuresofrural-urbanand
developmentalinequalities,thatis,abundantcapital,laborshortagesandhigh
wagesincertaincountriesorurbanareas,andtheoppositeinothercountriesor
ruralareas(Masseyetal.,1999).
Macroeconomically,neoclassicaleconomicssuggeststhatinternationalmigration
isrelatedtotheglobalsupplyanddemandforlabor(Sjaastad1962;Todaro1969).
Nationswithscarcelaborsupplyandhighdemandwillhavehighwagesthatpull
immigrantsinfromnationswithasurplusoflabor.Atthemicrolevel,migrationis
theresultofdecisionsmadebyindividual“rationalactors”whoweighupthepros
andconsofmovingrelativetostaying,orthereturnsagainsttheinvestmenttothe
humancapitaloftheindividual,basedonabundantinformationabouttheoptions.
Sjaastad(1962)interpretedtheresultsofthiscost-benefitcalculusasadecisionto
migratebasedonreturnstotheindividual’sinvestmentinhisorherhuman
capital.
31
2.1.3Householdeconomicstheory
Yetmostoftheseprevioustheorieshadleftoutthehouseholdcontextofmigration.
Migrationdecisionsmaynotbeindividuallymade,butarelikelytobejointlytaken
withintheambitofthehousehold,andfordifferentmembersofthehousehold
(King,2012).Sometimesthescaleofthedecision-makingunitmovesfurtherinto
theextendedfamiliesandevenwidercommunalgroups(Masseyetal.,1998);this
isparticularrelevantinafamilistandcollectivistculture,forexample,the
traditionsinruralChina.
Hence,thenewhouseholdeconomicstheoryshiftsthedecision-makingfromthe
individualandarguesthattheappropriateunitsofanalysisformigrationresearch
arefamiliesandhouseholds(Stark&Bloom,1985).Akeyinsightofthisnew
approachisthatmigrationdecisionsarenotmadebyisolatedindividualactors,
butbylargerunitsofrelatedpeople–typicallyfamiliesorhouseholds–inwhich
peopleactcollectivelynotonlytomaximizeexpectedincome,butalsotominimize
risksandtoloosenconstraintsassociatedwithavarietyofmarketfailures,apart
fromthoseinthelabormarket(Massey,1998).
Ithasbeenarguedthatespeciallywithinthecontextofdevelopingcountries,an
individualismotivatedtomovenotonlyforhis/herowngoalsbutalsoforthe
survivalofthehousehold(Lauby&Stark,1988).Theperspectiveofthenew
32
householdeconomicsbecomesespeciallyrelevantinthesecontexts,becausepoor
familiesindevelopingcountrieslackinstitutionalmechanismsofprivateinsurance
marketsandgovernmentalprogramsthatminimizehouseholdrisks,whichare
morecommoninmanydevelopedcountries.Thus,thereistheincentiveto
self-insurebysendingoneormorefamilymemberstoacityorabroadtoremit
earningsthatguaranteesfamilyincomeandreducesrisksincurredthroughcrop
failures,croppricefluctuations,andunemployment(Masseyetal.,1998).Alarge
proportionofrural-urbanmigrantsindevelopingcountriesremitasignificantpart
oftheirearningstotheleft-behindfamilymembersathome,therebyreducingthe
incomevarianceassociatedwithworkinagriculture(Lauby&Stark,1988).
2.1.4Segmentedlabor-marketandworldsystemstheory
Inaddition,fromamoreglobalperspective,segmentedlabor-markettheory(Piore,
1983)arguesthatFirstWorldeconomiesarestructuredsoastorequireacertain
levelofimmigration,primarilydrivenbypull,notpushfactors.Thistheory
suggeststhatdevelopedeconomiesaredualistic:aprimarylabormarketofsecure,
well-paidjobsfornativeworkers;andasecondarylabormarketoflow-skill,
low-wage,insecureandgenerallyunpleasantjobsinfactoriesandtheservice
sector,filledmainlybymigrantworkers.Totheextentthatthesecondarylabor
marketmaybesplitintoemploymentsubsectionsaccordingtogender,raceor
nationality,itbecomessegmented.Onthewhole,thecreationofthesejobs
precedesthemigrantswhofillthem(Samers,2010).Similarly,world-systems
33
theory(Sassen,1988)arguesthatinternationalmigrationisaby-productofglobal
capitalism.Contemporarypatternsofinternationalmigrationtendtobefromthe
periphery(poornations)tothecore(richnations)becausefactorsassociatedwith
industrialdevelopmentintheFirstWorldgeneratedstructuraleconomicproblems,
andthuspushfactors,intheThirdWorld.
2.2Theoriesinfamilystudies
2.2.1FamilyStressManagement
Theglobalandsocietalmigrationpatternsarethelargercontextsoftheprocesses
inwhichindividualfamiliesareinvolved.Migrationnotonlyinvolveseconomic
activities,butalsoaffectsfamilystructureandfunctioning.Absenceofthefather
and/ormotherhasdirectimpactontheirrelationshipsandinteractionswith
householdmembers,especiallytheirchildren.Additionalcaregivingrolesof
residentcaregivers,e.g.grandparents,mayfurtheraffectdynamicswithinthe
nuclearandextendedfamiliesacrossgeographicaldistance.Therefore,migrant
parentsandtheleft-behindfamilymemberscompriseafamilysystem,whichmay
beconceptualizedasaunitthatmanageslossorabsence,throughoutarangeof
relationshipswithchildren,parents,aswellasgrandparentsorothercaregivers.
Besidesphysicalseparationduetomigration,variousstressorswithinthefamily
networkmayinterferewiththeparent-childdyadsandtheirrelationship
dynamicsthatpromotechilddevelopment(Sroufe,etal,2005;Murray&Murray,
34
2010).Familystressisaprocessoffamilychangeratherthansimplyaneventor
situationthathappensinortoafamily(Boss,2002,p12;Walker,1985).Inthis
process,familycopingwasdefinedasthegroup’smanagementofthestressful
eventorsituation(McCubbin,1979).Boss(2002,p79)contendedthatafamilyas
agroupisnotcopingfunctionallyifevenonemembermanifestsdistress
symptoms,thustheexplorationofbothindividualandgroupindicatorsis
essentialinassessingfamilycoping.BuildingonLazarus’s(1966,1976,1977)
workonindividualcoping,Boss(2002,p79)developedtheconceptoffamily
copingasthecognitive,affective,andbehavioralprocessbywhichindividualsand
theirfamilysystemasawholemanageastressfuleventorsituation.Itis
noteworthythatwhileeconomicresourceisbeneficial,itisnotapredominant
factorinthisprocess,whichismorerelevanttotheoverallfunctioningofthe
familyanditsmembersinthepsychosocialdimension.
CopingbehaviorsaredefinedbyLazarus(1966,1976)as(a)directaction
behaviorsthatareusedtochangeastressedrelationwithone'sphysicalorsocial
environmentand(b)intrapsychicformsofcoping,whicharedefensemechanisms
(e.g.,detachmentordenial)usedtoreduceemotionalarousal.Theactionsand
thoughtscannotchangethesourceofthestress,andthusarereferredtoas
emotion-focusedcoping.Suchcopingbehaviorscanoftenbeidentifiedinmigrant
families,bothamongchildrenandadultfamilymembers.Forexample,children
maytendtoseekemotionalsupportfromtheirpeersorplaymatesinabsenceof
migrantparents;ortheymaytrytorationalizetheseparationdistressthroughan
35
understandingoftheoverallhouseholdbenefitsfrommigration.
Familiescanalsouseproblem-focusedcopingtomasterormanageastressorand
providechilddevelopmentsupport(Lazarus,1976;Papaliaetal.,2001).The
initialdecisiontomigratecanactuallyberegardedascopingwithfamily
economicstressors.Thenanotherscenariomaybethatmigrantparentsmove
backhomeorbringtheirchildrentolivetogetherinthecities,toresolvethe
stressduetoseparation.
Attachmenttheorycontributestoelucidatingthecognitive-emotionalprocessin
dyadsofparentsandschool-agechildren(tobefurtherdiscussedinsection2.3),
asattachmentstylesinyoungchildrentendtopersistovertime(Ainsworthetal.,
1978,p320;Main,1996).Yetattachmentbehaviorsarereferredtolessfrequently
duringadolescence,whenthebalanceofattachmentandexploratorybehaviors
shiftstowardsmoreindependence,especiallyinthephysicalabsenceofthe
parents(Allen,2008,p419).Asaresult,theoriesincopingbehaviors,whichcan
beappliedbothatindividualandfamilylevels,maybebettersuitedinexplaining
thebehavioralconsequenceandfamilysupportforolderleft-behindchildren.
2.2.2FamilyContext
Familycontextisalsocriticalinunderstandingtheprocessofstressmanagement.
Boss(2002,p15)developedherContextualModelofFamilyStressby
36
surroundingHill’s(1949,1958)ABC-XFamilyStressModel(Figure2.1).In
additiontostressevent(A),resources(B),perceptions(C),anddegreeofstress
orcrisis(X),sheproposedthreeinternalandfiveexternalcontextsthatwould
influencehowafamily’sstressprocessplaysoutovertime.
Theinternalcontextsinclude(a)structural,(b)psychological,and(c)
philosophicaldimensionsofthefamilysystemitself(Boss,2002,p45).The
structureofmigrantfamiliesinvolvesfamilycompositionaswellascaregiving
arrangements.Psychologicalcontextmayrefertothemigrantfamily’sinherent
mechanismsinperception,appraisal,definition,orassessmentofseparationand
loss.Philosophicalcontextindicatesvaluesandbeliefsatthemicrolevelfamily
environmentthatmaydifferfromthesocialculture,forexample,decision-making
onreturningtothemigrant-sendingcommunityfromacity,inconsiderationof
childeducationanddevelopment.
Thefiveexternalcontextsarecomponentsoverwhichthefamilyhasnocontrol
andyetinwhichtheyarefirmlyembedded:(a)heredity,(b)development,(c)
economy,(d)history,and(e)culture(Boss,2002,p69).Hence,themigratory
decisionsandfamilycopingstrategiesareshapedbythespecificcharacteristicsof
themigrant-sendingcommunityaswellasthedestinationcities,sincesupportive
orrestrainingfactorsatcommunitylevelmaydirectlyaffectfamilyenvironment.
Meanwhile,historicalandculturaldeterminantsinthecountryandgloballythat
arerelatedtomigrantfamiliesalsoneedtobeconsideredascrucialexternal
37
contexts.
Figure 2.1: The ABC-X Contextual Model of family stress (Boss, 2002)
2.2.3Socio-ecologicalapproachandfamilysystemstheory
Thesocio-ecologicalmodelonhumandevelopment(Bronfenbrenner,1979)
underscoresthefactthatfamiliesareembeddedinlargersocialsystemswhich
thechildmaynotseemtodirectlyfunctionin,butmaynonethelessbehighly
influential.Thereforetheentireecologicalsysteminwhichchildgrowthand
developmentoccursneedstobetakenintoaccount.Thefivelayersofecological
system(Figure2.2)includemanyextra-familialfactors,suchascommunity
settings,localpolitics,theworkplace,andthesocioculturalcontext.
38
Figure 2.2: Bronfenbrenner's (1979) ecological framework for human
development
Laterdevelopmentoftheoriginalframeworkpresentedmultipledimensions,asa
Process–Person–Context–Timemodel(Bronfenbrenner&Morris,1998).
Accordingtotherevisedmodel,humandevelopmenttakesplacethrough
processesofprogressivelymorecomplexreciprocalinteractionsinitsimmediate
familyenvironment,onafairlyregularbasisoverextendedperiodsoftime.
Further,theform,powerandcontentanddirectionofsuchprocessesvary
systematically,asajointfunctionofthecharacteristicsofthedevelopingperson
andbothimmediateandmoreremoteenvironment.Theperspectiveofthese
39
processesisadaptedinthisthesiswheninvestigatingtheconnectionsbetween
differentvariablesofthecontext,individualcharacteristicsandoutcomesof
interest.
Children’sattachmentstoparticularcaregiversinfluence,andareshapedby,the
broaderfamilycontext.Familysystemtheoryisconcernedwithfamilydynamics,
involvingstructures,roles,communicationpatterns,boundaries,andpower
relations.Itisrelativelymorefocusedonadultsandcurrentfamilyfunctioning
(Rothbaumetal.,2002).Whilefamilysystemtheoryrecognizesthatcoparenting
dynamicsaredistinguishablefromdyadicparent-childrelationship(Erdman&
Caffery,2013,p43),thesurrogateparentingbygrandparents,accompaniedby
distantparentingofmigrantparents,createsamorecomplexinteractionpattern
involvingavarietyoffactors,withinandoutsidefamilysystems.
2.3AttachmentTheory
2.3.1Theearliestattachment
Currentliteratureinfamilystudiesandchildpsychologydrawonattachment
theorytounderstandthelinksbetweenchildpsychologicalhealthandparental
absenceorpresence,throughthelensofchildren’srelationshipswithparents
and/orresidentcaregivers.Bowlby(1969)indicatedthattheearliestbonds
formedbychildrenwiththeircaregivershaveaprofoundimpactthroughoutlife.
40
Theinitialattachmentachildformsisgenerallytoitsmotherorotherhabitual
caregiver,althoughitmaybesupplementedovertimebyattachmentstoother
persons(Ainsworth,1969;Bowlby,1982,p40).Theactualorthreatened
disruptionofattachments,alsoknownasthe“strangesituation”,canleadachild
toexperiencedepression,anxiety,orangerwhenproximitycannotbemaintained
(Ainsworthetal.,1978,p116).
Thecaregiverorattachmentfigureprovidesasecurebaseforinfantsandchildren,
fromwhichtheyexploretheenvironment,formothersecurerelationships,seek
supportwhenneeded,anddrawstrengthsfromthesupport(Ainsworthetal.,
1978,p135).Youngchildrentendtoformattachmentrelationshipswithmore
thanonecaregiver(Ainsworth,1969;Schaffer&Emerson,1964).Howeverthe
principalattachmentfigureplaysthemostimportantroleinachild’semotional
wellbeing(Bowlby,1980,p308).Childrendevelopexpectations,or“internal
workingmodels”,ofthecaregiver’savailabilityandresponsivenessbasedontheir
previousinteractions(Bowlby,1980,p81).AsJulietHopkins(1990,p467)
explained:
Ifachildhasexperiencedreliablyresponsivecaregiving,he/shewill
constructaworkingmodeloftheselfascompetentandloveable,butif
he/shehasexperiencedmuchrebuffhe/shewillconstructamodelofthe
selfasunworthyofhelpandcomfort.
41
2.3.2Developmentanddisruptionofattachmentrelationship
Inveryyoungchildren,attachmentstendtobedependencyrelationships
characterizedbyadrivetomaintainphysicalproximitytoaprimarycaregiver.
Withage,childrencontinuallyreviseandexpandtheirinternalworkingmodelas
theircognitive,emotional,andsocialcapacitiesincrease,andastheyinteractwith
parents/caregivers,formingotherclosebondswithadults,siblings,andfriends
(Berk,2012,p421).Buildingupontheinternalworkingmodels,olderchildren
andadultscontinuetomonitortheavailabilityofattachmentfiguresaccordingto
theiractualbehaviors(Bowlby,1973,p52).Thismonitoringmayoccur
increasinglythroughdistantformsofcommunicationasayoungchildgrowsup
(Kobak&Madsen,2008,p24),whichlatertransformsintoadrivetomaintain
symbolicproximityvialessdirectcommunication,suchasthroughphonecalls,as
anindividualmatures(Vanoreetal.,2015).
Whenlinesofcommunicationaredisruptedbytheattachmentfigures’prolonged
absence,emotionaldisengagement,orsignalsofrejection,olderchildrencan
developnegativeemotionssimilartoyoungchildren’sreactionstophysical
separation(Kobak&Madsen,2008,p24).Interruptionsinattachment
relationships,orthedevelopmentofunresponsiveorunpredictableattachment
relationships,shakethesenseofsecurityanindividualderivesfromattachments
(Armsden&Greenberg,1987).Insecurelyattachedchildrenaremorelikelyto
respondtolossorstresswithdysfunctionalthoughtsandemotions(Kobak&
42
Madsen,2008,p26;Sroufeetal.,1999).Consequently,migrationofacaregiver,
andthedisruptiontophysicalproximityitmaybring,maycausesignificant
psychologicaldistressamongrecipientsofcare.Factorsthatcontributeto
attachmentrelationshipsincludedurationandqualityofcare,emotional
investment,socialcues,andpresenceofthecaregiver(Colin,1996;Cassidy,2008,
p15).
2.3.3Migratoryseparationandattachment
Althoughattachmenttheorymayapplytodifferenttypesofparent-child
separation,itisimportanttonotetheuniquefeaturesofmigratoryseparation
andthelimitationsoftheattachment-basedinterpretation,ascomparedto
parentalabsenceinotherscenariossuchasincarcerationorambiguousloss.
Separationduetoeconomicmigrationisaphenomenonthatemergedonan
unprecedentedscaleonlyduringthepastfewdecades,andinmostcasesthe
migrationisdrivenbyparents’desiretoimprovethecircumstancesand
wellbeingforthewholefamily,particularlychildren(Rapoport&Docquier,2006).
Consequently,olderchildrenwithincreasedcognitiveunderstandingareunlikely
toperceivetheseparationasintentionalrejectionandabandonment.During
migration,openlinesofcommunicationandfrequentreunionbetweenmigrants
andtheirchildrenarefeasiblewiththehelpofmoderntechnologyand
transportation,whichmaycontributetorelievingthestressduetotheprolonged
43
physicalabsence.
Mostchildandfamilypsychologystudiesusingattachmenttheorylargelyfocused
ontheseparationofparentsfromchildrenowingtocrisiseventssuchas
incarceration,death,ordivorce(Mazzucato,2014).Thesestudieshavefound
strongnegativerepercussionsofparentallossforchildren'semotionalwell-being,
particularlyincasesinwhichprolongeduncertaintyaboutthepermanenceofloss
oritscause(ambiguousloss)blockedappropriatecopingandstressmanagement,
resultinginthedeteriorationoffamilylifewhentasksandroleswerenot
reassigned(Carrolletal.,2007;Boss,2004).
2.4Summary
Thischapterreviewedtheoreticalperspectivesinrelationtomigration,family
systems,andchild-parentattachment,illustratingapreliminaryframework
structurethatentailsthesystemsorenvironmentsinwhichthechildlives,from
themacro-levelsocietytothemicro-levelhousehold.Althoughtheconceptual
frameworkneedstobefurtherdevelopedbasedonourempiricalstudies,existing
theorieshaveprovidedusefulinputsfromvariousperspectivesofthemechanisms,
betweenindividualwellbeingstatusandthelargersocialphenomenonof
migration.
Thereviewedtheoriessuggestthatthecoreoutcometobeinvestigatedinour
44
studies,childpsychosocialwellbeing,isdeterminedbytherelationshipswith
parentsandcarefromcaregivers,whichinvolvesacomplicatedprocessshaped
bythechild’svariousinteractionswithindividualsandenvironments,during
infancytoadolescence.Astheleft-behindchildgrowsup,thewellbeingstatusand
familyconditionsmayalsoevolveovertime,throughaseriesofdynamicsinthe
extendedfamily.Intheabsenceofmigrantparents,thespecificfactorsthatmay
influencechildwellbeingincludefamilycarearrangementsandavailabilityof
psychosocialsupport,aswellashouseholdsocioeconomicstatus.Theseaspectsat
familylevelarefurtheraffectedbythecorrespondingfactorsinthelargersocial
contexts,whichcontributetotheconfigurationofthefamilychildcarestrategies
andmigrationprocesses.Thespecificcommunity/societylevelfactorsinclude
generalmigrationpatterns,communitysolidarityandresources,andoverall
socioeconomicstatusinmigrantsendingandreceivingareas.
Factorsandmechanismsillustratedinthetheoreticalreviewalsohelpbetter
organizetheresultsoftheempiricalreviewinthenextchapter,andinterpretkey
variablesinthequantitativeandqualitativestudies.Thereviewofempirical
literaturewillcontinuetodeveloptheevidencebaseforthisthesis,withamore
close-upviewofchildrenandfamiliesaffectedbymigration.Byconfirming
relevantpartsofthetheories,thecontext-specificfindingsfromtheempirical
evidencewillalsostrengthentheinterpretationoftheresultsfromourcurrent
studies.
45
Chapter3 ReviewofEmpiricalLiterature
3.1Introductionandscopeofthereview
Thischapterprovidesanextensivereviewoftheempiricalliterature,including
bothquantitativeandqualitativestudies,onthewellbeingofLBC.Both
internationalmigrationandinternalmigration(withinacountry)were
consideredrelevantinthisreview.Thisisbecausethekeyissuesintheresearch
questionofthisthesis,suchasparent-childrelationshipoverdistance,andfamily
careandsupportenvironments,donotseemtobemarkedlydifferentbetween
internationalandinternalmigration.
Thetheoreticalreviewinthepreviouschapteridentifiedvariouspatternsof
migration,butthecorerationalesbasedonsocioeconomicdisparitiesremained
highlyconsistent.Asadisruptionorstresseventinthefamily,parentalabsence
duetomigration,irrespectiveoftheirdestination,alsofitswithboththe
attachmenttheoryandfamilysystemstheories.Manyofthemethodologies
employedintheempiricalliteraturearelikelytobeexchangeableacrossdifferent
settings.Additionally,resultsfromthisreviewmayhelpinterpretfindingsfrom
thecurrentstudiesbycomparingwiththeliteratureondifferentpatternsof
migration.Comparisonofthecontextsbetweenourempiricalstudiesandthe
extantliterature,especiallystudiesofinternationalmigration,willbepresented
inChapter5(Section5.4.2).
46
Forquantitativestudies,resultsarecategorizedbydifferentaspectsofchild
wellbeingoutcomes,andthenbydimensionsofassociatedfactors.Forqualitative
studies,thearticlesareclassifiedintoafewbroadkeythemes.Asummaryofthis
reviewispresentedattheendofthischapter,followedbythecontributionofthis
reviewtothecurrentqualitativeandquantitativestudiestobepresentedinthe
nexttwochapters.
3.1Methods
Asystematicsearchwasconductedtoidentifyempiricalstudiesonthe
psychosocialwellbeingofleftbehindchildrenglobally,publishedinbothEnglish
andChinese,usingthefollowingdatabases:PubMed/Medline,PsycINFOandWeb
ofKnowledge,andCNKI(ChinaNationalKnowledgeInfrastructure).Both
quantitativeandqualitativestudieswereincluded.Wedidnotlimitthe
publicationyearinordertoretrievemaximumnumberofrelevantarticles.
TheflowdiagramofthesearchandselectionprocessisoutlinedinFigure3.1.The
inclusioncriteriaareasfollows:(1)thestudysubjectsareLBC,i.e.,childrenliving
intheirhometownorcountryoforigin,separatedfromoneorbothmigrant
parents;(2)psychosocialwellbeingoutcomesofthechildrenwereexamined(3)
forquantitativestudies,acomparison/controlgroupwasincluded,and;(4)
studiespublishedinpeerreviewedjournals
47
Morespecifically,thefollowingexclusioncriteriaregardingmeasuresand
participantswerefurtherapplied:1)studiesonmigrantchildrenwho
accompaniedtheirparents,includingrefugeechildren;2)studiesthatonly
includedoutcomemeasuresinparentsand/orcaregivers,withoutreportingchild
outcomes;3)studiesthatonlyexaminededucationaloutcomesinchildren;4)
quantitativestudieswhichdidnotconsiderotherfactorsofchildpsychosocial
wellbeing,besidesparentalmigrationstatus;and5)quantitativestudiesinwhich
onlybivariatedescriptiveanalysiswasconducted.
Wedidnotrestrictthetypesofpsychosocialwellbeingmeasures,or
qualitative/quantitativemethodologiesinthisreview.Keytermsinpsychosocial
wellbeingwereselectedafterapre-searchoftheLBCliterature.Duetothe
variousexpressionsindefiningtheLBCandmeetingourreviewcriteria,four
groupsofkeywordswereusedasbelow.
#1:"leftbehind"and(childrenoradolescentsorteenagers)
#2:(childrenoradolescentsorteenagers)and(parent*orfather*ormother*)
and(leftorleav*orawayorabsenceorabsentorseparat*orunaccompaniedor
alone)
#3:migrat*ormigrant*oremigra*ortransnationalorimmigra*
#4:psychologicalorpsychosocialordepressionoranxietyorresilienceorneglect
oremotionalorlonelinessormentalorbehavior*orbehaviour*orstress
48
Thenweperformedafinalsearchinwhichgroup#1and#2werecombinedby
thepreposition“or”andthencombinedwithgroup#3and#4with“and”.Forthe
Chineseliterature,sincemoststudiesusedtheterm“留守儿童”whichalready
indicatedparentalmigration,group#3wasremoved.Otherkeywordswerealso
adjustedfromtheliteraltranslationfromEnglish,tobettermatchwiththe
academiclanguageuseinChinese.Thekeywordgroupsareasfollows(including
PinyinandEnglishtranslation):
#1:"留守 [LiuShou;left-behind]"and(儿童 [ErTong;children]or青少年[Qing
ShaoNian;adolescents]
#2:(儿童 or青少年)and(父母 [FuMu;parents]or父亲 [FuQin;father]or母
亲 [MuQin;mother])and(外出 [WaiChu;goingoutfromhome]or打工 [Da
Gong;workingawayinthecity]or迁移 [QianYi;migration]or进城 [JinCheng;
intothecity]
#3:心理 [XinLi;psychological]or抑郁 [YiYu;depression]or焦虑 [JiaoLv;
anxiety]复原力 [FuYuanLi;resilience]or弹性 [TanXing;resilience]or情感
[QingGan;emotion]or孤独 [GuDu;loneliness]or行为 [XingWei;behavior]
Thenthefinalsearchcombined#1and#2bythepreposition“or”,andthen
combinedwithgroup#3and#4with“and”.Thereferencelistsoftheidentified
English-andChinese-languagearticleswerealsohand-searchedforfurther
relevantstudies.Dataandliteraturewereextractedfromeachoftheincluded
studiesusingastandardformat:authorandyear,studylocation,sample,key
49
measurements.
Figure 3.1: Flow chart of article selection procedure
Records from initial search conducted in 3 databases:
n=1,342
Web of Knowledge:498 PubMed: 243 PsycINFO: 89
CNKI: 512
Abstract screening n=1,075
Papers retrieved for Full text screening
n=287
Papers that meet the inclusion criteria n=83
Papers included in the review: n=84
Duplicates excluded: n=267
Papers excluded: n=788
Reviews: 32 Not empirical studies: 453
Not relevant or no comparison groups: 303
Papers excluded: n=204
Reviews: 9 Not empirical studies: 49
Not relevant: 37 No comparison groups: 31
Simple descriptive quantitative study: 78
Papers included by hand searching the references:
n=1
50
Table 3.1: Summary of reviewed quantitative studies conducted in China and
published in English
Author, Year Location Sample Definition of
LBC
Age Key measures
Fan et al.,
2010
China
Hunan
(schools)
Case control
629 LBC
(including 102
past-LBC), 645
non-LBC.
Both parents
away for over
12 months
12.4 ± 2.2 SDQ
SES (Socioeconomic
status)
Education
Teacher engagement
Migration history
Relative/non-relative
caregiver
Gao et al.,
2010
China
Guangdong
(schools)
541 LBC, 2445
Controls.
One or both
parents away
for over 6
months
10 to 18 SES
Youth Risk Behavior,
Student Health
Gao et al.,
2013
China
Guangdong
(schools)
2609 (44% LBC;
30%
father-only;3%
mother-only;11%
both)
One or both
parents away
for over 6
months
11 to 19
13.8(SD1.1)
Past 30-day smoking,
Smoking Self-Efficacy
Questionnaire
He et al.,
2012
China
Hubei
(schools)
590 LBC, 285
controls
One or both
parents away
for over 6
months
9 to 14 Children’s Depression
Inventory, Family
Affluence Scale,
Multidimensional Scale of
Perceived Social Support
Hu et al.,
2014
China
Wuhan
(Schools)
3473 Children
(parent-report)
43% local, 35%
migrant, 22%
LBC.
One or both
parents away
for over 6
months
5 to 17 SDQ
(Ordinary Least Square
(OLS) regression
(nested))
Huang et al.,
2015
China (six
provinces)
Household
survey
(snow ball)
608 LBC, 755
non-LBC
One or both
parents away
for over 6
months
Mean 10 (2
to 18)
care-giver
report if
under 4
Pediatric Quality of Life
Inventory
51
Jia et al.,
2010
China
Shandong
(household
survey)
324 LBC, 282
non-LBC
One or both
parents away
11.4 (8 to 14) Pediatric Quality of Life
Inventory
Jia & Tian,
2010
China
Shandong
(household
survey)
324 LBC, 282
non-LBC
One or both
parents away
11.4 (8 to 14) Children’s Loneliness
Scale
Relationship with parents
Frequency of
communication with
parents/family gathering
Ling et al.,
2015
China
Hunan (1
village only)
268 LBC, 228
non-LBC
LBC definition:
standard
One or both
parents away
for over 6
months
11.9 (7 to 16) The Youth Self-Report;
Children’s Loneliness
Scale;
Depression Self-Rating
Scale for Children
Liu et al.,
2009
China
(three
provinces)
592 LBC One or both
parents away
10 to 17 Children’s Depression
Inventory; State-Trait
Anxiety Inventory
Liu et al.,
2010
China
Hubei
(random
towns -
schools)
Case control
230 LBC, 250
controls
based on
question -
whether grew
up with both
parents work
away
Mean 11 (8
to 15)
Piers-Harris Self-Concept
Scale;
Childhood Loneliness
Scale;
whether they like their
guardians/parents
time spent in physical and
leisure activities
Luo et al.,
2011
China (5
省)schools
1214 LBC (1024
F/O, 190 M/O),
1019 non-LBC,
aged 11-23
One or both
parents away
for over 6
months
11 to 23 State-Trait Anxiety
Inventory
The Rosenberg
Self-Esteem Scale
Ren &
Treiman,
2016
China
(nationwide)
1107 LBC, 3613
non-LBC, 342
migrant children
One or both
parents away
for work
Mean 12.2
(10-15)
Center for
Epidemiological
Studies Depression Scale
for Children; Self-esteem
and self-motivation
measures designed by
52
the authors
Su et al.,
2012
China
Jiangxi
501 LBC (218
one parent out,
283 both parents
away), 653
non-LBC.
One or both
parents
migrated to
cities for
employment
Mean 14.12 Frequency of speaking
with parents;
Life and school
satistaction (1 Q each);
Children's Loneliness
Scale; Oxford Happiness
Questionnaire
Tao et al.,
2014
China Anhui 650 LBC, 100
controls
One or both
parents away
for over 6
months
3 to 5 Family Support Scale;
Parenting Self-efficacy
Scale;
Preschool-aged Child
Mental Health Scale
Wang et al.,
2015a
China
Chongqing
4857 LBC One or both
parents left
for job over
the past 6
months
7 to 17 Children's Depression
Inventory
Wen et al.,
2015
China
Guangxi
(school)
397 LBC, 477
non-LBC
Born and
raised in the
countryside;
and one or
both of their
parents
migrated to
cities for
employment.
10 to 17 Qs on Five C's model of
PYD
self-rated health, school
grades, educational
aspiration, and problem
behavior; parental
monitoring
Wen & Lin et
al., 2012
China
Hunan
(village
snowball)
LBC 303 (153
both parent away,
108 father only,
42 mother only)
322 controls,
One or both
parents away
8 to 18 Satisfaction with life and
study, positive health
behavior, school
engagement,
Family socializing
process,
Social support,
self-efficacy,
Internal locus of control
53
Wu et al.,
2015
China
Guizhou
(schools)
363 LBC, 103
previously
left-behind, 158
non-left-behind
One or both
parents away,
including past
migration, for
over 6
months
Mean 12.45
(8 to 17)
Depression (Center for
Epidemiological Studies
Depression Scale for
Children); Family Social
Capital (parent-child
interaction; parental
monitoring); Community
social capital
Zhan et al.,
2014
China
Shaaxi
(schools)
4275 LBC (3111
both parents
away), non-LBC
450
Father or
mother
worked
outside of
village in the
past 8 months
Mean 15, 12
to 19
Rosenberg self-esteem
scale
Zhao et al.,
2009
China Anhui
(schools)
1390 LBC, 1131
non-LBC
One or both
parents away,
including past
migration, for
over 6
months
12.83 ± 1.93 P-H children’s
self-concept scale; mental
health diagnosis test
Zhao et al.,
2014
China Anhui
(schools)
1694 LBC, 1223
non-LBC
Based on
questiona -
parent(s) out
for work?
Mean=12.52;
7 to 17
The Social Anxiety Scales
for Children (SASC);
Family function (APGAR);
Quality of life; Neglect and
physical abuse
(Parents-Child Conflict
Tactics Scale)
Table 3.2: Summary of reviewed quantitative studies conducted in China and
published in Chinese
Author, Year Province Sample Definition of LBC Age Key measures
Chen et al.,
2011
Sichuan 340 LBC, 208
controls
Unspecified Mean=13.0
SD=0.9
Vaux Social Support
Record; Psychological
resilience scale; Chinese
Middle School Student
54
Mental Health Scale
Dong & He, 2014
Yunan 654 LBC, 589
controls
One or both
parents migrated
Mean=13.
35, SD=0. 84
Family Assessment Device;
Resiliency Scales for
Children & Adolescents;
duration of migration
Dong & Zhang, 2013
Yunnan 713 LBC, 632
controls
One or both
parents migrated
Junior
middle
school
students
(Year 7-9)
Resiliency Scales for
Children & Adolescents;
Child Loneliness Scale
Dong & Zhang, 2013
Yunnan 713 LBC, 632
controls
One or both
parents migrated
Junior
middle
school
students
(Year 7-9)
Resiliency Scales for
Children & Adolescents;
Eysenck Personality
Questionnaire for Children;
SES; duration of migration
Fan et al., 2013
Hunan Longitudinal
study:151
LBC, 56
controls
(parental
migration
status
unchanged
between
baseline and
follow-up)
One or both
parents currently
away
Mean=10.9,
SD=1.6
Family Atmosphere Scale;
self-report happiness; Affect
Balance Scale (negative
and position emotions);
Multidimensional Scale of
Perceived Social Support;
duration of migration
Gao 2010 Henan 116 LBC, 116
controls
Unspecified Primary
school
students
Self-report: Piers-Harris
Children’s Self-
Concept
Scale; duration of parental
migration; frequency of visit;
attitude to migration; sense
of hope
Caregiver-report: SES;
health status; reason for
55
migration; burden in
childcare
Han et al., 2012
Shaanxi 211LBC, 329
migrant
children, 210
controls
One or both
parents away for
over 6 months
Mean=13.9,
SD=1.4
The Healthy Kids Resilience
Assessment; Sense of
Security Scale; parent-child
closeness; wealth status
He et al., 2011
Anhui 893 LBC, 335
controls
Unspecified 6-14 Center for Epidemiological
Studies of Depres- sion
Scale for Children;
economic status; source of
491family income; care
style.
Hou et al., 2015
Henan
and
Shaanxi
1475 LBC,
209 children
living in
special
families
(parents
deceased,
incarcerated,
or divorced);
2355 controls.
One or both
parents migrated,
who visit home
every 3 months or
more
Mean=12.8,
SD=1.9
Self-Esteem Scale;
Children's Depression
Inventory; Seiffge-Krenke
Problem
Questionnaire (problem with
future)
Family type; mother’s
education; school class
characteristics;
neighborhood safety
(occurrence of crime)
Huang & Li, 2007
Jiangxi 570 current
LBC, 110
previous LBC,
133 never
LBC.
Mean=13.4,
9-18
Mental Health Test; type of
caregiver (grandparents,
one parent, adult relatives
from father’s generation;
relative from same
generation or no caregiver)
Jiang 2013 Fujian 963 LBC,
1520 controls
One or both
parents currently
away
Mean=13.9,
SD=0.8
SDQ; wealth status;
duration of migration;
frequency of communication
with parents; caregiver;
Jiang et al., 2013
Anhui 1694 LBC,
1223 controls
One or both
parents away for
over 6 months
Mean=12.6,
SD=2.2,
range 7 to 17
Children’s Depression
Inventory; Social Anxiety
Scale for Children; Trait
56
Coping Style Questionnaire;
Parents-Child Conflict
Tactics Scales; Children’s
Self-Esteem Scale;
Adaptation Partnership
Growth Affection Resolve;
SES
Lan et al.,
2009
Guangxi 270 LBC, 609
controls
One or both
parents currently
away
7-16 Eysenck Personality
Questionnaire; Center for
Epidemiological
Studies Depression Scale
for Children
Frequency of seeing
parents, self-rated
personality; satisfaction with
family; school performance;
SES; family history of
mental illnesses
Li & Tao, 2009
Anhui 334 LBC,
5506 controls
One or both
parents migrated
14-16 Suicide tendency
questionnaire; Self-Rating
Depression Scale;
Self-Rating Anxiety Scale;
Trait-Oriented Coping Styles
Questionnaire; parents’
education; ethnicity
Li et al., 2012 Anhui 810 LBC,
1329 controls
One or both
parents migrated
Junior
middle
school
students
(Year 7-9)
who were
under 14
Smoking; drinking (more
than once per week);
violence under alcohol
influence (more than once
per semester); time spent
online (over 1 hour/day).
Parents’ education status;
urban/rural residence;
ethnicity
Liu et al., 2011
Hebei 194 LBC, 352
controls
One or both
parents currently
away
Mean=12.3,
SD=2.6
Depression self-rating scale
for children; SES; primary
caregiver; care style; school
performance
57
Liu et al., 2014
Sichuan Longitudinal
study
T1 (Year 7):
322 LBC, 228
controls
T2 (Year 8):
276 LBC, 180
controls
T3 (Year 9):
217 LBC, 152
controls
One or both
parents away for
over 6 months
Mean=13.0,
SD=0.9
The Healthy Kids Resilience
Assessment; Social Support
Appraisal Scales; Chinese
Middle School Student
Mental Health Scale
Quality of relationship
between father and mother;
strictness of disciplining
from mother, father, and
caregiver
Lu et al., 2014a
Hubei 1031 LBC,
895 controls
One or both
parents away for
over 6 months
Mean=14.1,
SD=2.4
Chinese Middle School
Student Mental Health
Scale; duration of migration;
missing parents; attitude
toward migration
Lu et al., 2014b
Guizhou 250 LBC, 538
controls
One or both
parents currently
away
Mean=14.3,
SD=1.1
Mental Health Test; duration
of parental migration;
missing parents; attitude
towards migration; quality of
communication with
parents; frequency of
contact
Wang et al, 2011a
Anhui 1730 LBC,
1223 controls
One or both
parents currently
away for more
than 6 months
Middle
school and
Year 4-6
primary
school
students
Children's Depression
Inventory; Adaptation
Partnership Growth
Affection Resolve; attending
boarding school; single
child; SES; school
performance
Wang et al., 2011b
Henan 250 LBC, 212
controls
One or both
parents currently
away
Mean=14.0,
7-18
Friendship Quality
Inventory; Children’s
Loneliness Scale; Children’s
Depression Inventory;
economic status
Wang W et al., 2014
Liaoning 952 LBC, 952
controls
One or both
parents migrated
Primary
school
students
Symptom Check
List-90; Eysenck Personality
Questionnaire; Trait Coping
Style Questionnaire; Social
Support Rating Scale;
58
relationship with peers and
teachers; parents’
education; caregiver type
Yang et al.,
2013
Hunan 2110 LBC,
2026 controls
One or both
parents migrated
School Year
3,5,8, and11
(approximate
age 9-18)
Children's Hope Scale;
SDQ; duration of parental
migration; co-resident
household members
Yang, et al.,
2010
Anhui 416 LBC, 525
controls
One or both
parents currently
away
Mean=12.1,
SD=2.2
Children's Depression
Inventory; SES; single child
or not; parents’ divorce;
self-report body type
Zhang et al., 2007
Anhui 491 LBC, 393
controls
One or both
parents currently
away
Mean=10.54,
SD=1.57
Social Anxiety Scale for
Children; education style of
parents and teachers
(intellectual or not); amount
of pocket money; health
status; level of learning
difficulties
Zhao 2006 Sichuan 160 LBC, 154
controls
One or both
parents currently
away
Mean=11.8,
SD=0.8
Piers-Harris Children’s
Self-Concept
Scale; Eysenck Personality
Questionnaire for Children;
age at separation; duration
of separation
Zhao et al.,
2012
Ningxia 926 LBC,
1091 controls,
One or both
parents away for
over 6 months
Mean=10.05
(5-16)
Self-report: Piers Harris
Self-Concept; Child
Loneliness Scale; Social
Anxiety Scale for Children.
Caregiver report:
Caregiving behaviors, level
of care; closeness with
parents; child’s relationships
with peers and teachers.
Zhao et al., 2015
Henan 378 LBC, 351
controls
One or both
parents currently
away
Mean=13.7,
SD=2.3
Mental Health Test
; duration of parental
migration; parents’
education; relationship with
teachers, peers and
59
neighbors; intensity of
parental care;
communication with
caregiver; caregiver’s
attitude toward child
educaion
Zheng et al., 2014
Hubei 156 LBC, 123
controls
One or both
parents migrated
12-14 Social Support Rating
Scale; Mental Health Test;
Resiliency Scales for
Children & Adolescents
Zhou et al.,
2011
Sichuan 291 LBC, 430
controls, 670
urban migrant
children
Both parents
migrated
Mean=12.3
SD=1.53
The Healthy Kids Resilience
Assessment; SES
Zhu 2016 Sichuan,
Anhui,
Henan,
Guangxi
1251 LBC,
704 controls
One or both
parents away for
over 6 months
Mean=11.1,
SD=1.1
The Gansu Survey of
Children and Families
(Internalizing and
externalizing behaviors)
Composition of family and
caregivers; SES; parents’
occupation; house location;
migration prevalence in the
village
Zhu et al., 2014
Guizhou 545 LBC,
1024 controls
One or both
parents away for
over 6 months
Mean=12.71,
SD=1.85
Mental Health Test; Coping
Style Short Questionnaire;
ethnicity; school year
Table 3.3: Summary of reviewed quantitative studies that were not conducted in
China
Author, Year Location Sample Definition of
LBC
Age Key measures
Adhikari et al.,
2013
Thailand 519 LBC, 511
controls,
One or both
parents away
for over 6
months
aged 3-5 and
9-11
SDQ (parent-report)
Physical health
Mental health status
60
of primary caregivers
Adumitroatie et
al., 2013
Romania Case control
177 LBC, 107
non-LBC
One or both
parents away
Mean=16.82
(14-19)
Child Parental
Acceptance-Rejection
Questionnaire; The
Adult Personality
Assessment
Questionnaire
Aguilera-Guzman
et al., 2004
Mexico 106 LBC, 194
controls,
Father
periodically
absent for 6+
years
aged 11-14 Stressors and
compensators from
relationship with
absent father, peers,
family, and their
communities, and
essential facts about
the teenagers.
Crawford et al.,
2009
Inner city
New York,
USA
35 LBC, 731
controls
Children who
were
separated
from mother
for at least
one month
before age 5
Borderline personality
disorder (BPD)
symptoms,
Child and maternal
risk factors, Insecure
attachment
Graham &
Jordan, 2011
Indonesia,
the
Philippines,
Thailand,
and
Vietnam
1916 LBC
(including
farther/mother/both
working overseas
for 6+
months),1960
controls, aged 3-5
and 9-11
One or both
parents away
for over 6
months
3-5
9-11(Caregiver
report)
SDQ
Jordan &
Graham, 2012
Indonesia,
the
Philippines,
Thailand,
and
Vietnam
N=1523, about half
LBC half controls,
One or both
parents away
for over 6
months
aged 9-11
(Caregiver and
self report)
General Happiness
(1Q, incl parent
report), school
enjoyment (1Q),
school performance
(1Q, incl parent
report)
61
family functioning
(APGAR)
caregiver mental
health (SRQ 20)
Lahaie, et al.,
2009.
Mexico parents from 755
households of LBC
At least one
family
member who
had migrated
to the US in
the last 5
years, 754
controls.
Parents reported:
behavioral or
academic troubles at
school, whether
dropped out of
school, any emotional
problems
Mazzucato et al.,
2015
Ghana,
Nigeria,
Angola
(Schools)
1747 in Ghana,
1344 in Angola,
1690 in Nigeria
One or both
parents away
for 3 months
11 to 21 SDQ; care
arrangements, SES,
school performance,
health, and
psychological well-
being.
Pottinger, 2005 Jamaica Case control:
27 LBC; 27
controls
aged 9-10
Unspecified Academic
achievement
(diagnostic tests)
Self-esteem
(Piers-Harris
Self-Concept Scale)
Children’s thoughts
and experiences
regarding migratory
separation
Child behaviour
Senaratna et al.,
2011
Colombo,
Sri Lanka
Case control:
253 pairs of LBC
and controls
matched by age,
gender, and
neighbourhood,
mothers
working
overseas
minimum 12
months
aged 5-10. Behaviour (Child
Behaviour Check List)
Socio- demographic
and Risk-factors
Information
Questionnaire
School performance
and attendance
62
Smeekens et al.,
2012
Philippines 108 LBC, 97
controls, aged
13-18
One or both
parents
working
abroad
13-18 Physical health
(Physical Health
Questionnaire)
Missing parents or
not
Perceived stress (The
Homesickness
Questionnaire)
Loneliness
Avoidant coping
(Coping Scale for
Children and Youth)
Tomsa & Jenaro
et al., 2015
Romania
(Schools)
Case control
163 LBC, 163
non-LBC
One or both
parents away
for over 6
months
Mean=13.5,
12 to 15
The State-Trait
Anxiety Inventory for
Children; The Short
Mood and Feelings
Questionnaire; Anger
Expression Scale;
Children's Coping
Strategies Checklist
Vanore et al.,
2015
Moldova
(household
survey)
471 LBC, 1508
non-LBC
One or both
parents living
abroad for at
least 12
months
consecutively
at the time of
the survey
Mean=10.8 (4
-17)
Caregiver
report
SDQ
Wickramage et
al., 2015
Sri Lanka Case control
410 LBC, 410
comparison (some
under 5 and no
SDQ results)
One or both
parents away
for over 6
months
aged 1-17 SDQ,
Nutritional status,
parent SES and
mental health
3.2Quantitativestudies
63
Atotalof68paperswereincludedinthefinalreview;36ofthemwerepublished
inEnglishand32publishedinChinese.ThemajorityoftheEnglish-language
studies,i.e.22paperswerebasedonresearchconductedinChina(Table3.1).
Others(Table3.2)wereinSouth/SoutheastAsia(six),LatinAmerica(four),
EasternEurope(three),Africa(one,acrossthreecountries),andtheUSA(one).
AllChinese-languagestudieswereaboutLBCinChina.
Inafewcases,multiplearticleswerefoundtobefromthesamestudysample.
Threearticles(Adhikarietal.,2013;Jordan&Graham2012;Graham&Jordan,
2011)useddatacollectedintheCHAMPSEA(ChildHealthandMigrantParentsin
SoutheastAsia)project.Jiaetal.(2010)andJian&Tian(2012)werebasedonan
identicalsamplefromShandong,China,whilelookingatdifferentpsychosocial
measures.Similarly,apairofresearcherspublishedtwostudies(Dong&Zhang,
2013a&2013b)ondifferentChinesejournalsbasedonthesamesamplein
Yunnanprovincebutdifferentmeasures.Also,Gaoetal.(2010)andGaoetal.
(2013)useddifferentgroupsofthesamesampleinGuangdong,China.
3.2.1Characteristicsofthestudysamples
TheconceptofLBC,i.e.childrenlivingapartfromoneorbothmigrantworker
parent(s),hasbeengenerallyconsistentinallincludedstudies.However,the
specificdefinitionintermsofthedurationofmigrationvariedinidentifyingthe
LBCinthestudypopulations.Manystudiesexaminedthedurationofparent-child
64
separation(untilthetimeofsurvey),andsetaminimumeligibilitycriterionfor
LBC,rangingfrom6to12monthsexceptonestudy.Othersdidnotspecifythe
duration,butdefinedtheLBCgrouponlybyquestionsregardingthemigration
statusofparents,orwholivedorcaredforthechildren.SomeChinese-language
studiesdidnotevendefinetheLBC,astheterminChinese(留守儿童)already
indicateschildrenofmigrantworkers.Yetonlythreestudies(Huang&Li,2007;
Wuetal.,2015;Adhikarietal.,2013)lookedatchildrenwhowerepreviously
left-behindbutcurrentlylivingwithbothparents,andcomparedthemwithnever
left-behindchildren.
Moststudiesonlyincludedschool-agechildren,andcollecteddatabyself-report
instruments.Fourstudiesrecruitedpre-schoolagegroups(Crawfordetal.2009,
Graham&Jordan2011,Taoetal.2014,Vanoreetal.,2015),andused
caregiver-reportdata;twoofthese(Graham&Jordan2011,Vanoreetal.,2015)
triangulatedcaregiver-reportandself-reportresultsfortheolderchildreninthe
studies.
LBCwerealsogroupedbymother-only,father-only,orboth-parentmigration.
Threestudies(Senaratnaetal.,2011;Luoetal.,2011;Fanetal.,2012)includedin
thisreviewonlyfocusedononeofthesethreegroups,andonearticle(Graham&
Jordan,2011)excludedtheboth-parentawaygrouptocomparetheothertwo
groups.
65
Inmostcountries,fathersaremorelikelythanmotherstobecomethemigrantin
afamily.Parreñas(2008)specificallylookedintothepracticeof“transnational
fathering”andtherelationshipbetweenmigrantfatherandchildren.Yetthe
numberoffemalemigrantsfromcertainSouth/SoutheastAsiancountriesisclose
tomalemigrants,ifnotlarger.Inparticular,Senaratnaetal.’s(2011)studyonly
investigatedthewellbeingofchildrenleftbehindbymothers,since65%oflabor
migrantsfromSriLankaarefemale.Whenbothparentswereaway,grandparents
usuallybecamecaregiversoftheLBC,acrossallcountries.
Whilemoststudiesfocusedoncomparingmigrationpatternsbasedonwhich
parent,orwhetherbothparentshadmigrated,veryfewlookedintotheeffectsof
differentcaregiversonchildwellbeing.Theliteraturetendedtoassumethatifthe
motherdidnotmigrate,sheshouldbetheprimarycaregiver;orifbothparents
wereaway,onegrandparentshouldbeprimarycaregiver.However,care
arrangementsmaybemorecomplexthantheseassumptions,andthe
identificationofprimarycaregivermaydifferbetweendifferentrespondents.
Variousfactorsassociatedwithparentalmigrationandchildpsychosocial
wellbeingwereexaminedintheliterature,includingdemographiccharacteristics,
socioeconomicstatus,familyrelationships,functioningandsupport,aswellas
school,peer,andcommunityvariables.Thesecategoriesandthecommon
indicatorsareoutlinedinasummaryframework(Figure3.2).
66
Figure 3.2: Summary framework of the quantitative literature
Socioeconomic contexts
Parental migration Family relationships,
functioning and support
Parent-child relationship, family
dynamics, family function and
social support
School, peers, and community
School performance, school
engagement, peer relationship,
neighbor relationship, community
development.
Socioeconomic status
Household income, remittance,
wealth status, education level,
Demographic characteristics
Age, age at separation, gender,
family composition
Child psychosocial wellbeing
Quality of life, mental health
and resilience, psychosocial
strengths and difficulties,
self-concept and self-esteem
67
3.2.2ComparisonbetweenEnglish-languageandChinese-languagestudies
AlthougharticlespublishedinEnglishcoveredmanycountriesaroundtheworld,
themajorityofthemwereabouttheLBCinChina.StudiespublishedinChinese
journalsprovidedfurtherperspectivesabouthowtheissuesrelatedtoLBCwere
viewedandaddressedbytheChineseacademiaandpolicy-makers.
Ingeneral,studiespublishedinChinesewereoflowerquality,withlessrigorous
samplinganddatacollectionprocessesandlesssophisticateddataanalysis.Most
Chinese-languagestudiesadoptedasimplestudydesign.Manyofthemexplored
theoverallmentalhealthstatusofLBC,byemployingageneralmentalhealth
scaleormultipleinterrelatedpsychosocialoutcomevariables.Instruments
developedbyChineseresearchers(forchildreninChina)andinternational
researcherswerebothused.Thereportingwasgenerallyfocusedonvarious
descriptivecharacteristicsofthechildpopulation,ratherthanthemechanismsof
thechallengesfacingtheLBC.
TheEnglish-languagestudiestendedtobemorespecificregardingthe
dimensionsofwellbeingthatwereexamined.Theauthorsusuallylookeddeeper
intothesocioeconomicstatusandotherhousehold-levelfactorsthatmayaffect
childwellbeing,andfittedregressionmodelsthatincorporatedavarietyof
independentvariables.Theoriesfrommultipledisciplinessuchaschild
developmentandfamilystudieswereoftenreferredtointhebackgroundand
68
discussion,whereastheinterdisciplinaritywasoftenlackingamong
Chinese-languagestudies.
Nevertheless,theonlytwolongitudinalstudiesretrievedinthereviewwereboth
publishedinChinese.Theyofferedimportantinsightsabouttheimpactof
parentalmigrationoveracoupleofyears’time,atacrucialstageofearly
adolescence.Howeverthetwostudieshadasmallsamplesizeandhighdrop-out
rates.Thestudydesignwasalsosimplewithfewmeasuresaboutfamily
characteristics.
Inthefollowingsectionsofthischapter,whennecessary,thelanguageofthe
articlewasnotedifitwasChinese,inordertobettercomparewiththe
English-languagestudiesfromdifferentperspectivesintheLBCliterature.
3.2.3Happinessandqualityoflifemeasures
SeveralstudiesexaminedthehappinessandlifesatisfactionamongLBCin
comparisonwithnon-LBC(Jordan&Graham,2012;Grahametal.,2012;Wen&
Lin,2012;Suetal.,2012;Gaoetal.,2010;Jiaetal.,2010;Huetal.,2015),andit
wasgenerallyshownthatLBCtendedtobelesshappyandlesssatisfiedwithlife
andschool.AnimportantlongitudinalstudypublishedinChinese(Fanetal.,2013)
suggestedthattheself-reporthappinesslevelsignificantlyreducedamongLBC
overthe2.5-yearstudyperiod,whilethelevelinnon-LBCremainedunchanged.
69
However,theoutcomevariablesinthesestudies(Jordan&Graham,2012;
Grahametal.,2012;Wen&Lin,2012;Gaoetal.,2010;Jiaetal.,2010;Fanetal.,
2013)wereonlybasedonasinglequestionaskingthechildrenabouttheir
happinessorlifesatisfaction,whichmayhaveover-simplifiedrealexperiencesof
bothLBCandnon-LBC.
Moreover,Jiaetal.(2010)andHuangetal.(2015)lookedintoruralChinese
children’s“healthrelatedqualityoflife”,aconceptthatincludesdomainsrelated
tophysicalandmentaldisorders,throughhouseholdsurveysusingThePediatric
QualityofLifeInventory(PedsQL)scale.Amongchildrenaged2-18acrosssix
provinces,Huang(2015)foundthatLBCwereworseoffinPedsQLoverallscores,
withlowerphysical,emotionalandsocialfunctionings,ascomparedtootherrural
children.Jiaetal.(2010)showedsimilarresultsinasmallersampleofschool-age
children,exceptthatthephysicalhealthsubscalescoresdidnotshowsignificant
difference,indicatingpsychologicalchallengesintheLBC.
Whilethe“healthrelatedqualityoflife”maybeausefulindicatorofchild
wellbeingingeneral,itisnotcloselyrelevanttothechallengesfacedbytheLBC.
Similarly,measuresinthePedsQLinstrumentwerealsosupposedtoberelatedto
physicalandmentaldisorders,especiallyinpediatriccaresettings.Infact,there
hasbeennoclearevidenceregardingdisadvantagesintermsofphysicalhealth.
Theprolongedseparationfromparentsislikelytoaffectqualityoflifethatis
relatedtopsychologicalandsocio-behavioralchallenges,ratherthandiseasesor
70
disorders.Hencetheresultsfromstudiesusinghealthfocusedsurveytoolsmay
notbefullywarranted.
3.2.4Mentalhealthstatus
SeveralChinese-languagestudieslookedintothementalhealthstatusofchildren
inChina.A100-itempsychometricscalenamely“MentalHealthTest”(MHT),
designedbyChineseresearchers(Zhou,1991),wasoftenusedonmiddleschool
childrenintheChineseliterature.Itincludedeightcontentscales:learning
anxiety,interpersonalrelationshipanxiety,lonelinesstendency,self-accusation
tendency,sensitivenesstendency,physicalsymptoms,horrortendency,and
impulsiontendency.FivestudiesinthisreviewusedMHTandallfoundtheLBC
grouphadworsementalhealth(Zhaoetal.,2015;Luetal.,2014;Huang&Li,
2007;Zhengetal.,2014;Zhuetal.,2014).Inparticular,LBCshowedahigherlevel
ofanxietysymptomsthannon-LBCinallthesestudies;threestudiesindicated
thatLBCtendedtobemorescaredofstressfulsituationsthantheirpeers(Zhaoet
al.,2015;Luetal.,2014;Huang&Li,2007).
AnothercommonlyusedpsychometricstoolforChineseadolescentswas“Chinese
MiddleSchoolStudentMentalHealthScale”.Basedonthisscale,twostudies
(Chenetal.,2011;Liuetal.,2014)foundnosignificantdifferenceinmentalhealth
statusbetweenLBCandnon-LBC.Importantly,Liuetal.’s(2014)longitudinal
studysuggestedthatduringthe2-yearperiodbetweenthefollow-upandbaseline,
thementalhealthstatusofLBCandcontrolsshowedasimilaramountofdecline;
71
nosignificantdifferencewasidentifiedbetweenthetwogroupseitheratbaseline
orfollow-up.
Lingetal.(2015)usedtheYouthSelf-Reportscaleandidentifiedhigherlevelof
mentalhealthproblemsinaggressiveandexternalizingandinternalizing
behaviors,somaticdisorders,anxiety/depression,andsocialproblemsinLBC
thannon-LBC.BymetricsofPreschool-agedChildMentalHealthScale,Taoetal.
(2014)foundamongpreschoolersaged3-5thatLBCwerenotworsethan
controls.
DespiteanoveralldisadvantageamongLBC,mixedfindingsfromtheabove
studiesalsosuggestedthecomplexityofLBC’smentalhealthstatus,andrevealed
certainspecificdimensionsofpsychosocialwellbeingthatwereworthfurther
investigation.Thescreeningtoolsusedinthesestudiesprovidedanoverall
assessmentofchildren’smentalhealth,insteadoffocusingoncertaintypesof
disorders.However,despitethecomprehensiveness,themeasureswerestill
focusedonthesymptoms,insteadofdimensionsofpsychosocialwellbeing.The
resultsmaythusunderestimatetheimpactofparentalmigration,since
disadvantagesofLBCmaybedemonstratedfromspecificpsychologicaland
behavioralaspects,whicharenotrelatedtomentalhealthdisorders.Thetwo
mentalhealthscalescommonlyusedinChinawerenotusedinanyothercountry,
andthusthescoresandstudyfindingscannotbeusedtomakeinternational
comparisons.
72
3.2.5Resilienceandpositiveyouthdevelopment
Resiliencewaspointedoutinsomearticlesasapsychologicalperformancethat
hadapositivesignificanceforthementalhealthofchildren,despiteadverse
situationssuchasprolongedparentalabsence.Afewofthesestudiesattempted
tomeasureresilienceasakeyoutcomevariable.
IntheChinese-languageliterature,theHealthyKidsResilienceAssessmentwas
usedbythreestudies,toexploreexternalandinternalresilienceconstructs
associatedwithchilddevelopment.Withverysimilarstudydesign,bothZhouet
al.(2011)andHanetal.(2012)investigatedthreegroupsofchildreninsuburban
areas:LBC,urbanmigrantchildren,andchildrenlivingwithbothparents.Both
studiesfoundnosignificantdifferencebetweenLBCandnon-LBC.Incontrast,
resultsfromotherstudies(Dong&Zhang,2013a;Dong&Zhang,2013b;Dong&
He,2014;Zhengetal.’s2014)basedonquestionnairesdesignedforChinese
studentsdemonstratedlessresilienceinLBC.Liuetal.’s(2014)longitudinalstudy
measuredthesameoutcomeamongagroupofmiddleschoolstudentsatYear7,8
and9.TheresultsimpliedatrendofdecliningresilienceamongLBC,butonlyat
Year8didtheLBChavesignificantlylowerresiliencethannon-LBC.InSoutheast
Asia,Jordan&Graham(2012)articulatedresilienceasacombinedoutcomeof
multiplemeasuresofgeneralhappinessandwell-beingaspects,andfoundgreater
resiliencewasassociatedwithlongerdurationsofmaternalabsence.
TheinvestigationaroundresiliencewassimilartotheperspectiveofPositive
73
YouthDevelopment(PYD),implyingabeneficialeffectofdifficultsituations.Wen
etal.(2015)probedintothePYDaspectsofCompetence,Confidence,Caring,
Character,Connection,throughahostofquestionsthatweresimilartotheones
inSDQ.HoweverlittledifferenceinthesePYDoutcomeswasdetectedbyparental
migrationstatus.Nevertheless,thepositivepsychologyperspectiveisan
interestingdirectiontofurtherexploringLBC’soverallpsychosocialwellbeing,
takingintoaccountthebeneficialinfluenceofparentalmigration,forexamplevia
economicandsocialremittances.
3.2.6Psychosocialstrengthsanddifficulties
TheStrengthsandDifficultiesQuestionnaire(SDQ)wasusedasaprimarysurvey
instrumentinfourofthereviewedpapers.Itisawidelyusedchildpsychosocial
wellbeingquestionnaireandhasbeenvalidatedinmultiplecountries(Goodman,
2007).Itcomprisedof25itemsin5hypothesizedsubscales:emotionalsymptoms,
(somatic,unhappy,worries,nervous,fears)conductproblems(tempers,obedient,
fights,lies,steals),hyperactivity,peerproblems,andpro-socialbehaviors.
TheCHAMPSEAproject
AspartoftheCHAMPSEA(ChildHealthandMigrantParentsinSoutheastAsia)
study,GrahamandJordan(2011)usedtheemotionalandconductsubscalesof
SDQasmeasurementsofpsychologicalwellbeingamong3,876children(including
74
1,960LBC)inIndonesia,Thailand,Vietnam,andthePhilippines.Theresultswere
inconsistentacrosscountries.OnlyinIndonesiasignificantlyhigheremotional
distresswasfoundamongLBCbytransnationalmigrantsascomparedtochildren
livingwithbothparents,afterchildageandgenderwereaccountedfor.However,
inthePhilippinestherelationshipwasintheoppositedirection,thatisLBCwere
lessvulnerabletoemotionalproblems.Withregardtochildren’sbehavior,onlyin
ThailanddidtheresultsshowincreasedoddsofLBCexperiencingconduct
problemsascomparedtonon-LBC,whileFilipinoLBCwerelesslikelytohave
conductproblems.TheauthorsattributedtheadvantagesamongLBCtothestrong
civilsocietysupportorganizationsinthePhilippines.
Inaddition,Adhikarietal.(2013)examinedtheoverallSDQscoresamongthe
ThaichildrenintheCHAMPSEAstudy.Nosignificantdisadvantagewasshownin
LBC,althoughchildrencurrentlylivingwiththeirmotherswhohadreturned
homefrommigrationmayhaveexperiencedmorepsychologicalproblems.
TheCHAMPSEAstudieshighlightedthecontext-specificityofLBC’swellbeing,as
noconsistentcorrelationbetweenpsychosocialwellbeingandparentalmigration
wasfoundevenamongthefourcountriesinthesameregion.Also,some
limitationsmayweakentheconclusionsfromthisstudy.First,thechildmeasures
wereallreportedbycaregivers,whomayhaveunder-estimatedchildren’s
distressduetoparentalmigration.Thiswaspointedoutbythesameauthorsina
laterpaper(JordanandGraham,2012)basedonthesamesamplecomparingthe
75
self-reportandcaregiver-reportresults.Also,theCHAMPSEAsampleswere
selectedfromareasofhighout-migration,wheretheoverallsocioeconomic
developmentbenefitedsignificantlyfromoverseasremittances.Intheseareas
bettercopingstrategiesandstrongersocialsupportnetworksmayhavebeen
developedfortheLBC,becauseoftheleft-behindcommunity’sabilitiesin
mobilizingconsiderableamountofresources.Itwasalsoquestionablewhether
thestudysampleswererepresentativeoftheentireLBCpopulationinthe
respectivecountry.
OtherstudiesusingSDQ
FourstudiesinChinausedSDQtomeasurechildoutcomes;amongthemonlyJiang
(2013)didnotidentifysignificantdifferenceinpsychosocialoutcomesbetween
LBCandnon-LBC,basedoncut-offscoresthatdefinedemotionalandconduct
problems.Fanetal.(2010)demonstratedthatLBChadsignificantlylowerlevelsof
psychosocialwellbeingfromallfivesub-scalesofSDQ,whileonlyhyperactivity
andpro-socialbehaviorremainedsignificantlydifferentafteradjustingfor
demographicvariables.Similarly,Huetal.(2014)foundworseoutcomesinLBC
fromallSDQsub-scalesthroughbivariateanalyses,butnoneofthedifferences
weresignificantanymoreaftercontrollingforfamilyandschoolcharacteristics.
Yangetal.(2013)suggestedamongtheLBC,thosewhowereleftbehindbytheir
motherhadhigherlevelofdifficultiesthanthoseleftbehindbytheirfatherorboth
76
parents.Theauthorsalsofoundthatsenseofhopewasanimportantfactor
contributingtoLBC’swellbeing.
InSriLanka,Wickramageetal.(2015)obtainedresultsfromthreeversions(self,
parentandteacherreport)oftheSDQforeachschool-agechild.Thecomposite
scoresinthefivedomainswerecalculatedaccordingtoavalidatedpredictive
algorithmofSDQ,generatedfromalargecommunitysampleinBritain.Theresults
showedsignificantlymoreLBCchildrenwereinthe“borderline”or“abnormal”
categories(43.3%)thanthecomparativechildren(33.6%).Thedifferencewas
alsosignificantinemotionalandhyperactivitysubscales.
AstudyinMoldova(Vanoreetal.2015)onlyusedoutcomesfromtheemotional
andconductsubscalesinthecaregiver-reportedSDQ,andthescoreswere
convertedfromcontinuoustocategoricalvalues.Forboys,havingamigrantparent
correspondedtoahigherprobabilityofanabnormalconductproblemsscore,but
notforgirls.Emotionaloutcomesdidnotshowsignificantdifference.Parental
migrationdidnotbearasignificantinfluenceontheemotionalsymptomsscale.
Mazzucatoetal.’sstudy(2015)inthreeAfricancountriesfoundthatchildrenwith
oneorbothmigrantparentshadhigherlevelsofpsychologicaldistressthan
childrenlivingwiththeirparents.Therewasnosignificantdifferencebetween
internationalandinternalmigration,inGhanaandNigeria;butforAngola,
internationalparentalmigrationhasagreaternegativeimpactonchildren's
psychologicalhealth.
77
TheSDQtotaldifficultiesscorehasproventobeavalidmeasureofpsychosocial
wellbeingacrossvarioussettings,andthereforetheinstrumenthasbecome
popularinstudiesonLBC,aspecialbutgenerallyhealthypopulationwithout
clinicallyrelevantrisksformentalhealthproblems.TheSDQisausefultoolfor
comparingoutcomesbetweengroupsofchildren,buthasnotbeenwidelyusedfor
clinicaldiagnosisofmentalhealthconditions.TheinitialdesignersofSDQdid
provideacomputerizedalgorithmtopredictchildpsychiatricdiagnosesonthe
basisofSDQscores(Goodman,etal.2000).Howeveritrequiresthequestionnaires
tobecompletedbyparent,teacher,andthechildhim/herselfforeachindividual
case.Somestudies(Jordan&Graham2011,Wickramageetal.,2015)usedan
estimatedcut-offscoreofSDQoritssubscales,anddichotomizedtheoutcomein
theiranalyseswithoutcomparingmeanoutcomescores.Thisapproachmay
underminetherobustnessofSDQandlosecertainamountofinformation.
3.2.7Emotionalsymptoms
Loneliness
SevenpapersinChinausedtheChildren’sLonelinessScale(CLS)scoreasthe
outcomemeasure(FourinEnglish:Liuetal.,2010;Jia&Tian,2010;Suetal.,2012;
Lingetal.,2015;threeinChinese:Wangetal.,2011b;Zhaoetal.,2012;Dong&
Zhang,2013a);the16-itemscaleisoneofthemostreliablemeasuresofitskind
78
withaCronbachalphaof0.80orabove(Liuetal.2010).Allthesestudiesexcept
one(Wangetal.,2011b;whichfoundqualityoffriendshiptobeakeyfactorof
wellbeing)concludedthatLBCweremorevulnerabletolonelinessthanchildren
livingwithbothparentsinruralareas.SimilarstudiesinthePhilippinesreported
associationbetweenparentalmigrationandlonelinesswithothermeasurements
(Asis,2006;Smeekensetal.,2012).
AlthoughtheconcreteimpactonLBC’soverallwellbeingremainsunclear,it
appearsthatlonelinessisverycommonamongLBC,whichcanbeexplainedasa
directconsequenceofparentalabsence.Onespecificweaknessofthesestudies
wasthelackofconsiderationofotherfamilyandcommunitycharacteristics.
Particularly,familystructureandcarearrangements,aswellasmigration
prevalenceandsocioculturalcontexts,maybecloselyassociatedwithboth
migrationpatternsandloneliness.Aslonelinessisarelativelysubjective
experienceinsteadofamentalhealthdisorder,theinterpretationofitshouldbe
basedonvariouscontextualfactors.Associationswithothermentalhealth
symptomsmayrevealcertainmediatingeffects,aslonelinessappearstobea
directconsequenceofparentalabsence.
Depression
StudiesalsoexploredindicationsofmentalhealthconditionsamongLBCbyusing
variousscreeningtoolsforspecifictypesofmentaldisorders.Althoughtheresults
79
didnotnecessarysuggestclinicaldiagnosis,strongnegativeimplicationsof
parentalabsenceweredemonstratedinmoststudies.
Children’sDepressionInventory(CDI)wasthemostcommonlyuseddepression
scaleinamongthereviewedstudies.Animportantstudy(Heetal.,2012)showed
that15.6%ofleft-behindschoolchildrenhadadepressionrisk,significantlyhigher
thanthecontrolsafteradjustingfordemographicsandsocioeconomicstatus.The
studywascarriedoutinaruralcountyofChina’sHubeiProvince.Thedepression
symptomscoresdidnotdifferbetweenchildrenwithamigrantmother,father,or
bothparentsaway.SimilarstudiesinEnglish(Wangetal.,2015a)andChinese
(Houetal.,2015;Wangetal.,2011a;Wangetal.,2011b)thatappliedCDIalso
foundhigherdepressionprevalence(rangingfrom12.1%to24.8%)amongLBC
thancontrols.Thesamplesacrossthesestudiesincludeddiversegeographicareas
acrossmainlandChina.Additionally,alsobasedonCDI,Wangetal.(2015a)noted
thatboth-parentmigrationwasassociatedwithhigherlikelihoodofdepression
thanfather-ormother-onlymigrationinChongqingmunicipality,whereasWang
etal.(2011b)foundinHenanthatchildrenleftbehindbyoneparentwereless
likelytobedepressed,comparedtobothnon-LBCandchildrenleftbehindbyboth
parents.
Lingetal.’s(2015)studyadoptedanotherpopulardepressionscreeningtool,
DepressionSelf-RatingScale(DSRS)forChildren,inasmallsampleinHunan,
China.ThestudyfoundhigherdepressionscoreintheLBCgroupthanthecontrol
80
group,butthedifferencewasnotsignificant.MeanwhiletheAnxiety/Depression
subscalefromtheYouthSelf-Report,appliedinthesamestudy,indicatedworse
outcomesinLBC.Usingacut-offscoreofDSRS,aChinese-languagestudy(Liuetal.
2011),found27.1%ofLBChaddepressionrisk,muchhigherthantheproportion
amongnon-LBC(13.9%)inasmallsampleinHebeiprovince.
Arecentstudy(Ren&Treiman,2016)appliedtheCenterforEpidemiological
StudiesDepressionScaleforChildren(CES-D),andfoundnosignificantdifference
indepressionsymptomsbetweenruralLBCandruralchildrenlivingwithboth
parents.However,itisnoteworthythatthisstudyoversampledthemore
populatedvillages,whereavailabilityofresourcesrelatedtochildwellbeingmight
confoundtheresults.Withthesamedepressionscale,twostudiespublishedin
Chinese(Lanetal.,2009;Heetal.,2011)demonstratedhigherlevelsofdepression
inLBCthannon-LBC,afteradjustingforfactorsincludingsocioeconomicstatus.
Differentdepressionprevalencefromthesamesurveytoolsuggeststhevariation
ofthescaleofchallengeforLBC.Studiesmaynotbeabletoproperlyidentifythe
impactofparentalmigration,withoutusingtheinstrumentthatbestsuitedforthe
targetchildpopulation.Adjustingforcovariatesatmultipledimensionsisalso
importantinillustratingacomprehensivesituation.
Wuetal.’s(2015)studyprovidesanexcellentexampleoftheinvestigationof
familyandcommunitylevelvariables,amongbothcurrentandpreviousLBC.After
81
adjustingforfamilysocialcapitalandcommunitysocialcapital,childrenwhowere
currentlyleft-behindexhibitedsignificantlyhigherlevelsofdepression,than
nativeruralchildren(thosewhowereneverleftbehindormigratedwithparents).
Childrenwhowerecurrentlylivingwithbothparentsbuthadbeenleftbehindby
theirmigrantparents,showedsignificantlylowerlevelsofdepressionthanthe
nativeruralchildren.However,thesmallsamplesize(N=591)andthefive
categoriesofmigrationstatusresultedinverysmallandunequallysizedgroupsof
childrenintheanalysis,whichwasalimitationofthestudy.
Anxiety
Luoetal.(2011)usedtheState-TraitAnxietyInventorytoexaminethestate
anxiety(anxietyaboutanevent)andtrait-anxiety(anxietylevelasapersonal
characteristic)symptomsamongchildreninalargesampleacrossfiveChinese
provinces;theyfoundthatsubjectswhoexperiencedseparationwithfathershave
increasedstate-anxietybutnottrait-anxiety.Tomsa&Jenaro(2015)appliedthe
sameinstrumentinasampleinRomania,andfoundthatLBChadhigherlevelsof
bothstateandtraitanxiety.Anxietyscoreswerepredictedbycopingskills,
depression,andanger,controllingforthequalityoftherelationshipswithparent,
andsociodemographiccharacteristics.
Zhaoetal.(2014)foundincreasedsocialanxietyinLBCascomparedwiththeir
peersinruralAnhuiProvince,China.ThehigherSocialAnxietyScalesforChildren
82
(SASC)scoresinLBCwereassociatedwithlowerqualityoflife,poorerfamily
function,physicalabuse,beingfemale,havingmoresiblingsandminorities.Earlier
studiespublishedinChinese(Zhangetal.,2007;Zhaoetal.,2012)indicated
similarresultsfromthesamescale,showinghigherlevelsofsocialanxietyinLBC
thannon-LBC.Zhangetal.(2007)furthersuggestedthatLBCwhoweregirls,
younger,lackedpropercare,andwerelesshealthyweremorevulnerable.
Althoughthestudiesseemtoshowtheassociationbetweenparentalmigration
andanxiety,thespecificmechanismshavenotbeenclearlyelicited.Inparticular,
Zhaoetal.(2014)examinedvariousaspectsofwellbeinganditspotential
determinants,butthesefactorsareyettobeconnectedtoprovideaconvincing
explanation.Onereasonmaybethatanxietyaccompaniesmanymentalhealth
conditions,andmaynotbeagoodindicatorofLBC’spsychosocialchallengesdue
toparentalabsence.
3.2.8Behavioralissues
Studieshaveexaminedchildbehaviorsthatarerelatedtomentalhealthor
psychosocialwellbeing.Theirresults,inChina,SriLanka,andMexico,respectively,
suggestedthatLBChadmorebehavioralproblems.Lingetal.(2015)examined
behavioraloutcomesfromtheYouthSelf-Reportquestionnaire,andrevealed
higherrisksofexternalizingbehaviors,internalizingbehaviors,somaticdisorders,
anxiety/depression,socialproblems,andloneliness,inLBC.InSriLanka,
83
Senaratnaetal.(2011)reportedthatLBCweremorelikelytohavementalhealth
problems,basedonthescoresfromtheChildBehaviorCheckList(CBCL-S).
SimilarresultswerealsofoundinMexico(Lahaieetal.,2009)althoughthe
outcomemeasurementwasonlyaskingthecaregiverwhetherthechildrenhad
behavioralproblems.
Overall,thesestudiesshowedhigherprevalenceofbehavioralproblemsinLBC.
Yettheconnectionsbetweenparentalmigrationandbehavioralchallengesare
rarelydelineated.Lingetal.(2015)setagoodexampleofinvestingmultiple
dimensionsofbehavioralwellbeing,althoughotherfactorsinthechild’sfamily
andcommunityweremissedoutintheanalysis.Interpretationoftheunadjusted
effectsofparentalmigrationonchildoutcomesshouldalwaysbecautiousasmany
factorsmayinterferewithcorrelations.
NoneoftheChinese-languagepapersinthisreviewfocusedonbehavioral
problemsaskeyoutcomeindicators.Yetsomestudiesexploredbehavioralaspects
ofwellbeingmetrics.Zhaoetal.(2006)andGao(2010)reportedthatLBChad
higherlevelsofbehavioralproblemsthantheirpeers,accordingtothePiers-Harris
Children’sSelf-ConceptScale.Yangetal.(2013)notedthatamongLBC,senseof
hopewaspositivelyassociatedwithpro-socialbehaviors,basedontheresults
fromtheStrengthsandDifficultiesQuestionnaire.Thesefindingssuggestedthat
behavioralissuesinLBCmayentailavarietyofaspects,andwerecloselylinked
withotherdimensionsofpsychosocialwellbeing.
84
Additionally,someEnglish-languagestudiesconductedinChinahavefounda
higherprevalenceofhealth-relatedriskbehaviorsinLBC.Gaoetal.(2010)
examinedthedifferencesinhealthbehaviorbetweenLBCandnon-LBCinrural
China.BasedonaquestionnaireadaptedfromChineseandinternationalsurvey
instruments,theirresultsdemonstratedthatLBCwereathigherriskofskipping
breakfast,lowerratesofphysicalactivity,havingeversmokedtobacco,suicide
ideation,andbeingoverweight.Gaoetal.’s(2013)studybasedonasimilarsample
furtherindicatedthatwhilematernalmigrationincreasedtheadolescent’s
smokingrisk,paternalmigrationseemedprotectiveforsmoking;theimpactof
parentalmigrationonsmokingwaspartiallymediatedbyself-efficacy.Wen&Lin
(2012)alsofoundthatLBCtendtohavemoreunhealthybehaviorsbyasking
whetherthechildsmoked,usedillicitdrugsorbingedonalcoholinthelastmonth.
ThesestudiesindicatedthatLBCseemtohaveconsiderablehigherhealthrelated
riskbehaviors,althoughtheirphysicalhealthwasnotfoundtobeproblematic.Yet
themorein-depthquestionmaybeaboutthequalityofcareintheabsenceof
migrantparents,andtheexistingstudieslargelyfailedtoexploretheintermediate
mechanismsthatcausedriskybehaviors.
3.2.9Self-conceptandself-esteem
Self-conceptisbroadlydefinedasaperson’sperceptionofhim/herself,and
85
typicallyreferstohowonefeelsaboutone’sworthacrosstheevaluative
dimensionssuchassocial,academic,behavioralandphysicaldomains(Marsh&
Hattie,1996).ThePiers-HarrisChildren’sSelf-ConceptScale(CSCS)isa
self-reportquestionnaireaskinghowchildrenmayfeelaboutthemselves,asa
broadrangeofpsychologicalhealthassessmentsbysixdimensions:behavior,
intellectualandschoolstatus,physicalappearanceandattributes,anxiety,
popularity,andhappinessandsatisfaction.
FivestudiespublishedinChineseusedthisscale(Liuetal.2010;Zhaoetal.,2009;
Zhaoetal.,2012;Gao,2010;Zhaoetal.,2006),allofwhichsuggestedsignificantly
loweroverallself-conceptlevelascomparedtonon-LBC.Thetwo
English-languagepapersusingthesamescalesfurtherreportedthatLBChad
loweroverallself-conceptscoresinallsubscalesofPiers-HarrisCSCSexcept
physicalappearance(Liuetal.2010)andbehavioral/intellectualconceptscores
(Zhaoetal.,2009)intherespectivestudies.Additionally,someChinese-language
studiesindicatedthelowerself-conceptinLBCmaybeassociatedwithearly
separationfromparents(Zhaoetal.,2006)andlongerdurationofparental
migration(Gao,2010).HoweverinJamaica,Pottinger(2005)foundnodifference
inself-conceptbetweenLBCandnon-LBCaccordingtoPiers-HarrisCSCS
measurement,butthesamplesizewastoosmall(only54)todrawvalid
conclusions.
Asimilarbutlessgeneralpsychologicalinstrument,RosenbergSelf-esteemScale,
86
wasusedbyafewotherstudies.Zhanetal.(2014)indicatedthatLBCwhose
parentsbothmigratedhadlowerself-esteem,andparentalmigrationseemedto
havefurtherreducedself-esteeminchildrenwithinitiallowself-esteem.Luoetal.
(2011)indicatedthatchildrenleft-behindbyfatherhadlowerself-esteemthan
thoseleftbehindbymotherandthenon-LBC.Incontrast,Adumitroatieetal.
(2013)foundinRomaniathatself-esteemwashigherinteenagerswithboth
parentslivingabroad,ascomparedtotheteenagerswhoseparentswereinthe
country.
Althoughself-conceptdemonstratesanoverallwellbeingstatusofthechild,the
existingstudieshavenotseparatelyinvestigatedfactorsinfamily,school,and
communityenvironmentsthatmayaffectdifferentdimensionsself-conceptin
differentways.Self-esteem,whichcanbeseenasoneaspectoftheself-concept
measures,appearstobecontext-specificandmaybemoredependentonhow
migrationisviewedinthelocalcommunity.Thebiasduetoself-reportshould
alsobeconsidered,asmoststudiesdidnottriangulatestudyfindingswithchild
outcomesreportedbyotherrespondents.
3.2.10Factorsassociatedwithwellbeinginleft-behindchildren
Age(attimeofsurvey)
87
Childrenindifferentagegroupsrespondandadjustdifferentlytomajorlife
changessuchasparentalmigration.Theeffectofagemayvarydependingonthe
measuresofpsychosocialoutcomewhencomparingLBCwiththeircounterparts.
Thetransitionfromchildhoodtoadolescenceinvolveschangesinmultiple
domainsincludingphysical,sexualandsocialdevelopmentwithaconsiderable
rangeofindividualdifferencesintheageatwhicheachofthesechangesoccur.
Fanandcolleagues(2011)demonstratedthatyoungerLBCgenerallyhadmore
emotionalandbehavioralproblemsthanolderLBC(meanageofLBC=12.2±2.2
years)asismeasuredbytheSDQquestionnaire,evenafteraccountingforthe
durationofbeingleftbehind.AChinese-languagestudy(Zhangetal.,2007)also
foundthatyoungerLBChadhigherlevelofsocialanxiety,ascomparedtotheolder
ones.BothWuetal.(2015)andLanetal.(2009;publishedinChinese)foundolder
childrentendedtohavemoredepressivesymptomsthantheyoungerones.Liuet
al.(2014)furtherreportedthatyoungerchildren,bothLBCandnon-LBC,had
betteroverallmentalhealthstatus.
YetHuangetal.(2015)demonstratedthatolderagewasassociatedwith
improvementinhealth-relatedqualityoflifeamongLBC.Interestingly,Wangetal.
(2015a)revealedaU-shapedage-specificdepressionprevalencecurve:10–12
yearsagegrouphasthelowestrate,andtheyounger(middlechildhood:7–9years
group)orolder(adolescent:13–15yearsgroup,16–17yearsgroup)agegroups
haveahigherrate.
88
Ageasavariableintheabovestudiesispotentiallycorrelatedwithavarietyof
wellbeingfactorsthatemergeatdifferentstagesofchildren’slife.Asaresult,the
effectsofagemayvarydependingondifferentpsychosocialoutcomes.While
youngerLBCmightfeelparentalseparationsimplyasabandonment,someolder
childrenexperienceambivalentfeelingsofacceptance/respectand
grief/resentment,whileatthesametimebeingmoreawareofthefamily’s
situationandparents’effortsandsacrifices(Reyes,2007).However,asthe
transitionintoadolescenceinvolvesvariousphysicalandpsychologicalchanges,
childrenmayalsoencounterdifficultiesinadaptationthusbecomingmore
vulnerabletotheeffectsofparentalmigration(Heetal.,2012).
Ageatanddurationofseparation
Childageatwhichthemigrantparentsfirstdepartedhomewasfoundtobe
anotherimportantfactor,inrelationtomultipleaspectsofpsychosocialwellbeing,
especiallyintheempiricalstudiesconductedinChina.Thiswassimilartothe
theoreticalliteraturesummarizedinthepreviouschapter,whicharguedthat
separationfrommigrantparentsmaycauseadverseaffectsfromapsychological
developmentperspective.
Liuetal(2009)suggestedthatchildrenwhowereseparatedfromparentsata
youngeragehadmoresymptomsofanxietyanddepression,basedonasampleof
89
LBCacrossthreeprovincesinChina.Thiseffectwasespeciallypronouncedfor
childrenwhowereseparatedfromtheirmothersorfrombothparents.Luoetal.
(2011)foundinchildrenwithanabsentfathertheseparationfromthefatherfor
childrenaged7to12waslinkedwithgreaterstate-anxiety,comparedtootherage
groups,andgirlsseparatedfromtheirfatherbefore2years,andcurrentlyin
middle/highschoolshowedlowerself-esteem.Fanetal.(2010)alsodemonstrated
thatchildrenwhowereleft-behindatanearlieragetendedtohavemore
emotionalproblemsandlesspro-socialbehavior,accordingtoSDQscores.An
earlierstudy(Zhaoetal.,2006)publishedinChinesenotedthatyoungerageat
separationwasassociatedwithintrovertpersonalityandlowerself-conceptabout
physicalappearanceandattributes.AlongitudinalstudyconductedinNewYork
evensuggestedthatextendedmaternalseparations(definedasatleastone
month),beforeage5independentlypredictlong-termrisksforoffspring
borderlinepersonalitydisordersymptoms(Crawfordetal.,2009).
Thedurationofparentalabsencealsomatteredaccordingtothereviewed
literature.Fanetal.(2010)foundthatlongerseparationfromparentswas
significantlycorrelatedwithmorepsychopathology,especiallyemotional
symptoms.Similarly,Liuetal.(2009)showedthatseparationfromparentsbefore
7yearsoldwasassociatedwithmoresymptomsofanxietyanddepression,
especiallyforchildrenleft-behindbymothersorbothparents.Lingetal.(2015)
indicatedthatlongdurationofseparationhadanegativeimpactonthechild’s
psychologicaldevelopment,forchildrenabove6yearsoldatthetimeofseparation.
90
ForLBCunder6yearsatthetimeofseparation,theimpactofdurationinfluenced
onlyproblemswiththoughts.ManyChinese-languagepapersalsoincluded
durationofparentalmigrationasavariableintheirsurvey;mostofthemfound
longerseparationwascorrelatedwithworsepsychosocialoutcomes(Dong&
Zhang,2013b;Dong&He,2014;Luetal.,2014a;Zhaoetal.,2015;Luetal.,2014b;
Gao2010;Fanetal.,2013).Inparticular,alongitudinalstudy(Fanetal.,2013)
foundthatlongerdurationofparentalmigrationpredictedincreasednegative
emotionsattheendofa2.5-yearstudyperiod.
Yetsomepapersreportednosignificanteffectsfromthedurationofparental
migration(Yangetal.,2013;Jiang2013;Zhaoetal.,2006).Otherstudieseven
suggestedlongerseparationmightimprovechildresilience.Huangetal.(2015)
alsoshowedchildrenwhoseparentshadmigratedforalongerduration,scored
higherinqualityoflifemetrics.Jordan&Graham(2012)foundgreaterresilience
inchildwellbeingwasassociatedwithlongerdurationsofmaternalabsence.
Gender
AmongstudiesusingSDQ,bothFanetal.(2010)andHuetal.(2014)found
left-behindgirlsweremorelikelytohaveemotionalsymptoms,whereasboyshad
higherlevelsofproblemsinhyperactivityandpro-socialbehaviors,accordingto
resultsfromSDQscores.Huetal.(2014)andJiang(2013)comparedtheSDQ
scoresbetweenboysandgirls,andsuggestedthatboyshadhigherlevelof
91
difficultiesinconduct,peerrelationship,andprosocialbehaviors.Wickramageet
al.’sstudy(2015)inSriLankaalsorevealedhigherlikelihoodofanypsychiatric
diagnosisinleft-behindboysthangirls,basedonstandardizedcompositionSDQ
scoresreportedbyparent,teacherandself.
Manystudiescomparedthewellbeingoutcomesbetweenboysandgirls.Wuetal.
(2015)andZhaoetal.(2014)foundthatgirlshadhigherlevelsofdepressive
symptomsandsocialanxiety,respectively,thanboys.Similarly,Liuetal.(2009)
indicatedthatfemalestudentshadhigherlevelsoftraitanxietyanddepression
thandidmalestudentswhenbothparentsleftforthecities.Similarly,a
Chinese-languagestudy(Jiang2013)foundgirlswereworseoffinvariousmental
healthmeasures.Meanwhile,afewotherstudiespublishedinChinesesuggested
thatboysweremoredisadvantaged,intermsofsocialsupport(Zhengetal.,2014)
anddepressivesymptoms(Yangetal.,2010).
Someotherstudiesidentifiedwellbeingdisadvantagesrelatedtoparental
migrationamonggirls.Luoetal.(2011)foundfemalemiddle/highschool
studentswhowereseparatedwithfatherbefore2years,sufferedfromlower
self-esteem.Zhanetal.(2014)indicatedthatLBCgirlswhosefatherorboth
parentsmigratedouttoworkinothercitiesareinclinedtoscorelowerinthe
self-esteemtestthanboys.Incontrast,Zhouetal.(2011)andZhu(2016)found
amongboysthatLBChadworsementalhealthstatusthannon-LBC.Gaoetal.
(2010)reportedthatascomparedtonon-LBC,left-behindteenagegirlsinChina
92
weremorelikelytousesubstancesandhaveemotionalproblemsandexpress
sadness,whilstleftbehindteenageboysweremorevulnerabletobeing
overweightandaddictedtotheinternet.
StudiesconductedincountriesotherthanChinaimpliedthatthesocio-cultural
contextsplayedcertainrolesinthegenderdisparitiesofchildwellbeing,especially
thecommunitieswheretraditionalgendernormsstillprevail.Aguilera-Guzmanet
al.(2004)showedthatleft-behindboysinMexicofeelthepressureoftheirroleas
providers,andfemaleLBCsfeltthepressureofhavingtodomorehousework,such
ascookingorwashingclothes.Theauthorssuggestedthatgender,notfather's
absence,appearstobethemostimportantriskfactorforpsychologicaldistress.
Tomsa&Jenaro(2015)foundboysappeartousemoredistractionstrategiesto
copewithvarioussituations,whereasgirlsseemtoexperiencemoreanger.
Parentalmigrationmayaffectboysandgirlsindifferentways.Manystudies
elicitedasimilarpattern:girlsseemtodemonstratemoreemotionaldistressor
internalizingproblems,whereasboystendedtohavemorebehavioralor
externalizingproblems.Hencegendereffectsmayvarybetweendifferent
wellbeingoutcomes,yetmoststudieswereunabletodistinguishbetweenthem.
Additionally,socioculturalcontextsplayasignificantroleingendernorms,which
shouldbefurtherexploredindifferentsettings.
Patternsofparentalmigration
93
TheimpactofmigrationonLBC’swellbeingmayvaryacrossdifferenttypesof
migranthouseholds,i.e.familieswithfatherormotherorbothparentsmigrating,
leavingchildrenbehindinthecareofgrandparents,otherrelatives/friendsor
evenalone.
Manystudiescomparedtheeffectsofone-parentandboth-parentmigration.For
example,Suetal.(2012)foundchildrenwithtwoparentsmigratingreportedthe
lowerleveloflifesatisfactionthanchildrenwithonlyonemigrantparentandthe
non-LBCgroup.Gaoetal.(2010)demonstratedthatcomparedtothosewithone
migrantparent,childrenwithbothparentsabsentweremorelikelytoengagein
riskbehaviors,tohaveemotionalandmentalproblemsandtobeoverweightor
stunted.Incontrast,Li&Tao(2009)indicatedthatchildrenleftbehindbyone
parentweresignificantlymorelikelytohavesuicidethoughts,comparedto
childrenleftbehindbybothparentsaswellasnon-LBC.
Otherscomparedmother-onlyandfather-onlymigration,anddemonstratedmixed
results.Somestudiessuggestedthatchildrenlivingwithfathersandleftbehindby
mothersseemedmoredisadvantagedthantheotherwayaround.Gaoetal.(2010)
revealedthatchildrenwithamigrantmotheronlyweremorevulnerableto
behavioralrisks,comparedtothosewithfather’sabsenceonly.Wen&Lin(2012)
indicatedthatchildrenleftbymigrantmothers(butnotfathers)arethemost
disadvantagedtypeofparentalmigration,intermsofhealthbehaviorandschool
94
engagement.InRomania,Adumitroatieetal.(2013)foundteenagersleft-behind
onlybymothershadalowerlevelofpsychologicaladjustmentcomparedwithboth
thegroupsoffather-absenceandboth-parentabsence,especiallyintermsof
aggressiveness,self-esteem,self-adequacyandworldview.Asis(2006)described
howchildrenleftbehindbymothersinthePhilippineshadworseemotionalhealth
comparedtochildrenoffathermigrants,althoughnostatisticalsignificancewas
examined.Wen&Lin(2012)pointedoutthatthisdisadvantageseemspartly
attributabletodeficiencyinfamilycohesionandfamilysupportinmother-only
migranthouseholds,withchildrenparticularlyreluctanttoseetheirmothers
migrate.Inmanysocietiestheprevailinggenderscriptofmotheraschildren’s
nurturerandfatherasbreadwinnerscouldshapeLBC’sexperiencewithregardto
mothersabsence(Grahametal.,2011).
Ontheotherhand,therearealsofindingsindicatingthegreaterimpactoffather’s
absencethanmother’s.Luoetal.(2011)reportedthatafather’sabsencehada
greaterinfluenceonthestate-anxietyandself-esteemofleft-behindadolescents
thanthatofthemother’s.Vanoreetal.(2015)notedthatthemigrationoffathers,
butnotthemothers,resultedinworsepsychosocialoutcomesforchildren.Dreby
andStutz(2012)exemplifiedthat,byunderstandingthesacrificeoftheirmigrant
mothermakesinordertosupportthem,LBCmayseetheirmother’smigration
moreacceptableovertimeandderivehighereducationalaspirationinreturn.
Meanwhile,Heetal.’s(2012)studyondepressionandZhaoetal.’s(2014)studyon
95
socialanxietyalsoindicatednosignificantdifferenceintheirfindingsamongthree
subgroupsofLBC(father,mother,andbothparentsaway).Graham&Jordan’s
study(2011)basedonSDQresultsalsoconcludedthattherewasnoevidencein
anyofthecountrysamples(Indonesia,Vietnam,Thailand,andthePhilippines)
betweenthefather-migrant/mother-caregivergroupand
mother-migrant/father-caregivergroupinpsychologicaldisordersthanchildren
ofamigrantfather,althoughonlyoneofthesetwogroupsappearedsignificantly
betterofforworseoff,ascomparedtothenon-LBCgroupineachofthefour
countries.
Father,mother,orboth-parentmigrationmayalsoaffectdifferentaspectsofchild
wellbeing.Wenetal.(2015)foundfather-onlymigrationcorrespondedtohigher
educationalaspirationsandhighlevelsofnoproblembehaviors;mother-only
migrationwaslinkedtobetterself-ratedhealth,and;both-parentmigrationwas
associatedwithhigherschoolgradesandhighereducationalaspirations.Lietal.
(2012)suggestedthatmother-onlymigrationwaslinkedwithsmokingbehaviors
andviolenceunderalcoholinfluence,andboth-parentmigrationlinkedwith
frequentdrinkingandmoretimeontheinternet.
Studiesthatinvestigatedchildren’spreviousexperienceofbeingleft-behindalso
showedmixedresults.Adhikarietal.(2013)foundinThailandthatmother’s
earliermigrationhistoryhadasignificant,independentassociationwith
psychosocialdifficulties.Huang&Li(2007)showedbothcurrentandprevious
96
LBChadworsementalhealthstatusthannon-LBC.However,Wuetal.,(2015)
indicatedthatchildrenwhohadpreviouslybeenleft-behindtendedtoexperience
fewerdepressivesymptoms.
Giventhecomplexityofpatternsofparentalmigrationandfamilystructure,a
majorgapintheliteraturewasthelackofinvestigationontheprimarycaregiver
livingwiththechild.Whetherthefather,motherorbothareawaydoesnot
determinetheprimarycaregiverathome.Children’sopinionsaboutwhocaresfor
themthemostmayalsoreflectcrucialfamilycharacteristics.Meanwhile,a
variationacrossdifferentcontextsisobservedregardingthedifferencesbetween
typesofparentalmigration,whichsuggeststheimportanceofconsidering
socio-culturalsettingswhenanalyzingtheimpactofmigrationtypeonLBC.
Genderrolesandnormsmaypotentiallyshapetheparent-childrelationshipand
children’sexperiences.Localtraditionmayalsoleadtoadominanttypeofmigrant
(suchasmigrantfathers)andconsequentlyatypeofcommunitysupportnetwork
comprisedoftheremainingfamilymembers.
Caregiversandsiblings
InChinaevidencesuggeststhathavingoneparentathomeisbetterforthe
well-beingofLBCthanlivingwithgrandparents(Jia&Tian,2010;Zhaoetal.,2009;
Huang&Li,2007).ChildreninthecareofgrandparentsinChinaandSoutheast
Asiaappearedtobemorepsychologicallyadvantagedthanthosewhowerecared
97
byotherrelativesorfriends(Zhaoetal.,2009;Fanetal.,2010;Jiang,2013;
Graham,2011;Senaratnaetal.,2011),andoldercaregiversseemedmore
favorabletochildren’swellbeingthanyoungercaregivers(AdhikariR,etal.2013;
Fanetal.2010).MexicanLBCwhoweretakingcareofthemselveswerefoundto
bemuchmoredisadvantagedthanotherLBC(Lahaieetal.2009).
ItisalsonoteworthythatchangesofcaregiverscouldhavenegativeeffectsonLBC.
Senaratnaetal.(2011)foundinSriLankathatchangeofprimarycaregiverwas
twiceormoretimesassociatedwithbehavioralrisks.Mazzucatoetal.(2015)
foundthatchildrenlivingintransnationalfamilieswhochangedcaregiversoneor
moretimeshadworsepsychologicalwell-being,ascomparedtothatofchildren
livingwithbothparentsandwhohaveneverchangedcaregiver,inGhana,Nigeria,
andAngola.
Thispatternoftheeffectofdifferentcaregiversseemstobequiteconsistentacross
countriesfromthereviewedliterature.Agradientofpsychosocialwellbeingis
observedinrelationtocaregiverarrangements,withchildrencaredbyoneparent
atthetop,followedbygrandparents,otherfriends/relatives,andLBClivingalone
atthebottom.
Withregardtootherhouseholdmembers,Senaratnaetal.(2011)identifiedthat
nothavinganeldersiblingwasassociatedwithmorebehavioralhealthproblems
inLBCinSoutheastAsia.Inaddition,Adhikarietal.(2013)andZhanetal.(2014)
98
examinedtheeffectofsiblingsasacontrolvariableamongallchildren,andfound
moresiblingswascorrelatedwithbettermentalhealth.However,inChina,some
reviewedstudiesfoundbeingasinglechildtobeaprotectivefactorforchild
wellbeing.Zhengetal.(2014)andJiangetal.(2013)indicatedthatbeingasingle
childwasassociatedwithhighersocialsupportandgreaterresilience.Zhaoetal.
(2014)foundthatamongLBC,childrenwithfewersiblingstendedtohavelower
levelsofsocialanxiety.Additionally,Yangetal.(2010)andWangetal.(2011a)
suggestedthatwhethertheLBCwassinglechildornotdidnotaffecttheir
wellbeing.
DespitetheOneChildPolicy,manyLBCstillhadsiblingsathome,althoughthere
seemedtobemixedeffectsofhavinganysiblings,especiallywhenavailabilityof
resourcesforchilddevelopmentwasconsidered.Whilecaregiversandsiblings
comprisedLBC’sco-residenthouseholdmembers,moreinsightsregardingthe
relationshipsintheextendedfamily,includingthemigrantparents,were
necessarytobetterunderstandthedisruptedfamilystructure.
Familyrelationships
SeparationbetweenLBCandtheirparentsdoesnotnecessarilymeanthereisno
communicationbetweenthem.Moderncommunications,suchasinternetand
mobilephones,contributetokeepingtheabsentparents“virtuallypresent”.They
99
mayalsovisithometomaintainrelationshipswithchildrenandsustainthefamily
functionfromafar.
InChina,themostimportantdaysforfamilyreunionareduringtheChineseNew
Year,whenmostmigrantswillreturnhomeintheone-weekpublicholiday.Ina
studyincluding1024LBCwithmigrantfathersacrossfiveprovinces,Luoetal.
(2011)notedthatover50%offatherscouldcomehomeateveryChineseNew
Year,whiletherestvisittheirhomelessfrequently;22%ofLBCcontactedtheir
fatherseverydaybyphone,andmorethan65%contactedthemeveryweek.Suet
al.(2012)reportednearly60%ofLBCwithtwoparentsmigrating(n=283)would
“always”contacttheirfathersand70%ofthemwoulddosowiththeirmothers.
However,LBCintransnationalfamiliesinothercountriesmaynotbeabletosee
theirimmigrantparentsasoften.Forinstance,inthePhilippinesmigrantworkers
mayvisittheirfamiliesathomeeverytwoyearswhenrenewingtheircontractsor
lessoftenforthehigh-skilledworkers(Asis,2006),andmanyundocumented
LatinoimmigrantsinAmericadonotvisittheirhomecountryatall,sincetheyare
undocumentedandhencenotallowedtocrossbackintotheUS.
Nevertheless,thepresenceofmigrantparentsthroughcommunicationorhome
visitsseemstobenefitthepsychosocialwellbeingofLBCacrossdifferentcontexts.
Jia&Tian(2010)demonstratedthatLBCwhohadbadrelationshipandlow
frequencyofcommunicationwithparentswerepronetoencountermore
100
loneliness.Liuetal.(2010)examinedwhetherandhowLBCliketheirmigrant
parentsandcaregivers,andtheresultsfrombothquestionswerepositively
associatedwithchildren’sself-concept,aswellasthefrequencyofhomevisitsand
communication(formigrantparents).However,simplemeasuresof
communicationfrequencymaynotreflectthequalityofparent-childrelationship,
sinceitthechildmayunwillinglytalkwithmigrantparentsoverthephoneunder
pressureofco-residentcaregivers.
Wangetal.(2015a)furtherlookedintothecharacteristicsofcommunication
betweenLBCandtheirmigrantparents.Thestudyshowedtheprotectivefactors
fordepressioninLBCincludedhighfrequencyofparent-childcommunication,
communicationbytelephone,andtalkingaboutdailytrivia,learningexperience,
schoollife,andfeelings.
AfewChinese-languagestudiesexploredanotherperspectivebyaskingabouthow
closewasthechildwithmigrantparents,orhowmuchtheparentscaredforthe
child.Hanetal.(2012)showedthat,“veryclose”parent-childrelationshipwas
linkedwithhighersenseofsecurityinLBC,althoughtheeffectdisappearedafter
adjustingforresiliencemeasures.Zhaoetal.(2012)andZhaoetal.(2015)found
that,higherextentofcarethechildreportedtoreceivefromtheparentswas
associatedwithbettermentalhealthoutcomes.
101
Relationshipbetweentheparentsasoneorbothofthembecamemigrants,wasa
crucialcharacteristicofthefamilyenvironment,butveryfewstudiesincluded
parentalmaritalstatusasavariable.SomeauthorssuchasWuetal.,(2015)
excludedchildrenofdivorcedparentsintheirsample.Yanetal.,(2010)notedthat
childrenofdivorcedparentsweremorelikelytohavedepressionsymptoms.Liuet
al.(2014)examinedtherelationshipbetweentheparentsandfoundittobean
importantpositivecorrelateofchildmentalhealth.YetHuetal.(2014)found
divorcewasnotassociatedwithpsychosocialstrengthsanddifficulties.
Despitethelong-termphysicalabsenceofmigrantparents,theirrelationshipwith
thechildappearedtobeanimportantwellbeingfactor.However,fewstudies
examinedotherrelationshipsinthefamilythatmaybemeaningfulinchild
development,suchastherelationshipbetweentheparents.Therelationshipsthat
involvedcaregiverswerealsorarelyexamined.Alternatively,thefamily
environmentmightalsoberegardedasavariable,andtheoverallcharacteristics
ofthehouseholdwereexaminedbysomestudiesasiselaboratedbelow.
Familyfunctionandsocialsupport
Twostudiesexaminedchild-reportfamilyfunctionusingAPGAR,arapidscreening
toolwithquestionsaboutadaptation,partnership,growth,affection,andresolve.
Zhaoetal.(2014)showedpoorerfamilyfunctionwaslinkedwithhighersocial
anxietyinLBC.Jordan&Graham(2012)furthernotedthatchildassessmentof
102
familyfunctioning,anexplanatorymeasureacrosswellbeingmeasures,seemedan
importantfactorinvariationinchildwellbeingbeyondthemigrationeffects.The
familyfunctioningseemstobeanimportantaspectoffamilycharacteristics,yetits
definitionmayneedfurtherclarificationintermsoftherelevancetochildcare.
Inadditiontofamilyfunctioning,socialsupportmayprovideimportantresources
thatbenefitchildpsychosocialwellbeing.Severalstudiessuggestedsocialsupport
wasaprotectivefactorofLBC’swellbeing(Chenetal.,2011;WangWetal.,2014;
Zhengetal.,2014;Fanetal.,2013;Liuetal.,2014;Taoetal.,2014;Wuetal.,2015).
Forexample,Zhengetal.(2014)foundthatLBChadlowerlevelofsocialsupport
thanthecontrolgroup,andhighersocialsupportwasassociatedwithbetter
overallmentalhealth. Taoetal.(2014)showedthatlesssocialsupportforLBC
maybeanimportantriskfactorfortheirmentalhealth,althoughLBChad
significantlybetterfamilyeconomicstatusthannon-LBC.
Morespecifically,alongitudinalstudy(Fanetal.,2013)publishedinChinese
revealedthat,socialsupportparticularlymediatedthepredictivecorrelation,
betweenfamilyatmosphereatbaselineandemotionalwellbeingatthefollow-up
surveyafter2.5years.Wuetal.(2015)alsoconcludedthatparentalmigration
influencedchildren'smentalhealththroughthemediatingeffectsoffamilyand
communitysocialcapital.Thesestudiesillustratethepositionoffamilyandsocial
supportonthepotentialcausalpathwaysbetweenparentalmigrationandchild
103
wellbeing,andunderscoredtheimportanceofadequatesupportfromoutsidethe
nuclearfamilyinimprovingLBC’ssituations.
Socioeconomicstatus
Thehouseholdeconomicsituationandparent/caregiver’seducationallevelscan
positivelyaffectchildwellbeingandmaymediateorconfoundtheimpactof
migrationonLBC.Whilemigrationwasusuallyaimedathigherincome,the
relativesocioeconomicstatusofmigranthouseholdspriortoparentalmigrationis
unknown.Therefore,theLBCmaynotalwaysenjoyeconomicadvantages
comparedtotheirpeers.
MoststudiesinChinathatcomparedeconomicstatusbetweenLBCandchildren
livingwithbothparentssuggestednosignificantdifferences(Heetal.,2012;Jiaet
al.,2010;Liuetal.,2010;Suetal.2012;Wen&Lin,2012).YetWenetal.(2015)
notedthatthehighestfamilyincomelevelwasfoundintwo-parentmigration
familiesandthelowestfoundinnon-migrantfamilies.Taoetal.(2014)andGaoet
al.(2010)alsosuggestedhighersocioeconomicstatusinmigrantfamilies,using
indicatorsofself-reportannualincomepercapitaandthepossessionof
multi-storeyhouse,respectively.Zhaoetal.(2014)furthersuggestedparental
educationlevelswerelowerinLBCthannon-LBC,althoughthisdidnotshow
associationwithsocialanxietyinmultivariateanalysis.
104
YetFanandcolleagues(2010)foundthattheleft-behindfamilieshadlower
economicstatusthanothers,andtheLBC’semotionalandconductproblemswere
significantlycorrelatedwithlowerfamilyeconomicstatusandlowereducation
levelofparentsandcaregivers.Wenetal.(2015)showedthatfather'seducation
wasinfactthelowestinboth-parentmigrationfamiliesamongallmigration
types.
AfewstudiespublishedinChineseindicatedbettersocioeconomicstatuswasa
protectivefactorforwellbeingoutcomesamongLBC,takingintoaccountof
parentalmigrationandothercovariates.Jiang(2013)showedthatself-report
familyincomestatuswaspositivelyassociatedwiththechild’spsychosocial
strengths.Dong&Zhang(2013b)andWangetal.(2014)foundLBCwhose
motherhadhighereducationlevelweremorelikelytohavebetterpsychological
outcomes.Inaddition,Gao(2010)notedthathighereducationlevelofthe
caregiverwasalsolinkedwithhigherself-conceptoftheLBC.
WhilemoststudiesimpliedlinksbetweenhigherSESbetterpsychosocial
wellbeing,theimpactofparentalmigrationoftenseemedtobeindependentfrom
thedifferencesinSES.Wenetal.(2015)indicatedthatincreasedincomein
migrantfamilieshelpedexplainsomeoftheimpactpatternsofparentalmigration
onchildwellbeing.However,Lahaieetal.(2009)demonstratednoassociation
betweenmonthlyremittancesandanyoftheacademic,behavioral,oremotional
outcomesamongLBCintheirstudyinMexico.Jordan&Graham(2012)further
105
notedthattheyfoundnoevidenceofamoderatingeffectofhighersocio-economic
statusonparentalmigrationinSoutheastAsia,associo-economicindicatorsare
consistentlyassociatedwithvariationsofwellbeingregardlessofparental
migrationstatus.
Mostofthemeasurementsofsocio-economicstatuswererathercrudeinthe
reviewedliterature,whichmakestheresultsdifficulttointerpretorcompare.For
example,somestudiesonlyusedasingleproxyindicatortoindicateeconomic
status,suchashousingtype(Gaoetal.,2010),andchildreninapublicorprivate
school(Smeekensetal.,2012).Otherindicatorssuchasself-reportedhousehold
income(Jia&Tian,2010;Jiang,2013)andwealthlevel(Suetal.,2012;Liuetal.
2010;Wen&Lin,2012)werealsoover-simplifiedandsubjecttoreportingbias.
Thewealthindex,calculatedfrommultipleindicatorsmaybeabetterapproachto
demonstrateeconomicstatus(Graham&Jordan,2011;Fanetal.,2010).
School,peers,andcommunity
Outsidethehousehold,theextantliteraturealsolookedintoLBC’spsychosocial
wellbeingintheirschoolandcommunityenvironments.English-languagestudies
conductedinChinaindicatedthatLBCweremorelikelytodropoutofschool(Gao
etal.2010)orhavelowerschoolengagement(Wen&Lin,2012)comparedto
non-LBC.ThreeChinese-languagestudies(Lanetal.,2009;Liuetal.,2011;Wang
etal.,2011a)furthersuggestedthatinadditiontoparentalabsence,lowerschool
performancemayincreasedepressionriskinChinesechildren.YetJordan&
106
Graham(2012)foundinSoutheastAsiathatLBCshowednodifferenceinschool
performanceandenjoyment,althoughlongerdurationofparentalmigration
seemedtohavenegativeinfluenceonschoolperformance.
Incontrastwiththeabovestudies,astudyinGuangxi,China(Wenetal.,2015)
foundfather-onlymigrationcorrespondedtolowerincidenceofproblem
behaviorsandhighereducationalaspirations,comparedwithnot-left-behind
children.Both-parentmigrationwasassociatedwithhigherschoolgradesand
highereducationalaspirations.
Thesecontradictoryfindingsimplythecomplexityintheeffectsoffactors
associatedwithLBC’swellbeingintheschoolenvironment.Aninterestingstudy
publishedinChinese(Houetal.,2015)investigatedtheeffectsofschoolclass
compositionintermsoftheproportionsofLBC,childrennotlivingwithparents
forotherreasons,andchildrenlivingwithbothparents.Theresults
demonstratedthathigherproportionofchildrenfrom“normal”familieswas
associatedwithlowerself-esteeminLBC.
Interactionwithpeers,whetheratschoolorinthecommunity,mayreflectan
importantaspectofachild’ssociallife.Someotherstudieslookedintotheeffects
oftheoverallrelationshipwithpeersonchildwellbeing.ThreeChinese-language
papers(Zhaoetal.,2012;Wangetal.,2014;Zhaoetal.,2015)reportedthat
positiverelationshipwithpeerswasassociatedwithbetteroverallmentalhealth.
107
Wangetal.(2011b)morespecificallymeasuredqualityoffriendshipwithpeers,
andfoundbetterfriendshipwerelinkedwithlesslonelinessanddepression.
Outsidethechild’sfamilyandschool,twostudiesalsoincludedvariables
regardingcommunity-levelcharacteristics.Houetal.(2015)foundthatchildren
livinginalow-crimeneighborhood,whichwasusedasaproxyindicatorofbetter
communitydevelopment,werelesspronetodepression.Zhaoetal.(2015)
reportedthatbetterrelationshipwithneighborswasassociatedwithbetter
overallmentalhealthoutcome.However,thesestudiesdidnotincludeany
economicstatusmeasuresoranyothercommunity-levelvariables,andthusthis
linkmaybeconfoundedbymanyotherfactors.
108
Table 3.4: Summary of reviewed qualitative studies
Author, Year
Location Sample Age Main measures
Arnold
2006
African
Caribbean
women
(who migrated to
the UK)
20 women (retrospective
study)
35 to 50 In-depth interviews
using the Separation–
Reunion Interview
Schedule
Asis 2006 LBC in the
Philippines
Adolescent children of
migrants (n=11), left-behind care-givers
(n=8) and community
workers involved in
organizing families of migrants (n=4)
About 10 to 12 Focus group
discussions
Coe 2011 Ghana (a
suburb of
the capital city Accra)
24 LBC 8 to 22 Focus group
discussions
Dreby
2007
USA and
Mexico
44 migrant parents in the
United States and 60 LBC
and 37 of their caregivers in a number of sites in
Mexico.
Pre-adolescent
s, adolescents,
and young adults
In-depth interviews
Dreby &
Stutz, 2012
Mexico 51 children of migrants
and 30 of their caregivers in Mexico.
16 were aged
between five and eight, 12
between nine
and twelve, 10
between thirteen and
fifteen, and 13
sixteen or
older
Structured interview
with the younger children and a semi-
structured one with
the older children;
interviews with young adults were open
ended
109
Graham et
al. 2012
Indonesia
and
Philippines
32 LBC (16 in each
country)
10 to 12 Semi-structured
interviews
Hoang & Yeoh,
2015
Vietnam 15 LBC, 37 caregivers 12 to 15 (children)
In-depth interviews
Hoang et
al., 2015
Indonesia,
Philippines, Vietnam,
Thailand
50 caregivers and 16
children in each country
9 to 11
(children)
In-depth interviews
Liu 2009 Jiangxi,
China
38 LBC, 32 caregivers, 15
teachers
Under 18
(children)
In-depth interviews
Murphy
2014
Jiangxi,
China
41 caregivers (parents), 59
children (33 were matched
pairs), 12 teachers
not reported In-depth interviews
Pantea
2012
Romania 24 grandmothers not reported In-depth interviews
Parennas 2008
Philippines 69 young adult children of migrant parents
young adults Open-ended interviews
Pribilsky 2001
Ecuador At least two family members in each of the 15
households (e.g. a father
and a child, mother and a
child, husband and wife)
not reported Open-ended interviews
Ye & Pan,
2011
Anhui,
Henan,
Hunan,
Jiangxi, and Sichuan,
China
Mixed methods: qualitative
sample not reported
not reported Structured and
semi-structured
interviews; group
interviews
Zou 2012* Guizhou 3 fathers and 6
grandparents from 4
families
Not reported Semi-structured
interviews
Zhang et
al., 2015*
Hunan 28 LBC and their caregiver
and teacher
Students at
Year 8
Semi-structured
interviews
110
*:StudiespublishedinChinese
3.3Qualitativestudies
Atotalof16qualitativestudiesacrossavarietyofgeographicalregionswere
includedinthereview(Table3.3),onlytwoofthemwerepublishedinChinese
(Zou,2012;Zhangetal.2015),and14werepublishedinEnglish.Amongthe
English-languagestudies,threewereconductedinChina,threeinLatinAmerica,
fiveinSoutheastAsia(threewerebasedontheCHAMPSEAproject),oneinAfrica,
oneinEasternEurope,oneinWesternEurope(retrospectivestudy).
3.3.1Migrationdecision
Accordingtostudiesinmultiplecountries,mostchildrenwereabletounderstand
theirparents’decisionofmigrationasfulfillingtheirroleofsecuringbetter
economicstatusforthefamily.Theywereabletorecognizethattheincreased
incomewouldnotonlyimprovetheireconomicwellbeing,butalsocovertheir
schoolcosts.
Hencemostchildrenseemedtosupportthedecision,andsomeindicatedthatthey
thinktheparentswereforcedtoleavethembehind.Althoughchildrenwould
111
rathernotlettheparentsgo,theyhadtoacknowledgetherealitythattheparents
areaway,andlearntolivewithit.Murphy(2014)foundamongmanyChinese
left-behindgirlsthattheystressedthematerialbenefitsoftheirfathers’migration,
whilebeingemphaticthattheywantedtheirmotherstoliveathome.
Yetthedecisionstomigrateorreturnweremadealmostexclusivelybytheparents
onbehalfoftheirchildrenandfamily’sbestinterests.Children’svoiceswere
largelyabsentduringthedecision-makingprocess.Parentsalsomakedecisions
aboutwhetherchildrenshouldstayathomeormigratewiththem.Childrenare
oftenthelastinthelinktomoveabroad,inpartbecauseofparents’concerns
aboutthedangersandcostsofchildren’smigrationaswellastheirdesirethat
childrenbeeducatedintheirhomecountry(Hondagneu-Sotelo,1994;Levitt,1998;
Massey,Alarcon,Durand,&Gonzalez,1987).
AreturnmigrantfatherinVietnam,interviewedbyHoang&Yeoh(2015,p186),
smiledasheexplained“itisusastheirparentswhomakethedecision.Theyarejust
kidsandtheyallwouldsaythesamethingwhentheyhavetobeseparatedfromtheir
mom.”
Consequently,asHoangetal.(2015)alsofound,somechildrenputupresistanceto
theirparent’simminentdeparture,beggingthemtostayoraskingsomeoneelsein
thefamilytointervenesothatthemigrantwouldchangehisorhermind.Even
aftergraduallygettingusedtotheabsenceofparentsandacceptingtheeconomic
112
reasonsoftheirdecision,thechildrenwouldnotcomfortablyaccepttherationales
oftheextendedseparation.AsYe&Lu(2011,p373)foundinaChineseLBC’s
narrationsduringgroupinterviews:
I’malwayswonderingwhymyotherclassmatescanhaveparentallovewhileI
can’t.ThisquestiondrovememadandIfeelsoboredwithlife.Myparentshave
beenworkinginthecityfor11yearsandtheyonlycamebackduringSpring
Festival.SometimesIthinkmydadwon’twantmeanymore.
InMexico,accordingtoself-report,childrenhadlittleagencyontheirfamily
decision-makingprocesses,withparentstendingtomakethemostofthedecisions
(Esteinou,2004).Transnationalfamilies,attimes,evensentchildrenbackhome
whentheymisbehavedintheUnitedStates(Grasmuck&Pessar,1991).
Interestingly,Dreby’s(2007)qualitativestudysuggestedthatchildrenmay
manifestagencyinshapingfamilies’migrationtrajectories,byexpressingfeelings
ofpowerlessnesswithdisplaysofemotionalwithholding,especiallyamong
youngerLBC;indoingsotheypressuredparentstoreturntoMexico.Adolescent
childrenwereabletomakemoredirectclaimstotheirparents’limitedresources,
evenbyactingaggressivelytofamilymembers.Forexample,inordertofulfilltheir
ownmigrationaspirations,manyolderchildrenpressureparentstomigrateusing
thefamily’ssocialcapital.
113
3.3.2Gendernormsandsocio-culturalcontext
Similar to findings fromquantitative studies, qualitative interviewselicited that
migrant fathers and mothers play different roles in migrant families. This is
largelydeterminedby the gendernorm in the local community and society.On
one hand, the absence of the roles that parents used to play in the care
environmentmayhavevariouseffectsonthechild,acrossmigrantfamilieswith
differentstructures.Ontheotherhand,children’sviewsonthedepartureoftheir
parentsmayalsodependonthelocalsocio-culturalsetting,whichleadtovariable
experiences.
Inmostsocio-culturalsettings,mother’sroleisassociatedwithemotional
intimacy,domestichouseworkandmostofotherchildcaretasks.Theleft-behind
motherswithnopaidemploymentoftenassumeboththediscipliningofchildren
andproductiveroles,suchasagriculturaltasks,previouslyundertakenbytheir
husbands(Hugo,2002).Meanwhile,father’sroleisassociatedmostlywith
economicprovisionsforthefamily.Migrantfathersmayfurtherrepresentan
advancedmodernlifestylethatthechildcouldliveupto.Theyalsocommand
respectthroughprovidingresourcestomeettheirchildren’sneedsandaspiration
ofbeingasuccessfulmigrant,ratherthandirectlythroughstrictforce(Pribilsky
2001).
114
Theshiftsinparentalrolesduetomigrationmaybeinterpreteddifferentlyby
children.Graham(2012)notedthatinSoutheastAsia,children’sstrongersenseof
abandonmentaftertheirmotherslefttoworkabroadmaybeduetotheprevailing
genderscriptsformothers,as“lightsofthehome”or“nurturers”.Similarly,in
Mexico,Dreby(2012)suggestedthatFathers’periodicabsences,although
distressing,donotdisruptchildren’slivesasmuchaswhenbothparentsmigrate.
InChina,Murphy(2014)foundthatchildrenlearnedthatlifeworkinginthecities
is“bitter”forthemigrantfathers,andsoistheirmothers’livingathometaking
careofthefamily.Yet,thestudyshowedthatchildrenfrommother-onlymigrant
householdsdidnotshowasmuchappreciationtotheirco-residentfathers.Inthe
study(Murphy,2014)agirlwhosefatherwasalocalvillageaccountant,saidher
mothersacrificedmuchforherandhersisterwhileherfatherenjoyedaneasylife.
3.3.3Emotionalandbehavioralimpact
Evenifthechildreluctantlyacceptedthefactofparents’migration,theseparation
andlosswouldstilldisturbthechildrenverymuch,asmanybecameemotional
whenrecallingthemomentofdeparture(Hoang&Yeoh,2015).Qualitativestudies
elicitedvariousemotionalexperiencesofLBCinresponsetothedepartureand
extendedabsencefromhome.Manychildren,atallagelevels,demonstrated
negativefeelingsofloneliness,vulnerability,abandonment,andevensymptomsof
depressionandanxietyaftertheirparents’departures(Dreby,2007).
115
Astheemotionalbondsbecamefragileduetothelongdistance,childrenmay
perceiveastronglackofparentalloveandmanifesteddeficitsofhappiness.Asa
girlinLiu’s(2009,p78)studysaid:
IfeelthatIhavereceivedlittlemotherlovebecauseIhavelivedapartfrom
myparentsfromaveryyoungage.Inthedepthsofmyheart,Ithirst
desperatelytostaywithmyparents.
Grahametal.(2012)foundthattheabsenceoftheirmothermaycreate
vulnerabilitiesrelatedtoseekingemotionalsupport.Somechildren’snarratives
indicatedtheirresponsetofeelingsadwasnottotellanyoneandkeptitasa
“secret”,ortheywouldfeel“embarrassed”,evenbytalkingwithgrandmother.
Parreñas(2008)notedthatchildrenoftenvoicedtheirfeelingofan“emotional
gap”,orasenseofdiscomfort,uneaseandawkwardnessthattheyfeeltoward
migrantfathers.Thismaycapturetheambivalencethattheyfeeloverthe
unfamiliaritythathasdevelopedfromtheprolongedseparation.
Childrenwhoseparentshadbeenworkingabroadforalongtimemayhavebeen
usedtotheirsituation.Forthesechildren,thecycleofmigrantparents’departure
andreturncanbemorepainfulthantheprolongedseparationitself.Forexample,
Asis(2006,p61)foundinthePhilippinesthatchildrenofseafarersparticularly
116
sufferedfromemotionalturbulenceduringtheirparents’shortvisitsbackhome.
Onechilddescribedsuchexperiences:
Iwouldratherthatmyfatherdoesnotleave.Whenheleaves,Iwouldrather
beinschoolandnotseehimgosothatIdon’thavetothinkthatheisleaving.
WhenIgethome,theywouldaskme,‘Whydidyounotcomewithus?’Iwould
say,‘Ijustdon’twantto.’Idon’tlikegoingthroughthat.It’slikethatevery
year.
However,theexistingresearchdidnotdelineateconcreteaccountsofchildren’s
emotionalexperienceswhenmissingorthinkingabouttheirmigrantparents.
Somestudiesmentionedthattheinterviewshadtobepausedorsuspended,
becausechildrenstartedtocryortheiremotionsbecametoointensetobe
expressed(Coe,2011;Murphy,2014).Suchsituationsthemselvesimpliedthe
psychologicalvulnerabilityofLBC.
Thelinksbetweenparentalmigrationandthechild’sbehavioralwellbeingmaybe
morecomplex,andfewstudieswereabletoarticulatethemechanismsbywhich
parents’absenceaffectedchildbehaviors.Zhangetal.(2015,p58)arguedthat
discriminationagainsttheLBCatschool,especiallybyotherstudents,maycause
significantpsychologicalburden;thechild’sperceptionofandresponsetothe
discriminationwouldleadtoproblematicbehaviors.Accordingtoachild’s
narrative:
117
Inmyfirstyearofmiddleschool,peopleknewthatmyparentshaven’tbeen
livingathome,sotheolderstudentswouldbullymeandsaytheydon’tlikeme,
andslapmeintheface…Ididn’tdaretofightbackbecausetherewerealotof
them.ButI’mreallypissedoff,soafterschool,Ibroketheirbicycles.
3.3.4Communication
ItisimportanttosustaincommunicationbetweenLBCandmigrantparentsover
timeanddistance,asameanstocompensateforcaredeficitswhichleft-behind
childrenexperience,sincetheiremotionalbondswerealreadyfragileaftersucha
longphysicalseparation.
Ingeneralsuchdistantcommunicationismaintainedonaregularbasis,overthe
phone or via the internet. However, the quality of communication was often
reportedasinadequate,accordingtochildren,parentsaswellascaregivers.Many
timeschildrenacteduninterestedwhentheirparentscalledhome.Asamigrant
motherstatedinDreby’s(2007,p1055)study:
Theonlythingthatmakesmefeelbadlyaboutmysoniswhenhedoesn’twant
totalktome.Sometimesthereisalittlebitofdistancebetweenusandhe
doesn’twanttotalk.
118
Manyco-residentcaregiversalsoplayedanactiveroleinpromptingthechildto
talkmorewiththemigrantparentsoverthephone.Anaunt-caregivermentioned
(Dreby,2007,p1055):
Ididn’twantmysistertothinkthatIwastellinghimnottotalktoher.So
sometimesIwouldtrickhimtogethimtotalktoheronthephone...My
sisterwouldcrywhenhedidn’twanttotalktoher.
Howevercaregiver’sencouragementorevenmanipulationdidnotseemtobenefit
theparent-childrelationshiporchangechildren’sattitudestowardparental
migration.Graham(2012)notedthatoftentimesthecommunicationwas
ineffective,aschildrenwereunwillingtosharetheirproblemswithmigrant
parents.Thebarriersincludedfearofbeingreprimanded,thegeographical
distance,andconcernaboutaddingburdentotheirparents.
Children’spassivityintransnationalcommunicationwasalsodemonstratedbythe
factthattheylettheparentstaketheleadinphoneconversations.Parentsalso
madefrequenteffortstodisciplinetheirchildrenfromafar(Hoang&Yeoh,2015).
ContentoftheconversationstendedtofocusonthelifeofLBCratherthanallow
thechildtogettoknowabouttheirparents’livesawayfromhome.Suchrepetitive
phoneconversationswithoutfullyengagingthechildcanexacerbatetheemotional
distance,leadingtoasenseofdetachmentandaloofness.AsaVietnamesegirlsaid
ofhermigrantmother(Hoang&Yeoh,2015,p189),
119
Sheoftentalksaboutourfamily....Irarelyaskheranything,partlybecauseI
amshy.Also,Idon’tlikeaskingherquestions.Mymomalwaysaskstoomany
questionssoIneveraskherback.Sheaskstoomuch,onethingafteranother.
SheasksifIdowellatschool,howmydadandgrandmotherareandhowwell
ourrelativesdo,etc.’
Inaddition,theemotionaldistancewasevenmanifestedinchildrenwhentheir
migrantparentsreturnedhome.Parreñas(2001)elicitedasenseofsocial
discomfortandemotionaldistance’betweenFilipinomigrantfathersandtheir
left-behindchildren.Dreby(2007)suggestedalevelofemotionalwithholdingin
LBCintheirinteractionswithmigrantparents,bothwhentheywereawayand
backhomeforashortvisit.AChinese-languagestudy(Zou,2012,p839)
conductedinGuizhouprovincerevealedthatcertainpositivechangeduringthe
temporaryreunionwasunabletolastlong,accordingtoagrandparent
(translatedbytheauthorofthisthesis):
Whentheparentscamebackhomewithallthefancygiftslikebikeandtoys,
thekidbecameveryinterested.Butaftertheparentsweregone,thegood
impressionaboutthemquicklydisappeared…Whentheycalledhome,thekid
wouldn’tevenanswer,orjustsayafewwordsandthen“bye-bye”.
120
3.3.5Education
Remittancesfrommigrantparentsmaybenefitchildren’seducationprospects
withbothincreasedeconomicresourcesandheightenededucationalaspirations.
InEcuador,Pribilskly(2001)foundthatmigrationflowstotheUShavedrastically
improvedschoolattendance,whichwashistoricallyhamperedbyhigh
matriculationfeesandafamily’sneedforchildren’slabor.Education
opportunitiesmaynotonlypreparestudentsforlimitedopportunitiesinthelocal
economy,buthelpachievedreamsofpermanentlylivingintheUS.
Acknowledgingparents’sacrificeandeffortsworkingawayfromhome,some
childrenwereabletodrawonthelong-distanceencouragementmigrantparents
andkeepconsiderableself-disciplineonstudy.Forexample,aleft-behindgirlin
Hoangetal.’s(2015,p271)studywasconstantlythetopstudentinherclass.She
waswellawareofwhyherfatherhadtobeseparatedfromthefamily,andforher,
studyinghardatschoolwastheonlywayshecouldsupporthim.Accordingtoher
mother:
Sheisgladbecauseherfather’smigrationmeansearningmoney.Occasionally
shecomplainsthatshemisseshimandwantshimtocomeback...Shewishes
herfathercouldstaywithher...Sheisneverabsentfromschool.Shegoesto
schooleveryday,neverskippingclassesevenwhensheisill...untiltheteacher
ordershertostop.
121
Hearingtheircaregivers’speakabouttheirmothers’sacrificesmaymotivatethese
childrentostudyharderanddowellatschooltojustifytheirmothers’selflessness.
Agrandparentcaregiverdescribedherapproachforhergranddaughter(Dreby,
2012,p84):
Well,lastyearshefailed[waskeptbackayear]andIsaid,Cindy,youhaveto
studyandmakeaneffortatschoolbecauseyourmomisn’ttheresufferingand
workingsothatyoudonothinghere.Sheissacrificingherselftheresothat
youhaveabetterlifehere.Youhavetotakeadvantageofhersacrifice.
IncontrastwithchildrenfromlowerincomesettingsinLatinAmerica,thehigher
educationalaspirationofChinesechildrenleadsthemawayfromthemigration
pathsoftheirparents.Infact,forfamilieswithconsiderableeconomicchallenges,
childrenwouldonlycontinueschoolingiftheydowell,otherwisetheywouldhave
tojointhemigrationworkforcebeforehigherorevensecondaryeducation.This
createdconflictingfeelingsinsomechildren:eventhoughtheyacceptedthe
importanceofstudy,theyhadcometodoubtthattheirowngradeswouldbegood
enoughforthemtoavoidmigration(Murphy2014).Forexample,aboyfeltthat
hisgradeswerenotgood.Hesaidthathewaswaitingforayearwhenhewouldbe
oldenoughtomigrate.Hesaidthatschoolnowheldlittleinterestforhimandthat
onlytheinternetcafealleviatedhisboredom,whileregrettingthatthestudyroute
wasclosedtohim.
122
Theemotionalrepercussionsoffamilyseparation,andtheresentmenttheyfeel
abouttheirparentsbeingaway,canalsocauseloweracademicachievementor
educationalaspiration.Especiallyamongchildrenwhowereleftbehindbyboth
parents,manychangesinthechildren’sdailylivesaftertheirparentsleavecan
imposeemotionalburdens,aswellaspracticalburdensincludingmorehousework,
ontheireducationalachievements.
Dreby’s(2012)studyinMexiconotedatenthgrader,whosefathermigratedfirst,
saidthatafterhismotherleftwhenhewasintheseventhgrade,hehadahard
timeatschool.Hebecamereallyclosetohismothersincehisfather’sdeparture.
Sincebothhisparentsleft,hedidnotwanttostudybecausehemissedhismother
toomuch.Dreby(2007,p1058)alsoreportedseveralcasesinwhichtheLBC
droppedoutfromschoolafteraseriesofbehavioralchangesandactingoutafter
parents’departure.Aleft-behindmotherinthestudysaid:
Everythingwasdestroyed[afterthefatherleft].Everyonewassadanddidn’t
eatwellanymore.Someofthemwereinsolentandrudetome.Ithinkthat
they[herchildren]lackedapaternalfigure.
Coe(2011,p19)reporteda19-year-oldboywhoattributedhislackofacademic
successinsecondaryschooltohismotherleavingfortheUnitedStates:
123
Shehadallthetimeforallofus,especiallyourstudies.Iwasverygood.There
werethingssheusedtodotohelpme,wakingmeupatdawnandsittingby
me...Sincesheleft,everythingturnedupsidedown,especiallymystudies.I
wasn’tgettinganyhelpfromanyone,somyacademicperformancein
[secondaryschool]wasgoodthough,butnotwhatIexpected.
3.3.6Careenvironmentintheleft-behindfamilies
Theeffectsofparentalmigrationonchildrenareusuallydependentonthequality
ofcarethechildreceivesfromhissurrogatecaregiver(Hoang&Yeoh,2015).Most
studiesreportedgrandparentsastheprimarycaregiversintheleft-behindfamilies,
especiallyifbothparentswereaway.However,children’sexperiencesand
attitudestowardgrandparentalcarevariedconsiderably.Ingeneral,childrenwho
wereleft-behindatveryearlyage,evenbeforetheyhadanymemoriesabout
migrantparents,tendedtobecloserwiththeirprimarycaregiverathome.
Forexample,Arnold(2006)interviewedwomenwhowereleft-behindinthe
Caribbean.TheywereveryyoungwhentheirmothersmigratedtotheUK,and
typicallyittookmorethanadecadeformotherandchildtoreuniteintheUKasif
forthefirsttime.Onewomanreportedretrospectivelyabouthergrandmother,
(Arnold,2006,p164):
124
IwassowelllookedafterandIwashappy;mygrandmotherwasmyworld.I
rememberherasafascinatingwomanwholovedme.Itwasjustawarmclose
relationship.
Itisalsonotablethatmostofthewomeninthisstudywhohadstrongattachment
withtheirprimarycaregivers(mainlygrandmothers)spokeofthepainofthe
separationfromthem;yetmorepainfulwasthetraumaticexperienceofreunion
withtheirmotherswhowerestrangerstothem.
InGhana,Coe(2011)foundthiscloserelationshipwithgrandparentsevenin
difficulteconomicsituations.A16-year-oldgirlwhosemotherandgrandfather
migratedtoSouthAfricawasinterviewed.Shewasbeingraisedbyher60-year-old
grandmotherwhowasillmuchofthetime,partlybecausetotheeconomic
deprivation.Thegirlexplained(Coe,2011,p12):
[Ialsolove]mygrandmother.Eventhoughshedoesn’thaveanything,she
helpsme.Shestrugglestogetmoneytogiveme.Ifmygrandmotherdoesn’t
haveanythingatall,shewillgoaskpeopleforhelpsothatwecaneat.
Grandparents’maynothavethenecessarysupportinchildcare,duetotheir
reducedmobility,insufficientdisposableincomefromremittances,andthe
simultaneousmigrationofseveralfamilymembers.(Hochschild,2000;Silver,
2006;Piperno,2006,2007;McGuire&Martin,2007;Lutz,2008;Pham&Hill,
125
2008).Yettheymayneedtocopewithchildren’scontestationoftheirauthority,
relationshipwiththeco-residentparentespeciallydaughter-in-law,divorceofthe
parents,oremotionalepisodesofthechild(Pantea,2012).Consequently,some
studiesindicatedchildren’sadverseexperienceslivingwithgrandparents.
Dreby(2012)interviewedayoungwomanwhowereleft-behindduringlate
adolescence,whopreferredlivingwithhermotherandexplainedthat(Dreby,
2012,p85):
Itisn’tthesame[livingwithhergrandparents].Wehavesmallproblems....Itisn’tthe
sametypeofaffection.Wedon’tfeelthesameaswithourmother.
InArnold’s(2006)studyintheCaribbean,awomanwasleftwithhermaternal
extendedfamily,alongwithseveralotherchildren.Shesadlyrecalled(Arnold,
2006,p163):‘‘Mygrandmotherhadtoomanychildrentolookafter;therewasnot
enoughlovetogoaroundandIfeltleftout.’’
3.4Summary
3.4.1Overviewandgapsintheliterature
Ingeneral,mostquantitativestudiesinthisreviewindicatedworsepsychosocial
wellbeingoutcomesinLBC,ascomparedtootherruralchildren;fewidentified
benefitsofparentalmigration.Whilesomestudiesusedoverallmeasuressuchas
126
qualityoflifeandbehavioralproblems,thedisadvantageofLBCisalsoquite
consistentacrossindividualdimensionsofpsychosocialwellbeingandmental
health.
Afewgapswereidentifiedinthecurrentliterature.First,longer-termimpactof
parentalmigrationwasrarelyexplored,asonlyafewstudiesattemptedtolook
intothepreviousexperiencesorlong-termeffectsofbeingleftbehind,and
providedsomesuggestivefindings,yetlackofstronglongitudinalevidencemakes
itdifficulttoreachvalidconclusions.Second,therelationshipdynamicsbetween
migrantparentsandLBC,akeyvariablefollowingtheirseparation,remained
uncleardespiteoffindingsoncertaindescriptivecharacteristics;thequalitative
andquantitativestudiesdidnotpresentsolidevidenceonthemechanismsleading
towellbeingdisadvantagesinLBC.Third,howtheleft-behindfamilyandtheir
socialenvironmentsprovidecareandsupporttochildrenhasnotbeen
comprehensivelyassessed.Althoughalltheabove-mentionedaspectsweresubject
tomethodologicalandpracticalchallenges,thisthesisaimedtoprovidefurther
insightsintothesegapsintheliterature.
Giventhecomplexmechanismsbywhichparentalmigrationaffectschild
wellbeing,varioustypesofpotentialcovariateswereexploredintheliterature.
However,duetomethodologicalandpracticalconstraints,noconcrete,
comprehensiveframeworksofLBC’spsychosocialwellbeingwereproposedinthe
literature.Nevertheless,thereviewedarticlestogetherrevealedanoutlineof
127
potentialdeterminants.Similartowhatwassummarizedinthepreviouschapter
ontheoreticalperspectives,crucialfactorsofchildwellbeinghavebeenidentified
withinthechild’sfamilyandcommunity,althoughthedirectionandsizeofthe
effectsvariedacrossdifferentcontexts.Someofthesecovariateswerealsofound
topotentiallymediateormoderatetheimpactofmigration.
3.4.2Contributionofthereviewtothecurrentqualitativestudy
Theliteraturereviewedinthischaptercontributestobothempiricalstudiesin
thisthesis.Forourqualitativestudy,thereviewprocessformulatedquestions
thatactedasastepping-offpointduringinitialobservationsandinterviews.The
micro-levelinsightsfromthequantitativeliteraturetookaclose-upviewpointon
familiesandindividuals,andpinpointedsomeofthewellbeingconceptsand
factorstobeinvestigatedinthecurrentqualitativestudy.
Duringthereviewofqualitativeliterature,theauthordidnotattemptto
synthesizeorapplyanydeductivereasoningprocess,andpersistentlyavoided
impositionofpredeterminedunderstandingandexistingframeworksonthe
currentstudy,byreflexiveinterpretationandconstantcomparison.Hence,the
frequentlyoccurringthemesinthequalitativeliteratureweresimplycategorized
inthischapter.Nevertheless,thecrudesummaryalsodemonstratedthatthe
existingstudiesmostlyfocusedonindividualexperiencesandrationalesof
migrationfrompersonalperspectives,ratherthanconsideringthefamilyasaunit
128
ofanalysis.Therewasalackofinsightsintowellbeingoutcomesrelatedtofamily
relationshipsanddynamicsinthesocialenvironment(suchasreturnmigration),
whicharetobefocusedonthecurrentqualitativestudy.
Theuseofrelevanttheoriesinmigration,familystudies,andchilddevelopment,
wasalsoscarceandincomprehensive.Thecurrentqualitativestudythusaimedto
buildontheexistingknowledge,exploremoreintothegaps,andcontributeto
boththeoreticalandempiricalliteratures.Asdataanalysisinthecurrent
qualitativestudyproceeded,thereviewedliteraturecontributedtofurthering
conceptualandtheorydevelopment.Moredetailsabouttheseprocesseswillbe
clarifiedintheintroductionandmethodssectionsofChapter5,thequalitative
study.
3.4.3Contributionofthereviewtothecurrentquantitativestudy
Theliteraturereviewalsoinformedthecurrentquantitativestudy.Althoughthe
quantitativestudyinthisthesiswasbasedonsecondarydataanalysis,theauthor
washighlyinvolvedintheinitialquestionnairedesign,byreviewingthesurvey
instrumentsandscreeningtoolsintheextantliterature,andselectingtheonesthat
weremostpertinenttoLBC’spsychosocialwellbeinginourstudysitesinZhejiang.
HighrelevanceofthestudiesfromtheCHAMPSEAprojectcontributedtothe
decisionofselectingSDQasthecorecomponentofourquestionnaire.Anumberof
129
variablesfromtheliteraturewerereferredtobytheauthorindesigningother
questionnaireitems.
Considerablegapsstillexistedinthequantitativeliterature,despiteavarietyof
psychosocialwellbeingmeasuresandcontrolvariableshavingbeenanalyzed.
Aboveall,sincethepreviousstudiesrarelyentailedsolidtheoreticalgroundwork,
investigationinthecurrentqualitativestudywasguidedbytheoretical
perspectivespresentedinChapter2regardingmigration,familystudies,andchild
psychology.Inlightofacomprehensivetheoreticalunderstandingoftopicsrelated
toLBC’spsychosocialwellbeing,thecurrentqualitativestudylooksintotheareas
oftheissuewhichtheextantliteraturewasunabletofullyaddress,aswasshown
aboveintheliteraturesummaryframework(Figure3.2).
First,thepreviousexperienceofbeingleft-behindwasinvestigatedbyveryfew
studieswithvariouslimitations;itthusbecameafocusthroughoutthisthesisand
pervious-LBCwasanalyzedseparatelyfromcurrent-LBC.Second,inorderto
improvethedepthofdataanalysisintheliterature,thecurrentstudyaimedto
distinguishtheeffectsoffamilycharacteristicsotherthanparentalmigrationon
childpsychosocialwellbeing,andtherolesofsupportforLBCpotentiallyfromthe
socialenvironment.Interactioneffectsbetweenthekeycovariatesandparental
migrationwerefurtherexplored,toelucidateanyexacerbatingandmitigating
effectsonchildwellbeing.Third,asfindingsfromtheliteraturewaslargely
context-specific,thecurrentquantitativestudyattachedsignificantimportanceto
130
exploringcontextualfactorsthatwerespecifictothestudysites,andhowthey
mayhaveshapedthesocialandfamilysystemsandtherelationshipdynamics
relevanttoLBC’swellbeing.
3.4.4Puttingempiricalstudiesintocontexts
Despitesomecommonalityregardingparentalabsenceandlackofcare,findings
fromthereviewedstudieswerehighlycontext-specific,asthemigrationdecision
anditsimpacttendtoberootedinlargersocialenvironments.Manyspecific
characteristics,suchasculturalandsocioeconomicfactors,makeitdifficultto
extrapolatethefindingstoanothercountryofevenanothercommunityofthe
country,withoutdetailedunderstandingaboutlocalsettings.Also,aswas
discussedinthepreviouschapter,theoriesaroundthefamilyandsocio-ecological
systemsofchilddevelopmentdelineatetheimportantrolesofbroadercontexts
outsidethehousehold.Hence,thenextchapterexaminesthelocalsocial
backgroundofthestudysites,andpolicyandadministrationsystemsthatwere
relevanttomigrationandchildwellbeing,inordertobetterunderstandthe
settingsandrationalesoftheempiricalchapters.
131
Chapter4 Studycontexts
4.1Introduction
Theprevioustworeviewchapterspresentedhowparentalmigrationmayaffect
childpsychosocialwellbeingthroughmultiplemechanisms,inwhichvarious
potentialconfoundingormediatingfactorswereinvolved.Asidefromthe
characteristicsofthefamilyrelationshipsandhouseholdenvironment,the
considerablevariabilityoftheempiricalstudyresultsisalsoattributedtovastly
differentsocioeconomic,culturalandpoliticalcontextsglobally.Therefore,itis
necessarytolookintoourspecificstudycontextsbeforethenexttwoempirical
chaptersofthisthesis.
Thischapterprovidesadescriptionofthelargercontextsofourempiricalstudies,
withafocusonsocioeconomicdevelopmentandmigrationinthestudyareas,
andpolicyandadministrationsystemsthatarerelevanttothewellbeingofrural
familiesaffectedbymigration,fromnationalleveltolocalcommunitylevel.These
contextsprovidecrucialbackgroundinformationofthedesignand
implementationofbothqualitativeandquantitativestudiesinthisthesis,and
mayinformtheinterpretationofthestudyresults.
4.2Studyareasandpopulations
132
Datainthisthesiswascollectedfromsamplesofchildrenandfamilymembers,
acrossfivecountiesintwoChineseprovinces,ZhejiangandGuizhou.Thetwo
provincesrepresenttherichestandpoorestpartsofChina,respectively,and
thereforemayreflectdifferentpatternsofimpactonLBCduetosocioeconomic
developmentstatus.
EthicalapprovalwasreceivedfromboththeEthicsReviewCommitteeof
UniversityCollegeLondonandZhejiangUniversitypriortothefieldstudy.
Participantswererecruitedfrommigrant-sendingruralareasinZhejiangand
GuizhouProvinces,China.ThreecountiesinwesternZhejiang,includingKaihua,
Jiande,andJiangshan,andtwocountiesinsoutheastGuizhouProvince,Guiding
andLongli,wereincludedinthestudy.Zhejiangisawealthyeasterncoastal
provinceinChina,butitswesternmountainousareaisrelativelyunderdeveloped,
withahighlevelofout-migration.GuizhouisoneofthepoorestChinese
provinces,althoughthetwocountiesinthisstudy,locatedneartheprovincial
capital,rankatmediumlevelinGDPacrosstheprovince.Figure4.1presentsa
mapshowingthelocationsofthetwoprovinces.
133
Figure 4.1: Location of the study provinces
4.2.1SocioeconomicdevelopmentandmigrationinZhejiangProvince
ZhejiangisawealthyeasterncoastalprovinceinChina,andithasbeentakingthe
leadinthecountry’surbanizationandeconomicboomsincethe1980s.In2014
Zhejiangranked5thinGDPpercapitaamongall31provincialleveladministrative
regionsinMainlandChina,onlyafterthreemunicipalities(Tianjin,Beijing,and
Shanghai)andJiangsuProvince(NationalBureauofStatistics,2014).Yet,much
liketheentirecountry,significantdevelopmentgapsexistbetweendifferentareas
Guizhou
Zhejiang
134
oftheprovince.ThemountainousareainwesternZhejiang,whereourstudies
wereconducted,isrelativelyunderdeveloped,withahighlevelofout-migration.
Zhejiangisbothafeederandreceivingprovinceformigrantworkers;therearean
estimated17.8millionmigrantworkersinZhejiangaccordingtoa2005
populationcensus,around40%ofwhichareintra-provincialmigrants(Zhanget
al.,2008).AccordingtoDuanandcolleagues’(2013)estimatebasedonthe6th
NationalCensusin2010,approximately1.2millionchildrenofmigrantworkers
wereleft-behindinruralZhejiangin2010,accountingfor30%ofallrural
childrenintheprovince.
EconomicvitalityinZhejiangislargelyconcentratedinaboutsevenlargecitiesin
theeasternpartoftheprovincealongthecoast,whichisamajor
migrant-receivingarea.Incontrast,westernZhejiangisrelatively
underdevelopedandhasbecomeamigrant-sendingarea,withmassivenatural
reservesthatlimittheindustrialexpansion.Jiandeenjoysahugelakeareathat
provideshigh-qualitywaterformillionsofpopulation,andthereforemanylocal
industriesareunderconstraintsinordertoprotectwatersources.Similarly,the
sourceofQiantangRiver,themotherriverofZhejiang,isinKaihua,wherethe
governmentenforcesstrictpoliciestolimitthesizeofindustryandprotectthe
naturalenvironment.
ThethreecountiesincludedinourstudiesarealllocatedinwesternZhejiang,
135
farawayfromthedevelopedcoastalcitiesintheprovince.BothKaihuaand
JiangshanareunderQuzhouprefecture.Jiandeisundertheadministrationof
Hangzhouprefecture,yetitisnexttoQuzhouprefectureandover140kilometers
fromHangzhouCity.WhileKaihuaisthepoorestamongthethree,theoverall
socio-economicdevelopmentandmigrationpatternsaresimilaracrossthese
counties,andthusKaihuaistakenasanexample.Thisisalsobecausethe
qualitativestudy(Chapter6)wasexclusivelyconductedinKaihua.
4.2.3IntroductiontoKaihua,themainprojectsite
OurmainprojectsiteswereinKaihuaCounty(population350,000).Kaihuais
underthejurisdictionofQuzhoucityanditislocated300kilometersfrom
Hangzhou,theprovincialcapital,inthewesternmostportionofZhejiangonthe
borderwithbothAnhuiandJiangxiProvinces.Itspopulationincludesover99.5%
Hanethnicity(NationalBureauofStatistics,2008)whoseethno-cultural
backgroundissimilartoruralareasacrossthethreeprovinces.
KaihuaisoneoftheleastdevelopedcountiesinZhejiangProvinceandatypical
migrant-sendingareainthetri-provinceregion;itsGDPin2010rankedatthe
47thplaceamongthe58countiesinZhejiang.Thisislargelyduetothefactthat
Kaihua’sindustrialdevelopmentisrestricted,asthesourceoftheQiantangRiver
isinthecountyaswellasmanyecologicalreserves.About85%ofthecountyis
mountainousareas,andtheforestcoverageover80%.
136
AccordingtoasurveybyKaihuaWomen’sFederation(2013),about55%of
school-agechildreninthecountywereleftbehind.Two-thirdsofthesewere
left-behindbybothparentsandtypicallylivewithgrandparentswhoareilliterate.
Theotheronethirdlivewithoneparent,usuallythemother,whilethefather
migratestowork.Mostyoungparentshaveoneortwochildren,asasecondchild
wasallowedifthefirstwasagirl,thoughthispolicyhasbeenrelaxedrecentlyand
nowallcouplesareallowedtwochildren.
Inthe1990s,Kaihuausedtobeahighlyindustrializedcounty,anumberof
forestry,cement,papermaking,andchemicalplantsarealloverthetownships
andvillages.Consequently,theenvironment,especiallythewatersourcefora
massivepartoftheprovince,becamehighlypolluted.Intheyearof2000,
Kaihua’sleadershipsetupthestrategyof“acountybasedonecological
environment”,whichbecameaturningpointofKaihua’sdevelopmententering
thenewcentury.In2006,duringavisittoKaihua,Mr.XiJinping,thePartyChiefof
ZhejiangProvinceatthetime,highlypraisedtheecologicalenvironmentand
encouragedthecountytodeveloptheeconomybytakingadvantageofthe
well-reservedenvironment.
Inrecentyears,anewtypeofindustry,processingfactoriesofsuppliedmaterials,
emergedinKaihua,andbecameamajordriveofthelocaleconomy.These
processingfactoriesmainlyassemblesmallproducts,whichrequireverylimited
137
trainingandcauselittlepollution.Anumberoflocalworkersarenowemployed
bythisindustry,andconsequentlytheout-migrationflowsareheldbacktosome
extent.Inthemeantime,tourismisalsoboominginKaihuathankstothe
preservationofthenaturalenvironment.Thenumberoftouristsincreasedby30%
fromyear2013to2014;thecountyissettobeanecologicalfunctionalarea,asa
“nationalpark”system,withonly10%landsforindustrialuse(Li,2015).
4.2.2SocioeconomicdevelopmentandmigrationinGuizhouProvince
GuizhouisoneofthepoorestChineseprovinces.Itisalandlockedprovince
locatedinsouthwestChina.In2014GuizhourankedthelastplaceinGDPper
capitaamongall31provinces,municipalities,andautonomousregionsin
MainlandChina.Historically,Guizhou’sdevelopmenthasbeenconstrainedbythe
weakinfrastructureandthefrailecologicalenvironment.Thecurrent
urbanizationandindustrializationprocessisfacedwithmajorchallenges
includinghugegapsbetweenurbanandruralareas,andinadequatesocial
services.
Guizhouisamajormigrant-sendingprovince.Amongthe8.6million
out-migrationpopulation,over70%chosetomovetoanotherprovince.Thereare
overthreemillionleft-behindchildreninZhejiang,accountingfor40%ofallrural
childrenintheprovince.Children’seducationattainmentinGuizhouisof
particularconcern:about5%ofruralchildreninGuizhouarenotenrolledin
138
China’snine-yearcompulsoryeducation,whichisthe8thhighestproportion
acrossMainlandChinaprovinces;andLBCareslightlymorelikelytodropout
thannon-LBC.NeglectandlackofprotectionforLBChascausedsevere
consequencesinGuizhou.Afewhorribleincidentsthatoccurredoverthepast
fewyearsintheprovincehavedrawnattentionfromacrossthecountryregarding
thewellbeingofLBC.TheworstonewhichhappenedinJune2015involvedthe
deathsoffourLBCfromthesamefamily.Thefoursiblings,oneolderbrotherand
threeyoungersisters,committedsuicideathomeinBijie,Guizhouwherethey
werelivingwithoutanyadultcaregiver,bydrinkingpoisonouspesticide.
Governmentofficialswerealreadyawareoftheirsituationbutfailedtotake
effectiveactions.
Thetwocountiesinthisstudy,GuidingandLongli,rankatmediumlevelinGDPin
Guizhouprovince.Theyarelocatednexttoeachotheratsouth-centralGuizhou
neartheprovincialcapitalGuiyang.BothcountiesbelongtotheCentralGuizhou
EconomicZone,akeyareawithstrategicpriorityintheprovince’sdevelopment.
AccordingtoadevelopmentplanoftheEconomicZone,itaimstobecomeabase
ofemergingindustries.However,thecurrentsocioeconomicstatusoftheentire
areaisstilllaggingbehindinSouthwestChina,andthusalargenumberof
workingagepopulationhavemigratedtootherprovinces.
4.3Policyandadministrationcontexts
139
Migrationasasocialphenomenonisnotonlydrivenbyintheeconomic
developmentimpetusandgapsacrossurbanandruralChina,butalsoshapedby
theratheruniquesocio-politicalsystemsintheChinesesociety.Infact,the
economicadvantageintheeasterncoastalareascanbeseenasaresultofthe
ReformandOpening-upstrategyadoptedbytheCentralGovernmentsince1980s
(Caietal.,2002).
Thepopulationcontrolstrategiesstartedinthesameperiod,knownastheOne
ChildPolicy,hadfar-reachingimplicationsonthestructureofChinesefamilies
andthewellbeingoftheirchildren.Thetop-downimplementationofsuchpolicies
requiresastronggovernmentadministrationsystematgrassrootslevel,whichis
organizedindifferentpatternsacrossthecountryandoftenunfamiliartothe
outsideworld.Inthissection,theimplementationpatternofthefamilyplanning
policyandgovernmentadministrationsystemsinthestudyareaswillbe
introduced,withafocusontheWomen’sFederation,angovernmentbody
responsibleforchildwelfareandprotection.
4.3.1FamilyplanningpolicyinChina
AcommonmisunderstandingofChina’sOneChildPolicyisthatallfamiliesacross
theentirecountryareonlyallowedtohaveonechild.Inpractice,beforethe
recentmajorshiftedtoTwoChildPolicy,therewerealreadymultiplevariations
oftheOneChildPolicyacrossdifferentareasanddifferentpopulationgroupsin
140
China.Thelocalpoliciesmayevendifferwithinaprefectureoracounty.
Acommonrelaxedversionofthepolicywasthe“OneandHalfChildPolicy”.It
meansthatifthefirst-bornchildisagirl,thentheparentsareallowedtohavea
secondchild;butnotathirdchild.Beforetherecentpolicychange,someversions
oftheTwo-ChildPolicyhadalreadybeenappliedtomanyruralareasinChina,as
wellascertainpopulationgroups,includingparents(someareasrequiredone
parent,whilesomerequireboth)fromanethnicminoritygroup,orifboth
parentsareonlychild.Uptothreechildrenwereallowedforethnicminoritiesin
ruralareas(Guoetal.,2003).
InmostofprojectsitesinZhejiangandallsitesinGuizhou,theOneandHalfChild
Policyapplied.ThereareahighernumberofethnicminorityfamiliesinSouthern
Guizhou,whowereallowedtohavetwochildren.
4.3.2Ruralgovernmentadministrationsystem
China'sadministrativesystemcomprisesanestedhierarchyofauthorities.Under
thecentralstate,indescendingorder,thereareprovinces,prefectures(cities),
counties,townshipandvillagelevels(Chung,2008).
Townships(XiangZhen;乡镇)thathavelargereconomy,area,andpopulation
(especiallynon-agriculturalpopulation)arenamedasZhen(镇).Othersare
141
namedasXiang(乡).InKaihua,thereare18townshipsincludingnineZhensand
nineXiangs.Atownshipgovernsanumberofadministrativevillages,and
occasionally,Communities(urbanizedneighbourhoods).Thereareonlynine
CommunitiesinKaihua,fivearenearcountycenterinChengguanZhen,threein
HuabuZhenasanotherurbanizedcenter.
An“administrativevillage”iscomprisedofseveral“naturalvillages”.The
governmentbodyisonlyattheadministrativevillagelevel.Overthepastdecade,
thegovernmentadministrativeunitsatcommunitylevelinruralareashavebeen
undergoingamergingprocessinmanypartsChina,largelyduetothedecreaseof
ruralpopulation.InKaihua,fromtheendof2010toearly2011,thegovernment
wentthroughtheprocesstoreducethenumberofadministrativevillagesfrom
449to255.Similarly,thenumberoftownshipswasreducedfrom27to18inyear
2005(ZhejiangOnline,2011).
Thevillagegovernmentbodyisoftenreferredasthe“TwoCommittees”(两委):
VillagePartyBranchCommitteeandVillagersCommittee,whicheffectivelywork
asoneentityandisoftenreferredtoasVillageCommittee(村委会).Headsof
thesetwocommitteesrespectivelyareVillage(Party)Secretary,thetopleaderof
thevillage,andVillageDirector,thesecondincommand.Usuallythereisatleast
onefemalestaffamongmembersofthetwocommittees.
142
4.3.3Women’sFederation(WF)system
TheWFhasrepresentationatcounty,township,andvillagelevelinKaihua.The
countylevelWFhaseightstaffcurrently.EachtownshipgovernmenthasaWF
staff(TownshipWFPresident),butmostofthemtakeotherresponsibilities
outsideWFaswell,suchasfamilyplanning,healthandsanitation,orgeneral
administrativeaffairs;theworkofWFisnormallylessdemandingorconsidered
lessimportant,especiallyascomparedtofamilyplanning.
ThetitleoftheWFrepresentativeatvillagelevelisliterally“DirectorofVillage
ConferenceofWomenRepresentatives”(村妇女代表大会主任),or“Directorof
WomenRepresentatives”(妇女主任),oneineachvillage.AllWFstaffatvillage
levelworkpart-time.TheWFrepresentativeisnormallyafemalememberofthe
TwoCommittees,whoalsotakesvariousresponsibilitiesapartfromwomen’sand
children’saffairs.InmostvillagesinKaihua,theWFrepresentative’sprimary
appointmentisactuallyFamilyPlanningCommissioner,whoreporttothelocal
familyplanningdepartment,amuchmorepowerfulagencyatalllevelsofthe
Chinesegovernment,ascomparedtoWF.Insomenaturalvillagesthereisalsoa
“WFLiaisonOfficer”,whowouldpreviouslybethevillageWFrepresentative
beforethevillagemergingprocess.TheWFrepresentativeisresponsibleforchild
wellbeing,thussheusuallyoverseestheaffairsrelatedtotheChildren’sClubs.
143
4.3.4Rolesofgovernmentagenciesonchildprotectionandintervention
programs
Unlikemanyothercountries,civilsocietyinChinaisnotwelldevelopedandthus
unabletobeactivelyengagedinchildprotection.Consequently,government
agenciesortheiraffiliatedorganizationsplayanessentialroleinsafeguarding
children’swelfare.Inthisregard,navigationwithinsuchadministrativesystems
becomeshighlyimportant.BesidesbetterunderstandingofLBC’swellbeingbased
oninformationprovidedbylocalgovernmentsandcommunityleaders,effective
supportfromgovernmentagenciesisalsocrucialinplanninganddelivering
interventionprograms,inordertoimprovethesituationofthesechildren.
Thestudiespresentedinthisthesiswerecarriedoutwithsubstantialsupport
fromlocalgovernmentagenciesandvillagecommittees,especiallyWomen’s
Federationanditsstaffmembers.Basicknowledgeaboutthechildwelfare
governancesystemhelpedusbetterunderstandtherolesofstakeholders,in
ordertoeffectivelyobtainrelevantinformationandadministrativesupport.
Moreover,ourinterventionprogramaimedatsafeguardingLBC’swellbeingwas
carriedoutinclosecollaborationwithWomen’sFederations.Ononehand,at
countylevel,WFofficialswereinvolvedinprojectplanning,andourintervention
programwasgivencertainpoliticalpriorityatalllevelsofWFsbrancheswithin
thecounty.Onetheotherhand,thegrassrootsWFrepresentativescomprised
144
mostvolunteersinourChildren’sClubs,andbecameavigorousworkforcein
deliveringprogramservicestochildreninthecommunities.
4.4Summary
Understandingofthestudycontextsprovidescrucialinformationwhen
examiningwellbeingchallengescausedbyacomplexsocialphenomenon.Forthe
quantitativestudy,manyvariablesassociatedwithchildpsychosocialwellbeing
statusarerootedinthelocalsocioeconomic,culturalandpoliticalsettings.Forthe
qualitativestudy,experiencesofchildrenandperceptionsoffamilymembersare
alsocloselyrelatedtothelargerenvironmentsinthemigrant-sendingaswellas
themigrant-receivingareas.Specificcommunityandpolicycharacteristicsas
wereintroducedinthischapterinformedthedesignandimplementationofour
empiricalstudies,andmayhelpbuildlinksbetweenourfindingswiththosefrom
similarcontextstheexistingliterature.
145
Chapter5 Experiencesofchildren,andrelationships,careand
supportinthefamilieslivingapart
5.1Introductionanduseofextantliterature
Thischapterpresentsaqualitativestudybasedonin-depthinterviewsofleft
behindchildren,theirmigrantparents,andresidentcaregiversintheruralareas
ofKaihuaCounty,westernZhejiangProvince.
Inthepreviouschapters,thereviewofboththeoreticalperspectivesand
empiricalstudies,aswellastheunderstandingofthelocalsocio-politicalcontexts,
providedcrucialinsightsfrommacro-tomicro-levelsforthequalitativestudyon
thepsychosocialwellbeingofLBC.Yetthischapterisnotfocusedontesting
hypotheses,butratheraimedtodevelopanewtheoreticalframeworkgrounded
inempiricaldatacollectedinthefield,withlimitedinputsfromtheextant
knowledge.
AlthoughthereviewinChapter3onlyretrievedafewqualitativestudiesabout
China,thefirstissuesidentifiedintheliteraturehelpedclarifythesubjectmatter
ofthecurrentstudy:theexperiencesofchildrenandperceptionsoffamily
membersinrelationtochildpsychosocialwellbeingundertheimpactofparental
migration,aswellastheinterrelatedfactorsinthefamilyandsocial
environmentsandthemechanismsthroughwhichtheyaffectLBC.
146
Literaturewasusedasaneffectiveanalytictooltostimulatethinkingand
heightentheoreticalsensitivityinthequalitativedataanalysis.Memoswerekept
alongwithliteraturereviewprocesstorecordthenewfindingsandideasfrom
theempiricalstudies,aswellasthepropositions,valuesandcontextlinkedwitha
giventheory,includingthepossibleshortcomingsofthetheory.Thememos
chronicledthemannerinwhichtheauthor’sthinkinghadchangedasaresultof
accessingextantknowledge,andenhancedtheauthor’ssensitivitytothe
meaningswithoutforcingexplanationsonthedata(Strauss&Corbin,1998,p.
47).
Forinstance,migrationtheoriesbecameimportantreferenceswhenexamining
therationalesbehindmigrationasperceivedbyfamilyandcommunitymembers;
thisperspectivehasoftenbeenignoredbypreviousresearchonLBC’swellbeing.
Thesocio-ecologicalapproachofhumandevelopmentandfamilysystems
theoriesshedlightonthecomplexenvironmentsinwhichthechildlives,when
lookingintotherangeofdynamicsbetweenchildren,migrantparents,and
caregivers.Theauthorwasthusinspiredtoexaminethechild’sreciprocal
interactionswiththeimmediatefamilyenvironment,aswellastheimpactofthe
largercontextsatcommunityandsocietylevelsofthehumanecologicalsystem.
Thetiminganddurationoftheseprocessesinchilddevelopmentwerealso
consideredvariablesofwellbeing.
147
Inthelaterstagesofdataanalysis,theextantliteraturealsoassistedtheoretical
sampling,bydirectingtosituationsthattheauthormightnothaveconsidered
otherwise.Atachmenttheory,forexample,wasreferredtoinordertoidentify
multiplerelationshippatternsbetweenparentsandchildren,overdistanceand
followingareunion.
Inthemeantime,theauthoractuallybecamemoreawareofpossibleunhelpful
preconceptions,byconstantlyreflectingontheideasthattheauthorhasbeen
exposedtowhileengagingwiththeliterature.Arespectfulyetcriticalstance
towardsextanttheoriesandempiricalfindingswasadopted,beforeincorporating
anyextantconceptsintothenarrativeofthisstudy.
Althoughtheliteraturereviewwasconductedbeforethequalitativefieldwork,
bothchapters2and3werewrittenupafterthequalitativeanalysiswas
conducted.Infact,theauthor’sreflectionsandinsightsfromtheempiricaldata
considerablyinfluencedthestructureofthereviewchaptersanddiscussions
abouttheextantknowledge.
5.2Methods
Ingeneral,theauthorfollowedStraussandCorbin’s(1998)groundedtheory
approachofdataanalysis,withcertainmodificationstobetterfitwiththeaims
andconditionsofthisresearchproject.Thisapproachincorporatedboth
inductiveanddeductiveanalyses,asopposedtoGlaser’s(1992)classicgrounded
148
theorywhichpredominantlyfocusesoninductivereasoning.Furtherelaborations
ofthecodinganddataanalysisprocesseswillbeprovidedintheSection5.2.5
below.
5.2.1Studyarea
ThestudywascarriedoutinKaihuacounty,westernZhejiangprovince,atan
initialphaseofalargerprojectthatinvolvedmoreprojectsareas.Moredetailed
informationaboutthecontextsofthestudyareawasprovidedinChapter4.
AlthoughthequantitativestudyofthelargerprojectincludedGuizhouProvince,
wewereunabletocarryoutthequalitativestudyinGuizhou,duetosignificant
practicalandfundingconstraints.Thisqualitativestudywasonlysupportedbya
smallresearchgrantoveroneyear;onlyduringtheimplementationofthe
qualitativeprojectdidourteamwinalargergrantwhichwastosupporttheLBC
researchandaninterventionprograminbothZhejiangandGuizhouprovinces,
overafive-yearperiod.Inaddition,ourChinaprojectteammemberswereall
basedinZhejiang.
WecollaboratedwiththelocalKaihuacountybranchoftheWomen’sFederation
(WF),whichhadrepresentationattownshipandvillagelevelsaswell.Therewas
aWFofficialineachtownshipandaWFstaffmemberineachvillage.Thevillage
WFrepresentativewasalsoamemberofthevillagecommittee.KaihuaWFwas
consultedaboutthedesignandimplementationoftheproject.HowevertheWF
149
representativesdidnotinterferewiththeresearchprocess,soinvestigatorsin
thisstudyremainedindependentfromanygovernmentagency,withoutany
obligationorrestrictionindatareportinganddissemination.Moreinformation
regardingtheimplicationsofWF’sinvolvementwillbepresentedattheendofthe
methodssection.
5.2.2Sampling
Intheinitialstage,apurposefulsamplingapproachwasadoptedtomaximizethe
possibilitiesofobtainingdataandleadsformoredataontheresearchquestion.
Aspurposefulsamplingiscommoninallqualitativestudies,itmayalsobeusedat
theinitialstagesofthetheoreticalsamplingapproachingroundedtheory.
Specifically,Glaser(1978,p45)notedthatresearchersmaybeginbytalkingtothe
mostknowledgeablepeopletogetalineonrelevanciesandleadstotrackdown
moredata,andwhereandhowtolocatethemselvesforarichsupplyofdata.
Inthisstudy,wereachedouttovillageleadersandWFstaffmemberstoobtain
anonymousinformationaboutmigrantfamiliesinselectedvillages,inorderto
includecommunities,familiesandindividualsfromwhichdiversetypesof
experiencesmayemergeinrelationtorural-to-urbanmigration.Thesefamilies
andchildrenhadvariousbackgroundsintermsofhouseholdsocio-economic
characteristics,familystructure,caringarrangements,andchildageatthe
beginningofmigration.
150
Asthestudyprogressed,inthelaterstagesofsampling,wecontinuedtochoose
children,familiesandcommunitieswhichwouldmaximizeopportunitiestoelicit
dataregardingvariationsalongdimensionsofcategories,andthatdemonstrated
whathappenedwhenchangesoccured.Ifnewcategorieswerediscovered,
additionalparticipantswererecruitedtoreflecttheseparticulardimensions.
Whenanoverallstorylinestartedtobeformulated,somefurthersampleswere
addedtoverifythestorylineandrelationshipsbetweencategories,andtofillin
poorlydevelopedcategories.
Whenselectingspecificcommunitiesforthestudysitesandpopulations,a
two-stageclustersamplingmethodwasemployedtoachievecertaingeographical
representativenessandpopulationcoverageinKaihua,byselectingonevillage
fromeachselectedtownship.Itisnoteworthythatweaimedatconductingfamily
groupinterviewsincludingallmembersineachfamily,althoughinmostcases
thiswaspreventedforpracticalreasonsbecausenotallfamilymemberswere
abletoparticipateduringtheshortfieldvisitperiodinthevillage(often1day),
especiallyatChineseNewYearwhenmanycommunityactivitiesandhousehold
choreslimitedpeople’savailability.Eventuallyinterviewswereconductedwith
25children,17parents,and13grandparents,from29families,across12villages.
Theageofthechildrenrangedfrom7to14(mean=10.93,SD=1.77).Thefamily
compositionandmigrationpatternsofallinterviewedfamiliesareshowninTable
5.1,withindividualsonthesamerowfromthesamefamily.
151
Table 5.1: Demographics, family composition, and migration pattern of
interviewed migrant families
Index child Co-resident adult Migrant parent
Child age when
left-behind Sex Age Sibling
Girl 10 1 elder sister Mother Father only 2
Boy 12 1 younger sister Mother Father only 10
Girl 11 None Other relatives Both parents 5
Girl 11 None Grandparents Both parents 2
Girl 9 1 elder sister Mother & Grandma Father only Right after birth
Girl 10 1 younger brother Mother & Grandparents Father only Unknown
Boy 10 None Grandma Both parents 3
Girl 14 1 elder sister Mother Father only Right after birth
Boy 12 None Grandparents Both parents 4
Boy 13 1 younger brother Grandparents Both parents 1
Girl 8 1 younger brother Grandparents Both parents Right after birth
Boy 14 None Grandparents Both parents 1 (with parents
from 5 to 7)
Boy 11 None Grandparents Both parents A few months
Girl 11 None Grandparents Both parents 3
Boy 11 None Grandpa Both parents 4
Girl 12 1 younger brother Mother Father only 6
Girl 10 1 elder sister Mother Father only 5
Girl 9 None Grandparents Mother only 9
Girl 7 1 elder sister Grandparents Both parents 3
Boy 12 None Grandma Both parents 5
Girl 10 None Grandparents Both parents 4
Boy 12 None Grandparents Both parents 5 or 6
Boy 13 None Grandma Father only* 11
Boy 13 None Grandparents Both parents 3
Boy 10 None Mother & Grandma Father only 1
Boy 12 None Mother Father only Unknown
Girl 12 1 elder sister Mother & Grandparents Father only Right after birth
Boy 9 None Grandma Both parents 4
Girl 13 None Grandparents Both parents 7
Boy 8 1 elder brother Grandparents Both parents 5
152
5.2.3Interviewpreparations
Interviewswereconductedduringtwoperiods.Thefirstperiodwas2nd
Septemberto18thNovember2013,rightafterschoolsstarted,followingthesome
children’sreturnfromvisitingtheirparentsinthecitiesduringthesummer
holiday.OneofthemostimportantChineseholidaysforfamilyreunion,
Mid-AutumnFestival,fellon19thSeptemberin2013.Thesecondperiodwasfrom
20thJanuaryto16thFebruary2014aroundChineseNewYearthatfellon31st
January2014.Mostmigrantworkerswouldreturnhomeduringthisholiday
season,andsomemaystayuntiltheLantern’sFestival,the15thdayintothe
ChineseNewYear.Thesetwoperiodswerechosenmainlyinordertorecruit
moreparticipants,bothchildrenandmigrantparents,inthequalitativestudy.
TheauthorwasintroducedtotheleadersofKaihuaCountyWFatthestartofthe
fieldwork,byaco-investigatorofthelargerLBCprojectwhowasaffiliatedwith
ZhejiangUniversityandwasinvolvedinotherresearchprojectswithinthecounty.
Inthefollowingweek,theauthorhadafewmeetingswiththeWFleadersto
introducethequalitativestudywithinthelargerproject.Theauthorrequested
logisticsupportfromtheWFanddiscussedaboutsomespecificfieldwork
arrangements.OneoftheVicePresidentsofKaihuaWFwasappointedasthefocal
personforthisstudy,andshecontinuedtoprovidecoordinationsupportforthe
authorafterthesemeetings.
153
TheVicePresidentofferedalistofnamesandcontactinformationofalltownship
WFofficialsinKaihua,andbrieflyinformedthemabouttheoverallaimsofthe
studyandthefieldwork.Onedaypriortoeachfieldtrip,theauthorwouldcontact
thelocaltownshipWFofficialbyphone,brieflyexplainthepurposesofthetrip,
andcoordinatewithspecificinterview-relatedlogisticarrangements.
Aswasmentionedintheprevioussectiononsampling,weapproachedleadersas
wellasmembersofthecommunity,atthestartofeachfieldtrip,oneortwodays
beforechildrenandfamilieswererecruitedinthevillage.Particularlytheauthor
wouldmeetwiththeWFofficialandoneortwoothermembersoftheVillage
Committee.Inthesemeetingstheauthorwouldrequestgeneralinformation
aboutthevillageanditsfamilies,suchaskeysectorsinthelocaleconomy,the
scaleofout-migration,theoverallwellbeingstatusoftheLBC.Theauthorwould
alsobrieflygooverthesamplingmethodology,andselecttheparticipantstobe
approachedfortheinterviewsindependentlyorwiththehelpofvillageleaders,
basedontheirinformationregisteredatthevillagecommittee.
Indoingso,theauthorconductedapreliminaryinvestigationintochildren’s
householdandphysicalenvironmentsinthecommunity,andhowlargersocial
systems,suchaslocalindustryandsocialsafetynet,wereinteractingwiththe
child’simmediateenvironment.Thiswaspartlyinspiredbythesocio-iconological
approachofBronfenbrenneretal.’s(1998),astheauthoredrecognizedthe
importanceofexploringthelargerecologicalsysteminwhichchilddevelopment
154
occurs,andelicitingtheconnectionsbetweendifferentvariablesofthecontext,
familyandindividualcharacteristics,andchildwellbeingoutcomes.Moredetails
relatedtothesocio-ecologicalmodelofhumandevelopmentwillbepresentedin
theresultsanddiscussion.
5.2.3Interviewprocedure
Homevisitappointmentsweremadeadaypriortotheinterview.Theselected
familieswerevisitedtorecruitchildren,parentsand/orcaregiverswhowere
present.Childrenwhowerenotathomewererecruitedatthelocalcommunity
centerwheretheyusuallyspenttimeafterschool.
Thestudypurpose,theinterviewprocessandtopicswereexplainedtoall
participants,andinformedconsentfromallintervieweeswasobtainedbeforethe
interview.Additionally,aparentorcustodialcaregiverofallinterviewedchildren
alsosignedaconsentform.Foryoungerchildren,theconsentwasobtainedfirst
fromtheirparentorcaregiver;thentheauthorexplainedthestudytothechildin
moreplainlanguage,inthepresenceoftheparentorcaregiver,alongwithwhom
theauthortooktheopportunitytoexplainanythingthatwasnotcleartothechild.
Ifnoparentorcaregiverwerepresentattheinterviewsite,afamilyvisitwas
conductedtodelivertheconsentform.Aftercaregiverconsentwasobtained,an
adultvolunteeratthecommunity,usuallyateacherorcommunityleaderwho
knewthechildwell,wouldaccompanythechildandprovidesupportduringthe
155
interviewprocess.Participantswereguaranteedthatwithdrawalfromthestudy
wouldnotaffectanybenefitfromthecommunityorgovernment,andthe
interviewwouldstopatanytimeifthechild,anyparent/caregiverorthe
accompanyingvolunteerfeltuncomfortabletocontinue.
Wealsohadmadearrangementstoprovidealocalcounselingserviceforany
children(orparents/caregivers)whofelttheywantedtofurtherexploreareasof
difficulty.Theseparticipantsmaymakeappointmentswiththehelpoftheauthor,
usingtheresearchteam’sconnectionsatthePsychologicalCounselingand
TherapyCenteratQuzhouThirdHospital.Theteamwouldfacilitatethe
transportationandotherlogisticsinaccessingtheservice.
InterviewswereconductedinMandarinChinesewithsomehelpfromtheWF
personneliftheinterviewees,mostlytheelderly,usedlocaldialect.TheWF
representativestendedtobeawell-trustedcommunitymember,thustheir
presencemostlyfacilitatedthecommunication.Interviewswereaudio-recorded
andtranscribedverbatimwithspecificconsentfromtheparticipants.
Observationalnoteswerealsotakenoninterviewees’behaviorsandfacial
expressions,aswellasinteractionswhenmultiplefamilymemberswere
interviewedconjointly.Ofparticularnote,afewchildrenbecameupsetand
tearfulduringtheinterview.Inthesecases,wesuspendedtheinterviewand
attemptedtocomforttheparticipant.Wecontinuedonlyifthechildandanadult
familymemberexpressedthewishtocontinue;ifthechildstillappearedsilent
156
andsadafterafollowingquestion,theinterviewwasthenterminated.Ifthechild
expressedneedstotalkwithpsychologicalserviceproviders,theauthorwould
offerhelpinaccessingtheservice,followingtheproceduredescribedinthe
previousparagraph.
Eachinterviewfloweddifferentlyastheresearcherandtheparticipant
dynamicallyco-constructedtheinterview.Interviewswithchildrenwere
semi-structuredandbeganwithquestionsongeneraldemographics(age,school
year,familymembers,primarycaregiverathome),andparents’migrationstatus.
Thenquestionscoveredtopicsincludingfeelingstowardseparationwith
parent(s),communicationwiththemigrantparent(s),carefromandrelationship
withthecaregiverathome,schoolperformanceandafter-schoolactivities.
Questionswereadaptedtothechild’sage,posingmoresimpleanddirect
questionstoyoungerchildrenandrelativelymorein-depthquestionswitholder
children.Interviewswereconductedwithparentsandgrandparents,using
similarbutrephrasedquestionstoexploreadultfamilymembers’perspectives.
Additionally,weaskedparentsandgrandparentsaboutincentivesandconcerns
aboutmigration,itspotentialimpactontheirchildren’swellbeing,andwe
exploredtherationaleofmigratorydecisionsintermsofinternalandexternal
familycontexts.
5.2.4Rolesofthelocalcollaborator
157
KaihuaWFwasconsultedaboutthedesignandimplementationoftheproject.
HowevertheWFofficialsdidnotattempttointerferewiththeresearchdesign
andfieldworkprocess.TheauthorwouldrequesttheaccompanyingWFstaff
membertostepoutfromtheinterviewroomifpossibleafterthegreetingand
introduction,unlessspecificrequestsforfacilitationorinterpretationweremade
bytheinterviewees,inwhichcasetheWFrepresentativewasoftenafamily
friend.
Ingeneral,theinvolvementofWFanditsstaffmembersdidnotaddtothebiases
inparticipantresponsesorcomplicatetheethicalimplicationsinthestudy.First,
theco-InvestigatorwhointroducedtheauthorusedtoworkwiththeKaihuaWF
onotherresearchprojectsofZhejiangUniversity;heandtheWFleadershad
establishedastrongrelationship.Theauthorwasintroducedasamemberofthe
ZhejiangUniversityresearchteam,andthusquicklygainedtrustduringtheinitial
meetings.Second,unlikesomeofthemoresensitivesocialchallengesthatmay
directlycausesocialinstabilityandresistance,theChinesegovernmenthad
largelytakenastancetoidentifytheissuesrelatedtoLBCandadvocateonthe
needsofeffectivesolutions,despitethelackofconcreteactions.TheStateCouncil
andmultipleministrieshadreleasedstatementsandguidelinesonstrengthening
thecareandsocialprotectionforLBC(GovernmentofChina2013;StateCouncil,
2016).TheKaihuaWFwereunlikelytobeconcernedaboutKaihuaLBC’s
psychosocialdifficultiesbeingdiscovered,whichweremuchlesstroublingthan
thehorribleincidentsinvolvingsuicideormurderthatwerereportedfromother
158
moreremotepartsofthecountry.Third,theWFstaffmemberdidnotgainaccess
toanyconfidentialorsensitiveinformationabouttheinterviewedfamilies
throughtheauthor.
Lastbutnotleast,theKaihuaWFhadalargelyneutralattitudetowardsthe
potentialimpactofourresearchoutcomesontheirwork.AlthoughWF’sfuture
collaborationwithourteamespeciallyontheinterventionprojectmayaddtothe
meritsoftheagencyandbenefititsleaders’performance,theofficialsalso
indicatedthattheywereunwillingtobemuchactivelyengagedtacklingtheLBC
issues.Thiswasbecauseoftheirlimitedresourcesonanumberofpriorityissues,
andtheirconcernthatmoreparentsmaybecomeusedtorelyingonthe
governmenttosupportthecareofLBC,withouttakingresponsibilitieswhen
followingthedecisionofmigration.
5.2.5Analysis
Theanalysisoftheinterviewdatafollowedtheprinciplesofgroundedtheory
approach(Strauss&Corbin,1998),asithelpedtoconstructatheoretical
explanatoryunderstandingofthesocialphenomenonofmigratoryseparation.
Experiencesandperceptionsoffamilymemberswereinvestigated,including
relationshipdynamics,actionsandreactionsinmanagingthesituationof
parent-childseparationandchildcareprovisionintheleft-behindfamily.
Perspectivesofmultiplemembersofthesamefamilyweretriangulatedtohelp
159
overcomethelimitationsofself-reportsandtoimprovetheunderstandingofthe
relationshipandpsychosocialsupportmechanismsintheextendedfamily.
Dataweregatheredandanalyzedsimultaneously,andcontinuallycheckedand
revisedthroughouttheresearchprocess,includingchangestointerview
questions,asnewtheoreticalideasemerged.Memosweretakentogenerate
explanationsoftheemergingconcepts,andfurtherdevelopthekeycategoriesas
wellastherelationshipsbetweenthem,whichhelpedtheauthorcompare
findingsfromthecurrentstudywiththeliterature,inorderto“becomesensitive
tomeaningwithoutforcingourexplanationsondata”(Strauss&Corbin,1998,p.
47)
Wefollowedtheprocedureofopen,axial,andselectivecoding,accordingtothe
methodsproposedbyStraussandCorbin’s(1998,p96)groundedtheory
approach,whichreliedmoreondeductivereasoningthananotherbranchof
classicGroundedTheory(Glaser,1992).Dataanalysisinthisstudyemphasized
theuseofdeduction,verificationandvalidationonemergingquestionsand
patterns,inadditiontothekeyinductionprocessesintheinitialandlaterstages
ofcoding.
Inthefirstphaseoftheanalysis,allofthetranscriptdata,primarilyfrom
interviewswithchildrenandparents/caregivers,wereopencodedinorderto
nameanddefinethepreliminarycategoriesandtheirdimensions,astheoretical
160
conceptswereallowedtoemergeinductively.Maincategoriesthatemergedfrom
experiencesandnarrativesoffamilymembersincludedrationalesofmigrationin
thefamilycontexts,emotionalconnectionsinattachmentrelationship,careand
supportintheleft-behindfamily,andreturnmigrationduetoconcernsabout
childcareanddisciplining.
Thenextphaseaxialcoding(aprocessthatisuniquetoStraussandCorbin’s
approach)embodiedacombinationofinductiveanddeductivereasoning.Data
wasputbacktogetherafterbeingfracturedinopencoding,bymakinglogicaland
meaningfulconnectionsamongthecategoriesthatweremostsalientin
participants’lives.Theemergingconceptsweretestedagainstandfittedintoa
pre-determinedparadigmmodel,aswasproposedbyStrauss&Corbin(1990,p.
96).Theadaptedmodelinthisstudyfocusedonthecontextthatgaveriseto
migration,theconditionsinwhichmigrationandparent-childseparationwas
embedded,theinteractionalrelationshipsandstrategiesinresponsetothe
separationandfamilyconditionsfollowingparentalmigration,andthe
consequencesofthestrategiestodealwiththesesituations.Thismodelwas
furtherdevelopedastheconceptualframeworkofthisstudy,toincorporate
qualitativefindingsthatcomprisedtheessentialcomponentsofthestorylineof
thisthesis.
Inthefinalstageofthecoding(knownasselectivecoding),weselectedand
organizedthemaincategories,andconceptualizedhowthesubstantivecodes
161
mayrelatetoeachotherashypothesestobeintegratedintothetheory(Holton,
2007,p.255).Dataanalysisandmemo-writingbecameincreasinglyconceptual,
throughconstantcomparisonbetweenincidentsandcategoriesfromthedata,
andwithcertaininputsfromtherelatedtheoreticalperspectivesintheliterature.
Acorevariable,themechanismsofmanagingtheimpactofmigration,eventually
emerged,asitspropertiesbecamedenseandtheoreticallyintegrated,and
accountedformostofthevariationoftheleft-behindexperience.
5.3Results
5.3.1Overviewofmigrantfamilies
MostfamiliesinKaihuacompliedwiththe“one-and-a-half-childpolicy”,atypical
variationoftheOneChildPolicyinruralChina:ifthefirstchildisagirl,thena
secondchildisallowed,butnotathird.SoatypicalKaihuafamilycompriseda
singlechild,oranelderdaughterplusayoungerson,ortwodaughters.Amongall
interviewedfamilies,ifbothparentslivedawayfromhome,childrenwereusually
caredforbygrandparents,especiallythegrandmother(exceptonechildcaredfor
byheraunt);theco-residentparentinoursamplewasalwaysthemother.Some
co-residentmotherswerealsoprovidingdailycareforchildren’sgrandparents
whoweretoofrailtoliveindependently;intwocases,themotherdidnotmigrate
becauseofherownphysicaldisabilityorchronicconditions.
162
Grandparentswerewillingtotaketheresponsibilityofcaringforgrandchildren.
Whenbothparentswereaway,onegrandparentwouldnaturallybecomethe
primarycaregiver.Infacttoprovidesupportforchildcareisavirtualexpectation
ofgrandparentsinChineseculture(Shortetal.,2001).Theyalsohavemoretime
forchildrenthanworkingparentsevenwhentheparentsliveathome.However,
theeducationlevelofgrandparentswasgenerallylowinKaihuaandall
grandmothersweinterviewedwereilliterateorsemi-literate.InruralKaihua,
manyfemaleelderlyweredeprivedofeducationopportunities,andtheyhaveled
shelteredlivesoftennotventuringfarfromtheirvillages.
BecauseZhejiangisoverallawealthyprovince,mostpeoplefromKaihua
preferredmigrationtocoastalrichcitieswithintheprovince,about200to300
kilometersaway.Othermigrantsalsochoseworkinsmallercitiesthatarecloser
toKaihua.Mostchildrenseemedtohavethechancetoreunitewithmigrant
parents,betweentwoandfivetimesayear.ChineseNewYearisthetimewhen
almostallfamilieswouldgettogether;thesummerholidayisanotherchancefor
themigrantstotaketimeoffandspendmoretimewiththeirchildren.When
possible,someolderchildrenwouldalsovisittheirmigrantparentsinthecity
duringschoolholidays.
5.3.2Householdsocioeconomicrationalesofmigration
Thedecisionofmigrationwasmadebasedonvariouscomponentsofthefamily
163
context.Ingeneral,ourinterviewresultssuggestthat“toearnmoremoney”was
theprimaryincentiveformigration,duetobetteremploymentopportunitiesand
higherincomeinlargecities,comparedtoKaihuawherefarmingisthe
predominantlivelihoodactivityandthescaleoflocalindustryisverylimited.Yet
thedecisionsaboutmigrationwereusuallynotmadeindividually,butjointly
takenbyandforthewholehousehold.
Manyintervieweesmentioned“everybodyhasgonetothecitiestowork”and“we
havetomakealivingandsupportthefamily,especiallyafterhavingachild”.
Somealsosaidthat“wehavetoleavebecausetherearenojobshereexcept
farming”.TheoverwhelmingmajorityofChinesepeasantshavenopension(Wang,
2006).Parentsoftenmentionedtheneedtoensuresomecapitalfortheiroldage.
Forthebestchanceofsurvivalandpossibleprosperityamonghousehold
members,notleastthechild,parentsfelt“there’snootherway”buttomigrateto
thecities.Aresidentmotherilluminatedthisrationalevividly:
Franklyspeaking,peoplehavetoleadalife,andlifeneedsmoneytobeled.If
onlytherewassomewherecloserwecouldearnsomemoney,wewouldn’t
leave...It’slikewegooutandearnmoneyjustforthekids.Iftheygotohigh
schoolorcollege,youhavetosaveupthemoneyforthem.Andwe’renot
youngnow,andwewon’tbemakingmuchmoneywhenwegetold.
Iftheeconomicsituationhadallowedthemtostay,themigrantswouldclearly
164
prefertolivewiththechildrenintheirhometown.Itseemedthattomigratewas
essentiallyadecisionbythewholehousehold,onethatwasnotmadelightly.
Moreover,itwasnoticeablethatevenaftertheinitialmove,theeconomicgains
frommigrationwereoftenweighedagainstotherintangiblelossinthefamily
relationships,astheykeptmonitoringsituations.Aleft-behindmotherstated:
Her(thedaughter)dadrarelycomesbackhome,justbecausehewantsto
makemoremoney.Hergrandmaisover80yearsoldandsomeonehastotake
careofher.Wehavetwokids–theyalsoneedmoneytogotoschool.Of
courseherdadwantstoseethemmoreoften;buthealwaysgoesoutin
JanuaryandcomesbackinDecember(lunarcalendar).Twoyearsagowe
visitedhergrandpainWenzhouonce,otherthanthatwedon’twanttospend
extramoneyontraveling.
Olderchildrenusuallyhadpositiveperceptionsofthereasonsfortheirparents’
migration,whiletheauthoroftendidnotreceiveclearresponsesfromyounger
children.A12-year-oldboysaid,“Idomisshim(father).Iwishhecouldcome
homemoreoftenandplaywithmeandspendmoretimetogether,butIdo
understandthathehastogoandearnmoney.”Inanothercase,themigrant
motherreportedherhowhersoncopedwiththeharshrealityandbecamemore
considerateofit:
165
Yetheknowsthatthefamilycouldn’tsurvivewithoutthemoney,sowehave
togo...Heunderstandsthisandhasbecomemature.Sometimeshesays,“Mom,
nextyearcanyoumovebackhome?”Butafterafewwordstosoothehim,he
isfinewithit.Especiallywhenwestayedtogetherforawhile,hewould
becomemorerelaxedandeasiertocommunicatewith.
AlthoughthenetincomeinurbanareasisusuallyhigherthanruralKaihuafor
individualfamilies,interestinglytherewasnocleardistinctionineconomicstatus
betweenmigrantandnon-migrantfamiliesfromruralKaihua,accordingtothe
communityleadersweinterviewed.Firstly,richermigrantparentsoftenhad
takentheirchildrentoliveinthecitywheretheywork.Secondly,parentsin
impoverishedfamiliestendedtobeforcedtomigrate,inordertomeetfamily
survivalneeds;thesefamilieswereactuallypoorerthanmostnon-migrant
families.Thirdly,localsmallenterprisesinruralKaihuahadprovidedjob
opportunitieswithcomparableincomewithmigrantjobsincities;somelocal
businessmenalsohadbecomewealthierthanmigrantworkers.
Foryoungcouples,constructionand/orpurchaseofanewhouseforthemselves
andtheirfurtherchildren,emergedasanessentialobjectiveofeconomic
activitiesintheextendedfamily.Thenewhouse,oftennotfarfromthevillage
theygrewup,appearedtohaveacriticalculturalsignificanceintermsofgaining
basicrecognitionofindependenceandgrown-upstatusinruralKaihua.Thecost
ofbuildingahouseinKaihuawasestimatedataboutCNY250,000to750,000by
166
theinterviewees,ahugeexpenseforalmostallfamilies.Althoughtheirparents
wouldprovidesignificantfinancialsupport,thecoststilloftenledtoheavydebts
equivalenttoseveralyears’incomeoftheextendedfamily.Thenaftergivingbirth
toachild,thefamilyeconomicburdenwasfurtherincreased.Agrandmother
describedherfamily’ssituation:
Tobuildthehouse,wesoldourfarmland,butstillhad190,000yuanofdebts.
Mysonhadmembranitiswhenhewaslittle,socouldn’thandelsophisticate
work,andhestilltookajanitorjobinafactoryandhiswifeworksatthe
sameplace.Yettheydon’tearnenoughmoney,andgrandpa(herhusband),at
70yearsold,stillhastoworkfortheforestryonthemountains.
TheseeconomicconcernsinatypicalfamilyinKaihuaweredrivingmanyyoung
peopleoutofKaihuaintothecities,notonlytomeettheircurrenteconomicneeds,
buttoachievebetterchanceforprosperityamongallfamilymembers,
particularlytheirchildren.Inthisregard,parentsalsoexpressedviewsabout
securingbettereducationopportunitiesfortheirchildreninthefuturethrough
migration,asthequalityofschooleducationinKaihuaismuchlowerthanurban
areas.
LikemanypartsofruralChina,theschoolsysteminKaihuahadundergonea
processwherebyprimaryschoolsinmanyremoteareasweremergedintolarger
schoolsinthepopulatedareas,usuallynearthetownshipcenter.Schoolsinsmall
167
villagessimplydidnothaveenoughstudentstorunclasses,becauseanincreasing
numberofchildrenjoinedtherural-to-urbanmigrationflows,andtheoverall
populationsizecontinuedtoreduceduetoyearsofOneChildPolicy.Itwasalso
increasinglydifficulttoretainteachersattheseschools.Asaresult,thecommute
toschoolbecametoolongformanychildren,andsomechosetoliveattheschool
dormitoriesonweekdays.Aprincipalofthetownshipschoolexpressedhis
seriousconcernsabouttheeducationdisadvantagesthatsystematically
underminedchildreninKaihua,orgenerallysimilarruralareasacrossthe
country:
Ourschooldoeshavemoreresourcesatourdisposalthanbefore–butwe
needtospendmoretimeandenergyonchildren.Nowteachersaresobusy
thattheyspendlittletimeoneachkid;andobviouslytheleft-behindonesneed
moreattentionfromtheteachers.Inanycase,tobehonest,thesechildrencan
hardlyclimbupthesocialladderbystudyingatourschool.Inthenational
exams,theywon’tbeabletocompetewiththoseeducatedinthecities.And
sadlyinthissociety,theexamswouldprobablybetheonlywaytosecurea
betterfutureforourkids.
Accordingtoothercommunityleadersandparentsweinterviewed,theeffects
oftheseschoolmergersonchildwellbeingwerecomplex.Infact,thechild
populationhasbeendecreasinginruralKaihua,bothduetothestrict“oneand
halfchild”policyandfamiliesthatmigratedwiththechildren.MostLBChad
168
neverlivedinthecitywithparents;butthosewholivedinabetter-offfamily
maycountontheirparentsinbringingthemthecityinnearfuture,where
boththeeconomiccostandqualityofeducationisusuallymuchhigherthanin
Kaihua.
5.3.3Childwellbeinginabsenceofmigrantparents
Thissectionpresentstheexperiencesandperceptionsregardingtheseparation
duetomigrationandtheparent-childrelationshipoverdistance,withafocuson
theconceptsthatdemonstratechildpsychosocialwellbeingstatus.Asadirect
consequenceoftheabsenceofparentsandthedisruptedconnections,emotional
distresssuchaslonelinessanddepressivesymptomsinchildrenwasreportedby
themselves,andparents,andcaregivers.Theadultfamilymembersoftenclosely
monitoredthechildren’sexperiencesandwellbeingstatus,withawarenessofthe
positiveandnegativeeffectsofmigration.
Emotionalattachmentsustainedoverdistance
Themajorityofchildrenexpressedthattheymissedtheirparents,andoftenthink
ofthemwhiletheywereaway.Despiteofthelengthyabsenceoftheparentsin
theirlife,somechildrenmaintainedstrongemotionattachmentwiththeirparents
throughoutacomplexmigrationprocesswhichinvolvedmanyoccasionsof
separationandreunion.Thefeelingsseemedthemostintensewhenmigrant
parentswereabouttoleavehomeagainafterabriefreunion.
169
A13-year-oldboyZhuangwasseparatedfromhisparentssoonafterbirth.The
parentswereworkinginShanghaiandbroughthimtothecityforkindergarten
whenheturned4.Zhuangreturnedhomeatage7duetobarrierstoaccessing
publiceducationinShanghai.Yetthefamilywouldreuniteinthecityduringthe
summerholidayeachyear.Zhuangreportedthathe“feltpainful”or“wouldcry”at
hisparents’departure,and“wishedtheycouldspendmoretimeathome”withhim.
Hismotherdescribedthesituationas,“Whenheseesushe’sreallyhappy,and
whenwehavetoseparateagainhebecomessadandupset.”
Anotherfamilyhadamuchdifferentmigrationpattern,andtheparentsshared
detailedaccountsabouthowtheeffectsofmigrationwereperceived.Themigrant
father,Qiang,hadbeenawayfortwoyears,leavingtwochildrenbehindwiththeir
mother,Huan.Thefamilywasinterviewedtogetherwhenthefatherreturnedfor
ChineseNewYear.Theparentsspecificallyexplainedaboutthechallengesforthe
parent-childrelationships.
Huan:Kidsalwaysyearnfortheirdadtocomehome;Xingalwayssaysonthe
phone“dad,whenareyoucominghome?Imissyousomuch.”Jing(the
daughter)isabitquietthough.Isaytothemthebestthingyoucandoisto
studyharder.Yourdadis(workingfor)yourfuture.Ifyoustudyhardthenyou
don’thavetoworkfarawayfromhome;ourfamilywillbehappilygathered
aroundthesametable.
170
Qiang:WhenImissthekidsIcallandchat;whenit’snotbusyatworkIvisit
home-Icomebackfourorfivetimesayear.Thekidsmissmeafterawhile.
Althoughwetalkonthephone,they’renotthatclosetoyouwhenyoudon’t
comehomeoften–theyonlyreportthegoodnews,andI’malsocautious
aboutwhatIsaybecausemywordsmayhurt.Whenathomewe’remore
frankwitheachother.
Boththeson,Xing,andthedaughter,Jing,brieflyexpressedmoderateemotional
distressduetoparentalmigrationinthefollowinginterviews,whichmayalsobe
attributedtotheshortdurationofseparation.Themigrantfatherandthe
co-residentmotheralsodidwellincommunicatingwiththeirchildren,about
dailylifeatschoolaswellastherationalesoffather’smigration.
Whereassomeboysopenlyexpressedtheirfeelingsandevennegotiatedwith
parentsonmigratorydecisions,left-behindgirlstendedtoexperiencesadness
andinternalizeddistresswhencopingwiththeseparation.Rui,an11-year-old
girl,liveswithheraunt,andbothherparentsworkintheprovincialcapital.She
wasinterviewedsoonafteratemporaryreunionduringthesummerholiday
whenshewenttoseeherparentsinthecity:
171
MyparentswenttoXiaoshan(AHangzhousuburb)whentheywere18,and
hadmewhentheywere24.IwasbornandraisedinXiaoshanuntilIfinished
kindergarten-thenIcamebacktoKaihua.
Imissmyparentsverymuch.Especiallyatthistimewhentheschooljust
started;itfeelslikeeverythingIseeremindsmeofthetimewhenIwaswith
them.NowwhenIthinkofthemIsometimescryabit.AndthenifIreally
wanttocry,I’dreadabookorsomethingtohelpmestopthinkingabout
them.
Despitethestrongemotionalattachment,Ruishowedcertainresilienceincoping
withtheabsenceofbothparents.Shedidnotshowmuchsadnessevenwhenshe
wastalkingaboutherdifficultfeelings.UnlikemostLBC,Ruigrewupwith
parentsinthecitybeforeelementaryschool,whichmayhavestrengthenedher
relationshipwithparentsasanLBCandhelpedhercopewiththechallengesof
separation.
However,itwasnotablethatfourchildrenactuallycriedduringtheinterview.
Theyseemedtobeaffectedbyeitherdifficulthouseholdeconomicsituationor
parents’divorce,orevenboth.Nine-year-old,Hong,wasinterviewedwithher
mother,Lian.Thefamilywasverypoor;Lianworkedasadoorkeeperforalocal
factoryandhadtotakecareofherparalyzedmother-in-law.Thefathermigrated
toaprovinceinNortheastChinathousandsofkilometersaway,andrecentlyhad
172
anaccidentatwork.Hongstartedweepingassoonasherfatherwasmentioned
andshewasn’twillingtotalk,despiteencouragementfromhermother.Lianwas
unabletocopewithherfamily’sfinancialsituationandattemptedtoseek
assistancefromthelocalcommunity.Whilecomfortingherdaughter,Liansaid:
She’sjustmissingherfathertoomuch;butsheisanobedientgirl.Shesaidshe
wantstovisitherdad.Itoldheritwastoofarandwouldtakeafewdaysto
getthere.Iwouldn’tbeabletofindhimeitherasI’malmostilliterate;plusher
grandmaneedsmeathome.Thenshedidn’tsaymuch–andsometimesshe’d
justcry.
Meanwhile,somechildrendidnotdemonstratestrongemotionswhentalking
abouttheabsenceoftheirmigrantparents,andaccordingtotheco-resident
caregivers,nosignificantproblemsinparent-childrelationshipwasobserved.For
example,Wenhui,aneight-year-oldgirlwasinterviewedwhosebothparentsleft
whenshewasthree.Accordingtoherteacherandgrandmother,Wenhuiisa
well-behaved,thoughtfulandintrovertgirl.Sherespondedtoourquestionsonly
inbriefwords:
Interviewer:Doyoumissyourparents?
Wenhui:Yes.
Interviewer:Doyouoftenthinkofthem?
Wenhui:Notreally.
173
Interviewer:Areyouoftenincontactwiththem?
Wenhui:Notmuch…justafewphonecallsaweek.
Interviewer:Wouldyouseekhelpfromthemwhenyouhavesometroublesor
difficulties?
Wenhui:No.
BasedonWenhui’sresponses,itremainsunclearhowtheparent-child
relationshipswereaffectedbymigration.Afterall,itwouldbechallengingtoelicit
theprobablycomplexemotionsandrelationshipsfromsomebriefdescription
aboutone’sfeelings,especiallyforaquiet,youngchild.Morein-depthexploration
ofthecareenvironmentinthefamilywouldbenecessary,andWenhui’scaseisto
bedelineatedfurtherinthefollowingsections.
Emotionalwithholdinganddetachmentovertime
Incontrasttothosewhofeltstronglytheabsenceoftheirparents,somechildren’s
feelingsabouttheirmigrantparentsseemedtobeambivalent,evenshowing
indifferenceduringtheinteractionswiththeparents.Mostofthesechildrenwere
leftbehindearlyinlife,evenduringinfancy;theemotionalconnectionsand
attachmentrelationshipwerelikelytobeconsiderablyweakenedbytheextended
separation.Also,childrenwithlimitedfamilycareresources,suchasthosewho
werelivinginextremepoverty,appearedtobeparticularlyvulnerabletothe
disruptionofparent-childrelationship.
174
Childrenwhodidnotdemonstrateacloserelationshipwithmigrantparents
tendedtobeunwillingtoanswerphonecallsfromthem.Oftentimesthe
co-residentfamilymemberswouldmakeeffortsonencouragingthechildtobe
cooperativeandcommunicatewiththeparentswhentheycalled.Amigrant
father,Shengli,describedthepatternsofcommunicationwithhis12-year-oldson,
andthechangesafterhevisitedhome.Shenglilefthomeshortlyafterhisson’s
birth;then4yearslater,hiswifejoinedhiminthecity,leavingthechildbehind
withgrandparents.
Whenweareawaymysondoesn’treallycommunicatewithmealthoughI
callhimquiteoften.Sometimesgrandmahastomakehimpickup.Overthe
phonewe[heandhiswife]justurgedhimtostudyhard,itcertainlyannoys
himthatwetalklikethatallthetime.SonowIgetthissituationmoreasI’ve
beenreflectingonmywaysofdisciplininghim,andeventhoughtaboutmy
ownexperiencewhenIwasakid–It’dprobablybetternottopushhimtoo
hard.
Iusuallycomeback3or4timesayear-whenI’mhomehegetsreallyclingy
andtellsmemoreabouteverything–especiallythingsthatworryhim.My
wifeactuallycomeshomeandstayswithhimduringthesummerholiday.Or
sometimesinholidaysmysonwouldgotoHangzhou[whereShengliworks]
forseveraldaysandcomebacktogetherwithmywifeandme.
175
Givenhisveryearlydeparturefromhisson’slife,Shengliseemedtohavemade
strongeffortsinmaintainingtherelationshipthroughphonecallsremotely.The
effortsencounteredconsiderablechallenges,andwereoftennotwellreceivedon
theotherendofthephone.However,importantly,hewasawareofthissituation,
anditappearedthathegraduallybecameabletomakemoreemotional
connectionsandevenprovidepsychologicalsupporttohisson,whichmighthave
contributedtotheclosenesswiththechildduringtheshortreunionsandthe
upsurgeinattachmentbehaviors.Thisshowedtheambivalentorcontradictory
attitudesandfeelingstowardsseparationfromparents.Shenglilateraddedthat
as“economicstatusisgettingbetter”,heandhiswifewereplanningtomovetoa
smallcitynexttoKaihuaandprobablyacceptalowerpaidjob,tobeabletospend
moretimewiththechild.
Someotherchildrenseemedmoredetachedfrommigrantparents,andeven
reunionswerenotnecessarilydesirable.ShortlyafterZhimin’sbirth,hisparents
migratedtoNingbo,arichcityineasternZhejiangProvince,leavinghimwithhis
twograndparents.TheparentsownedasuccessfulgrocerystoreinNingbo,and
theyactuallyhadarrangedforZhimintogoandlivewiththeminthecity.The
10-year-oldwasinterviewedtogetherwithhisgrandmother.
Interviewer:Howoftendoyoutalkwithyourparentsoverthephone?
Zhimin:Abouteverytwoweeksorso.
Interviewer:Doyoulikechattingwiththem?
176
Zhimin:No.Becausetheyjustmakeafuss.
Interviewer:Whatwouldyouguystalkabout?
Zhimin:Juststudy.
Interviewer:Doyoumissyourmomanddad?
Zhimin:SometimeIdo,sometimesIdon’t.
Grandma:ActuallyhisparentstookhimtoNingbowhenhewasreadyfor
kindergarten-thetuitionwasevenpaidalready.Buthejustcriedandcried
andhismotherhadtobringhimbacktous.It’sjustthatIwastheonewho
broughthimupsincehewasborn.Andhewasalsoreallyattachedtohis
grandfatherandwouldn’tlethimoutofhissight.Evennowhedoesn’treally
misshisparents,anddoesn’tseemclosetothemevenwhentheyarehome
onceortwiceayear.
BytakinghimtoNingbo,Zhimin’sparentsappearedtohaveforceduponhim
closerparent-childrelationshipsthathedidnotwant,aftergainingabetter-off
economicstatus.Lateron,theparents’communicationstylewasgenerally
didactic,withoutmuchsensitivityabouttheirson’spsychosocialwellbeingor
anythingexcepthisperformanceathomeandschool.Notsurprisinglytheboy
wouldturntohisgrandparentsforhelpanddailycare.
AlthoughZhimin’scaseillustratedthatricherfamiliesinKaihuadidnot
necessarilycopewellwiththeearlyseparation,itseemedpoorfamilieswith
complexproblemswouldencounterseriousdifficultiesinmanagingthe
177
relationshipstressduetomigration.AnexamplewasXia,aged13,whosemother,
Fang,re-marriedafterherfirsthusbanddiedandhadXiainherlate30s(verylate
inruralChina).Xia’sfatherhadbeenamigrantworkersincehewasateenager,so
shewascaredforbyFangfrombirth.AccordingtoFang,Xiahadfrequent
episodesofpsychosomaticabdominalpainanddizziness,andhadbeenoffschool
formonthsasaresult.
Interviewer:Doyoumissyourdadashe’salwaysaway?
Xia:[Shakeshead]
Interviewer:Whynot?
Xia:I’mnotsure.It’sjustIusedtogetalongverywellwithmydadwhenIwas
little;itwasliketwopersonsinone.ButasIgotolderwegrewapart.Wejust
don’tcommunicateanymore.
Fang:Theydotalkoverthephone.Buttheyjustcan’thaveaproper
conversation.IaskedherDadtocallupeveryweekandhaveachatwithher.
Buttheyhavenothingtotalkabout.ShehasnothingtosaytoherDad.Even
whenhe’sbackathomeitdoesn’tgetmuchbetter.
Xia:I’vejustgotnothingtodiscusswithhim.Iusuallyhangupwithinone
minute.
Fang:It’salsoherdad–hedoesn’tliketalkingverymuch,soshedoesn’t
feelinglikingtalkwithhimeither.
Interviewer(toXia):Doyoufeelstressedorlonelybecauseyourdadisnot
home?
178
Xia:It’sjust…Iusuallydon'ttalkmuchtopeople,orhangoutwithpeople…I’d
juststayinthehouse.
Interviewer:Whynot?
Xia:BecauseIcan’tfindanyone[smilessadly].
Interviewer:Really?Isitjustyourdad’snothereoryoudon’thavemany
playmatesorfriendsaround?
Xia:It’sjust…
Fang:She’sveryintroverted.
Xia:Also…theydon’tseemtounderstandwhatI’vebeenthrough.
Adverseeffectsfromfather’smigrationandcomplexdisruptionsinfamily
relationshipsprobablycontributedtothesignificantmentalhealthsymptoms,
includingpeerrelationshipproblemsthatXiawassufferingfrom.Importantly,the
negativeimpactduetothelengthyseparationmayhavebeenaugmentedover
time;despiteoftheclosenesswithherfatherduringherchildhood,atpresentthe
distantrelationshipseemedlargelydysfunctionalwithlittleemotional
connectionsofeffectivecommunicationbetweenwithherfather.Pooreconomic
statuscertainlyaddedtotheburdenofthefamily,possiblytothepsychological
burdenofXiaaswell.Yetbesidesthematerialresources,psychosocialsupport
fromtheextendedfamilyoreventhecommunitywouldhavehelpedtorelievethe
challengingsituationsinthisfamily.
179
5.3.4Carearrangementsandpsychosocialsupportintheleft-behindfamily
Thissectionexaminestheeffectsoffamilycareandpsychosocialsupporton
safeguardingchildwellbeing.Asidefromeconomicremittances,thenon-material
resourcesmaytakevariousformsandcomefromdifferentsources.Most
importantly,extendedfamilycarecomprisedessentialsupportandguidancefor
overallchilddevelopment.Sincethepsychosocialresourcestookvariousforms
andcamefromdifferentsources,manyfactorsotherthanparentalmigrationin
thesocio-ecologicalsystemmayinfluencethecareandsupportthechildreceived.
Asignificantresultofparentalmigrationwasthatfactthatmanychildrenbecame
muchclosertocaregivergrandparentsthantothemigrantparents.AWFstaff
memberquotedtheChineseproverb“closenessbetweeneveryothergeneration”
todescribethissituation.ForexampleinZhimin’scasefromtheprevioussection,
thegrandparentswereclearlythemostimportantcaregiversintheattachment
hierarchy,whereastheparentsonlyplayedasubsidiarycaregivingrole
throughouthischildhood.
Althoughthegrandparent-childclosenesswaslikelynotthefullequivalentof
parent-childbondsinmanyotherfamilies,somechildrendidappeartobehappy
withgrandparents’carewhichreplacedtheusualresponsibilitiesofparents.A
boysaidintheinterview:“IfeelOK(aboutmyparents’absence)becausemy
grandparentstakecareofmeverywell,basicallythesameasmyparents-although
180
Ilikemyparentsthemost,becauseItheyunderstandbetteraboutwhatIusually
thinkaboutandwhatIusuallydo.”
Afterall,whenbothparentsmigrated,theco-residentgrandparentswereclearly
themostessentialsourceofcareandsupportforalmostallchildrenleftbehindin
Kaihua.Thefactthatabsenceofparentalcaredidnotaffectthesechildren
accordingtotheirnarratives,impliedthatsomegrandparentswerecapableof
activeengagementinchildren’shomelifeandprovisionofeffectivesupport,
despitethechallengesinmaintainingemotionalconnectionsandhealthy
interactionsinchilddisciplining.Morefindingsaboutthechallengesin
grandparentsfulfillingtheircaregivingroleswillbepresentedinthenextsection.
Inaddition,siblings,andsometimescousins,werealsoincloseinteractionsand
mayprovidepsychosocialbenefitstoeachotherinthesharedcareenvironment.
Althoughmanychildrenreportedthattheywouldnotfeellonelybecauseoftheir
brotherorsisterathome,fromtheinterviewsitwasunclearwhetherthechild’s
wellbeingbenefitedfromhavingasibling,partlybecausethefamilieswithmore
childrentendedtobepoorerthanthesinglechildfamilies.
Butitwasparticularlynoteworthythatinthefieldvisits,theauthorfoundseveral
householdswherecousinswerelivingtogetherandcaredforbytheircommon
grandparents,duetothesmallnumberofchildrenineachnuclearfamily.
AlthoughthecaregivershadtolookaftermorethanoneLBC,thiscareapproach
181
wassustainedbyresourcesfrommultiplenuclearfamilies.A12-year-oldboy
namedJie,waslivinginthistypeofhousehold,andappearedtobehappywithit:
Ienjoylivingwithmycousininmygrandparents’house,andIoftenhangout
withalotoffriendsaswell.MygrandparentsaregreatbutIdomissmy
parents.SointheschoolholidaysIusuallygotoHangzhoutobewiththem,
exceptthattheymaycomebackhomeinthesummerandduringChineseNew
Year.
ApparentlyJie’sparentswereabletoaffordthetimeandmoneyspentonthe
family’sfrequenttravelsbetweenHangzhouandKaihua.Theauthorlaterlearned
thathisgrandparentsalsohadstableincomefromlightworkathomeforthelocal
processingindustry,andwerephysicallyverycapableoftakingcareofthe
children.Moreover,asJieandhiscousinwerebothonlychildren,the
concentratedresourcesfromtheirparentsensuredareasonablelivingstandard
forthesechildren.Theoverallfamilyatmospherewasfilledwithfrequent
interactionsamongthechildrenandbetweenthechildrenandgrandparents,and
nosignsoflonelinesswereobservedintheinterviews.Itseemedthatboth
childrenbenefitedfromeachother’spresenceinthefamily.
Insomeotherfamilies,thecarearrangementsandfamilysupportstructurewere
morecomplex,asmembersofextendedleft-behindfamilywerelivingapart,and
grandparentcaregiverswereseparatelycaringfordifferentchildren.The
182
previouslymentionedquietgirl,Wenhui,whowasmainlylookedafterbyher
grandpa,talkedmoreaboutherregulardays.ShewasinYear2inanelementary
boardingschoollocatedatthetownshipcenter.Shehadanoldersister,whogoes
toanotherschoolandliveswiththeirgrandmainthecountycenter.
Interviewer:Wouldyoubeleftathomealonesometimes?
Wenhui:Notreally,becausemygrandmaandsisterandcousincomehomeon
weekendstoo.GrandmaliveswithmysisterinKaihua(countycenter);my
cousinisinhighschoolandcomeshomeeveryotherweekend.
Interviewer:Isyourhomefarawayfromschool?
Wenhui:Yes.GrandpawouldcomepickmeuponFridays.Orsometimeson
Wednesdayifitgetstoohot(totakeashowerathome).
Interviewer:Sowhatdoyoudoonweekends?
Wenhui:Domyhomework.Andplaywithmysisterandcousin.
Interviewer:Wouldyouhangoutwithotherfriendsaswell?
Wenhui:Notreally.Imostlywanttohangoutwithmyfamily.
Interviewer:Wouldyouhangoutwithotherfriendsalotwhenyouareat
school?
Wenhui:Yeah,kindof.
AlthoughWenhui’scousinwasunabletobeapproachedfortheinterview,the
authorlearnedshewasalsoleftbehind,withhermotherathomewhowasher
primarycaregiver.Withinanextendedfamilystructure,thesecarearrangements
183
seemedtobeabletoovercomethechallengethatthetwosistershadtogoto
differentschoolsfarawayfromeachother,andprovideadequatecareand
necessarysupporttoeachofthem.Livingatschoolwithotherchildrenmayalso
haveplacedWenhuiintoapro-socialenvironment,althoughitseffectswere
unclearfromtheinterview.
Yetwhenpartsofthenuclearfamilysystembecamedysfunctional,grandparental
carewasmoreofthechild’slastresortinsteadofanintentionalarrangement.
Particularly,thedivorceofmigrantparentsmaycausesignificantdisturbanceto
thefamilydynamics,anddiminishthechanceforthechildtoreceivepropercare
andotherresources.Theparentsofa13-year-oldboynamedDongyanggot
divorcedwhenhewaseight.Hisfatherwasaconstructionworkerwhowould
oftentraveltonearbyruralareas,whilehismotherhadmigratedtoacitynextto
Kaihua.Dongyangstartedtocryatthebeginningoftheinterview,sotheauthor
stoppedtheinterviewandlatertalkedtoaneighborofhim,whoknewhisfamily
wellandoccasionallycaredforhim.Theneighborreported:
It’shis80-year-oldgrandmawhomainlycaresforhim(grandpapassed
away),afterhisdadwonthecustody.Hesometimesvisitshismaternal
grandparentsaswell.Buthis(paternal)grandmaisabittoofrailasa
caregiver,andthemoneysentfromhisdadisbarelyenough.Rumorhasit
thatthedadoftengoesgambling,whichruinedthemarriage.
184
Noweventhekid’sbedisbrokenandhasn’tbeenfixed.Hesometimescan’tget
holdofaphonetocallhisdad.Heoftenhastogobuyandeatinstantnoodles
–youcanseehe’squiteshortforhisage.SometimesIinvitehimtoeatatour
house,asheoftenplayswithmygrandson.Hismomheardaboutthe
problemsandnowoftengetshimtostayatherplaceonweekends.Butthen
hedoesn’thaveanyfriendstoplaywiththere.
ThelocalWFwasawareofDongyang’ssituationandwasabouttooffercertain
financialandsocialassistance.Dongyang’sexperiencessuggestedthatwhenthe
parentswereaway,thechild’scaregiversmightnotevenhavethecapabilityto
providebasicresourcestomeetessentialdevelopmentneeds,eventhoughthey
werewillingto.Divorcewasasignificantfactorwhichnotonlyresultedin
psychosocialdistressinthechild,butmayhaveweakenedimportantrelationship
tiesinthefamilyincoordinatingchildcareandhouseholdgoalstoprioritizethe
needsofthechild.Underthiscircumstanceandwithverylimitedmaterial
resourcesavailableinthefamily,thechildwouldhavetoresorttoextrasupport
fromthesocio-ecologicalsystem,includingallrelatives,neighbors,and
communityofficials,whomayormaynotbeabletohelpovercomethe
difficulties.
Insomefamilies,adultrelativesotherthangrandparentsmayalsobetheprimary
caregiverofthechild.Rui,theabove-mentioned11-year-oldwhowouldreada
booktodivertthoughtsofherparents,hadbeenlivingwithheraunt(her
185
mother’ssister)anduncle,aswellastheirson(hercousin)whojustgraduated
fromcollege.Ruitoldusmoreaboutherexperiences:
Myauntietakescareofmeprettywellactually.Shecooksevenbetterthanmy
dadandmom.Iwouldjusteatdinnerafterfinishingmyhomework,andthen
takeashower,readabookorsomethingonmybedbeforesleep.ButIcould
talkinMandarinwithmyparentsaboutthings,withthem(aunt’sfamily),
theytalkintheirlocaldialectandsometimesIdon'treallyunderstand,and
yetIdon’twanttointerruptthem…IthinkIwouldgetalongbetterwithmy
parents.
Ruiseemedtobeorganizingherdailylifequitewellinadifferenttypeof
extendedfamily,despitetheabsenceofbothherparentsandheremotional
thoughtsofthem.Rui’sexperienceoflivinginthecitybeforeelementaryschool
formulatedastrongerattachmentrelationshipwithhermigrantparentsthan
manyotherLBC,whichwouldbenefitherpsychosocialwellbeingafterthe
separation.Itseemedthatshewasabletopeacefullyacceptchangesinhersource
ofcare,anddevelopedtheabilitiesinadaptingintoanothersupportiveyetless
homelikeenvironment.
5.3.5Careanddiscipliningconcernsandthereturnofmigrantmother
Thissectionfocusesonhowthechallengesinmaintainingparent-child
186
relationshipsandsupportivecareenvironmentweremanifested,inthe
experiencesofthechildandperceptionsoftheparentsandcaregivers.The
discipliningissuesandotherconcernsaboutchildwellbeingrelatedtochildcare
provision,demonstratedinthesecases,leadtocertainchangesofpriorities
consideredinthemigrated-relateddecision-makingrationales.Themostexplicit
changewastheactualorintendedreturnmigrationamongthemigrantmothers.
Itshouldbeclarifiedthat“returnmigration”hereisreferredtoasthepermanent
decisiontocomebackhomeafterquittingjobinthecity,ratherthanatemporary
returnforareunionthatusuallyoccurredduringholidays.
Naturallyastheprimarycaregiverofthechild,aleft-behindmotherwouldplay
criticalrolesinsafeguardingthewellbeingofLBCathome,whiletakingchargeof
thevastmajorityofchildcareactivities.Manychildrenwhosefathermigrated
showedstrongattachmentwiththeirco-residentmother.Asagirlsaid,when
talkingaboutfeelingsaboutherfather’sdepartureafterareunion,“It’salright
becausemymotherishome.”However,remittancesfromonlyonemigrantmay
notmakeenoughfamilyincome,somanyyoungmothersstillhadtoworkintheir
village,township,orcounty,whichofteninvolvedlongcommute.Insuchsituation,
withoutstrongsupportfromthegrandparents,adequatechildcarefortheLBC
maybecomechallenging.
Forinstance,Yilin,motherofan11-year-oldboy,livesinasmallmigrant-sending
community.Herhusbandmigratedjustaftertheirsonwasborn.Yilinherselfleft
187
homewhenhersonwassixyearsold,butrecentlyquitherjobinthecityand
movedbacktoworkinalocalfactory.Theauthoraskedher,“Doeshe(herson)
misshisdad?Doyouthinkhefeelslonely?”Heranswersdepictedsomevivid
scenarioswithhersonathome.
WhenI’mhomeit’sbetter…Hewouldhaveachatwithdadonthephone
sometimes,butafterallthey’vebarelylivedtogether…Hisdadwenttothecity
whenhewasafewmonthsoldandrarelystaysathome.Justbecause…the
twoelders(parentsofYilin’shusband)onlyhaveoneson,sohehastogo.I
usedtobeawaytoobutcamebackthisyear–havetotakecareofmyson
nowashe’sgrowingup.
Butstillheiscertainlylonely,beingonhisown.Hewouldoftensay,“I’mso
boredbeingalone,Mom”.Iwouldask,“Whydon’tyougofindsomeonetoplay
with?”Hesaystheyallhavetheirownplaymates;hereallyonlyplayswiththe
grandsonofthatman[pointingtoaneighbor].Sometimewhenhegoesout
afterfinishinghomework,theotherkidsarealreadygoneandhecan’tfind
anyone…thenhehastocomehome,saying“sobored”again.Actually
sometimesIinvitesomekidstocomeoverandhangoutwithhim.Theyplay
heresothatIcanheadouttowork;otherwiseI’dhavetoherechatwithhim,
orelsehe’dgetbored.
Thegrandfatherwassaidtobetoofrailtotakepartinchildcare.Thegrandma
188
lateraddedinthesameinterviewthatshejustcouldnothandlehergrandson,
whowouldnotlistentoher,andwouldoftenmakeanexcusetogooutside.
DespitetheresponsibilitiesYilindecidedtotakebyreturninghometolookafter
herchild,sheseemedunabletospendenoughtimeonhimandprovideadequate
accompaniment.Althoughthechildexplicitlyexpressedhisfrustrationdueto
loneliness,thefamilymembersseemedtobeequallyfrustratedbecausethey
couldn’tfindaviablesolutiontothesituation,largelybecauseofthelimited
capabilityofgrandparentsinfulfillingsupportiveroles.
Ingeneral,grandparentsseemedtoshowamorerelaxedattitudetowards
discipliningchildren,inparticularlesslikelytousephysicaldiscipline.Oneofthe
communityleadersweinterviewedattributedthistotheconcernthatbeatingthe
childbygrandparentswouldmaketheparentsunhappy,althoughother
intervieweesfeltthatparentswouldwantstandardsofdiscipline(including
physicaldiscipliningintheChinesecontext)tobemaintainedintheirabsence.
Inaddition,aswefound,somegrandparentswerequitefrailandinpoorhealth,
simplylackingthenecessaryphysicalcapabilitytoprovideadequatecare.Their
burdensbecamemuchheavierifthereweremultiplegrandchildrentobecared
for.Forexample,inthecompanyofaWFstaffmember,weinterviewedthefamily
ofMrsLiu,62,whotookcareofthreegrandchildreninherhome.
189
Interviewer:Howwouldyoucompareyourcarewiththeparents’careforthe
kids?
WFstaffer:Grandparentsusuallytendtospoilthemabit.
MrsLiu:Theyaretheyoungestgeneration–wegrandparentswouldn’tbeat
themforsure.
Interviewer:Aretheiranydifficultiesforyoutotakecareofthemall?
MrsLiu:Astodifficulties,mybodystillworks,andweareusedtoharshlives
intheruralvillage[laughs].Butitisreallytoughworktakingcareofthe
threetogether.Kidsarejustnaughty–asyouseethehouseisamessallday
long.
Tingting,avivacious11-year-oldgirl,wasoneofMrsLiu’sgrandchildrenwho
wasathomeduringourinterview.Howeveritwasdifficulttoinitiatea
conversationwithherbecauseshewashappilyrunningaroundorplayingwith
her3-year-oldcousinanddidnotsitstilllongenoughtorespondtous.Her
mother,amigrantworkerwhoalsohappenedtobeathomeforaholiday,toldus
more:
Tingtingisreallytoolively…Hergrandparentsdon’tdisciplinehermuch;
they’renotstrictwithheratall.Theoldergeneration–theythinkvery
differentlyaboutlotsofthings.IworrythatTingtingwillstartpubertysoon,
andhergrandparentswillstillusetheoldvaluesandstandardswhencaring
forher.ActuallymyhusbandandIalreadydecidednottogobacktothecity
190
afterthismid-autumnday[theholiday]tolookafterher,asshe’sbecominga
biggirl.
ThemotherlatermentionedthatTingting’sgrandpausedtobeaschoolteacher,
soshewasnotworriedaboutthelackofhelpwithherstudies.Concernsabout
inadequatediscipliningonTingtingseemedtobethecentralfactorofthe
mother’sreturn.Itappearedthatnoneofthefamilymemberswereindeed
responsiblefortheproblem.Rather,someintervieweesattributedthehuge
discrepanciesbetweenthevaluesystemsofgrandparentsandchildrentothe
so-called“generationgaps”,asaresultofdrasticsocialchangeinChinaoverthe
pastdecades.Accordingtoacommunityleader,“Thegrandparentsaretwo
generationsolder–therearealotofnewthingsthatthekidsareinterestedbutthey
don’thaveaclueabout.Sooftentimestheycan’treallycommunicatewiththeir
grandchildrenortellthemwhattodo.”
Someothergrandparentsappearedtoadoptstrictermeasuresinacrudewayin
supervisingthechild.Forexample,afewteenagechildrenreportedthattheir
grandparentswouldnotallowthemtogooutunsupervisedforfearofaccidents,
sotheywerepermittedminimaltimeoutsideevenforschoolactivities.Whether
grandparentaldiscipliningstrategiesappearedtoberelaxedorstrict,theydidnot
seemeffectiveinmanycases.Somechildrencaredforbygrandparents,especially
boysleft-behindfromanearlyage,mayhavedevelopedexternalizingbehavioral
problems.Thiswassuggestedbycommunityleaderswhoreportedsome
191
left-behindboyscommittedvandalismorsufferedinjuriesinminoraccidents
fromriskyplayandactivities.Duetoconcernsaboutherchild’sbehavioral
change,amigrantmothernamedXiaojinsaid,whileherhusbandwouldcontinue
workinginthecity,shewouldquitthecityjobandcomebackhomenextyearto
takecareofthe11-year-oldson:
It’sjusthistemperthatI’mreallyworriedabout.He’sbecomeveryrebellious
inthelastfewmonths.Hewouldn’tlistentohisgrandparentsatall;always
answersback.WhenI’mhomeit’sbetter.Thisdoeshavesomethingtodowith
usbeingaway,becausegrandparentscan’tcontrolhim,andwecan’tcontrol
himfromadistance.
Inadditiontoconcernsaboutchildwellbeingathome,otherimportantevents
thataffectedtheentirefamilydynamicsmayalsocontributetothedecisionof
returnmigration.Forexample,thearrivalofanewbornoftenforcedthemother
toreturnhomeforthepostnatalperiod,andsometimesmuchlongerduetothe
needsforchildcare.Jiepingwasamotherwhomigratedsoonafterherdaughter’s
birth.Shedecidedtoreturnhomepermanentlyfollowingthebirthofhersecond
child,aboy,whenherdaughterwas8yearsold.Specifically,Jiepingemphasized
theinterestingchangeinherdaughter:
Howtosay…it’slikeIdidn’tspendmuchtimewithmydaughterwhenshewas
little,andweweren’tatallclose.Iknewshewaslonelyathomejustwithher
192
grandmother.ButnowthatI’mstayinglongerathomewithher,she’sdoing
muchbetter.Sheusedtogooutallthetimetotrytofindplaymates.Nowthat
herlittlebrothermakesthefamilylively,andsheactuallylikesstayingat
home.Becausebothofus[theparents]wereawaywhenshewasyoung,she
becamequiteintroverted,tobehonest.Butnowsheseemstobecomingoutof
herself.Everythingisgettingbetter–exceptwe’llneedtofigureaboutmoney
asthelittleguygrowsup.
TheauthorwasunabletoapproachJieping’sdaughterandthusfurther
informationaboutthechild’swellbeingwasunavailable.AsJiepingpointedout,
resourceconstraintsmayposecertainadverseeffectsonthelong-termwellbeing
ofthechildren,duetothereducedfamilyincomeafterherreturntoKaihua,and
theadditionalexpensesonthenewborn.
Theactualeffects,especiallythelongertermimpact,ofthemother’sreturn
remainedunclearamongthecasespresentedinthischapter,despitetheinitial
intentionofreturningtoprovidebettercareanddisciplining.Returnmigration
wasmoreofareflectionoftheproblemsthatthemigrantparentsandco-resident
caregiverswereabletodetect,andtheseproblemsbecametoosignificantforthe
migrantparentsnottotakeanyactions.
5.3.6Challengesineducationattainmentandcommunityenvironment
193
Manyparentsconsideredsatisfactoryschoolrecordaveryimportantobjectiveof
raisingtheirchildren.Whileacademicachievementitselfisanimportant
indicatorofchilddevelopment,itmayalsodirectlyaffectthepsychosocial
wellbeingofthechild.Aswasfrequentlymentionedduringtheinterviews,
childrenwhodidnotperformwellatschoolexamswereoftenundersignificant
pressurefromthefamilytostudyharder,inawaytoreciprocatethehardworkof
theirmigrantparents.MostintervieweesbelievedthatinruralareaslikeKaihua,
thereexistfewpathwaysforchildrentoemergeandsucceedotherthandoing
wellatschoolandpursuingqualityhighereducation.
Howevernomigrantparentsreportedconcernsaboutschoolperformanceasthe
primaryreasonofreturnmigration.Thismayberelatedtothebeliefthat
migrationmightpotentiallybringbettereducationopportunitiesfortheLBCin
thefuture,withthehelpofincreasedfamilyincomeandprospectsofmoving
permanentlytothecity,aswasnotedinanearliersection.Inthemeantime,
becauseofthephysicaldistanceaftertheirdeparture,migrantparentsbecame
unabletocloselymonitortheLBC’sschoolperformance.InKaihua,youngparents
wereusuallyeducatedatprimaryschoollevelatleast,yetmostpeopleinthe
grandparents’generationwereilliterate.Consequently,afterparentsmigrated,
manychildrenwouldreceivemuchlessacademicsupportandsupervision,evenif
onlyoneparentwasaway.Forexample,accordingtoXing’smotherHuan,the
boy’sschoolperformancehadbecomemuchworseafterhisfatherleft.The
parentsexplainedtheirworriesabouttheirbothchildren:
194
Qiang:Althoughwedon’tknowmuchabouthishomework,hewouldbehave
betterifIcouldbewithhim.Whenyousay(overthephone),“Xing,doyour
homework!”Hereplies“OK!”anddoesn’targue,buthestilldoesn’tdoit
properly.HoweverifIweresittingherehe’dhavetodoit–noescape.
Huan:Iusedtowatchthekidswhilemyhusbandwascookingdinner.Now
thatIhavetocook,Ican’twatchthem–IalsocaughtJingcopyingsomeone
else’shomeworkseveraltimes.Itoldher,thisisnotOKatall-ifyoudon’t
knowtheanswer,youhavetofigureitoutbyyourself.She’sagoodgirland
wouldfeelbadifshedoesn’tdowell.Butcheatingisnottheway.
Havingrealizedtheproblemandattributedittomigration,theparentshadno
intentiontoreturnhomepermanentlyyet.Afteralltheirpresencecouldnot
replacethechild’seffortsatschool,andtheywereonlyhopingforthebest
whilesparingnoeffortsonpushingandmonitoringhim,athomeorremotely.
Afewchildreninthisstudywentthroughcertainpreschooleducationinthe
citywherebothmigrantparentswereworking.Afterexperiencingearly
educationinthecityunderparents’care,theyhadtoreturnhometoattend
primaryschoolduetorestrictionsofschooladmissioninurbanareas.For
example,Zhuanghadexperienceof3yearsinakindergarteninShanghai,and
hermotherdescribedthechallengesofchildeducationforher:
195
He(Zhuang)wenttokindergarteninShanghai,sohe’sseenmoreoftheworld;
he’sgotagoodpersonalityandtalkssweetlywithpeople.Butbefore
elementaryschoolwehadtosendhimback,asitwasdifficulttogetinto
schoolsinShanghai.ThenfromGrade1wesenthimtoaboardingschoolin
Kaihua;hisgrandmawasgettingtoooldtotakecareofhimeveryday.Sohe
onlycomeshomeatweekends.Hehasbecomequiteindependentintaking
careofhimself.Butnowhisschoolrecordisjustfair,Idon’tknow,maybe
medium.Becausenobody’sabletohelphim;it’sallonhisown.
SincebothofZhuang’sparentswereworkinginthecity,theywereableto
jointlyprovideareasonablelivingstandard,andpreschooleducationforthe
childthatwasofhigherqualitythanwhatwasavailableinKaihua.Thisperiod
ofco-residencewiththeparentinZhuang’searlychildhoodalsohelpedhim
buildmoresecureattachmentrelationshipswiththem.Howeveritwas
unclearhowthischangeofschoolsettingsmightaffectthechild,andlike
manyotherLBC,theystillhadtofacethechallengesregardinglackof
academicguidanceandsupportathome.Grandparentswereapparentlynot
ashelpfulinacademicsupportandsupervisionasparents,whichraiseda
considerableconcerninthefamily.
Forchildrenlivinginremoteareas,longdistanceandlackoftransportationfrom
hometoschoolmaycausefurtherdisadvantageintheireducation.Someofthem
hadtoliveatthenearestboardingschoolinthetownship,andcomehomeonly
196
onweekendsandholidays,startingfromthefirstyearofprimaryschool.Other
families,morespecificallyatleastonegrandparentandthechild(suchas
Weihui’sgrandmaandhersister),chosetomoveclosertotheschoolandliveina
rentedaccommodationonschooldays,toavoiddifficultiesofcommuting.Yetnot
allfamilieswereabletoaffordthisstrategy.Anothergrandma-caregiver,Mrs
Zhao,elaboratedthesituationwithherandhergrandson:
Theschoolinourtownshipisabitfarfromthebusstop-Ihavetopickupthe
kidifitrains,whichsometimesisnoteasyforme.Theschoolinthe
neighboringtownshipisfartherfromhome,butit’srightnexttothebusstop
soIdon’thavetopickhimup.Sohegetsupat5:30inthemorningandcomes
homeat5pm.
Asthepopulationisdecreasinghere,manyschoolsmergedtogether–people
oftengoalongwaytoschoolorrentaplacenearbytolookafterthe
grandkids.Weusedtosendthemtoaboardingschoolabout20kilometers
awayforfirstandsecondgradeyears,buthecouldhardlytakecareof
themselves.Andhecouldn’ttakeshoweruntilFridaynightathome,and
wouldsmellterribleafterrunningaroundallthetime.
IngeneralspecificchallengesregardingLBC’seducationwerelargelyareflection
ofthemissingrolesofparents.Theco-residentgrandparentshadlimitedphysical
strengthsandacademicabilitiesthatwerenecessarytohelpchildrenwiththeir
197
homeworkorachievetheirintellectualpotentials.Especiallyunderdifficult
socio-economiccircumstancesorcommunityconstraints,childrenmightbecome
morevulnerableintackingtheadverseconditionsduetothedeficiencyofhelp
andsupport.
5.3.7Psychosocialsupportfromthecommunityenvironment
Besidestheschool,thelargercommunitywasacrucialcomponentinthechild’s
physicalandsocialenvironments.Communityenvironmentinawaydetermined
theavailableandaccessiblematerialresourcestomeettheneedsoffamiliesand
children,aswellastheintangibleresourcessuchassocialcapitalthatmaybenefit
childdevelopment.Jointeffortsofcommunitymembers,especiallyunder
organizedinitiatives,mayhelprelievesomeofthedifficultiesfacingthe
left-behindfamiliesthatthemigrantparentswereunabletotackleeffectively
overdistance.
Insomehighout-migrationcommunitiesthatweredenselypopulatedwith
left-behindfamilymembers,asenseofnormalizationofbeingleft-beindappeared
tocompensatethefeelingoflossamongtheincompletefamilies.Childrenwould
easilyfindplaymateswhosharedthesimilarexperiencesintheneighborhood,
andtendedtofeellesslonelyathome.Thecommunitycohesioninthese
relativelyisolatedsocio-ecologicalsystemsinKaihuamayoffersome
psychologicalrelieftoleft-behindfamilies,evenwithoutprovidinganyconcrete
198
benefits.ThepreviouslymentionedmigrantfatherShenglidescribedsuch
situationofthecommunitywherehisleft-behindfamilylives:
Ourtownshipasawholeisaremote,underdevelopedarea–almostevery
youngadultisawayandtheirchildrenarealllefthere,somysonfeelsfine.
Butifhewereamongjustafewleftbehind,hewouldprobablyfeelterrible.
Incontrast,manyparentsorcaregiversreportedthat,itwasnoteasyforthechild
tomakefriendsintheirvillagebecause“thereareonlyaboutfourkidsofhisage
inourvillage,asthewealthierfamiliesusuallytaketheirchildrentothecity”.
Duringthefieldvisits,itwasnotuncommontocomeacrossahousestanding
alonewithonlyafewadjacenthousesinsight.Suchfamilieswereoftenpoorer
thanotherswholiveatabetterlocation,andwereunlikelytohaveanyworking
ageyoungadultathome.
Intervieweesfurtherreportedcertaincommunity-leveleffortsthatweremadeto
providecareandsupportfortheLBC,asallcommunitymemberstheirleaders
reachtheconsensusthatout-migrationleftbehindagenerationofchildrenthat
neededextrasupport.TheChildren’sClubsprogram,forexample,wasoperating
insomevillagesinKaihua.TheClubsprovidedaphysicalspaceandsome
necessaryresourcesforout-of-schoolactivities,runbylocalcommunity
volunteers,aswellasuniversitystudents(duringschoolholidayeventsonly).
Children’sClubswereoftenusingpubliclyownedspaceandfacilities,including
199
thecommunitycenterorelderly’scenter,wherechildrenandadultswouldgather
forvariousactivities,inadditiontospecificchild-focusedprogramming.This
programseemedtobefunctioningwellinthecommunitiesandwaswelcomedby
manyfamilies.Acommunityleadersaid:
Thechildrenwouldoftenspendtimeintheclassroomsinthecommunity
center,withWFrepresentativewatchingthem.Wehavesomelargerevents
likedancingaswell,onthesquareoutsidethecommunitycenter,which
childrenalsojoinin.Weeveninvitedadanceteacher;somefamiliesfromthe
neighboringvillagesalsocomeby.Wehaveabasketballcourt,somephysical
exerciseequipment,andalibrary–thekidscanplaythereaswell.
Insomebetter-developedvillages,othercommunity-basedprogramsand
initiativeswerealsoorganizedtosupporttheleft-behindfamilies.Examples
includedpartnershipswithlocalbusinessand/orgovernmentagenciesin
financialsupportandsomespecificcampaignssuchaspost-cardcharitysale,and
a“surrogatemom”projectthatrecruitedvolunteerstoforgeone-on-one
relationshipwiththeLBC.Yetthesewereonlyachievedinasmallnumberof
villages,wherevarioustypesresourcesweremadeavailable,accordingtoa
townshipWFleader:
Financialsupportisakeyaspect,forexampleaVillageCommitteewasableto
offer30thousandyuantoopenitsChildren’sClub;theyalsoapplyformore
200
fundsfromtheupper-levelgovernments.Anotheraspectissupportfrom
villageleadersbytheirconcreteactions.Someoftenplaywiththekidsand
attendallmajoreventsaboutchildren–thismakesabigdifference.Withthe
supportfromtheVillagesCommittee,wecanimplementalotoffavorable
policiesforchildren’sbenefitsthroughdifferentprograms.Ofcourseyoualso
needgoodpeopletotakecareofthekidsattheClubs–mostlytheeducated
retireesorWFrepresentatives.
Manyvillagesdidnothavetheseprerequisitesandwereunabletoprovideextra
supportfortheleft-behindfamilies,especiallythoseinthemoreremote,poorer
areasofKaihua.Someadditionalchallengesrelatedtogeographicalisolation
madecommunitysupportmoreunlikelyinthesevillages,asanothertownship
WFleaderreported:
Ourpopulationandvillagesareveryscattered–forexampleanearby
“administrativevillage”hasnine“naturalvillages”,locatedacoupleof
kilometersfromeachother,andmanydon’tevenhavecellphonereception.
Undersuchcircumstanceit’squitedifficulttoorganizeanyprogramsor
activities...unlesstheyareextremelyattractivetothefamilies,whichwecan’t
reallyaffordthough.
Communityentailscriticalcomponentsofthesocio-ecologicalsystemandkey
variablesthatdeterminedrural-to-urbanlabormigration.Althoughthe
201
community-levelresourcesmayormaynotbeabletoturnintoeffective
psychosocialsupport,theintervieweesreportedcertainpositiveprospectsfor
population-basedinterventionstocontinuetobenefitchildren.
5.4Discussion
5.4.1Summaryofkeyfindings
Thisqualitativestudyprovidesacomprehensiveanalysisoftheimpactof
migrationonchildren,byelicitingcommunitycontextsandcausesofmigration,
consequencesoftheseparation,familycopingstrategies,andsocio-ecological
conditionswithregardtothisimpact.Overall,thelengthyseparationposes
considerabledifficultiesonmanychildren’spsychosocialwelling,especially
emotionaldistress,primarilythroughdisruptedattachmentrelationships.These
effectsmaynotbeeasilyrestoredevenifmigrantparentspermanentlyreturned
home.Thepsychosocialsupport(andlackof)fromtheco-residentfamily,aswell
asfactorsregardingschoolperformanceandcommunitycohesion,alsomodifyor
intensifythechildwellbeingoutcomesundertheimpactofparents’absence.
Thesubstantiveconceptsthatemergedfromourdata,andtheconnections
betweenthem,formulatedaconceptualframework(Figure5.1)withreferenceto
thepredeterminedgroundtheoryparadigmthatwasnotedintheMethods
sectionofthischapter.Thecontextsandconditionsofpsychosocialwellbeingin
202
thisframeworkcoincidedwiththesocio-ecologicalmodelofhuman
development’semphasisonconnectionsbetweendifferentlevelsoffamilyand
socialsystems.TheABC-Xfamilystressmodelwasalsoreferredto,inrelationto
thedelineationofemotionalcopingprocessesandstrategiesthatwasinitially
triggeredbyastressevent.
Figure 5.1: Conceptual framework of key factors in the impact mechanisms of
parental migration on child psychosocial wellbeing (Adapted from Boss, 2002)
Similartomanymigrant-sendingareas,fortheKaihuafamilies,thedecisionto
migratewasinitiallymadetoimprovesocio-economicstatus.Despitethe
Event/Situation
Migratory separation
Parent-child
Attachment relationships
Left-behind
family
Income, care & support
Child wellbeing
Internal contexts
School and community
external
External contexts
Urban-rural disparities
Time
203
remittancesandpotentialsocio-economicbenefits,prolongedseparationfrom
parentsmarkedlyalteredthefamilydynamics,andcausedconsiderable
challengesinpsychosocialwellbeingamongmanychildren,primarilythrough
disruptedparent-childrelationship.Asadirectconsequenceofparent-child
separation,thedisruptionoftheconventionalattachmentrelationshipsthus
becameamainfocusofthisstudy.
Theemotionaldistressappearedtobethemostpronouncedpsychosocial
difficulty,oftenmanifestedaspersistentthoughtsoftheirmigrantparentsand
certaindepressivefeelingswhenmissingthem.Childrenwithdysfunctional
attachmentrelationshipstendedtowithholdtheexpressionofemotionsintheir
communicationandinteractionwithmigrantparents.Behavioralandsocial
functioningproblemswerelessprevalentyetsignificantlyaffectsomechildren.
Familycarearrangementsandpsychosocialsupportseemedtocompriseacrucial
factorinsafeguardingthewellbeingofLBCinabsenceofmigrantparents,in
additiontotheavailablematernalresources.Sourcesofsupportincluded
caregiversaswellaspeersandcommunitymembers.Concernsaboutthechild
accumulatedovercriticalchilddevelopmentperiodsthroughvariousinteractions
andfeedbacks,duringwhichthegrandparents‘physicalstrengthscontinuedto
decline.Correspondingstrategieswereadoptedsometimesbymigrantsadjusting
thefamilyobjectivesandreturninghomepermanently,yettheeffectsofthe
returnremainedquestionable.Children’sschoolaccessandperformance,
204
correlatedtobothparentalabsenceandchildpsychosocialwellbeing,appearedto
confoundtheeffectsofmigration.Communitycharacteristicsentailedimportant
contextualfactors,whichtosomeextentdeterminetheamountofsupport
availableoutsideofthefamilyfortheirchildren.
5.4.2Comparisonofstudycontextswithextantinternationalliterature
Therural-to-urbanmigrationinChinainvestigatedinthisstudyoccurredwithin
thecountry’sborder,incontrasttothevastmajorityofexistingstudiesconducted
inothercountries.Whilefindingsfromtheinternationalmigrationliteraturein
relationtothisstudyshouldbereferredtowithcaution,thesimilaritiesbetween
internalandinternationalmigrationseemedtoprevailoverthedifferences,in
examiningthepsychosocialwellbeingoftheLBC.
Theincentivesofmigrationforbetterincomeprospectsandhumandevelopment
opportunitieswereverysimilar,whetherornotanationalborderwascrossed.
Yettheinitialmoveininternationalmigrationofteninvolveshighercoststhan
internalmigration,whichmayhaveblockedsomepoorerhouseholdfrommaking
thedecision.Entrypoliciesofthemigrationdestinationandareusuallystricter
forforeigners,andmorerestrictionswouldbeappliedforthemtoobtainsocial
securityandwelfare,especiallyfortheirchildren(Valtolina&Colombo,2012).
ButitisnoticeablethatduetothehouseholdregistrationsysteminChinaandthe
urban-ruraldivideineducationandsocialprotectionsystems,migrantsinthe
205
citieswouldalsoencountervariousconstraintsthatpreventedthemfrom
bringingtheirchildren.
Thetiminganddurationofmigrationwascomparable;manychildreninKaihua
wereleftbehindsoonafterbirth,asweremanyLBCinothercountries.
Dependingontheborderentrypolicies,sometransnationalmigrantsmaynotbe
abletovisithomeasfrequentlyastheChinesemigrants,particularlythe
undocumentedimmigrantsintheUSduetothepossibilityofbeingdeported.But
inSoutheastAsiaorwithinEuropeanUnion,forexample,thetripsbackhome
mayalsobeunchallenging.
Meanwhile,takingamorein-depthpointofview,prospectsforachildtomoveto
anothermoredevelopedcountrymighthavedifferenteffectsthantheexpectation
ofaurbandestination.InruralChina,excellenceinschoolperformancewould
enableruralchildrentothriveinthecities,withoutthehardshipenduredbytheir
migrantparents(Murphy,2014),andthereforethemotivationtostudyhardand
evenbehavewellmaybecamemuchstrongerthantheinternationalLBC.Onthe
otherhand,forachildlookingtojointheirparentsinadevelopedcountry,the
immigrationprocessdoesnotseemtorequirethemtobewell-performed
studentsandthustheschoolperformancefactorislessrelevant.
Anotablefindingaboutthesocio-economiccontextofthisstudywasthat,
rural-to-urbanmigrationdidnotnecessarilyresultineconomicadvantageover
206
otherruralfamilies.Thismaybeadistinctfeaturecomparedtocommunitiesin
lowerincomemigrant-sendingcountries.Thesheerproportionofremittancesas
partofthenationaleconomywashugeincountrieslikethePhilippines(10%of
GDP)andMoldova(23%ofGDP)(WorldBank,2015).
InsomeoftheCHAMPSEAstudycountries,especiallythePhilippines,the
overseasincomeadvantageandstrongconnectionsbetweenmigrantsandtheir
homecommunities,contributedtothedevelopmentofsocialsupportnetworks
forLBC(Graham&Jordan,2011).Incontrast,mostrural-to-urbanmigrantswere
stillstrugglingtomakealiving,andmovingoutofthehomecommunitywiththeir
childrenwouldcertainlybetheirprioritywhenpossible,ratherthancontributing
tothenextgenerationlivingintheirhometown,wherelittlefuturecanbeseen.
SinceveryfewcivilsocietyorganizationsexistinChina,effortsonsafeguarding
LBC’swellbeingwerelargelymadebythegovernmentandvillagecommittee
leadership,withtheirlimitedlocalresources.
5.4.3Attachmentrelationshipandemotionaldistress
Whiletheimpactfromparentalmigrationinvolvescomplexmechanisms,the
phenomenonthatimmediatelyfollowsthemoveisthephysicalabsenceofone
endoftheparent-childrelationship.Hencehowfamilyrelationshipsfunction
afterthischange,especiallybetweenmigrantparentsandthechild,may
representthedirectimplicationofmigrationonthechild.Therelationship
207
dynamicsinvolvehowmigrantparentsfulfilltheabsence,bykeepingvirtually
presentandplayingacaregiverroleremotely,andhowthepsychosocial
experiencesofthechildrespondtothedisruptionofparent-childrelationship
bonds.
Itisworthnotingthatbecausetherelationshipvariableislargelyanindependent
consequenceofmigration,itislikelytorepresenttheeffectsizeofmigrationafter
adjustingforotherfactorsrelatedtotheleft-behindfamilyandcommunity
environments,suchastheavailabilityofsupportandsocioeconomicstatus.The
specificcasualmechanisms,includinghowmigrationmayinteractwithother
variablesassociatedwithchildwellbeing,willbeexploredinourqualitativestudy
inthenextchapterofthisthesis.
Whilesomeexistingqualitativestudieselicitedthenegativefeelingssuchas
lonelinessandvulnerability,orsymptomsofdepressionandanxietyinLBC
(Dreby2007;Grahametal.,2012),theaccountsoftenillustratedmoreaboutthe
psychologicalstatus,ratherthanstronglypointingtotheexperienceofparental
absenceperseorfeelingsaboutthemigrantparents.Somequalitativestudies
(Dreby2007;Grahametal.,2012;Parreñas,2001;Hoang&Yeoh,2015;Zou,
2012)foundcertaincommunicationstylesbetweenmigrantparentsandchildren
mayunderminechildpsychosocialwellbeing.Ourresultsfurtherindicatethelack
ofsupportiveemotionalconnectionsoverthedistance,evenwithrelatively
secureattachmentbond.
208
Therelationship-emotionprocesses
Inthisstudy,theexplorationintotherelationshipsaroundthechildprovided
furtherinsightsabouttheimplicitemotionalprocessesinLBC,basedon
children’saccountsabouthowandwhethertheymissorthinkoftheirmigrant
parents,aswellasthechild’scommunicationandbondingpatternswithmigrant
parentsandco-residentcaregivers.Accordingtotheattachmenttheorists,with
attachmentfigures’long-termabsence,childrenregardlessoftheiragegroupmay
developemotionaldistresssimilartoyoungchildren’sreactionstophysical
separation(Kobak&Madsen,2008,p24),duetotheshakensecurebasethatwas
supposedtoderivefromattachments.
Ourfindingsillustratethatmanychildrenmanifestvulnerableemotionsintheir
persistentthoughtsoftheirmigrantparents,includingdepressivefeelings
especiallywheytheyexperiencevulnerability.Yet,thisappearstobepartofthe
routinemechanismsintheirfamilydynamics,whichhavemixedimplicationson
thechild.Despitetheheavyexperiencesexpressedbymostchildren,somecases
suggeststrengthenedresilienceaschildrenovercomethedifficultiesduringthe
lengthyseparation.Inparticular,thosewhousedtolivewithmigrantparentsin
thecityhaveforgedstrongerattachmentbonds,anddemonstratecertain
resiliencetonegativeemotionalimpactaftertheywereseparatedagainfrom
parents.
209
Yetotherchildrenrespondtotheabsenceofparentsinawaythatshowsamore
dysfunctionalrelationship.Specifically,someofthosewhowereleft-behindvery
earlyduringinfancy,suchasZhiminandXia,tendtowithholdtheexpressionof
anyemotionsintheirinteractionsandcommunicationwithmigrantparents.
Thesecasessuggestthatearlyseparationcanbemoredamagingtothe
attachmentbonds,sincethemigrantparentswereabsentwhentheattachment
relationshipandinternalworkingmodelofthechildwereinitiallyformed,aswas
suggestedbyattachmenttheory.
Communicationoverdistance
Ontheparents’end,thecommunicationoverthephoneisthemostimportant
waytobevirtuallypresentinthechild’sdailylifeandmanagetheirrelationship
bonds.Ourresultsshowedthatwhileparents,withthehelpofcaregivers,usually
makesureoffrequentcontactswiththechildonregularbasis,somechildrenare
notwillingtotakethephonecallsatall.Hoang&Yeoh(2015)identifiedsimilar
issueinruralVietnamthatreflectedthechild’sintentionallydistancing
themselvesfromtheparents.Thepsychologicalmechanismsinwithholdingthe
connectionsappeartobecomplex,involvingambivalentattitudestowardsthe
parents’absence,andtheirambiguousrolesasaremotecaregiver.Thepressure
fromco-residentcaregiversthatpushesthechildtospeaktomigrantparentsalso
seemsunlikelytoimprovetheirrelationshipbonds.
210
Negativeemotionscanserveassignalsthatpromoteopencommunicationin
secureattachmentrelationships(Ainsworthetal.,1978,p152).Althougholder
childrenappeartorecognizeandacknowledgeparentalmigrationasaneffortto
improvefamilywellbeing,ratherthananintentionalseparation,itseemsdeeper
thoughtsfromdailylifeoraboutparent-childrelationshipswererarely
communicated,andcontentofthephonecallswasmostlydidacticismfromthe
parents.Throughtheineffectivecommunication,adultfamilymembersmaynot
beabletoperceivemeaningfulemotionsandthoughtsfromthechild.Whenthe
painfulemotionscouldnotberelievedbyresponsiveandaccessiblecare,children
afterprolongedseparationmayattempttodownplayattachmentfeelings,and
diminishtheirexpectationsofthemigrantparents(Bowlby,1973,p56).This
couldmakeitevenmoredifficulttoinitiatefunctionalcopingstrategies.Hence,
migrantparentsshouldfocusmoreontheemotionalexpressionofchildrenand
improvedistantcommunication,andmakeeffortsonprovidingmoreconcrete
responsestothedailythoughtsandeventstheirchildgoesthrough.
Asfundamentalelementsofsecureattachmentbondsseemtoremaininthe
knowledgethattheparentcanreunitewiththechildifnecessary(Kobak&
Madson,2008,p33),theoccasionalreunionsbecameacrucialopportunityto
remedyattachmentrelationship.Meanwhile,theclosenessduringthereunions
mayalsobeasignofthedefensivesuppressionofattachmentfeelingswhilebeing
apart(Kobak&Madson,2008,p30),sothepsychologicalmechanismsbehindthe
211
positiveresponsetothebriefreunionmaybecomplex.Whatismanifestedduring
thereunionsisinsufficienttoinferthatpsychosocialwellbeingstatuswillbe
favorablefollowingthepermanentreturnofmigrantparents.
5.4.4Availabilityofsupportfromfamilyandsocio-ecologicalenvironments
Therationalesofmigrationinvolvetheobjectivesonsecuringmoreeconomic
resources,forthemigrantsthemselvesandtheirleft-behindfamily(King,2012).
Besideshouseholdincomestatus,theintangibleresourcestomakethefamily
thriveandbenefitchildhooddevelopmentarealsoindispensible.Effectivecare
arrangementsandadequatesupportintheintheleft-behindfamilyandthelarger
socio-ecologicalsystemshelpmitigatethelossofphysicalproximitywithchildren
anddisruptionofattachmentbonds.Meanwhile,lackofsupportivefactorsmay
exacerbatethedisadvantagesandevenleadtocrisesinchilddevelopment.
Ourfindingsindicatethatpsychosocialsupportfromcaregiversbecomesacrucial
typeofnon-materialresourcesforchilddevelopmentintheleft-behindfamilies.
Grandparents,whousuallybecometheprimarycaregivers,makegreateffortsto
ensurechildren’soptimalfutureandcompensatethelossduetoofparental
absenceinthefamily.Quantitativestudieshaveshownthegrandparentsare
bettercaregiversthanotheralternativecaregiversinfulfillingthegapsleftby
migrantparentsindailycare(Zhaoetal.,2009;Fanetal.,2010;Jiang,2013;
Graham,2011;Senaratnaetal.,2011).Similartosomefindingsinthequalitative
212
literature(Arnold,2006),severalchildreninourstudyareabletoforgestrong
attachmentrelationshipwithgrandparentsandseemtobelivinginahealthy,
supportiveenvironment,demonstratinglittledifferencethannormalnuclear
families.
However,giventhefactthatthegrandparents’physicalconditionsandabilities
relatedtochildcarevaried,lessonscanbedrawnfromdifferenttypesandamount
ofsupportthatwereavailabletobythechild,aswellasthesituationsand
arrangementsthatwereabletoenhancethechild’sresiliencetoparentalabsence.
Compositionoftheleft-behindfamilybesidesgrandparentsindicatescertain
significanceinthisstudy.Despitelittleevidencefromtheexistingqualitative
literature,theoreticalperspectivesoffamilystudiesandsocio-ecologicalsystems
elucidatedthepeereffectsintheco-residenthousehold(Berk,2012;
Bronfenbrenner&Morris,1998).Ourresultssuggestthatpeeraccompaniment
relievesthelonelinessfrombeingleftbehindandmayimprovetheoverallchild
developmentenvironment.Inthisregard,onespecifictypeofcarearrangement
inthecontextofChina’sOneChildPolicy(anditsvariants),i.e.capable
grandparentslookingaftermultiplecousins,seemstobeapatternthatisableto
offeradequatepsychosocialresourcesforchildren,withmoreconcentrated
economicresourcesthanmulti-siblingfamiliesinwhichonlyonenuclearfamilyis
providingremittances.Itispossiblethattheinteractionsbetweencousins
involvedapeerrelationshipthatdiffersfromrelationshipbetweennuclearfamily
213
members,whichmaystrengthenthesocialfunctioningofthechild.
Anothertypeoffamilyinwhichthechildonlyliveswiththeleft-behindmother,
appearstobealessfavorableenvironment;themotherassinglecaregiver,such
asFeng,haslimitedtimeandeffortsonchildcareespeciallywhenshealsohasto
work,withafamilyincomethatislikelytobelowerthantwo-migrantfamilies.
Furthereffectsofcaregiverarrangementwillbeexaminedinthequantitative
study.
Thegrandparents’caregiverrolebecomesparticularlyimportantforchildren
livinginextremelydisadvantagedsituations,asCoe(2011)suggestedinthestudy
inGhana.InKaihua,someofthesechildrenarefacingenormousdifficulties,
especiallywhenproperpsychosocialsupportisunavailablefromgrandparentsor
alternativesources.Tomeetthehousehold’sbasicsurvivalneeds,manypoorest
parentstendtohavenochoicebuttomigrate,becauseofthelackofemploy
opportunitiesintheirhometownthatmaysignificantlyrelievetheireconomic
distress.Also,somefamiliesarebrokendownbecauseoftheparents’divorce;the
situationleavesthechildwithonlyunstablecareandlivelihoodprovision
arrangements.Whenparents’divorceleadstothechild’smoveofresidence,
disadvantagesinpeerandsocialsupportandevenschoolattendancemayfollow
inanunfamiliarenvironment.
Thevulnerablechildreninthesefamiliesareindireneedofextrahelp;andthe
214
communitymaybecometheirlastresort.Dongyang’scasesuggeststhat
neighborsmayalsoplaycaregiverrolesinextremecases.Theresultsalsoshow
connectionsbetweenLBCandcommunityenvironment,suchasthebenefitsfrom
peeraccompanimentinafavorablecommunityenvironment,andevencertain
typesofcommunitycarethroughprematuremodelsthathavebeenestablishedin
Kaihua.Theseappeartobebeneficialcomplementofthefamily-wideresources
forchilddevelopment,whereasisolated,remotecommunityenvironmentmay
addtothedifficultiesinchildwellbeing.
5.4.5Theprocessesandeffectsofreturnmigration
MigrationtheoriesreviewedinChapter2underscoredthefunctionalityof
migrantsandtheirhouseholdsasrationalactorsthataredrivenbyreturnson
economicresourceandhumancapital,againstthecostofmigration(Sjaastad,
1962).InKaihua,ourfindingssuggestthattheprimaryconsiderationwhen
parentsmadethedecisiontomigratewastocopewiththefirst-orderstressin
familyeconomicstatus.However,similaraswhatBoss(2002,p48)notedinthe
familystressmodel,althoughmigrationisacopingmechanismleadstoincome
gains,italsobringsnew,attendingstressorstothefamilymembers,especiallythe
child.
Mostofthesestressorswererelatedtotheinsufficientqualityofchildcare
grandparentswereabletoprovide.First,frailtybecauseofoldagelimitedsome
215
grandparents’physicalstrengths,mobilityandenergy,whichbecomes
considerableconstraintsespeciallyinthedailycareofyoungerchildren,orolder
childrenwhoneedextraattention.Also,“generationgaps”makesitdifficultfor
theelderlytosustaineffectivecommunicationandinteractionswithchildren.
Thesechallengesmayevenprogressivelyunderminetheircapabilityinsoothing
children’semotionaldistressfordiscipliningtheirbehavioralproblems.Asa
result,thepsychosocialdifficultiesinchildrenmaycontinuetoexacerbate,when
thenecessaryextrasupportandsupervisionfromcaregiversremainsunavailable.
Hence,itisimportanttorecognizethepossibleaugmentationovertimeofthe
negativeimpactfromparentalabsence.ExistinglongitudinalstudiesinChina
alreadysuggestthecontinuousdeclineofpsychologicalwellbeingduringearly
adolescence,whichalsoappearstobeassociatedwithlongerdurationofparental
absence(Dong&Zhang,2013b;Dong&He,2014;Luetal.,2014a;Zhaoetal.,
2015;Luetal.,2014b;Gao2010;Fanetal.,2013).Someinterviewsinthecurrent
studyindicatedsimilarworseningsituationovertime,forexample,Xiausedtobe
muchcloserwithhermigrantfatherinherearlychildhoodbutthey“grewapart”.
Oftentimesthefamilydidnottakecorrespondingactionsintimebeforethe
problemsworsen,sincemigrantparents,andevenco-residentcaregiversdidnot
detecttheearlyworryingsignsofchildwellbeingandbecomeawareofthe
situation.Inthesecases,theparent-childinteractionsviadistantcommunication
appeartobeunproductiveinachievingopen,in-depthconversationsoreffective
discipliningofbehavioralproblems.
216
Itmayalsobeextremelychallengingtoreversethenegativepsychosocialimpact
causedbyparentalmigration,evenwithfulleffortsfromparentsand
grandparents.Whenthemigrantmothereventuallydecidestoreturnhome
permanently,itisstillunclearfromourresultswhetherthismovewillbe
followedbysignificantpositivechangesinthechild.Infact,aquantitativestudy
(Adhikarietal.,2013)oftheCHAMPEAprojectfoundinThailandthatmother’s
earliermigrationhistoryhadasignificant,independentassociationwith
psychosocialdifficulties.
First,aswasnotedintheresultssection,inourstudythedecisiontoreturnitself
indicatesthatthechallengesinchildcareanddisciplininghavebeensosignificant,
thateconomicsoflabormigrationtheorynolongerappliesinthesefamilies.
Migrantparentshavetoadjustthefamilygoalsandpriorities,byshiftingthe
focusfromeconomicprospectsbackontochilddevelopment.Second,economic
statusisstillakeyfactorinensuringadequateresourcesforthechild.Becauseof
thereducedincomeafterquittingthejobinthecity,thereturnmigrantsarelikely
toseekotherincomesourcesathome,andthereforetheymaynothaveenough
timeandenergytofocusonchildcare,astheywereintendingto.Empirical
evidencealsoindicatesthatmotherswhoarewageworkersspendsignificantly
lesstimeinchildcare,basedondatafromeightChineseprovinces(Shortetal.,
2012).
217
Lastbutnotleast,thereturnmigrationmayindeedbringnewchallengesforthe
familyandchildren.Althoughmostinfantsbecomeattachedtomorethanone
familiarperson,thereisastrongtendencyforthemtopreferaprincipal
attachmentfigureforcomfortandsecurity(Bowlby,1969,p308).Followingthe
mother’sreturn,theprimarycaregiverroleofoneofthegrandparentsisboundto
bechallenged.Becauseofthedisturbanceinthecontinuityofcarefromtheinitial
attachmentfigure,attachmenttheoristssuggestthechildispronetodistressdue
totheinsecuritydemonstratedintheunstablerelationshipbonds(Bowlby,1973,
p35).Previousempiricalresearchhasalsoshownthatfrequentchangeof
caregiversleadstonegativeeffectsonLBC’swellbeing(Senaratnaetal.,2011;
Pottinger,2005).
5.4.6Limitations:studysample,age,andgender
ThesmallsampleinthisstudywasrecruitedfromarichprovinceinChinaand
hencecannotbenationallyrepresentative,consideringthemassive
underdevelopedruralareasacrossthecountry.Alsomostparentsinthisstudy
migratedwithintheprovinceandthusweremorelikelytovisithome,compared
tothosewhomigratedtoanotherprovince,comprisingapproximatelyonethird
oftotalmigrantsinChina(Qiao&Huang,2013).Despiteoureffortsinrecruiting
familieswithdiversecontextsintermsoffamilystructure,income,andmigration
timing,thesmallsampledidnotallowustocomprehensivelyinvestigateall
variablesinfamilycontextsanddistinguishtheirindividualeffectsonchild
218
wellbeing.Wewerealsounabletocomparemigrantfathersandmigrantmothers
inattachmentrelationships,partlyduetothefactthatbothmigrantparents
wouldusuallyjointhephonecommunicationwiththechildandvisithome
together.
Relatedly,althoughcertaintheresultsshowedcertainsignsthatgirlsweremore
likelytointernalizingproblemsandboysmorepronetoexternalizingbehaviors,
thedatahasnotbeenstrongenoughtomakevalidconclusionregardinggender
disparitiesofchildwellbeing.Similarly,theagevariationsinthisstudywas
insufficienttoinferdifferentiatedpsychosocialwellbeingattributesbetweenage
groups.Withamuchlargersample,thequantitativestudypresentedinthenext
chapterwillinvestigatethechilddemographiccharacteristicsinmoredetail.
5.4.7Studyimplicationsandconnectionswiththequantitativestudy
VeryfewqualitativestudiespublishedinEnglishorChinesehaveaddressedthe
wellbeingofleft-behindchildreninChina.Thisstudyprovidesinsightsfroma
comprehensiveperspectivethatincorporatesbothmigrantsandleft-behind
familymembers,throughtheexplorationofarangeofrelationships,functions
andmechanismsinrelationtofamilycareandpsychosocialsupport,withinthe
familyandcommunitysystems.Withcertaincaveats,ourfindingsmayeven
informtheresearchonfamilieslivingapartduetoreasonsotherthanmigration,
suchasincarcerationandmilitaryservice.Interventionprogramsshouldbe
219
developedtoimproveparent-childcommunicationinthefamilycopingprocess;
communitycareandsupportprogramsmayalsobenefitchildandfamily
wellbeing.Furtherresearchimplicationswillbediscussedinthelastsectionof
thefinalchapter.
Duetopracticalconstraints,thequalitativedataanalysisdidnotfinishwhenthe
quantitativedatacollectionstarted.Nevertheless,theinitialqualitativefindings
informedtheoverallframeworkandobjectivesofquantitativestudy.Buildingon
theoriginalgeneralresearchquestiononhowparentalmigrationmayaffectchild
wellbeing,qualitativeresultsshedlightsonthepotentialcausalpathways
betweenmigrationandchildoutcomes,byidentifyingtherelevanthousehold
characteristicsandotherenvironmentalfactors.
Aboveall,variousaspectsofchildpsychosocialwellbeingidentifiedinthe
qualitativeresultscontributetotheselectionofoutcomemeasuresand
questionnaireinstrumentsinthequantitativestudy,toinvestigatecritical
wellbeingdimensions,particularlyemotionalsymptoms,peerand
socio-behavioralproblems.
Althoughpreviouslyleft-behindchildrenwerenotincludedinthequalitative
studysample,thefindingsonmother’sreturnsuggestthatreturnmigrationmay
notbeabletoresolvethechallengesfacingtheirchildrenandmayevencreate
newstressors.Someresultsimpliedtheadverseimpactofparentalmigration
220
mayaccumulateandprogressivelyunderminechildwellbeing.Earlymigration
afterchildbirthandparentalabsenceduringcriticaldevelopmentperiodsalso
appearedtobealarming.Asaresult,thepsychosocialwellbeingofpreviousLBC
whoseparentsreturnedhomepermanently,orthelong-termimpactofparental
migration,becomesakeypointofinterestinthequantitativestudy,andwillbe
examinedthroughoutthenextchapter.
Accordingtothequalitativefindings,themostdirectimpactfollowingmigration
isembodiedintheparent-childrelationship,andhelpsinterprettheeffectsof
parentalmigrationthatareindependentofothercovariatesinthequantitative
study.Inotherwords,evenafteradjustingforcharacteristicsoffamilyandsocial
environmentsthatwereexploredinthequalitativestudy,theremaining
correlationbetweenmigrationandchildpsychosocialdifficulties,especially
emotionaldistress,isstillmeaningful.
Thequalitativeresultsalsocontributetotheinclusionofcovariatesinthe
quantitativestudy,whicharemoreindirectlyrelatedtobothmigrationandchild
experiences.Carearrangementsandfamilycompositionemergedasimportant
conceptstobefurtherexploredusingquantitativedata.Whoaretheco-resident
familymembers,andwhoplaystheroleofprimarycaregiver,seemstobe
importantvariablesthatmayreflecttheconsequencesofmigrationandthe
wellbeingstatusofchildren.Parents’divorce,shownasastrongundermining
factorofLBC’swellbeinginDongyang’scase,isalsotobefurtherexaminedinthe
221
quantitativestudy.
Thisstudyshowshowpsychosocialsupportfrommultiplesources,including
familyandcommunitymembers,providesessentialresourcesforthechildin
absenceofparents.Theavailabilityofsupportthusbecomesamajorfactortobe
examinedinthequantitativestudy.Importantly,whenlookingintothemost
disadvantagedfamilies,qualitativeresultsindicatechildrenwithlesssupport
fromfamilyandsocialenvironmentstendtobedisproportionatelyaffectedby
parentalmigration,eveninaprogressivemannerovertime.Thispromptsthe
investigationoninteractioneffectsbetweenavailabilityofsupportandparental
migrationinthequantitativestudy.
Duetothelackofvalidmeasuresofwealthstatusinthequalitativestudy,the
quantitativechapterwillfurtherinvestigatethehouseholdeconomicmechanisms
inrelationtomigrationandchildwellbeing,asinfamilycareresourcesandthe
largersocioeconomiccontextsoflabormigration.Thequantitativestudyisbased
onamuchlargersampleacrosstwoprovinces,sochilddemographicsincluding
ageandgenderwillbetakenintoaccountsintheanalysismodelsinthenext
chapter.
222
Chapter6 Impactofparentalmigrationonchildren’s
psychosocialwellbeingmeasures:aquantitativesurvey
6.1Introduction
Thischapterpresentsthequantitativestudyofthisthesis.Usingacross-sectional
self-reportquestionnaire,thisstudyfurtherinvestigatesthecentralresearch
questionofthisthesis:howparentalmigrationaffectschildpsychosocial
wellbeingandwhataretherolesofotherfactorsintheLBC’sfamilyandsocial
environments.
Datawascollectedinthisstudyaspartofalargerproject.Alocalprojectteam
basedinZhejiangaffiliatedwithZhejiangUniversitycollecteddatafromallstudy
sites,andthusresultsinthischapterarefromthesecondarydataanalysisthat
theauthorconducted.However,theauthorparticipatedinthedesignofsurvey
instrument,basedonthereviewofbothempiricalandtheoreticalliteratures.
Connectionsbetweenthereviewandquestionnaireinstrumentappliedinthis
studywillbeelaboratedintheMethodssection.
Thequantitativeanalysisaimedtoinvestigatethecurrentandlong-termimpact
ofparentalmigration,ontheemotional,behavioralandsocialfunctioningaspects
ofchildwellbeingthatmayreflectconsequencesofparentalabsence.Inparticular,
thestudyfocusedonfamilycarearrangementsandsourcesofpsychosocial
223
support,inadditiontomaterialresourcesforthechildren.Thesefactorswere
exploredinordertoindicatethepotentialimpactmechanismsorcausalpathways,
fromparentalmigrationtochildpsychosocialwellbeingoutcomes.
Morespecifically,weexamined:1)theimpactonchildpsychosocialwellbeingdue
tolengthyseparation(formorethan6months)frommigrantparents,including
bothcurrentandpreviousexperiencesofbeingleftbehind;2)characteristicsof
availableresourcesforoverallchilddevelopmentinabsenceofmigrantparents,
includingfamilycareprovision,psychosocialsupport,andwealthstatusthatmay
confoundthedirecteffectsofparentalmigration;3)howchildwellbeingfactors
(aswerementionedintheprevioustwopoints)wereassociatedwithspecific
dimensionsofpsychosocialwellbeing,ascomparedtotheoverallwellbeing
metric;4)whethertheeffectsofchildcareriskfactors,suchasparentaldivorce,
andlackofsocialsupport,aregreaterinleft-behindchildrenthannever
left-behindchildren.
6.2Methods
6.2.1Studypopulations
DatainthisstudywascollectedfromthethreecountiesinZhejiangandtwo
countiesinGuizhou.ThethreecountiesinZhejiangwereselectedinthepoorest
regionofZhejiang,andthetwocountiesinGuizhouwereamongthericherrural
224
areasintheprovinceneartheprovincialcapital.
Officialsattherelevantdepartmentsofcountyortownshipgovernmentswere
interviewedtoidentifycommunitieswithhighproportionsofLBC,andto
understandtheeconomicstatusofthesecommunities.Twentymigrant-sending
townshipsinwesternZhejiang,andtensuchtownshipsinsoutheasternGuizhou,
wereincludedinthisstudy.InZhejiang,tomakethesamplemorecomparableto
Guizhou,twovillageswithhigherout-migrationandlowereconomicstatusin
eachtownshipwerefurtherselectedassamplingunits.Ineachselectedtownship,
twoschoolswheremostLBCgotowereincludedinthestudy,withthe
permissionoftheschoolprincipalandlocalDepartmentofEducation.Tobe
eligibleforthisstudy,studentsshouldbeinYear4toYear9fromtheselected
schools;andinZhejiang,studentsshouldalsocomefromtheselectedvillages.
6.2.2Datacollection
Aswasstatedintheintroductionsection,quantitativedatausedinthischapter
frombothZhejiangandGuizhouprovinceswascollectedbyaprojectteaminthe
field,aspartofalargerproject.Theprojectteamcomprisedofresearchersand
graduatestudentsatZhejiangUniversityandZhejiangNormalUniversity,andthe
authorconductedsecondarydataanalysis.
Alleligiblestudentswereprovidedwithadetaileddescriptionofthestudydesign
225
andaninformedconsentform.Thosewhoagreedtoparticipatewereaskedto
completeaself-administeredquestionnaireintheirclassroom,withoutthe
presenceofanyteachersorschooladministrators.Participantswereassuredthat
theycouldwithdrawfromthestudyatanytime.Theywerealsotoldthatthere
werenorightorwronganswers,andtheiranswerswouldremainconfidential;no
oneexcepttheresearcherswouldhaveaccesstoinformationtheyprovidedinthe
questionnaire.
Trainedresearchassistantshelpedyoungerstudentsfillinthequestionnaireby
explainingthequestionsifrequested.Eachparticipantwasbrieflyinterviewed
afterthequestionnairewascompleted,toconfirmthequestionsrelatingto
parentalmigrationhistory.Correctionsweremadeaccordingtotheinterviewfor
inconsistentanswers.Allcompletedquestionnaireswerecollectedandputintoa
boxthatwasthensealedtodemonstratethatconfidentialitywasguaranteed.
6.2.3Measures
Questionnairedesign
Thesurveyinstrumentwasdevelopmentbasedonreviewsofbothempiricaland
theoreticalliteraturesconductedbytheauthor,withfurtherinputsand
modificationsfromthelargerprojectteam.Thetheoreticalperspectives
elucidatedinChapter2suggestareasoffocuswhendesigningsurveyquestions.
Argumentsaccordingtoattachmenttheoryhighlightedthefocusareasof
226
emotionalandbehavioralissueswhenexaminingchildwellbeing.Thefamily
studiestheoriespointedoutthecrucialeffectsofhouseholdcharacteristics,as
wellaslargercontextualfactorsintheleft-behindfamily’ssocio-ecologicalsystem.
Migrationtheoriesillustratedcomplexmechanismsthatdrivetheactionsrelated
tomigration,whichinvolvedimportantfactorsintheleft-behindfamilyand
communitiestojustifytherationalesofmigration(orreturnmigration).
Particularly,householdeconomicsperspectiveinformedtheinvestigationand
discussiononfamilywealthstatusinthisstudy.
AswasshowninChapter3,instrumentsandquestionnaireitemsintheempirical
studieswerecriticallyreviewed,todesignthesurveyinstrumentthataddresses
thecentralresearchquestionbymeasuringoverallandspecificdimensionsof
wellbeing,aswellasimportantcovariates.Mentalhealthscaleswereexcluded
becausethisstudydoesnotfocusonpsychopathology,butratherthe
comprehensivepsychosocialmechanismswhichinvolvebothpositiveand
negativeaspects.Emotionalandbehavioralwellbeingbecamethefocusofchild
outcomemeasures,whichmaybebetterlinkedwithparentalabsenceaccording
tothetheoreticalreview.Theapplicabilityinthenon-clinicalsettingsinrural
Chinawasalsoconsidered,andcertainlanguageinthequestionnairewas
modifiedaccordinglytoavoidmisinterpretation.
Psychosocialwellbeing
227
Theself-reportversionofStrengthsandDifficultiesQuestionnaire(SDQ)was
appliedtomeasurepsychosocialwellbeingofchildren.TheSDQ(Goodman,2007)
comprises25itemsofpsychosocialwellbeing,infivedimensionsincluding
emotionalsymptoms,(somaticsymptoms,unhappy,worries,nervousinnew
situations,andmanyfears)conductproblems(tempers,obedient,fightsorbullies,
lies,steals),hyperactivity(restless,fidgety,easilydistracted,thinksbeforeacting,
goodattention),peerproblems(solitary,hasgoodfriend,generallyliked,bullied,
betterwithadultsthanwithchildren),andpro-socialbehaviors(considerate,
helpfulifsomeonehurt,sharesreadily,kindtoyoungerchildren,often
volunteers).Eachitemwasscoredfrom0to2(nottrue,somewhattrue,and
certainlytrue).Eachdimensionwasmeasuredbythesummedscoreofthefive
itemsasasubscale,withvaluesrangingfrom0to10.Allbutthepro-social
subscalewerethengroupedtogethertogenerateatotaldifficultiesscore,ranging
from0to40,withhigherscoresindicatingmoreseveredifficulties.Alldependent
variablesfromSDQwerecodedascontinuousinouranalysis.
Table 6.1: Frequency distribution of responses in each SDQ item
Subscale Questionnaire item Frequency
(%) Emotion
I get a lot of headaches, stomach-aches or sickness Not true 53.4 Somewhat true 36.0 Certainly true 10.6
I worry a lot
Not true 16.5 Somewhat true 58.7 Certainly true 24.8
228
I am often unhappy, down-hearted or tearful
Not true 52.6 Somewhat true 34.8 Certainly true 12.6
I am nervous in new situations. I easily lose confidence
Not true 32.2 Somewhat true 45.0 Certainly true 22.8
I have many fears. I am easily scared
Not true 50.8 Somewhat true 34.2 Certainly true 15.0
Conduct I get very angry and often lose my temper
Not true 36.5 Somewhat true 49.0 Certainly true 14.5
I usually do as I am told
Not true 4.5 Somewhat true 58.4 Certainly true 37.1
I fight a lot. I can make other people do what I want
Not true 70.3 Somewhat true 25.9 Certainly true 3.8
I am often accused of lying or cheating
Not true 78.7 Somewhat true 17.4 Certainly true 3.9
I take things that are not mine from home, school or
elsewhere
Not true 85.7 Somewhat true 6.1 Certainly true 8.3
Peer relationship
I am usually on my own. I generally play alone or keep to myself
229
Not true 63.3 Somewhat true 26.5 Certainly true 10.1
I have one good friend or more
Not true 6.6 Somewhat true 19.6 Certainly true 73.8
Other people my age generally like me
Not true 12.7 Somewhat true 46.8 Certainly true 40.5
Other children or young people pick on me or bully me
Not true 62.7 Somewhat true 28.5 Certainly true 8.8
I get on better with adults than with people my own age
Not true 31.6 Somewhat true 43.7 Certainly true 24.7
Hyperactivity I am restless, I cannot stay still for long
Not true 31.5 Somewhat true 53.6 Certainly true 14.9
I am constantly fidgeting or squirming
Not true 46.4 Somewhat true 41.4 Certainly true 12.2
I am easily distracted, I find it difficult to concentrate
Not true 31.9 Somewhat true 50.4 Certainly true 17.7
I think before I do things
Not true 8.8 Somewhat true 53.2 Certainly true 38.0
I finish the work I'm doing. My attention is good
Not true 14.7
230
Somewhat true 63.0 Certainly true 22.3
Prosocial I try to be nice to other people. I care about their feelings
Not true 3.0 Somewhat true 58.4 Certainly true 38.7
I usually share with others (food, games, pens etc.)
Not true 6.2 Somewhat true 45.0 Certainly true 48.8
I am helpful if someone is hurt, upset or feeling ill
Not true 4.2 Somewhat true 43.0 Certainly true 52.8
I am kind to younger children
Not true 4.4 Somewhat true 32.3 Certainly true 63.3
I often volunteer to help others (parents, teachers,
children)
Not true 5.3 Somewhat true 57.5 Certainly true 37.3
231
Figure 6.1: Frequency distribution of SDQ total difficulties score
232
Figure 6.2. Frequency distribution of SDQ subscale scores
TheSDQhasprovenitsreliabilityandvalidityacrossdifferentculturesand
settings,andhasbeenvalidatedinChinesepopulation(Kouetal.,2007).Some
previousstudiesonLBChaveusedSDQoritssubscalesasprimaryoutcomesof
childwellbeing(Mazzucatoetal,2015;Vanoreetal.,2015;Wickramageetal.,
2015;Fanetal.,2010),whichcanprovidescorestobecomparedwithourresults.
Parentalmigrationstatus
Parentalmigrationstatuswasdeterminedaccordingtothetwoquestions“has
yourfather(andmother)takenajobawayfromyourhometownandbeenabsent
foroversixmonths?”Theoptionswere“yes,currentlyabsent”,“yes,previously
absent”,and“no,never”.Ifoneorbothparentswerecurrentlyabsent,thechild
wasdefinedasa“current-LBC”;ifnot,andifoneorbothparentswerepreviously
absent,thechildwasdefinedas“previous-LBC”;andifneitherparentwasever
away,thechildwas“never-LBC”.
Householdcharacteristics
Householdcharacteristicsincludedeconomicstatus,primarycaregiver
(grandparents,father,mother,orothers),parents’marriagestatus(marriedor
divorce),andanysiblings(yesorno).Economicstatuswasmeasuredbythe
233
numberofhouseholdappliances,amongairconditioner,refrigerator,washing
machine,television,andcomputer.Thevariablewasthencodedaspoor(zeroto
oneitem),fair(twotothreeitems),andwealthy(fourtofiveitems).Theprimary
caregiverofchildrenwasidentifiedbasedontwoconsecutivequestions,“Who
areyoucurrentlylivingwith?”and,“Amongthem(wholivewithyou),whotakes
careofyouthemost?”Caregiversweregroupedintofourcategories:
grandparents,father,mother,andothers(includingrelatives,friends,siblings,
andnocaregiver).Theanswerswerethencross-checkedwithparentalmigration
status.Ifthechildidentifiedamigrantparentatthetimeasprimarycaregiver,
thisparent,whencodingthevariable,wasreplacedbyanadultlivingwiththe
child,usinganorderofprecedence:mother,father,grandparentsandthenothers.
Childsocialsupportandschoolperformance
Socialsupportwasmeasurebyascaleofsixitemsadaptedfromthe
MultidimensionalScaleofPerceivedSocialSupport.Weexcludedthewordings
thatspecifiedfamilyorfriendsinordertoexploreifsupportthechildcanget
fromanyone.Wealsoonlyfocusedontheaspectsthatbetterapplytorural
childreninChina.Thequestionsaskedwere:whetherornotthereissomeone
wouldhelportalkwiththechild,whenthechild1)hastroubleinstudy,2)
worriesaboutproblems,3)isteasedorbullied,andwhetherthereisanyonethe
childcan4)sharehappinesswith,5)sharesorrowwith,and6)seekhelpor
guidancefromwhenthingsgowrong.Theanswer“yes”wascoded1and“no”as0.
234
Childrenwhoscored6,i.e.“yes”inallsixitemswerecategorizedashighsocial
support.Childrenwhoscored3to5werecategorizedasmedium,and0to2as
lowsocialsupport.Schoolperformancewasmeasuredbythequestion,“In
general,whatisyouracademicperformancelevelinyourclass?”Theoptions
were“top,upper-middle,middle,lower-middle,low”,codedfrom1(low)to5
(top).
Table 6.2: Overall sample characteristics
N / Mean % / SD
Parental migration status
NLB 701 19.8 PLB 907 25.6 CLB 1930 54.6
Age 12.36 2.08 Sex
Male 1674 47.45 Female 1854 52.55
Province Zhejiang 1352 38.21 Guizhou 2186 61.79
Primary caregiver Grandparent 1630 46.24
Father 435 12.34 Mother 1307 37.08 Other 153 4.34
Parents divorced No 3114 88.72 Yes 396 11.28
Any Sibling No 761 21.53 Yes 2774 78.47
School performance Very poor 290 8.22
Poor 659 18.68 Fair 1516 42.97
Good 752 21.32
235
Very good 311 8.82 Family/social support score
0 14 0.40 1 60 1.71 2 124 3.54 3 235 6.71 4 451 12.89 5 678 19.37 6 1938 55.37
Household items: Air conditioner
No 2676 75.70 Yes 859 24.30
Washing machine No 598 16.92 Yes 2937 83.08
Television No 120 3.39 Yes 3415 96.61
Refrigerator No 721 20.40 Yes 2814 79.60
Computer No 2447 69.22 Yes 1088 30.78
Motorcycle No 1700 48.09 Yes 1835 51.91
Cellphone No 162 4.58 Yes 3373 95.42
Car No 2864 81.02 Yes 671 18.98
Wealth score 6.2 2.76 Age at separation 4.52 3.34 Migrant parent(s)
Both 1602 56.45 Father 860 30.30 Mother 376 13.25
236
6.2.4Dataanalysis
Chi-squaretestandanalysesofvariancewereconductedtocomparechildand
householdcharacteristics,acrossthreegroupsofchildrenwithdifferentparental
migrationstatus.Thenanalysesofvarianceandcovariancecompareddependent
variablesacrossthesegroups.Tukey-Kramer(TK)testwasappliedinposthoc
analysesthatcomparedpsychosocialoutcomesacrossthreeparentalmigration
groups.TheTKmethodconductsthesetofallpairwisecomparisons
simultaneously,andisrelativelyconservativewhendetectingsignificant
differencesbetweengroupswithunequalsamplesizes(Jaccard&Guilamo-Ramos,
2002),asisthecaseinthisstudybetweenparentalmigrationstatusgroups.
Multiplelinearregressionmodelswerefittedtoexaminetheassociationsbetween
thepsychosocialoutcomesandcharacteristicsofchildrenandfamilies.The
baselinemodelincludedparentalmigrationstatus,themainvariableofinterest,
andchilddemographics:age,genderandprovince.Thesecondmodeladded
householdcharacteristics,includingeconomicstatus,caregiverandsibling,and
parents’marriagestatus.Inthethirdandfourth(full)model,twocovariates,social
supportandschoolperformancewereaddedsubsequently,toexploretheir
potentialmediatingeffect.Forhouseholdcovariatesthatremainedsignificant(i.e.
p≤0.05),theirinteractiontermswithmigrationstatusweretestedwithinthefull
modeltoestablishwhethertheeffectofparentalmigrationdifferedacrosslevelsof
237
eachcovariate.Theadjustedmeanswereestimatedforeachsubgroupofthe
interactiontermandcomparedpairwise.
6.3Results
6.3.1Sampledescriptivestatistics
Ofthe3,632studentswhowereeligible,3596childrenagreedtoparticipateinthe
study.NinechildrenwhofailedtocompletetheSDQsection,andeightchildren
withoneorbothparentsdeceasedwereexcluded.Thefinalstudysampleincluded
3579participants,including1930current-LBC,907previous-LBC,701never-LBC,
and41(1.1%)missingdataonparentalmigrationstatus.
238
Table 6.3: Sample characteristics by parental migration status
CLB PLB NLB Statistics P
Mean (SD)/N(%) Mean (SD)/N(%) Mean (SD)/N(%)
Age 12.3 (2.1) 12.5 (2.1) 12.3 (2.1) 4.25 0.014 Sex
1.492 0.474
Male 899 (46.7) 443 (49.1) 332 (47.5) Female 1028 (53.4) 459 (50.9) 367 (52.5)
Province
96.46 <0.001 Zhejiang 693 (35.9) 282 (31.1) 377 (53.9) Guizhou 1238 (64.1) 625 (68.9) 323 (46.1)
Primary caregiver
625.9 <0.001
Grandparent 1254 (65.2) 252 (27.9) 124 (17.8) Father 150 (7.8) 147 (16.2) 138 (19.8)
Mother 390 (20.3) 486 (53.7) 431 (61.8) Other 128 (6.7) 20 (2.2) 5 (0.7) Parents divorced
54.75 <0.001 No 1637 (85.5) 812 (90.3) 665 (95.6) Yes 278 (14.5) 87 (9.7) 31 (4.5)
Any Sibling
5.51 0.064 No 434 (22.5) 170 (18.8) 157 (22.4) Yes 1494 (77.5) 736 (81.2) 544 (77.6)
Household wealth 18.68 <0.001 Poor 185 (9.6) 62 (6.8) 53 (7.6) Fair 1181 (61.3) 564 (62.3) 291 (41.5) Wealthy 562 (29.2) 280 (30.9) 357 (50.9) Social support 18.63 0.003 Low 122 (6.4) 38 (4.3) 38 (5.5) Fair 780 (40.8) 346 (38.8) 238 (34.2) High 1010 (52.8) 507 (56.9) 420 (60.3) School performance 18.68 0.017 Very poor 155 (8.1) 77 (8.5) 58 (8.3) Poor 375 (19.5) 167 (18.5) 117 (16.7) Fair 849 (44.2) 390 (43.1) 277 (39.6) Good 400 (20.8) 188 (20.8) 164 (23.4) Very good 144 (7.5) 83 (9.2) 84 (12.0)
239
Table6.1presentsthefrequencyofanswersfromeachofthe25itemsoftheSDQ.
Figure6.1andFigure6.2demonstratethedistributionsoftotaldifficultiesscore
andscoresfromthefivesub-scalesofSDQ.Table6.2showstheoverallsample
characteristicsacrossallchildreninthestudysample.
Table6.3presentsthesocio-demographiccharacteristicsofchildrenbytheir
parentalmigrationstatus.Overallthereweremoregirlsthanboysinthestudy
sampleandthegenderdistributiondidnotdifferacrossthethreegroups.
Previous-LBCwereslightlyolder(meanage12.54,SD2.1)thantheothertwo
groups.Theproportionsofchildreninthetwoleft-behindgroupswerehigherin
GuizhouthaninZhejiang(68.9%versus31.1%).Withregardtofamily
characteristics,nearlytwothirdsofcurrent-LBCwereprimarilycaredforby
grandparents,whereasrespectiveproportionsforprevious-LBCandnever-LBC
were28%and18%.Parentsusuallytooktheprimarycaringroleinprevious-LBC
andnever-LBC,althoughtheproportionwasslightlylowerinprevious-LBCthan
innever-LBC.
Parentswhohadmigratedweremorelikelytobedivorced.Current-LBC’sparents
wereabout3times,andprevious-LBC’sparentstwotimesmorelikelythan
never-LBC’sparentstobedivorced.Approximatelyonefifthofthechildrenwere
singlechild,acrossthethreegroups.
Fiftypercentofnever-LBCreportedthattheywerefromwealthierhouseholds,
240
comparedwith30%fromthetwoleft-behindgroups.Infactcurrently-LBCwere
significantlypoorerthannever-LBCinbothZhejiangandGuizhouprovinces
(p<0.01).Besideshouseholditems,weexaminedthechild-reportrelative
economicstatusinourquestionnaire.Thesensitivityanalysiswithhousehold
itemmeasurereplacedbytherelativewealthvariableshowedconsistentresults,
furtherconfirmingthedisadvantagedeconomicstatusamongthetwoleft-behind
groups.
Ingeneral,current-LBCalsohadlowersocialsupport(p<0.001)andschool
performance(p<0.001)thannever-LBC;andprevious-LBCwerebetweenthe
levelsoftheothertwogroupsinthesetwomeasures.Mostchildrenwere
left-behindbybothparents,andfathersweremorelikelytomigratethanmothers.
About20%ofallLBCwereseparatedfrommigrantparentsbytheageofone;
morethanhalfwereleft-behindbyschoolage.
6.3.2Parentalmigrationstatusandpsychosocialdifficulties
Table6.4showsthedifferencesbetweenthethreegroupsofchildren,intermsof
thekeypsychosocialoutcomesfromtheSDQ,includingtotaldifficulties,andthe
fivesubscales.Bothcurrent-LBCandprevious-LBChadhighertotaldifficulties
meanscoresthannever-LBC.Previous-LBCappearedslightlyworseoffthan
current-LBCinallmeansubscalescoresexceptemotionalsymptoms.After
controllingforage,sex,andprovince,posthocanalysisconfirmedthatboth
previous-LBCandcurrent-LBChadsignificantlyhighertotaldifficulties,aswellas
241
worseemotionalsymptoms,peerrelationshipproblems,andhyperactivitythan
never-LBC.Nodifferenceswereidentifiedbetweentheprevious-LBCand
current-LBCafteradjustingforage,sexandprovince.
Adjustedbetween-groupdifferenceswerealsofoundineachprovince.In
Zhejiang,posthocanalysisshowedsignificantlyhigherlevelofdifficultiesin
current-LBCthannever-LBC,andinGuizhoubothleftbehindgroupshadhigher
difficultyscoresthannever-LBC.
242
Table 6.4: Psychosocial outcomes by parental migration status
Note: Adjusted with age, sex, and province.
Parental migration Unadjusted Adjusted
Never (1) Previous (2) Current (3)
Mean (SD) Mean (SD) Mean (SD) F P Post hoc F P Post hoc
Total difficulties 12.21 (4.71) 13.16 (5.06) 12.94 (5.05) 7.31 0.001 (1,2;1,3) 4.03 0.0179 (1,2;1,3)
Emotion 3.51 (2.05) 3.83 (2.22) 3.83 (2.18) 5.73 0.003 (1,2;1,3) 3.42 0.0329 (1,3)
Conduct 2.20 (1.51) 2.35 (1.61) 2.22 (1.51) 2.32 0.098
1.43 0.2393
Hyperactivity 2.70 (1.65) 2.98 (1.68) 2.89 (1.66) 2.95 0.053 (1,3) 3.01 0.0493 (1,3)
Peer 3.80 (1.88) 4.00 (1.93) 4.00 (1.99) 5.24 0.005 (1,2;1,3) 2.93 0.0536 (1,2)
Pro-social 7.17 (1.90) 7.21 (1.84) 7.19 (1.90) 0.09 0.913
0.23 0.7978
243
6.3.3Effectsofhouseholdcharacteristics
Table6.5presentsthemultipleregressionresultsofchildtotaldifficultiesscore.
Parentalmigrationstatus,bothpreviousandcurrent,wasassociatedwith
significantincreasesinpsychosocialdifficultiesinallmodels.Thestrongimpact
ofpreviousparentalmigrationwasalmostunchangedfromthereducedmodel
(B=0.81,p=0.001)tothefullmodel(B=0.75,p=0.003).Increasesofpsychosocial
difficultiesincurrent-LBC(B=0.75,p=0.001)appearedtohaveslightlyreduced
butremainedsignificant(B=0.57,p=0.017),afteradjustingforhousehold
characteristics,socialsupportandschoolperformanceinthefullmodel.The
meanincrease(regressioncoefficient)inthecurrent-LBCgroupdroppedfrom
0.85to0.67whenhouseholdcharacteristicswereadjusted(Model2),andfurther
downto0.58whensocialsupportvariablewasadded(Model3).
244
Table 6.5: Regression coefficient (SE) for total difficulties score on parental
migration status with and without adjustment for household characteristics,
social support, and school performance
Model 1 Model 2 Model 3 Model 4 B SE P B SE P B SE P B SE P
Parental migration (ref: never)
Previous 0.81 0.25 0.001 0.78 0.26 0.002 0.77 0.25 0.002 0.75 0.25 0.003 Current 0.75 0.22 0.001 0.67 0.25 0.007 0.58 0.24 0.017 0.57 0.24 0.017
Sex (ref: male)
Female -0.33 0.17 0.051 -0.34 0.17 0.051 0.00 0.17 1.000 0.29 0.17 0.086
Age 0.02 0.04 0.564 0.03 0.04 0.530 0.10 0.04 0.015 0.03 0.04 0.410
Province (ref: Zhejiang)
Guizhou 0.87 0.18 0.000 0.78 0.21 0.000 0.8 0.21 0.000 0.92 0.21 0.000 Household wealth (ref: fair)
Poor
0.64 0.31 0.042 0.60 0.31 0.052 0.46 0.30 0.132 Wealthy
-0.09 0.22 0.662 0.11 0.21 0.619 0.28 0.21 0.181
Primary caregiver (ref: Grandparent)
Father
0.19 0.29 0.496 0.22 0.28 0.424 0.23 0.28 0.398 Mother
0.23 0.21 0.284 0.25 0.21 0.233 0.33 0.21 0.107
Other
0.56 0.43 0.193 0.31 0.42 0.459 0.29 0.42 0.490 Divorced parents
1.29 0.28 0.000 1.14 0.28 0.000 1.00 0.27 0.000
Any Sibling
0.09 0.22 0.684 0.17 0.22 0.439 0.11 0.21 0.607 Social support (ref: high)
Low
3.46 0.37 0.000 3.03 0.37 0.000 Medium
1.53 0.18 0.000 1.32 0.18 0.000
School Performance (ref: fair)
Very poor
2.66 0.32 0.000 Poor
0.88 0.23 0.000
Good
-0.90 0.22 0.000 Very good
-1.39 0.30 0.000
245
ChildrenfromGuizhouweremorelikelytohavehigherpsychosocialdifficulties.
Boysappearedtobemarginallymorevulnerableinthereducedmodelwith
demographicsandhouseholdcharacteristics(B=0.67,p=0.051),butthe
associationdisappearedwhensocialsupportisadjusted.Agedidnotshowany
effectuntilsocialsupportwasaddedinModel3,whereolderchildrenseemedto
havemoredifficulties;butthesignificancewaslostafterschoolperformanceis
controlledfor.ChildreninGuizhouhadmarkedlyhigherpsychosocialdifficulties
thanthoseinZhejiang(B=0.92,p<0.001).
AsisshowninModel2,comparedwithmiddle-incomefamilies,childrenfrom
poorerfamilieshadhigherdifficultiesscore;butchildrenfromwealthier
householdsdemonstratednodifferencesfromthoseinthemediumlevelincome
families.Nevertheless,theeffectofhouseholdwealthdisappearedinthefull
model.
Parents’divorceshowedastrongassociation(B=1.00,P<0.001)withhighertotal
difficultiesafteradjustingallcovariates.Withregardtohouseholdmembers,who
wastheprimarycaregiverandwhetherthechildhadanysiblingdidnotappearto
affectpsychosocialoutcome.Additionally,wefoundthattheco-residentparentof
currentLBCwasnotnecessarilytheprimarycaregiver.Basedontheprimary
caregiverthatwasdeterminedaccordingtoourmethods,unadjustedanalysis
246
showedlowesttotaldifficultiesscoresinchildrencaredforbymothers(12.81),
andhighestinthosecaredforbyotherrelatives(13.61).
6.3.4Effectsofsocialsupportandschoolperformance
Socialsupportwasanimportantcovariatewhichhadastrongpositiveeffecton
childhoodpsychosocialwellbeing.Theeffectofcurrentseparationfrommigrant
parentsstayedalmostunchanged,whereasthecoefficientforpreviousparental
migrationreducedbyabout13%.Schoolperformancesalsoshowedhighly
negativeassociationswithtotaldifficultiesscore;strongeffectswereindicatedat
eachoftheirordinallevels.
6.3.5Differentdimensionsofpsychosocialwellbeing
Table6.6presentstheregressionresultsofSDQsubscalescoresthatshowed
significantbetween-groupdifferencesinTable6.2.Afteradjustingforall
covariates,bothcurrentandpreviousabsenceofparentswerestilllinkedto
emotionalsymptoms.Inparticular,previous-LBCseemedtobemore
disadvantagedintermsofpeerrelationship,comparedtonever-LBC.
247
Table 6.6: Regression coefficients for emotional symptoms, peer relationships,
and hyperactivies subscales on parental migration status, household
characteristics, social support, and school performance
Emotional symptoms Peer problems Hyperactivities
B P B P B P
Parental migration status
(ref: never)
Previous 0.29 0.010 0.19 0.020 0.15 0.134
Current 0.31 0.004 0.09 0.257 0.16 0.094
Sex (ref: male)
Female 0.84 <0.001 -0.12 0.031 -0.07 0.332
Age 0.01 0.513 -0.11 <0.001 0.15 <0.001
Province (ref: Zhejiang)
Guizhou 0.4 <0.001 0.19 0.005 0.13 0.118
Household wealth (ref: fair)
Poor 0.2 0.133 0.35 0.001 -0.1 0.379
Wealthy 0.12 0.205 -0.17 0.015 0.21 0.013
Primary caregiver
(ref: Grandparent)
Father -0.14 0.124 0.00 0.973 -0.08 0.306
Mother -0.09 0.472 0.04 0.642 -0.02 0.878
Other 0.11 0.577 -0.21 0.145 0.08 0.626
Divorced parents 0.34 0.004 0.17 0.061 0.25 0.018
Any Sibling 0.02 0.846 0.00 0.967 0.03 0.704
Social support (ref: high)
Low 0.88 <0.001 1.14 <0.001 0.64 <0.001
Medium 0.39 <0.001 0.41 <0.001 0.44 <0.001
School Performance (ref:
fair)
Very poor 0.55 <0.001 0.57 <0.001 0.83 <0.001
Poor 0.08 0.423 0.29 <0.001 0.32 <0.001
Good -0.18 0.064 -0.19 0.009 -0.4 <0.001
Very good -0.21 0.120 -0.14 0.175 -0.88 <0.001
248
Age,sex,andeconomicstatusshowedsignificantassociationswithdifferent
subscaleoutcomes,inbothpositiveandnegativeways.Femalechildrenwere
morelikelytohaveemotionaldifficulties,whereasboysweremoresusceptibleto
peerrelationshipproblems.Olderchildrentendedtohavelesspeerrelationship
problemsbutweremorehyperactive.Meanscoresforemotionalsymptomsand
peerproblemsubscaleswerehigherinchildrenfromGuizhouthanthosefrom
Zhejiangprovince,whiletherewasnodifferenceinhyperactivityscore.Although
householdwealthwasnegativelyassociatedwithpeerrelationshipdifficulties,
childreninricherfamiliesmaybemorelikelyhavehyperactivityproblems.
6.3.6Resultsincurrentlyleft-behindchildren
ThesubsequentanalysisinTable6.7onlyincludedthecurrent-LBCgroupto
explorefactorsrelatedtocurrentleft-behindexperiences.Incontrarytocommon
conception,childrenwhoidentifiedmotherasprimarycaregiverhadhigher
psychosocialdifficulties(meanscore13.40)thanthosecaredforbygrandparents
(meanscore12.91)inthefullregressionmodel.Meanwhile,furthersubgroup
analysis(notshowninregression)showedifagrandparentwasprimarycaregiver
andthemotherwasalsolivingathome,thecurrent-LBChadaverylowlevelof
totaldifficulties(meanscore11.45).Left-behindgirlshadworsepsychosocial
outcomesthanboys,incontrasttowhatthereducedmodelthatincludedall
childrenappearedtoindicate.Anewlyaddedvariableinthisanalysis,childage
249
whenfirstseparatedfromamigrateparent,wasnegativelyassociatedwithtotal
difficultiesscore.
Table 6.7. Regression coefficients (SE) for total difficulties score on parental
migration status, household characteristics, social support, and school
performance for currently left-behind children only
B SE P
Sex (ref: male)
Female 0.57 0.25 0.020
Age 0.07 0.06 0.219
Age at separation -0.11 0.04 0.001
Province (ref: Zhejiang)
Guizhou 0.93 0.31 0.003
Household wealth (ref: fair)
Poor 0.68 0.41 0.097
Wealthy 0.25 0.32 0.435
Primary caregiver
(ref: Grandparent)
Father 0.45 0.44 0.308
Mother 1.07 0.30 <0.001
Other 0.64 0.47 0.17
Divorced parents 1.1 0.34 0.001
Any Sibling -0.16 0.3 0.599
Social support (ref: high)
Low 3.11 0.5 <0.001
Medium 1.68 0.25 <0.001
School Performance (ref: fair)
Very poor 2.58 0.46 <0.001
Poor 0.8 0.32 0.013
Good -0.7 0.31 0.022
Very good -1.47 0.46 0.001
250
6.2.7Interactioneffects
Nosignificantinteractioneffectsofparentalmigrationandkeycovariateson
psychosocialdifficultieswerefound.Themagnitudesoftheimpactfromparental
migrationonpsychosocialdifficultieswerehighlyconsistentbetweenZhejiangand
Guizhou(Figure6.3),despitetheconsiderablechildwellbeinggapsbetweenthe
twoprovinces.However,theimpactsofparents’divorceandlowersocialsupport
appearedtobehigherinthetwoleft-behindgroupsthantheneverleft-behind
group,whencomparingadjustedtotaldifficultiesscoresbetweenthedivorcedand
undivorced,andbetweenlowandhighsupportgroups,withineachtypeof
parentalmigrationstatus.(Figures6.3;Table6.8).
Figure 6.3: Adjusted total difficulties score by parental migration status and
covariates
251
Table 6.8: Between-subgroup comparisons of adjusted mean total difficulties
score by key covariates and parental migration status
Never-LBC Previous-LBC Current-LBC
Adjusted
Mean F P
Adjusted
Mean F P
Adjusted
Mean F P
Province
5.55 0.019
5.65 0.018
13.44 <0.001
Zhejiang 13.4
14.2
13.9
Guizhou 14.3
15
14.9
Divorce
0.03 0.854
5.58 0.018
9.88 0.002
Divorced 13.5
15.4
14.9
Married 13.4
14.1
13.9
Social support
9.04 0.003
16.01 <0.001
59.11 <0.001
Low support 13.7
14.5
14.6
High Support 12.6
13.2
12.9
Note: Subgroup means were adjusted for all covariates in the full model (model 4) in
Table 6.5.
6.4Discussion
6.4.1Summaryofkeyfindingsandcontributionstotheliterature
Ourfindingsindicatethatparentalmigrationisassociatedwithindependent,
long-lastingadverseoutcomes,especiallytheemotionalandsocialdimensionsof
childpsychosocialwellbeing.Relationshipbetweennuclearfamilymembers,care
252
arrangements,andavailabilityofsupportinfamilyandsocialenvironmentsseem
tobeimportantcovariatesinthecausalmechanismsthatconnectchildwellbeing
outcomeswithparentalmigrationstatus.Schoolperformancealsoaccountsfor
someofthevariationsinchildren’spsychosocialwellbeing.Yetsocioeconomic
statusdoesnothaveanoveralleffectonpsychosocialwellbeing,andbetter-off
wealthconditionisnotnecessarilyassociatedwithmigrationstatus.
ThepsychosocialdifficultiesinLBChavebeenidentifiedinanumberofprevious
studies,aswasshowninChapter3.However,mostofthemfocusedonlimited
perspectives.First,theoutcomevariablesinmanystudiesfailedtopresenta
comprehensivepictureofvariousaspectsofpsychosocialwellbeing.Aswas
discussionintheempiricalreviewchapter,someauthorsappliedmentalhealth
diagnostictoolstoscreenforsymptoms,withoutmuchconsiderationonthetraits
mentallyhealthychildrenthatmayalsodistinguishthedisadvantagesinLBC,
especiallyinsocialinteractionsandrelationships.Asingletypeofsymptomsuch
asdepressionandanxietywasoftenthecentralresearchaim,whichdoesnot
provideenoughinsightsregardingthecommunityandsocialimplicationsand
potentialinterventions.
Also,theextantliteraturelacksafullaccountofthehouseholdandsocial
environmentsofLBC,whenexaminingchildwellbeingoutcomes.Thefamily
relationshipdynamics,inadditiontothephysicalabsenceofparents,particularly
needin-depthexplorationinordertorevealthemechanismsinchilddevelopment.
253
Theinteractioneffectsofbetweenthesecovariatesandparentalmigrationstatus
arerarelyexamined.Akeyfindingofourstudyisthathavingdivorcedparentswas
agreaterpsychosocialriskfactorforcurrentandpreviousLBCthanforneverLBC,
andsimilarly,socialsupportwasmoreprotectiveincurrentLBCandpreviousLBC
thaninneverLBC.Thispatternsuggestsfamilyrelationshipcrisesmaycause
extraordinaryvulnerabilityinLBC,andsuggeststheimportanceofpsychosocial
supportinsafeguardingthemoredisadvantagedLBC.
Nevertheless,inconsistencewithourresults,thevastmajorityofexistingstudies
conductedinChinaandgloballyhaveshownevidenceacrossmanyprovinceson
thedisadvantagesofpsychosocialwellbeinginLBCcomparedtochildrenliving
withbothparents.ThesimilarconclusionacrossdifferentprovincesinChinaalso
corroboratesourfindingthatmigrationaffectschildreninsimilarwaysacross
ZhejiangandGuizhouprovinces.Meanwhile,aswasdiscussedintheprevious
qualitativechapter,internationallabormigrationmayinvolvedifferent
mechanismsandconditions,sothelocalcontextsregardingtheeconomic
resources,careprovisionandsocialsupportshouldbeconsideredwhen
interpretingtheirfindings.
Forinstance,amongthefourCHAMPSEAProjectcountries,childrenofmigrant
fathersinIndonesiaandVietnamwerefoundtohavepooreremotionalwell-being,
comparedtochildreninnon-migranthouseholds,whereasinThailandmixed
resultswereindicated(Adhikarietal.,2013;Graham&Jordan2011).Inthe
254
Philippines,acountrywheretransnationallabormigrationhasbeenlong
established,LBCactuallydemonstratedadvantagesinpsychosocialwellbeing.This
maybeattributedtothegovernment’sactiverolesinprotectingtherightsofits
migrants,andstrongcivilsocietysupporttothoseleftbehind(Graham&Jordan
2011),asopposedtotheChinesecontexts.Thenormalizationoftransnational
familiesmayalsobeprotective,especiallyinthesehighout-migrationareas,from
whichthesamplewasdrawn.YetintheChinesesetting,lackofsocialcapitalinthe
migrant-sendingcommunitymaypartiallyexplainthedisadvantagesfacingLBC
(Wen&Lin,2012).Additionally,whileinthePhilippinesthecostsofmigration,
especiallytocountriesinanothercontinent,maypreventpoorerhouseholdsfrom
consideringtransnationalmigration,inVietnam,somecommunesofferloansto
facilitatemigrationofpoorerparents,whichmakesthesocio-economicvariability
inVietnamesestudysamplemoresimilartoruralChina.
6.4.2Independenteffectsofparentalmigrationonpsychosocialdifficulties
Inthequantitativeliterature,avarietyofpsychosocialwellbeingmetricswere
applied,butfewwereabletoreflecttheindependenteffectofthedisrupted
relationships.Inthecurrentstudy,thestrongeffectofmigrationontotal
psychosocialdifficultiesafteradjustingforallcovariatesimpliesthattheimpactof
parentalmigrationisindependentofkeycharacteristicsinthefamilyandsocial
environmentsofLBC.Aswasdiscussedinthepreviouschapterofqualitativestudy,
thisdirectimpactislikelytobeembodiedintheproblematicrelationshipswith
255
migrantparentsfollowingthemigration,andthelengthyprocessofmaintaining
theconnectionsremotelyandthepossibledetachmentfromtherelationships
bonds.ThisisinfactsupportedbyHuetal.’s(2014)study,alsobasedonSDQ
outcomemeasures,whichfoundabinaryvariable“harmoniousfamilyrelationship
(ornot)”accountedforalargeamountoftheadverseeffectduetoparental
migration.
Emotionalsymptomsandsocio-behavioralproblems
Particularly,parentalmigrationwasfoundtohavethelargestimpactonemotional
wellbeing,amongalldimensionsoftheSDQmeasures.Thiscorrespondswiththe
majorqualitativefindingthatmanyLBCexperienceemotionaldistressduetothe
absenceofparentalrolesintheattachmentrelationship.Resultsfromthis
quantitativestudysuggestthescopeofthisemotionalimpactisfarbeyond
parent-childrelationship;rather,theemotionalsymptomsmaybepresentin
variousaspectsofthechild’sdailylife.
ConductproblemsdidnotshowdifferencebetweenLBCandnon-LBC,which
suggestsparentalmigrationmaynotnecessarilycausemisbehaviors.Thisisalso
consistentwiththequalitativeresultsthatlittleevidencedirectlypointstoany
significantbehavioralchallengesinLBC,eveninthereturnmigrationcases;yetthe
concernsexpressedbysomeparentsaboutchilddiscipliningdoindicatethe
hyperactivetraitsintheirchildren.Peerrelationshipissuesseemtoreflectthe
256
inadequatesocialfunctioningintheLBC,whichmaybesubjecttotheinfluenceof
familyrelationshipdynamicsaswell.Familiescomprisedofmultiplecousins,as
describedinthequalitativechapter,mayinawaystrengthenthesocialfunctioning
withthechild’speers.
Impactfrompreviousparentalmigration
Ourresultssuggestthat,previousexperienceofprolongedseparationfrom
migrantparentshadsimilarnegativeimpactonchildwellbeing,tothecurrent
absenceofparents,afteraccountingforotherfactors.Previousstudiesthat
investigatedchildren’spreviousexperienceofbeingleft-behindshowedmixed
results.Adhikarietal.(2013)foundinThailandthatmother’searliermigration
historyhadasignificant,independentassociationwithpsychosocialdifficulties.
Huang&Li(2007)showedbothcurrentandpreviousLBChadworsemental
healthstatusthannon-LBC.However,Wuetal.,(2015)indicatedthatchildren
whohadpreviouslybeenleft-behindtendedtoexperiencefewerdepressive
symptoms.
Sincethevastmajorityofmigrationflowsarewithinthecountry,itisnot
uncommonthatmigrantparentsreturnhomeafterlivingapartfromtheirchildfor
anextendedperiod.Infactthequalitativestudydiscussedthecasesofafew
formermigrantmothersindetail,anddiscussedtheconsiderablychallenging
scenariosthatmayfollowthemother’sreturn.Althoughthereturnofthemothers
257
inthequalitativestudydoesnotchangethecurrent-LBCstatusbecausethefather
isstillaway,mostofthesechallengesstillapplywheninterpretingtheresultsof
theimpactofpreviousmigrationfromthisquantitativestudy,particularlythe
possiblehighseverityofpsychosocialsymptomsaccumulatedovertimethatledto
theparent’sreturn,andtheconsequencesfollowingthechangeofprimary
caregiver.Thepossibledecreaseoffamilyincomeisnotreflectedinthe
quantitativefindingsincethereturnmothersarestillcountedasinacurrent-LBC
family.
Asacross-sectionalstudy,wewereunabletocomparechildwellbeingbeforeand
afterparentalmigration.Yetitisprobablethatthemigrationcausedthewellbeing
problems,insteadoftheotherwayaround.Fromthemigrationtheories,itcanbe
inferredthattheparentswouldhavenotleavehomeandpursuehigherwellbeing
status,iftherewereconcernsaboutthechild’spsychologicalordevelopmentrisks
thatmaybeaggravatedbytheirdeparture.Relatedly,emotionalorbehavioral
problemsinprevious-LBCthatemergedduringparents’absence,mayhave
contributedtothedecisionofreturninghome.Thismaybearationaletoexplain
thedisadvantagesinprevious-LBC.Thedisruptionofparent-childcommunication
andattachmentrelationshipinearlychildhoodmayhavelong-termnegative
psychologicalimpacts(Kobak&Madsen,2008,p26),andthereturnofmigrant
parentsisunlikelytoresolvethecomplexconsequencesoftheirprolonged
absenceinthechild’slife.
258
Asafewpreviousstudiessuggested,theirreturnmightevencreatenewchallenges
inthechild’slifeduetothechangeofprimarycaregiver(Senaratnaetal.,2011;
Pottinger,2005).Whiletheongoingseparationandpastseparationseemtoaffect
emotionalwellbeingtoasimilarextent,thisstudyidentifiedpeerrelationship
problemsweremoreserveinprevious-LBC,butnotcurrent-LBC.Thismayimply
thenewchallengesregardingsocialfunctioningaftermigrantparents’return,
whilefurtherinvestigationsareneededtoelaboratetherisksthatarespecifically
facedbyprevious-LBC.
6.4.3Familycareandsocialsupport
Carearrangementsandrolesofprimarycaregiver
Asourqualitativestudyemphasized,theroleofprimarycaregiverisessentialin
thecarearrangementsforLBC.Inadditiontowhichparenthasmigrated,onlya
fewpreviousquantitativestudies(Vanoreetal.,2015;Suetal.,2012)examined
whoistheprimarycaregiverasanindicatoroffamilystructureandcaregiving
pattern.Theirresultssuggestthatgrandparentsprovidebettercarethanother
non-parentcaregiversforLBC.However,morein-depthanalysisofcare
arrangementsisneededtodisclosefurtherinformationaboutthefamily
environmentandpatternsofcareandsupport.
Inthisstudy,weidentifiedtheprimarycaregiveraccordingtochildren’s
self-report,astheonewholooksafterthemthemost,insteadofaskingadult
259
householdmemberstoidentifytheprimarycaregiver.Thisisbecausechildren’s
views,albeitsubjective,shouldbemoremeaningfulinreflectingtheir
relationshipswithco-residentadults,sinceitistheirexperiencesthatmatterfor
theirwellbeing.InChina,theconceptof“primarycaregiver”isnotasestablished
asitisintheWesterncontext;althoughthephrase’sChinesetranslation(ZhaoGu
Zhe;照顾者)iseasytounderstand,itisnotcommonlyused.Asaresult,the
childrenmayrelymoreontheirsubjectivefeelingswhennominatingtheZhaoGu
Zhe.
Interestingly,wefoundthatsomechildrenidentifiedamigrantparentnotliving
withthemastheirprimarycaregiver.Thisindicatesthechild’srelativelyclose
relationshipandmoreimportantly,emotionalattachmenttotheparents.Yetthe
realitiesofmigrationandthesubsequentlengthyphysicalseparation,havecreated
aninevitabledisruptiontothisinitiallystrongattachment;theclosenessbetween
parentsandthechildmayleadtoextraemotionalcost,whichwasshownas
depressivefeelingswhenmissingparentsinourqualitativefindings.Also,thefact
thatthesechildrendidnotnameaco-residentialadultasprimarycaregiver
indicatedthelackofemotionalconnectionswiththemandpossiblyinadequate
qualityofcare,whichalsohavenegativeeffectsonchildwellbeing.
AdultcaregiversofLBC,mostlygrandparentsandparents,havedifferentrolesin
supportingchilddevelopmentandinsafeguardingthepsychosocialwellbeingof
LBC.Perhapsthemoststrikingfindinginthisstudyisthattheco-residentmother
260
doesnotseemtoprovidebettercarethangrandparents.Current-LBCprimarily
caredforbytheirmotherappeartobesignificantlymorevulnerablethanthose
caredformainlybygrandparents.Inparticular,aleft-behindworkingmotherin
theruralareaislikelytohaverelativelylimitedincomeandtimeonchildcare,
whichmaybeasignificantconstraintofchildpsychosocialwellbeing.Astudy
usingdatafromtheanationallyrepresentativesurveyinChina(Chang&Dong,
2010)foundlabormigrationhassignificantlyincreasedtheleft-behindmarried
women’sworkburdensandtheirtimespentonbothfarminganddomesticwork
includingcareprovision.Meanwhile,otherempiricalevidencesuggested
co-residenceofgrandparentsdecreasesmothers'childcareinvolvement(Chenet
al.,2000),andinmanycasesgrandparentcareforthechildenablesparentsto
participateinwagework(Grayetal.,2005).Hence,multiplecaregiversincluding
grandparentsmayhelprelievetheburdensofthemotherintheleft-behindfamily
care.
Itisalsopossiblethatsomeofthemothersreturnedfrompreviousmigration,and
subsequentlytakingtheprimarycaregiverrolefromthegrandparents.Thenthis
scenariocorrespondswiththediscussionsinthequalitativechapteraboutthe
considerabledifficultiesfollowingthereturnofmigrantmother.Comparedto
otherfamiliesinwhichgrandparentsprovidegoodcareforthechild,grandparents’
inadequateabilitiesinchildcaremayalsoinfluencedthemotherdecisionofnot
migratingandkeepingtheprimarycaregiverrole.
261
Ontheotherhand,theresultssuggestthatsomegrandparentsmaybecapable
primarycaregivers,especiallywhenthemotheralsolivesintheleft-behindfamily
andissupportingchildcare,ascenariofoundinasmallgroupofcases.Itis
possiblethattheprimarycaregiverneverchangedduetomigration,althoughthe
mothermaybeareturnmigrant.Oneadvantageofthesefamiliesmaybethestable
carestructurethatleadstowell-maintainedrelationship-emotiondynamicsinthe
extendedfamily.Yetfurtherresearchisneededformorein-depthinquiriesabout
thiscarearrangement.
Psychosocialsupportinthecareenvironment
Besidesprovidingmaterialresourcestomeetthechild’sphysicaldevelopment
needs,acriticalfunctionofLBC’scareenvironmentisthecompensationofthe
psychologicallossduetoparentalabsenceinthechild’sdailylife.Inorderto
safeguardthewellbeingofLBC,caregiversmustbeabletomakedeepconnections
withthechildandprovidesupport,guidanceandsupervisionwhenmostneeded.
Inthisstudy,family/socialsupportwasmeasuredspecificallyaswhetheranyone
wouldbeusuallyavailabletoofferessentialsupporttothechildinvarious
scenarios.Thisvariableisnotonlyrelatedtothecarequalityandarrangementsin
theleft-behindfamily,butreflectstheresourcesforbetterpsychosocialwellbeing
fromthelargersocialenvironmentsthatmayinvolvefriendsatschoolorother
communitymembers.
262
Relatedly,conceptssuchasfamilyfunctioningandsocialsupporthavebeenwidely
examinedintheliteratureascrucialpropertiesofthefamilyandsocio-ecological
systems(Zhaoetal.,2014;Jordan&Graham2012;Chenetal.,2011;WangWetal.,
2014;Zhengetal.,2014;Fanetal.,2013;Liuetal.,2014).Thesequantitative
studies,includingalongitudinalstudyconductedinChina(Fanetal.,2013),
identifiedfamilyorsocialsupportasanimportantmediatoroftheimpactfrom
parentalmigrationonchildwellbeing.Resultsfromthelongitudinalstudy
confirmedthatlackofavailablesupportisastrongcorrelateofbothparental
migrationandhigherpsychosocialdifficulties,whichmoderatestheeffectof
migrationonchildren.Inotherwords,thesupportappearstobeeffectivein
relievingthedistressinchildrenduetoparentalmigration,especiallyfortheLBC
whoarecurrentlyseparatedfrommigrantparents.Previous-LBCseemtosuffer
lessfromlackofsupportthancurrent-LBC,andtheeffectsizeofmigrationon
previous-LBCremainedalmostunchangedaftersocialsupportwasaccountedfor.
Thismayimplythelonger-termrepercussionsofparentalabsenceonemotional
andbehavioralwellbeingthatisnotreflectedbythecurrentsupportlevel(Cassidy,
2008,p6).
Theinteractioneffectanalysistakesfurtherstepandindicatesthatavailabilityof
supportisprotectiveagainstpsychosocialdifficultiesespeciallyamongLBC.The
resultshighlighttheimportanceofnon-maternalresourcesinLBC’scare
environment;thechildcarearrangementsandstrategiesmustbeabletorelieve
thevulnerabilityofLBCandremedythelossofcaringrelationshipsduetothe
263
migrantparents.Also,policiesandinterventionsforimprovingsocialsupport
networksshouldbeputinplaceinthelargersocio-ecologicalenvironmentofthe
children,includingtheirschoolandneighborhood.Additionalpsychosocial
resourcesfromthesesettings,aswassuggestedinafewcasesinourqualitative
study,mayalsobenefittheLBCandespeciallyhelppreventthedeteriorationof
theirwellbeinginanunfavorablefamilyenvironment.
Parent’sdivorce
Parent’sdivorceappearstobeaspecificandimportantunderminingfactorinthe
careenvironment.Ashortcomingoftheextantquantitativeliteratureisthatmost
studiesexcludedchildrenofdivorcedparentsfromtheirstudysample;our
findingsshowthatitisimportanttotakeintoaccountthisfactor.Thenegative
impactonpsychosocialwellbeingissignificant,whichexacerbatestheeffectof
beingcurrentlyleft-behind,accordingtoourstudy.Infactourresultsimplythat
migrationmayevenbeassociatedwithhigherdivorcerate.Althoughthedivorce
mayhavetakenplacebeforemigration,itismorelikelythatmigrationescalated
relationshipproblemsbetweenthesecouples.
Parents’divorceoftencausesdamagetothefamilyfunctioningandrelationship
dynamics,andleadstoadverseeffectsonchildren’spsychologicaldevelopment
(Kobak&Madson,2008,p34).Thetwoparents’collaborationonchildcaretendsto
bemuchlessclose,leadingtoreducedabilitiesforthemtoprovidetimelyand
adequatesupportwhenthechildisinneed.Theseparationbetweentheparents
264
mayalsoaffectthecareprovisionstructure,followingthelikelydisconnection
witheachother’sparents-in-lawwhomaybethechildren’sco-residentcaregivers.
Thencoordinationbetweenparentsandcaregiversoverdistancebecomesmore
difficult,whichexacerbatestheexitingchallengesinchilddiscipliningand
providingresponsivecareduetomigration.Theallocationofremittancesmayalso
becomelesseffective,duetothelesscooperativeco-residentcaregiversandlackof
communicationaboutthechild’sneeds.
Siblingsinleft-behindfamilies
Theempiricalliteraturereviewinthisthesisshowedmixedfindingsaboutthe
effectofhavinganysiblingsonthewellbeingofLBC,suggestingmultiplefactors
maybeatplayinrelationtosiblingpresence.WhileZhaoetal.(2014)indicated
thatbeinganonlychildwasprotectiveformentalhealth,otherstudiesshowed
oppositefindingsinChina(Zhanetal.,2014)andSoutheastAsia(Senaratnaetal.
2011;Adhikarietal.,2013).Additionally,Yangetal.(2010)andWangetal.
(2011a)suggestedhavingsiblingsornotdidnotaffectchildwellbeing.
InruralChina,althoughtheoddsappearedtobeagainsttheLBCwhowerealso
singlechildwithoutsiblingcompanionship,ourstudydidnotshowsignificant
associationsbetweennumberofsiblingsandpsychosocialoutcomes.Onepossible
reasonisthatresourcesforchildcarebecomemoreconcentratedinsinglechild
families,ascomparedtomultiple-siblingfamiliesinwhicheconomicbenefitsfrom
265
migrantparentswouldbedistributedamongallchildren.Also,theco-resident
caregiverwouldbespendingtheirlimitedtimeandenergyonasmallernumberof
children.
Additionally,asourqualitativefindingsshowed,someleft-behindfamilies
comprisechildrenwhoarecousinsandcaredforbytheircommongrandparents.
Duetopracticalconstraintsandthetimingofthesurveydesign,thequantitative
questionnaireforthisstudydidnotincludeanitemaskingaboutthepresenceof
cousins.Howeveraswasdiscussedinthepreviouschapter,thisliving
arrangementpotentiallyintensifiedtheresourcesforchildcarefrommultiple
migrantfamilies,andmaybeapreferableoptionifthegrandparentsarecapableof
caringformultiplechildren.
6.4.4Familywealthstatusandchildwellbeing
Abetterhouseholdeconomicsituationwerefoundtopositivelyaffectchild
wellbeingandmaymediatetheimpactofmigrationonchildreninsomeprevious
research(Graham&Jordan,2011;Fanetal.,2010).Howeverinthesestudieslittle
wasdonetomeasuretheoverallpsychosocialsupportfromthefamilyinaddition
toeconomicresources.Ourstudyshowsthattheeffectofhouseholdwealthon
children’soverallpsychosocialoutcomeseemsonlysignificantbeforethe
availabilityofsupportwasaccountedfor.Thisindeedcorroboratesthecausual
pathwayframeworkproposedinthequalitativechapter,wherebyincomeand
266
familyandsocialsupportcomprisetheresourcesforchildwellbeing,asidefrom
therelationshipimpactduetoparentalabsence.Althoughtheeconomicstatus
priortomigrationremainsunknown,additionalmaternalresourcesfrom
remittancesmaycontributelittletothereliefofadversepsychologicalexperiences
inchildren.
Nevertheless,analysesinthisstudyindicatedifferentiatedwealtheffectsacross
dimensionsofpsychosocialwellbeinghavecomplicatedtheoveralleffectoffamily
wealthstatus,whichveryfewstudieshaveinvestigatedin.Firstofall,emotional
wellbeingseemslargelyunrelatedtowealthstatus.Althoughpreviousstudiesin
ChinasuchasHeetal.(2012)indicateemotionalsymptomsareworseinpoorer
families,theeffectmayreflecttheloweroverallpsychosocialsupportfromthe
poorerfamilies.Againourfindingsupportsthepossibilityofindependentimpact
fromparentalmigrationthrougharelationship–emotionmechanisminthefamily,
aswasdiscussedinthequalitativestudy.Thisalsoimpliesthatthefinancial
assistanceschemesinsomecommunities,albeitnecessaryfortheextremelypoor
households,doesnotcompensateforthechild’slossofanattachmentfigure.
Yetinhigher-incomefamilies,childrentendedtoencounterfewerpeer
relationshipproblems.Itispossiblethatbetter-offfamiliescanaffordmoresocial
activitiesforchildren,includingsomeafterschooltutoringclasses,aswellasmusic
andartclasses,eitheravailableatschoolorinthecountycenter.Also,aswas
suggestedinsomequalitativeinterviews,childreninwealthierfamiliesaremore
267
likelytotraveltothecitiestheparentsworkatandsocializewithmorepeople,
whichmaybenefittheirsocialfunctioning.Additionally,accordingtothe
qualitativefindings,better-offfamiliestendtoliveinmorepopulated,newly
developedneighborhoods,wherechildrencanmeetandplaywitheachother.
Somefamiliesaretoopoortobuildanewhouseandmove,eveniftheir
neighborhoodhasbecomelargelyempty.
Whileparentalmigrationdoesnotseemtoincreasebehavioralproblems,better
economicstatusmaycontributetochildren’shyperactivities.Asparents
frequentlynotedinthequalitativeinterview,inpoorareasinruralChina,tostudy
hardandbehavewellatschoolisusuallythemostpracticalwayforachildtoget
outofpoverty.Wealthierchildrenmayhavelesspressurefromlivelihood
concerns,andtheircaregiversmaybemorelenientandofferthemmorefree
choicesinlife.Hencesomechildrenappeartoactoutunderthemoreflexible
environmentandpossiblylackofcare,toanextentwithoutdramatic
delinquenciesthough.
Itshouldbeclarifiedthatthehouseholditemsincludedinourmeasureof
householdwealthdonotnecessarilyindicatetheincomeofmigrantparents.This
waspartlybecauseofthechallengesincreatingareliablemeasure,fromeither
migrantsorcaregivers,duetopracticalconstraints.Yetsuchitemsmaybetter
reflectwhetherthecareenvironmentofLBCsecuredmorematerialresources
frommigrantparents,astheincreasedincomeduetomigrationdoesnot
268
necessarilyleadtotheimprovedeconomicwellbeingorhumancapitalbackhome.
Infact,arecentstudyexaminedthecompositionsharesoffamilyexpenditure,and
foundthatremittance-receivingfamilies,ascomparedtootherruralfamilies,
spendonly2%higheronnon-durablegoods(mainlyfoodandclothing),but20%
loweronchildeducation;othermajorexpendituresinleft-behindfamiliesareon
familybusinessanddurablegoods.Suchspendingpatternsseemunlikelyto
providesubstantialbenefitsforchilddevelopment(Démurger&Wang,2015).Itis
alsopossiblethatevenifthehouseholdwealthstatusdidimprove,thechildmight
notbeabletofeelthedifference,asissuggestedbyouranalysesusingchild-report
relativewealthstatus.
6.4.5Householdeconomicsoflabormigration
Inourstudysample,economicstatuswasthehighestinnon-migrantfamiliesand
lowestamongthecurrentmigrantfamilies,inbothZhejiangandGuizhou,
accordingtobothhouseholditemscoresandrelativeeconomicstatus.Ingeneral,
familiesofLBCseemtobestillinalowereconomicclass,andevenpoorerthan
otherruralfamilies.Althoughthisisincontrarytomanystudies,Fanetal.(2010)
alsofoundthattheleft-behindfamilieshadlowereconomicstatusthanothersin
HunanProvince,China.Duetothelimitationofcross-sectionaldata,thefamily
incomelevelpriortomigrationwasunavailable.
269
Aslabormigrationisprimarilydrivenmyincreasedincomeintheory,LBCwhoare
facingmanydisadvantagesinruralChinamayreflectaparticularphaseofthe
largersocialphenomenon.Lowcostofrural-to-urbanmigrationenablesalmost
everyonetoworkinthecity,butonlythebetter-offmigrantsareabletobringtheir
childrenover.Inadditiontothemuchhigherlivingexpensesinthecity,children
arealsofacedwithconsiderablebarriertoaccessingqualityeducation.Becauseof
theruralresidence,childrenofmigrantfamiliesusuallyneedtopayaconsiderable
amountofextrafeestobeadmittedtothepublicschoolsinthecity.
Yettherehasbeenamarkedincreaseofmigrantchildrenlivinginthecities,
whereasthenumberofleft-behindchildrenremainedlargelystableinruralareas
(UNICEF,2014b).Theout-migrationflowsmayhavecontributedtoreducingthe
incomegapsinruralareas,asaresultofricherfamilies’departure,andthe
remittancesfrompoorermigrantstotheirleft-behindfamilies.Asanewroundof
urbanizationhasexpandedtosmallercity-townsclosertotheruralcommunities,
theinitial“primarylabormarket”ofsecure,well-paidjobshavealsoreachedinto
moreruraltownshipcenters(Zhanetal,2015).Meanwhile,inpartsofheurban
areas,asecondarylabormarketoflow-skill,low-wage,insecurejobsstillexistsin
themanufacturingandservicesectors,filledbymanyruralmigrants.Hence,
rural-to-urbanmigrationisnotequivalenttosocioeconomicadvantages.The
incomegainsmaycontributelittletorelievingpsychosocialdifficultiesofchildren,
especiallythosesufferingfrommultipleaspectsofdeprivationbesideseconomic
status.
270
UnlikemostotherstudiesconductedinChina,amajorstrengthofthisstudyisthe
comparisonbetweentwoChineseprovinceswithdistinctsocioeconomic
developmentlevels,consideringthehugevariabilityacrossthecountry.While
childreninGuizhoufaredmuchworsethanthoseinZhejiang,ourresults
demonstratedhighlyconsistentimpactsfromparentalmigrationonchildren
betweenthetwoprovinces,suggestingthatlocalsocio-economicdevelopment
leveldoesnotcomplicatethechallengesfacedbyLBC.
6.4.6Schoolperformance
EducationoutcomeofLBCisacommontopicoftheLBCliterature,yetthelinks
betweenschoolperformanceandpsychosocialwellbeingarenotwellestablished.
Althoughitisnotacentralissueinthisthesisasafactoroutsideofthefamily
environment,academicfunctioningintheschoolsettingreflectscertain
disadvantagesofLBCinthefamilysystem.Ourqualitativestudyindicatedthe
negativeeffectsofmigrationonchildren’sschoolperformanceaswasperceivedby
theparents,whichpossiblywasonereasonforsomeofthemtoreturnhome
permanently.Quantitativeresultsfromthisstudydemonstratethatpoorschool
performanceisstronglyassociatedwithbothparentalmigrationandpsychosocial
difficulties.Lackofparentalguidanceandsupervisionmaycausethelowerschool
performance,whichmayfurtherleadtoincreaseddistressinchildren.InChinaas
inmanysocieties,educationisthekeyvehicleforsocialmobilityandprosperity
271
(Lu,2007;Morooka&Liang,2009),particularlyforruralchildrenwhoare
generallydisadvantagedineducationalandcareeropportunities.Murphy’s(2014)
qualitativestudyinChinafurthersuggestedthathighereducationwasconsidered
bytheLBCastheonlywaytoavoidmigration;undersuchpressuresomechildren
withlowergradesstartedtogaveinandmisbehave.
Theinclusionofschoolperformanceinthisquantitativestudyisanattemptto
incorporateotherfactorsoutsideofthefamilysystem,whendelineatingthe
potentialcausalpathwaysbetweenmigrationandchildpsychosocialwellbeing.
Howeverfuturestudieswithmorerigorousandideallylongitudinaldesignare
needed,inordertoidentifythespecificeffectsandcomplexinteractionsrelatedto
factorsfromtheschoolsetting.
6.4.7Ageandgenderdisparities
Overall,agedidnotshowasignificantimpactonchildwellbeingafteraccounting
forallotherfactors.However,olderadolescentsseemtohavemorehyperactive
problems.A“Year8Phenomenon”hasbeenfoundinChinesestudents.Itrefersto
arangeofpsychologicalandbehavioralchallengesrelatedtopubertythatemerge
duringchildren’ssecondyearofmiddleschool(Year8,oneyearbeforeentering
highschool,ages13-15).Thisagegroupindeedcorrespondswithsome
adolescentsinthequalitativestudy,whosemigrantparentsexpressedconcerns
abouttheirdiscipliningissuesandschoolperformance.Whentheseolderchildren
272
areleftbehindbyparents,thepsychosocialproblemsaswellasdifficultiesin
studyandschoollifewhilepreparingforthehighschoolentryexam,becomeextra
challenging(Wang&Zhang,2004).
Olderchildrenalsoseemtohavebetterpeerrelationships,andbenefitmorefrom
socialsupportthanyoungerchildren.Ourqualitativeresultsalsoshowedthat
friendsbecameanimportantpartofLBC’scopingmechanisms.Itappearsthat
psychologicalchallengesthatonlyemergedinadolescencewerepartiallyrelieved
bypeersupport,ratherthanparentalorgrandparentalcare.Accordingtoasurvey
acrosssixprovinces,onlyinthe8th-yeargroupofleft-behindmiddleschool
studentsconsideredtheirpeersasprovidingthemostsocialsupport,evenmore
thanparents(Wang&Zhang,2004).Aswassuggestedinthequalitativechapter,
grandparentcaregiversmayhavetroubleunderstandingandcommunicating
aboutchildren’srealfeelings,partlyduetothemassivechangesinChinesesociety
andpeople’slivesoverthepastdecades.Theadolescentsmaythereforebecome
morewillingtosharethoughtsandtalkaboutconcernswiththeirpeers,inthe
absenceofparents.
Thequantitativeresultsconsiderablyexpandedthequalitativefindingsongender
disparitiesatpopulationlevel.SexratiointheChinesepopulationhasbeen
markedlyhigh,duetosonpreferenceandsex-selectiveabortionpracticesunder
thestrictfamilyplanningpolicies,overthepastfewdecades.Incontrast,ourstudy
foundthatmoregirlsthanboysareleftbehindinbothZhejiangandGuizhou,
273
whichimpliedalargenumberofboys,ratherthangirls,hadmigratedwiththe
parentstothecities.Itseemsthatwhenparentsmakethedecisionofmigrating
withtheirchildren,boysarestillmorelikelytobepreferredthangirls.
HoweverourfindingsarecontradictorywiththereportfromAll-ChinaWomen
Federation(2013)basedoncensusdatain2010,whichshowedmoreleft-behind
boysthangirlsnationally.Thismightbeduetothelowereconomicstatusinthe
studycounties.Onepossiblereasonisthatforbetterofffamilies,itwouldbeeasier
tobypasstherestrictionsfromthelocalvariationofOneChildPolicy,through
simplypayingthepenaltychargeorotherinformalapproaches.
Genderalsohasdifferenteffectsacrosswellbeingdimensions.InChina,studies
(Fanetal.,2010;Huetal.,2014)indicatedleft-behindgirlsweremorelikelyto
haveemotionalsymptoms,whereasboyshadhigherlevelsofbehavioralproblems.
Althoughnodifferenceswereshownintermsofbehavioralissues,ourresults
indicatethatgirlsaremuchmorevulnerabletoemotionaldistress,whichwasalso
demonstratedinthequalitativestudyofthisthesis.Girlsalsoseemtohavebetter
peerrelationships.InruralChina,similartomanyothercultures,girlsweremore
likelythanboystoberesponsibleforyoungersiblingsandhouseholdchoresafter
theirparentsleft(Parreñas,2005;Aguilera-Guzmanetal.,2004).Duringthe
absenceofmigrantparents,girlsmaythusbecomeamorethoughtfulandcaring
familymember,whichpotentiallyhelpsthemgetalongwithpeersoutsideofthe
familyaswell.
274
6.5Strengthsandlimitations
Thequantitativestudyhasthefollowinglimitations.First,duetopractical
constraintsinrecruitingmigrantsandlackofliteracyinsomegrandparents,we
wereunabletocollectdatafromparentsandcaregivers,thusallthedataisfrom
childself-report,withouttriangulationofresponsesfromotherfamilymembers.
Second,asacross-sectionalstudy,casualpathwayscannotbefullydelineated,
betweenthemigrationstatusandchildwellbeing,andthusthefindingsshouldbe
consideredexploratory.Third,educationallevelsofparentsandcaregivers,and
specificindicatorsofqualityofcareandrelationships,suchasthefrequencyand
qualityofcommunication,werenotassessedinthisstudy.
Despitethelimitations,ourstudystronglysuggeststhatleft-behindchildrenhave
markedlyhigherpsychosocialdifficulties,basedonasampleselectedfromoneof
therichestandoneofthepoorestChineseprovinces.Thequestionnairesurvey
achievedahighresponserate,andcollecteddetailedinformationaboutcurrent
andpreviousexperiencesofleftbehind,aswellasvariouschildwellbeing
correlatesthatallowedustoexplorethecharacteristicsofthechild’sfamilyand
socialenvironment.Inparticular,ourstudypointsoutthepotentialnegative
impactofpreviousleft-behindexperiences,whichareoftenneglectedinacademic
studies,policiesandinterventions.Wehavealsoidentifiedmajorfactorsthatare
associatedwithpsychosocialwellbeingintheseruralchildren,especiallyparental
275
divorceandavailabilityofsupport.Ourfindingsprovidecrucialinsightsregarding
theimpactofparent-childseparation,aswellasdifferenttypesofcare
arrangementsandresourcesforpsychosocialwellbeingintheabsenceofmigrant
parents,andleadthedirectionoffurtherresearchonqualityoffamilycareand
relationships.
276
Chapter7 Discussionandconclusion
7.1Overallsummaryofstudyresults
Althoughthestudysamplesinthisthesisarenotnationallyrepresentative,our
resultsfromaverywealthyprovinceandthepoorestprovinceinChinaprovided
crucialevidenceabouttheimpactmechanismsofparentalmigrationfrom
multipleperspectivesofchildwellbeing.Drawingontheconceptualframework
elaboratedinChapter5,thisfinalchapterdelineateshowourstudieselicitedthe
potentialcausalpathwaysbetweenparentalmigrationandpsychosocial
symptoms,acrossmultiplelayersofthesocio-ecologicalsystemofchild
development.
Overall,parentalmigrationtendstoresultinlong-termadverseeffectson
children’spsychosocialwellbeing.Theprimarymechanismleadingto
psychosocialdifficultiesisthroughdisruptedparent-childrelationship,
particularlycausingemotionaldistress.Thesecondarymechanismaffectschild
wellbeingthroughcarearrangementsandpsychosocialsupportintheleft-behind
familyenvironmentandcontext,includingdiscipliningfromcaregiveraswellas
socialinteractionswithpeersandcommunitymembers.Additionalfactors
includehouseholdeconomicstatuswhichprovidesresourcesforchild
developmentandjustifiesthemigrationincentive,aswellasschoolperformance
andcertaincommunitycharacteristics.Importantly,ourfindingsalsoilluminatea
timedimensionacrossthesemechanisms,indicatingapersistentaccumulation
277
processofsocio-emotionaldifficulties,especiallythroughachild’scritical
developmentalperiods.
Specifically,thischaptersummarizestherelationship-emotionmechanisms
showingtheimpactofmigration,othernon-emotionalaspectsofchildwellbeing
inrelationtoparentalmigration,aswellasthetimedimensionoftheimpactfrom
migration.Majorfactorsonthepotentialpathways,includingcarearrangements
andpsychosocialsupport,householdeconomicresources,andsocioeconomicand
communitycontexts,arealsoelaboratedintermsoftheirrespectivefunctions
andtheconnectionsbetweenthem.Ineachsubsection,keyfindingsareoutlined,
followedbymorein-depthdiscussiononthefindingtheme.Finally,the
conclusionsectionelaboratestheimplicationsofthecurrentstudies,aswellas
potentialstrategiesandsolutionstoaddressthechallengesfacedwithLBC.
7.2Keyfindingsandcontributions
7.2.1Therelationship-emotionmechanismsshowingtheimpactofmigration
Findingsoutline
Ingeneral,quantitativeandqualitativestudiesinthisthesisdemonstratedsimilar
patternsofpsychosocialdifficultiesexperiencedbyLBC.Emotionaldistresswas
themostpronouncedchildwellbeingchallengethatstuckoutfrombothfield
interviewsandtheself-reportquestionnaires.Thisisconsistentwithmanyofthe
278
extantempiricalstudiesandtheoreticalperspectivesinfamilysystemsand
attachmenttheory.However,resultsfromthecurrentstudieshighlighted
parent-childrelationshipasacruciallinkbetweenparentalmigrationand
psychosocialsymptoms,whichembodiesindependent,directconsequencesof
parentalmigrationandcrucialattributesofchildpsychosocialwellbeing.
Multiplepatternsofemotionaldistresswereillustratedinthequalitativestudy,
fromthenarrativesthatinawayportrayedtherelationshipdynamics.“Missing
mymother/fathertoomuch”wasacommonthemeinthequalitativedata,as
thesechildrenshowedstrongemotionalattachmentwithmigrantparent(s)over
distance.Thepsychologicalvulnerabilitywasnotableassomechildren’sdailylife
wasaffectedbythoughtsoftheparents,accompaniedbyfeelingsoflossand
sadness.Yetcertainotherchildrenappearedmuchmoreindifferenttoparental
absence.Forexample,they“don'tfeelliketalkingwith”themigrantparentseither
overthephoneorathome,anddidnotdemonstratepositiveemotional
connectionswiththem.Ambivalentattitudestotheparent-childrelationshipsuch
as“sometimesImissthem,sometimesnot”inawaysuggestedmorecomplex
intrapsychicconflictsinthesechildren,whichmaycauseadditionalemotional
distress.
Characteristicsoftheparent-childrelationshipwerealsoreflectedintheir
communicationpatterns.Inmanyfamiliestheparentandchild“dotalkoverthe
phone,butwithout“aproperconversation”,aswasreflectedbyamother.Most
279
Chineseparentsinourstudywerenotgoodatinitiatingemotionallysensitiveor
in-depthconversationsaboutthechild’srealfeelings,andoften“makeafuss”,
accordingtoaboy,aboutthebehavioralorschoolperformance.Itwascommon
thatthegrandparentshavetourgethechildtopickupthesephonecallsagainst
theirwill.Whilesomeoccasionalreunionswitnessedthechild’sclinginesswith
theparent,extrasadnessinthechildwasalsoreportedinsomecaseswhenthe
migrantparenthadtoleaveagain.
Inthequantitativestudy,emotionalsymptoms(includingunhappyfeelings,
worries,nervousness,fears,andsomaticsymptoms)wastheonlySDQsubscale
thatshowedsignificantincreaseinbothcurrent-andprevious-LBCgroups,
comparedtonever-LBC;andtheeffectsizeonbothleft-behindgroups,measured
asthecorrelationcoefficient,wasalsothelargestamongfivesubscales.The
associationbetweenparentalmigrationandemotionalsymptomswas
independentofallindividual,familyandenvironmentalcharacteristics,which
indirectlyimpliedthattheunmeasuredfactor,parent-childrelationship,wasa
criticalfactor.
Overalldiscussion
Asanintuitiveresponsetoparentalmigrationandtheirlengthyabsence,
emotionaldistressinchildrenstuckoutfrombothfieldinterviewsandthe
self-reportquestionnaires.Ourfindingsareconsistentwiththeoretical
280
perspectivessuchasthefamilystressconceptualframework(Boss,2002,p40)
andtheunderpinningsofattachmenttheory(Bowlby,1969).Thekeychallengeof
parentalmigrationistheprolongedseparationfromtheirchildren,ofwhichthe
immediaterepercussionwouldbedisruptedparent-childrelationship.
Attachmenttheoryemphasizestherolesofsecureattachmentfromthebeginning
ofachild’slifetoadolescence.Accordingtoourqualitativefindings,theemotional
distressinchildrenwasmainlyderivedfromtheinsecureattachmentwith
migrantparents,whichisalsocharacterizedbydysfunctionalcommunication
overthedistance.
Careandsupportfrommigrantparentswasmaintainedmainlythroughdistant
communicationoverthephoneorinternet,sincefrequentphysicalreunionswere
impractical.Yetourfindingssuggestthatthephoneconversationsorinternet
chatsoftenturnouttobedidacticdiscipliningfromtheparents,ratherthan
heart-to-heartcommunicationaboutrealfeelingsandthoughtsfromthechild.As
aresult,itisunlikelyforthechildtoseekemotionalandsocialsupportthrough
distantcommunication,especiallyconsideringsomechildrenneverexperienceda
closerelationshipwithmigrantparentsevenpriortotheirdeparture.
Theextrapsychosocialrisksduetoearlyseparationfromparents,asiselicited
fromexistingliterature(Liuetal.,2009;Fanetal.,2010)andourqualitativeand
quantitativefindings,needparticularattention.Eventhoughrationalesbasedon
attachmenttheorymaynotperfectlyexplainLBC’sexperiencesofseparation,
281
theyareparticularlyrelevantforthe“earlyleft-behindchildren”.Lackofasecure
baseevenduringinfancymaypartlyexplaintheextradisadvantageinthese
children.Theinsecurityisrelatedtotheinadequatecareandsupportinthe
left-behindfamily,andperhapsthetransferofprimarycaringrolesto
grandparentsthatdisruptedtheoriginalparent-childrelationship.
7.2.2Non-emotionalaspectsofchildwellbeing
Findingsoutline
Incomparisontoemotionalsymptoms,behavioralproblemsorexternalizing
disordersappeartobelessofanissueinLBC.Accordingtothequantitative
surveyresults,onlycurrent-LBCappeartobeworseoffinhyperactivebehavioral
issues(restless,fidgety,easilydistracted,don’tthinkbeforeacting,lackof
attention),butneithercurrent-orprevious-LBChadshowedanydifferencethan
non-LBCinthemoreseriousconductproblemsordelinquencies(tempers,
obedient,fightsorbullies,lies,steals).Qualitativeinterviewsalsoelicitedsimilar
scenarios,wheresomeLBCseemedtobemischievous,suchasdisobedienceof
caregiverorspendingtoomuchtimeplayingoutside,butdidnotmisbehaveina
waythatcausedserioustroubles.
SocialfunctioningdisadvantagesinLBCwerealsosuggested,especiallyinthe
quantitativeresults.Theadjustedanalysisshowedincreasedpeer-relationship
problems,particularlyinprevious-LBCcomparedtonever-LBC,whilethe
282
increaseincurrent-LBCbecamenon-significantafteraddingcontrolvariables.
Thequalitativestudypresentedsomeindirectevidencerelatedtosocial
functioning,includingchildren’slonelinessandunwillingnesstosocializewith
peers.Emotionaldistresssuchasdepressivefeelingsalsoseemedtosuppress
dailysocialbehaviorsinsomechildren.Theisolatedsocialenvironmentinasmall,
remotecommunityalsonegativelyaffectedsocialinteractionsaccordingto
severalinterviewees.
Overalldiscussion
Unlikeemotionalsymptoms,behavioralproblemsarelesslikelytobethedirect
consequenceofparentalmigration.Particularlyforyoungerchildren,theparents
wouldprobablyhavemadesolidcarearrangements,inordertomakesurethat
theirchildrenaresupervisedanddisciplinedbyadultcaregivers,mostly
grandparents.Hence,theexternalizingbehavioralproblemsareunlikelyto
emergedrasticallysoonaftertheirparentsleft.AtschoolsinruralChina,
children’sbehaviorsarealsousuallycloselymonitoredanddisciplinedbythe
teachers,especiallywhenbehavioralissueswerelinkedwithadeclineofthe
student’sschoolperformance.
InmanypartsofruralChina,traditionalChinesevaluesofobedience,piety,and
educationforsuccessstillprevail.Suchvaluesystemsandsocio-culturalcontexts
mayhelpkeepchildrenandadolescentsinschoolandfollowtheguidanceand
283
supervisionofauthoritativefigures,andpreventthemfromactingout.Yetatrend
ofgivingupschoolandjoiningthemigrantforcehasbeenobservedinruralLBC
(Murphy,2014;Démurger&Wang,2015),andrisingjuvenilecrimeratesamong
thesemigrantchildrenhavebecomeanemergingconcern(Robsonetal.,2008).
Furtherimprovementinfamilyandcommunitycareshouldtakeintoaccountthe
functionofsocialcontexts,inordertoaddressthechallengesfacedwithLBC.
7.2.3Atimedimensionoftheimpactfrommigration
Findingsoutline
Anotherfocusofthisthesisisthedelineationoflong-termimpactofparental
migration,throughmultipleperspectivesincluding:progressivedeclineofchild
psychosocialwellbeingovertime;effectsofparentalabsenceatcritical
developmentperiods;challengesforchildrenafterparents’permanentreturn;
overallandspecificimpactsofparentalmigrationonprevious-LBC.
Certainqualitativefindingsindicatedthechildandmigrantparentdyad“grew
apart”afteryearsofseparation.Insomecasesthebehavioralproblemsshoweda
suddenincrease“inthepastfewmonths”duringearlyadolescence.Thisisrelated
tothefindingsthatparentalabsenceduringcriticaldevelopmentperiods,
particularlyinfancyandpuberty,seemedtobedetrimentaltosomechildren’s
wellbeing.Emergingandaccumulatingconcernsaboutthecareofsomeextra
284
vulnerablechildrenpromptedsomemigrantmotherstoreturnhome
permanently.However,thesereturnparentsfacedchallengesinjoiningthecare
environmentagainwhileearningsufficientincomeforthefamily.
Inourquantitativestudysample,uptoaquarterofallruralchildrenhadpast
experiencesofbeingleftbehind.Multipleregressionresultsindicatedhighly
significantimpactofparentalmigrationonthepsychosocialwellbeingof
previous-LBC.Besidesemotionaldistress,previous-LBCalsohadhigherlevelsof
peerrelationshipproblems,whichappearedtobeslightlyworsethan
current-LBC.WhileLBC’swellbeingseemedtobesignificantlyunderminedby
lackofsupport,thenegativeeffectswerelargerontheprevious-LBCgroupthan
current-LBC.
Overalldiscussion
Aphenomenonthatiscloselyrelevanttoourstudiesisreturnmigrationfromthe
citiesbacktoruralhometowns.Findingswithregardtotheimpactofreturn
migrationonchildrenareamongthemostcrucialcontributionsofthisthesis,
sincethelong-termimpactofpreviousparent-childseparationduetomigration
hasnotbeenverywellstudiedglobally,andchildrenofreturnedparent(s)have
beenlargelyignoredbythevastmajorityofstudiesonLBC,exceptafewones
(Wangetal.,2015a;Wuetal.,2015;Adhikarietal.,2013).Inadditiontochanges
infamilygoalsandstructures,returnmigrationshouldalsobeviewedwithin
285
largersocialandcommunitycontexts.
Returnmigrationreflectschangesinfamilygoals,therationalesbehindthe
migrationdecision,orontheperceivedeffectsofparent-childseparationonchild
wellbeing.Giventheprofoundimpactonfamilystructureandfunction,the
migrationdecisionisusuallymadewithintheambitofahousehold(King,2012).
Thedecisiontomigrateorreturnwasmadeinconsiderationofsocioeconomic
environmentsandfamilyandindividualwellbeing.Howeverthishousehold
decision-makingprocess,especiallyregardingreturnmigration,hasrarelybeen
examined.
Ourqualitativestudyexploredhowfamiliesevaluatedthecostsandbenefitsof
migration,focusingonthereturnmigrationcaseswhichprovidedcrucialinsights
intothedecision-makingprocess.Asthefindingsillustrated,tosimplyinterpret
migrationasincome-drivenwouldbemisleadingwheninvestigatingitsimpacton
children.EspeciallyinChina’sfamilistculture,migrationisoftenconsideredasa
moveforthebenefitoftheentirefamily,includingchildren(Esping-Andersen,
2000).Parentsmayhaverecognizedtheconsequencesofprolongedseparation
fromtheirchildren,yettheinitialmigrationdecisionwasbasedonthe
contemplationthatthegainsfromworkinginthecitieswouldoutweightheloss
fortheirchildren’swellbeing,especiallyinlongerterm.However,oncethe
situationhaschanged,especiallyasthechildgrowsup,thisjustificationmayno
longholdvalid,whichleadstothedecisionofreturninghome.
286
Meanwhile,ourquestionnairesurveyresultssuggestthatpreviousparental
migrationhassimilarpsychosocialimpactascurrentmigrationonchildwellbeing.
Relatedly,consistentwithLiuetal.(2008)hasfound,ourresultsshowedthat
childrenwhowereseparatedfromparentsatayoungerageweremoreproneto
psychosocialdifficulties.Futurestudiesmayincludeotherin-depthmeasures
regardingthecyclicpatternofmigration,suchasdurationofseparationand
reunion,frequencyofhomevisits.Thesemayturnouttobecomeimportant
factorsthatinfluencethefamilystructureandfunction,andrelationship
dynamicsbetweenmigrationparentsandtheirchildren.
7.2.4CarearrangementsandpsychosocialsupportFindingsoutline
Findingsregardingfamilycarestructure,suchasthechallengesofthemotheras
singlecaregiverandpotentialbenefitsofmulti-cousinfamilies,markedsomekey
contributionsofthisthesis.Thesearrangementsoffamilycare,aswellasthe
communityenvironment,providepsychosocialsupportforLBC,acrucialfactor
onthemechanismsthatsafeguardthewellbeingofLBCfromadverseeffectsof
parentalmigration.
Oneparticularlysurprisingfindinginthisthesiswasthat,LBCwhowere
primarilycaredforbythemothertendedtohaveworsepsychosocialwellbeing,
287
comparedtochildrencaredforbygrandparents,accordingtothequantitative
subgroupanalysis.Qualitativefindingssuggesteddifficultiesforaworking
motherinlookingafterherchild,whosometimeshadto“invitesomekidstocome
overandhangoutwithhim(theson)”.Othersneededto“figureoutaboutmoney”
becauseofthereducedincome.Theoftenseriouschilddiscipliningproblems,
whichpromptedreturnmigration,furtherexacerbatedthechallengesforthe
mothertotackle,forexampleoneofthemwasworriedthathersonhad“became
veryrebellious”.
Childrenofcapablegrandparentsreportedthey“takecareofmeverywell,
basicallythesameasmyparents”,yetsomegrandparentsexpresseddoubtsabout
theirphysicalstrengthsforchildcare.Parentsareconcernedthegrandparents
“thinkdifferentlyaboutlotsofthings”.Withthis“generationgap”,grandparents
tendedtobeeither“notstrictatall”ortoostrict,whilesomechildrenwereoften
notcooperative.Childreninaparticulartypeofhouseholdstructure,where
capablegrandparentscaredforchildrenwhoarecousins,seemedtoenjoyit
especiallythepeeraccompaniment.
Inourquantitativestudy,thenovelanalysisofinteractioneffectsidentified
parent’sdivorceandlackofsupportaskeynegativecorrelatesofallaspectsof
childpsychosocialwellbeing,andbothcorrelatesshowedexacerbatingeffects
particularlyamongtheLBCgroupsasopposedtonon-LBC.Onecaseinthe
qualitativestudyillustratedachildofdivorcedparentslivingwithafrail
288
grandmothersufferingfromhungerandneglect.Whilethequantitativemeasure
didnotidentifythesourceofsocialsupport,thequalitativeresultsshowed
neighborsandcommunityactivitiesofferedcrucialcontributionstotheLBCin
need.
Overalldiscussion
Carearrangementsandpsychosocialsupportinvolvevariousfamily
characteristicsaswellasthelargersocialenvironmentofthechild.Theprimary
caregiver,usuallyeitheraco-residentmotherorgrandmother,apparentlyplaysa
significantrole.Somestudiesshowedthatchildrenprimarycaredforbyaparent
farebetterthanthosecaredforbygrandparentsorotherrelatives(Gaoetal.,
2010;Suetal.,2012).Ourqualitativefindingsindicatedvariouslevelsof
capabilitiesamonggrandparentsinprovidingadequatecareandsupervision.The
resultsfromourquantitativesurveyshowednooveralldifferencesbetween
caregivers,althoughamongLBC,motherasprimarycaregiverseemedtohave
morenegativeinfluencecomparedwithgrandparents.
Theco-residentmother,evenifnotidentifiedasprimarycaregiver,mayprovide
essentialsupporttothegrandparentwhotakescareofthechildthemost.Onthe
otherhand,whenthemothertakestheprimarycaringrole,itisquestionablethat
whetherthegrandparentsareabletoprovideadequatecareforthechild.Unlike
somestay-at-homeparentsintheruralareas,thereturnmigrants,whowereable
tofindajobinacity,arelikelytobestillworkingfulltimeinornearbytheirrural
289
hometownaftertheirreturn;theseleft-behindmothers’non-childcareburdensin
thehouseholdmaystillbehigh,whereasthequalityofcareandamountoftime
theyspendwithchildrenbecomelimited(Chang&Dong,2010).Itislikelythat
noneofgrandparentsweresupportiveenough,whowouldotherwisehavebeen
identifiedbythechildastheprimarycaregiver.Ifthemotherusedtobeamigrant,
herreturnmayresultinthechangeofprimarycaregiver,whichwasfoundto
haveadverseimpactonthechildpsychosocialwellbeing(Senaratnaetal.,2011;
Mazzucatoetal.,2015).Despiteherpresenceathome,newdisturbanceinfamily
relationshipsandunclearcaringrolesmayleadtochildren’sambivalentfeelings
andlastingemotionaldifficultieswithparents,withoutfullyresolvingtheloss
fromseparation.
Childrenalsohavemultiplesourcesoffamilyandsocialsupport.Siblingsand
cousinsplaycrucialrolesinchilddevelopment(Brody,2004).Although
quantitativestudydidnotshowsignificanteffectsfromhavinganysiblings,the
qualitativeinterviewselicitedthatchildrenbenefitedfromlivingwithsiblingsor
cousins,aswellassocialactivitieswithfriendsintheneighborhoodoratschool.
Infactitwasnotuncommonforachildtolivewithcousins,underthecareof
grandparents;andchildrenoftendevelopcloserelationshipswithneighborsat
theirsimilarageinthecommunity.Inthesecases,peersupportmaybeableto
compensatethelossduetoparentalabsence,andbecomeanimportantsourceof
socialsupport,accordingtoourquantitativeresults.Sourcesofsupportfrom
community,school,andpeersprovideafavorablesocialenvironmentoutsideof
290
thefamily,whichisalsocrucialinsafeguardingchildwellbeingespeciallyforLBC.
Ourquestionnairesurveywasnotrestrictedtosupportfromwithinthefamily
andincludedvariousscenariosinwhichchildrenareinneed.Aparticularnotable
findingwastheexacerbatingeffectofsocialsupportonLBC’spsychologicalrisks,
whichhighlightstheneedsofsupportivefamilyfunctionandsocialnetworkfor
thechildren.
Thecareenvironmentalsoentailsrelationshipsbetweenthecaregivers,including
migrantparents,co-residentparentsandgrandparentsofthechild.Both
quantitativeandqualitativefindingsshowedthatparents’divorceconsiderably
exacerbatedthestressinthecareenvironment,anddiminishedchildren’s
abilitiesinseekinghelpandemotionalsupportfromthefamily.AsChina’scrude
divorceratehasdoubledduringthepastdecade(Davis,2014),withmigrationas
apossiblecontributor,theconsequencesofparentaldivorceinLBCare
particularlyconcerning. Migrationanddivorcealtogethermakeinteractionsand
coordinationextremelydifficultinthenuclearfamily,andresultinlargely
dysfunctionaldynamicsforeffectivechildcare.
7.2.5Householdeconomicresources
Findingsoutline
Ourfindingsprovideddeepinsightsintotheeconomicfactorsinmigrantfamilies
291
andofferedmorespecificresultsthanmostexistingresearch.Unlikefamilycare
andsupport,householdeconomicresourcesdidnotseemtobesignificantly
associatedwithemotionalandoverallpsychosocialwellbeinginchildren.
However,ourquantitativestudyshowedthatwealthstatusmayaffectspecific
aspectsofpsychosocialdifficulties.Inaddition,thequalitativestudyindicated
certainsignsofcorrelationbetweenthatsocialfunctioningandeconomicstatus.
Theeffectofhouseholdwealthonchildren’soverallpsychosocialoutcome
becamenon-significantaftertheavailabilityofsupportwasaccountedfor.
Althoughemotionalwellbeingseemedlargelyunrelatedtowealthstatus,lower
economicstatusledtomorepeerproblems,butlesshyperactivesymptoms.In
thequalitativestudy,achildfromabetter-offfamilywhohas“agoodpersonality
andtalkssweetlywithpeople”wassaidtohavebenefitedfromhisearlier
experiencelivinginShanghaiwithparents.Accordingtoanotherchild,suffering
frompovertywaspartlythereasonwhyherpeers“don’tseemtounderstandwhat
I’vebeenthrough”.
Overalldiscussion
Despiteoftherelativelylowfamilyincomeinthestudyareasascomparedtothe
restofChina,economicstatusmaynotbeakeyfactorofchildren’s
socioemotionaldistressinthisstudysample,regardlessoftheirparental
migrationstatus.YetamongtheLBC,lowerfamilyincomeseemstoberelatedto
292
certaininternalizingproblemsintheirsocialinteractions,whichmayincidentally
beprotectiveagainsthyperactivities.
Ourfindingsimplythateconomicdisadvantagesareoftenaccompaniedwiththe
lackofothernon-materialresourcesforchilddevelopment,butitisthe
non-economicfactorsthatdeterminedchildren’psychosocialwellbeing.However,
thecurrentliteraturefailedtoelicitthemultipleburdensfacedwiththeLBCin
impoverishedmigrantfamilies,whoareoftenhighlyvulnerableboth
economicallyandpsychologically.Thepovertyviciouscycleofteninvolves
multipledimensionsofdeprivation;parentalabsenceundersuchcircumstances
wouldsignificantlyhinderchilddevelopment.Asbothourqualitativeand
quantitativeresultsindicated,thesepoorestLBCareinpressingneedsofboth
economicandpsychosocialsupport.Henceinadditiontotheexistingfinancialaid
programsforchildreninpoverty,inhighout-migrationareasmoreinitiativesto
addressthepsychologicalaspectsofwellbeingshouldbedeveloped.
7.2.6Socioeconomicandcommunitycontexts
Findingsoutline
Qualitativefindingsinthisthesiselicitedthehouseholdsocioeconomicrationales
behindthemigrationdecision.Ourquantitativeresultsfromthese
migrant-sendingareasindicateddisadvantagedeconomicstatusinLBC’sfamilies
293
intheircommunity.Giventheavailabilityofcertainresources,thecommunity
organizationsandmembersinsolidaritywiththesefamiliesmayhelprelieve
someofthedifficultiesfacingtheLBC.Whenthesacrificefromtheparent-child
separationseemedtooutweighthesocioeconomicgains,rationalesoftheinitial
migrationdecisionwereoverturnedandsomemothersdecidedtopermanently
returnhome.
To“earnmoremoney”inthecitywhiletheywerestillatarelativelycompetitive
ageinthelabormarket,wasauniversalincentiveformigrationamongtheyoung
parents.Theothersideoftheissuewasintheruralvillage,“therearenojobs
exceptfarming”forthosewhohadfewskillsorconnections.Indeednon-migrant
familieshadmorewealththanpreviousmigrantfamilies,andcurrentmigrant
familieswerethepoorest,inbothZhejiangandGuizhouprovinces.Meanwhile
apparentlythemigrationinawaywas“justforthekids”tohaveliveabetterlife.
Theprospectforthechildrentogotoschoolinthecityand“secureabetterfuture”
wasalsoanincentiveofthemigrantfamily.
Howeverchildren’scurrentsituationliesintheirhomecommunity.When
“everybodyhasgonetothecities”yetthecommunitystayswell-connected,those
childrenleftbehindmayactually“feelfine”becauseofthenormalizationof
migration,asopposedtothecontextofalightlypopulatedremotevillage.
CommunityactivitiessuchastheChildren’sClubprogramalsohelpedLBCreceive
extracareandsupport.Thesecommunity-leveleffortsweredependentonthe
294
sufficientfinancialsupportandstrongvillagecommitteeleadership.
Overalldiscussion
Likeinmanyotherdevelopingcountries,migrationinChinahasbeen
predominantlydrivenbyeconomicprospectsandhumandevelopment
opportunitiesinbetter-developedareas.UnderChina’srapideconomicexpansion
anddrasticsocialchange,theinitialmotivationformigrationisalsoreflectedin
thecommunitycontexts.Whilealargeliteraturepinpointedtheeconomic
benefitsfortheleft-behindfamilymembers(McKenzie&Sasin,2007),the
cost-benefitassessmentinthedecision-makingprocessmaybehighly
context-specificandfairlysophisticated.Ourstudieshighlightedtheimportance
ofcautiousinvestigationintotheeconomiccontextsrelatedtomigration.
Ashigherdevelopmentlevelinurbanareasisanessentialpullfactorofmigration,
deprivationinruralareasiscertainlyapushfactor(Zhang&Song,2003).The
lowercostsofmigrationwithinChina,ascomparedtointernationalmigration
fromotherdevelopingcountries,haveenabledthemassiveflowsof
underprivilegedpopulations.Ononehand,manyfamilieshadtospendmuchof
theirincomeonbuildingahouseintheirhometown,whichleftlimitedliquid
assetsthatcandirectlybenefitchildren.Ontheotherhand,oncefamiliesbecome
richenough,theytendtoleavethecountrysideandmovetothecitytogetherwith
theirchildren.Hence,itwouldbeimportantnottoassumeeconomicadvantages
295
inleft-behindfamilies,andchildreninthesefamiliesmaybestillinlackof
materialresources.
Themagnitudeofthechallengescausedbyparentalmigrationisalsomanifested
bythenumberandproportionofLBCinruralChina.Inthetypical
migrant-sendingareasinthetwoprovinces,ourquantitativestudyfoundover
halfofthechildrenhadoneorbothparentsawayatthetimeofsurvey,much
higherthantheproportioncalculatedbasedonnationalcensusdata(37.7%)
(Duanetal.,2013).Consideringthetremendousvariabilityofsocio-economic
developmentstatusacrossthecountry,bothwithinandacrossprovinces(Lietal.,
2013),itislikelythatruralout-migrationisamuchmorecommonphenomenon
insomeplacesthanothersinChina.
Yettheexistingliteraturerarelytakesintoaccountthescaleofmigrationasa
specificcommunitycharacteristic.Accordingtoourfieldinvestigation,even
withinonecounty,migrationratesmaydiffersignificantlybetweendifferent
townshipsorvillages.Communitysocialcapitalhasbeenfoundtobestrongly
associatedwithbetterchildmentalhealthoutcomes(Wuetal,2015).Although
quantitativesurveywasnotabletoexaminecommunity-levelvariables,our
qualitativefindingssuggestmigrationratemaybeanimportantcontextofchild
wellbeing.Highout-migrationcommunitiestendtodeveloptargetedsupport
mechanismsandprogramsforchildrenandfamilies.Also,livinginacommunity
withalargegroupofsimilarfamiliesandchildrenmayhelpsomeLBCdealwith
296
thechallengesofparentalabsence.Tothecontrary,stigmatizationofbeingLBC
appearedtobemorelikelyincommunitieswithlessout-migration.Higher
populationdensitymayalsorelievethelonelinessofLBC,assomeinterviewees
noted,whichhasnotbeenproperlyexaminedbytheliteratureeither.Further
researchshouldaimtoprovideacommunityscopingoftheissue,andelicitthe
effectsoftheproportionanddensityofleft-behindfamiliesinthecommunityon
childwellbeing.
Soonafterthe2008financialcrisishitChina,massiveflowsofmigrantsreturned
totheirruralhometownsduetotherisingurbanunemploymentrates.This
phenomenondidnotlastforlongbeforetheeconomystartedtorecoveraround
2010.TherecentslowdownofChineseeconomy,however,mayalsopotentially
resultinadeclineoflabordemandsinthecities.Thescaleoflocaljobmarketin
mostpartsofruralChinaislargelydeterminedbymediumandsmallbusinesses,
whichrelyonavarietyoffactorstothrive.Aswefoundinthequalitativestudy,
thelocalmanufacturingindustriesinKaihuahaveretainedafairlylargelabor
force,especiallywomen.Theongoingnewphaseofurbanizationfocusingon
similarsmalltownsChinamayfurthercreatemorejobsintheformerlyrural
areasandattractmoremigrantstoreturnandworknearhome(Wangetal.,
2015b).Meanwhile,asthenumberofruralworkingmothersincreases,qualityof
careforchildrenofmigrantfathermayfacenewchallenges.
Childwelfareandchildprotectionsystemsinthemigrant-sendingcommunities
297
areanothercrucialaspectofsocialcontext.Particularly,policiesandprograms
thatprovidetargetedsupportforLBCmayplayanimportantroleinsafeguarding
thewellbeingchildrenandtheirfamilies.Wuetal.(2015)hasfoundinGuizhou
provincethatcommunitysocialcapitalplaysasignificantroleinmitigatingthe
depressionrisksamongLBC.Inlessdevelopedcountriesandcommunities,the
localsystemsareoftenunabletoidentifychildreninneedofpsychological
supportandtoprovideadequateservicesforthesechildren(Pateletal.,2008).
Hence,targetedsupportprogramsfortheLBCinmigrant-sendingsocietiesand
communitiesbecomeanimportantfactorofchildwellbeing.
TheChildren’sClubinterventionprogramhassetamodelofcommunitysupport
forresource-limitedsettings.Similarmechanismsinothercountrieshavenot
beenwellestablished,exceptoneprograminMexicobuiltuponcommunity
networks.Effectiveimplementationandfavorableoutcomesofcommunity
supportmaybelinkedwithothercontext-specificinstitutionalarrangementsand
socioeconomicsituations.Communitysolidarityontacklingissuescausedby
out-migrationmayalsoraiseawarenesswithinthefamilies,onthechallenges
facedbychildren,andstrengthenthesupportfrombothco-residentandmigrant
familymembers.Itisnoteworthythatdespiteofthesignificantcontributionto
childpsychosocialwellbeing,communitysupportwillnotandshouldnotreplace
theresponsibilitiesoffamilycaregivers.Familyenvironmentstillprovidesthe
mostimportantcareandresourcesforchilddevelopmentinabsenceofmigrant
parents.
298
7.3Conclusion
7.3.1SystemssafeguardingthewellbeingofLBC
Findingsfromourstudiesbasedontheoverarchingframeworkprovidecrucial
evidencetoimprovethecurrentchildwelfareandchildprotectionsystems
safeguardingthewellbeingofLBC.Institutionsandagenciesshouldaimto
identifyareasthatarehighlyaffectedbyout-migration,andspecificallytargetthe
vulnerablechildrenandfamilies.Strategicmappingoftheissuerequires
persistentmonitoringoflocalfamilynetworksandpopulationmovements,with
coordinationacrossmultiplegovernmentandcommunityagencies.
ThemassivemigrationinChinaisrelatedtothefundamentalsocial
transformationthatcamealongwiththecountry’srapideconomicgrowth,and
thereforegovernmentagenciesandthepublicservicesectorareprobablythe
mostaccountableandcapable,intackingthedifficultiesimposedonfamilies
livingapart,especiallytheLBC.Ideally,supportiveinstitutionsinurbanandrural
areasshouldbeabletocollaborate,andcoordinatetheireffortsinservingspecific
migrantpopulationsandtheirfamiliesathome.
However,theuniversalbasicsocialservicesprovidedbythegovernment’ssocial
policysystemsarestillratherweakinChina.Recentlawsandpoliciesdesignedto
protectchildren(notjustleft-behindones)havebeenpracticallyunenforced
299
(Shang&Katz,2014).Astheinstitutionalwelfareprovisionisnotfunctioning,
familynetworksoftenfillthestatevoidinprovidingprotectiontotheLBC
(Valtolina&Colombo,2012).Consequently,familiesindisadvantagedsituations
areinevitablysubjecttoinadequateprovisionofcareandresourcesforthe
childrenleftbehind.
7.3.2Stableandprotectivefamilyenvironment
Astablefamilyenvironment,includingbothmigrantparentsandco-resident
caregivers,isessentialforthewellbeingofindividualLBCinthemigrant-sending
communities.Astomigrantparents,childwelfareandprotectionagenciesshould
makesurethattheytakebasicresponsibilitiesinchildcareevenaftertheir
departure;especiallythearrangementsforthechildtoreceiveessentialcareand
supportathomemustbemade.Communityrepresentativesandsocialworkers
shouldhelpmaintainandreinforcetheconnectionsbetweenthechildand
parents.Forinstance,somedonorshavefundedreuniontripsforthefamilies
livingapart.Parentingprogramsmayalsobeorganizedtohelpmigrantparents
bettercommunicatewithchildren,andprovideguidanceandsupportover
distance.
Ontheruralleft-behindfamilyside,varioustypesofsupportmayalsobe
enhancedbyfurthereffortsfromthegovernmentandcommunityorganizations.
Itisimportanttorecognizethatmigrantsarenotguaranteedwithbetter
300
economicstatus.Infact,theyarerathervulnerabletoeconomicshocks,since
mostofthemprefertoinvestinnewhousingratherthansaveforliquidassets
(Janson,2014).Financialaidshouldbereinforcedforchildreninneed,andthe
LBCineconomicallydeprivedfamiliesshouldbespeciallyidentifiedandtargeted.
Ingeneral,governmentorcommunityagenciesshouldconstantlymonitorthe
welfareofLBC,tooffertimelysupportoranyessentialresourcesthatarein
demand.Localpublicinstitutionssuchasschoolsandhospitalsmustalsodomore
tonotifytheauthoritiesofcasesofabuseorneglect.Programslikethe“Children’s
Club”mayintegrateresourcesatthecommunitytoprovidecareandsupervision,
aswellassocialinteractionplatformsthatwouldbenefitchilddevelopment.
7.3.3Implicationsforpolicyandinterventions
Ourresultsmayinformfuturestrategiesandinterventionprogramsinaddressing
thepsychologicalanddevelopmentalrisksfacedbytheLBC.Firstofall,thetarget
populationofpoliciesandinterventionsshouldnotberestrictedtothecurrently
left-behind.Childrenandfamilieswhopreviouslyexperiencedtheprolonged
separationshoulddrawmoreattentionfromthepolicymakersandsocialservice
providers.Meanwhile,childreninthemostdisadvantagedleft-behindfamiliesare
inpressingneedsofbettercare,withregardtobothfinancialaidsand
socioemotionalsupportfromthecommunity.Incombinationwithexistingsocial
protectionsystemsandprograms,LBCwhohavedivorcedparents,incapable
caregiverand/orliveinaremotearea,shouldbeidentifiedandspecifically
301
targeted.Registrationandrecordsofthesechildrenshouldbeestablished,inorder
toprovideconstantmonitoringandsupportforchilddevelopment.
Well-developedinterventionsandprogramsshouldalsodistinguishthedifferent
dimensionsofchildwellbeing,andprovidespecificcontentandactivities
accordingtoindividualcharacteristicsofthechildandfamily.Forinstance,
emotionalsupportanddevelopmental-behavioralinterventionsmaybecarried
outviadifferentactivitiesandatdifferentplaces,includingathome,school,and
communitycenter,dependingonthespecificaimsoftheprogram.Further
experimentalresearchneedstoexplorethepertinentapproachesthataddress
overallandindividualdimensionsofthewellbeingofLBC.
Tostrengthenthesupportiverelationshipsinchildcareenvironmentisanother
priorityinimprovingthewellbeingofbothcurrentlyandpreviouslyleft-behind
children.Thisinvolvesparent-childcommunicationoverdistance,
grandparent-childinteractionsathome,aswellastheconnectionsbetweenthe
fatherandmotherandtheirjointeffortsonchildcare.Parentingprogramsmaybe
designedformigrantparentsandco-residentcaregivers,respectively.Ifthe
couplesdonotmigratetogether,certainrelationshipormarriagecounseling
servicesmaybepiloted,tostrengthenthesolidarityofthenuclearfamily
especiallyoncareandsupportforthechildren.
302
Thelocalcommunityalsoplaysanimportantroleinprovidingcareandsocial
supporttotheLBC.Communitycareprogrammaybeorganizedbylocalofficials
andcommunitymembersforthechildren,notonlytosharetheresponsibilitiesof
caregivers,buttocreateaplatformforchildrentospendtimetogetherinavariety
ofactivities.Suchprogramsmayalsoincludeacademictutoringtohelpwith
children’sschoolperformance.Indoingsothecommunitywouldbeabletooffer
criticalsourcesofpeerandsocialsupporttothelocalLBC.
7.3.4Solutionstokeepfamiliestogether
CertainlytheultimatesolutiontotheLBCissuesistokeepfamiliestogether.In
themigrant-receivingcities,governmentsandlargeemployersshouldstriveto
incorporatethemigrantsintotheurbansociety,byreducingthebarrierstourban
citizenship,andimprovingtheirworkingconditions,benefitsandwelfare.
Enhancedeffortsfromnon-governmentalorganizationsandcorporatesocial
responsibilitiesarealsoexpected,forbetterservicestothechildrenofmigrant
workers.Ontheruralside,morelocalresidentsshouldbeabletoachievebetter
livingstandardsbyworkingneartheirhometown.Inthelatesturbanization
process,thisreliesonincreasedinvestmentsandtargetedpoliciesinsmaller
townswithinruralareas,particularlyonstart-upbusinessesormid-and
small-scaleindustries,inordertoattractalargerlocalworkforce.The
governmentandcivilsocietyshouldcontinuetheeffortsonensuringtheessential
welfareforchilddevelopmentandqualityeducationattainmentforallchildren,
303
andbuildupfavorable,supportiveenvironmentsinruralcommunitieswhere
childrencanthriveandachievetheirfullpotential,withoutprolongedseparation
fromtheirparents.
304
ReferencesAdams,R.H.,&Page,J.(2005).Dointernationalmigrationandremittancesreducepovertyindevelopingcountries?WorldDevelopment,33,1645–1669.Adhikari,R.,Jampaklay,A.,Chamratrithirong,A.,Richter,K.,Pattaravanich,U.,&Vapattanawong,P.(2014).Theimpactofparentalmigrationonthementalhealthofchildrenleftbehind.JournalofImmigrantandMinorityHealth,16(5),781-789.Adumitroaie,E.,&Dafinoiu,I.(2013).Perceptionofparentalrejectioninchildrenleftbehindbymigrantparents.Revistadecercetaresiinterventiesociala,42,191.Ainsworth,M.D.S.(1969).Objectrelations,dependency,andattachment:Atheoreticalreviewoftheinfant-motherrelationship.Childdevelopment,969-1025.Ainsworth,M.D.S.,Blehar,M.,Waters,E.,&Wall,S.(1978).Patternsofattachment:ApsychologicalstudyoftheStrangeSituation.Hillsdale,NJ:Erlbaum.Aguilera-Guzmán,R.M.,deSnyder,V.N.S.,Romero,M.,&Medina-Mora,M.E.(2004).Paternalabsenceandinternationalmigration:StressorsandcompensatorsassociatedwiththementalhealthofMexicanteenagersofruralorigin.Adolescence,39(156),711.All-ChinaWomen’sFederation(2013).Nationalsurveyreportonthesituationofleft-behindchildrenandmigrantchildreninChina:Beijing.Allen,J.P.(2008)Theattachmentsysteminadolescence.InJ.Cassidy&P.R.Shaver(Eds.),Handbookofattachment:Theory,research,andclinicalapplications(2nded.,pp.419-435).NewYork:Guilford.Armsden,G.C.,&Greenberg,M.T.(1987).Theinventoryofparentandpeerattachment:Individualdifferencesandtheirrelationshiptopsychologicalwell-beinginadolescence.Journalofyouthandadolescence,16(5),427-454.Arnold,E.(2006).SeparationandlossthroughimmigrationofAfricanCaribbeanwomentotheUK.Attachment&humandevelopment,8(2),159-174.
305
Asis,M.M.(2006).Livingwithmigration:Experiencesofleft-behindchildreninthePhilippines.AsianPopulationStudies,2(1),45-67.Bian,Y.(2002).Chinesesocialstratificationandsocialmobility.AnnualReviewofSociology,91-116.Berk,L.E.(2012).ChildDevelopment,9thEdition.NewYork:Pearson.Boss,P.(2002).Familystressmanagement.ThousandOaks,Calif.:SagePublications.Boss,P.(2004).Ambiguouslossresearch,theory,andpractice:Reflectionsafter9/11.JournalofMarriageandFamily,66(3),551-566.Botezat,A.,&Pfeiffer,F.(2014).TheImpactofParentsMigrationontheWell-BeingofChildrenLeftBehind–InitialEvidencefromRomania.ZEW-CentreforEuropeanEconomicResearchDiscussionPaper,(14-29).Bowlby,J.(1969).Attachmentandloss:Vol.1.Attachment.NewYork:BasicBooksBowlby,J.(1973).Attachmentandloss:Vol.2.Separation:Anxietyandanger.NewYork:BasicBooks.Bowlby,J.(1980).Attachmentandloss,Vol.3:Loss,sadnessanddepression.NewYork:BasicBooks.Bronfenbrenner,U.(1979).Theecologyofhumandevelopment:Experimentsbydesignandnature.Cambridge,MA:HarvardUniversityPress.Bronfenbrenner,U;Morris,P.A.(1998).W.Damon&R.M.Lerner,ed.TheecologyofdevelopmentalprocessesinHandbookofchildpsychology,Vol.1:Theoreticalmodelsofhumandevelopment(5thed.).NewYork:JohnWileyandSons,Inc.pp.993–1023.Bryant,J.(2005).ChildrenofInternationalMigrantsinIndonesia,Thailand,andthePhilippines:Areviewofevidenceandpolicies.InnocentiWorkingPaper2005-05,UNICEFInnocentiResearchCentre,Florence.
306
Butler,F.R.,Zakari,N.(2005)Grandparentsparentinggrandchil-dren:assessinghealthstatus,parentalstress,andsocialsupports.JournalofGerontologyNursing31:43–54 Cai,F.,Wang,D.,&Du,Y.(2002).RegionaldisparityandeconomicgrowthinChina:Theimpactoflabormarketdistortions.ChinaEconomicReview,13(2),197-212.Carling,J.,CeciliaM.,&Schmalzbauer,L.(2012).CentralThemesintheStudyofTransnationalParenthood.JournalofEthnicandMigrationStudies,38:191-217.Cameron,L.,Erkal,N.,Gangadharan,L.,&Meng,X.(2013).Littleemperors:behavioralimpactsofChina'sone-childpolicy.Science,339(6122),953-957.Carroll,J.S.,Olson,C.D.,&Buckmiller,N.(2007).Familyboundaryambiguity:A30-yearreviewoftheory,research,andmeasurement.FamilyRelations,56(2),210-230.Cassidy,J.(2008).Thenatureofthechild’sties.InJ.Cassidy&P.R.Shaver(Eds.),Handbookofattachment:Theory,research,andclinicalapplications(2nded.,pp.3–23).NewYork:Guilford.Cerami,A.(2006).SocialPolicyinCentralandEasternEurope:TheEmergenceofaNewEuropeanWelfareRegime,LITVerlag,Berlin.Chen,H.,Hu,B.,&He,F.(2011).Influencingfactorsonpsychologicalhealthbetweenleft-behindchildrenandnotleft-behindchildreninjuniormiddleschool.ChineseJournalofSchoolHealth,4,012.Chen,F.,Short,S.E.,&Entwisle,B.(2000).TheimpactofgrandparentalproximityonmaternalchildcareinChina.PopulationResearchandPolicyReview,19(6),571-590.Chung,H.(2008).StateregulationandChina'sadministrativesystem:Aspatialperspective.ChinaReview,201-230.Coe,C.(2011).Whatislove?ThematerialityofcareinGhanaiantransnationalfamilies.InternationalMigration,49(6),7-24.Colin,V.L.(1996).Humanattachment.NewYork:McGraw-Hill.
307
Cong,Z.,&Silverstein,M.(2011).Intergenerationalexchangebetweenparentsandmigrantandnon-migrantsonsinruralChina.JournalofMarriageandFamily,73,93–104.Constant,A.F.,&Zimmermann,K.F.(Eds.).(2013).Internationalhandbookontheeconomicsofmigration.EdwardElgarPublishing.Crawford,T.N.,Cohen,P.R.,Chen,H.,Anglin,D.M.,&Ehrensaft,M.(2009).Earlymaternalseparationandthetrajectoryofborderlinepersonalitydisordersymptoms.Developmentandpsychopathology,21(03),1013-1030.delaGarza,R.(2010).Migration,developmentandchildrenleftbehind:Amultidimensionalperspective.SocialandEconomicPolicyworkingpaper.NewYork:UNICEF.Daly,S.L.,&Glenwick,D.S.(2000).Personaladjustmentandperceptionsofgrandchildbehaviorincustodialgrandmothers.JournalofClinicalChildPsychology,29(1),108-118.Ding,G.,&Bao,Y.(2014).Editorialperspective:assessingdevelopmentalriskinculturalcontext:thecaseof‘leftbehind’childreninruralChina.JournalofChildPsychologyandPsychiatry,55(4),411-412.Dollar,D.(2014).SinoShiftChina’srebalancingopensnewopportunitiesfordevelopingAsia.FinanceandDevelopment,51(2),10-13.Dong,Z.&He,Z.(2014)Relationshipbetweenfamilyfunctionandmentalelasticityinleft-behindchildreninminorityareas.MaternalandChildHealthCareofChina29(6),920-923.Dong,Z.&Zhang,D.(2013a).Relationshipbetweentheresilienceandalonenessamongleft-behindchildreninminoritynationalityregions.ChinaJournalofSchoolHealth,34(7),827-829.Dong,Z.&Zhang,D.(2013b).Studyontheresilienceoftheleft-behindchildreninminoritynationalityregionsinYunnanprovince.ChinaJournalofChildHealthCare,21(9),920-922.
308
Dorigo,G.,&Tobler,W.(1983).Push-pullmigrationlaws.AnnalsoftheAssociationofAmericanGeographers,73(1),1-17.Dreby,J.(2007).ChildrenandpowerinMexicantransnationalfamilies.JournalofMarriageandFamily,69(4),1050-1064.Dreby,J.,&Stutz,L.(2012).Makingsomethingofthesacrifice:gender,migrationandMexicanchildren'seducationalaspirations.Globalnetworks,12(1),71-90.DuanC.R.,LvL.D.,Guo,J.,&Wang,Z.P.(2013).SurvivalandDevelopmentofLeft-behindChildreninRuralChina:BasedontheAnalysisofSixthCensusData.PopulationJournal,199(03),39-51.Erdman,P.,&Caffery,T.(Eds.).(2013).Attachmentandfamilysystems:Conceptual,empiricalandtherapeuticrelatedness.Routledge.Esping-Andersen,G.(2000).AWelfareStateforthe21stCentury:AgeingSocieties,Knowledge-basedEconomies,andtheSustainabilityofEuropeanWelfareStates.ReporttothePortugesepresidencyoftheEuropeanUnion.Lisbonsummit.Fan,F.,Su,L.,Gill,M.K.,&Birmaher,B.(2010).EmotionalandbehavioralproblemsofChineseleft-behindchildren:apreliminarystudy.Socialpsychiatryandpsychiatricepidemiology,45(6),655-664.Fan,X.,Fang,X.,Lin,D.,&Zhu,D.(2013).Unfavorablefamilyatmosphereandleft-behindchildren’spsychologicaladaptation:mediationeffectofsocialsupport.HunanSocialScience,(5),128-131.Gao,J.(2010).Analysisontheself-consciousnessofruralmigrantworkers'left-behindchildrenandtheinfluencingfactors.ChinaJournalofHealthPsychology,18,58-62.Gao,Y.,Li,L.P.,Kim,J.H.,Congdon,N.,Lau,J.,&Griffiths,S.(2010).TheimpactofparentalmigrationonhealthstatusandhealthbehavioursamongleftbehindadolescentschoolchildreninChina.BMCpublichealth,10(1),56.Gao,Y.,Li,L.,Chan,E.Y.,Lau,J.,&Griffiths,S.M.(2013).Parentalmigration,self-efficacyandcigarettesmokingamongruraladolescentsinSouthChina.PloSone,8(3),e57569.
309
Givaudan,M.,&Pick,S.(2013).Childrenleftbehind:howtomitigatetheeffectsandfacilitateemotionalandpsychosocialdevelopment:supportivecommunitynetworkscandiminishthenegativeeffectsofparentalmigration.Childabuse&neglect,37(12),1080-1090.Glaser,B.G.(1978).Theoreticalsensitivity:Advancesinthemethodologyofgroundedtheory.SociologyPress.
Glaser,B.(1992).BasicsofGroundedTheoryAnalysis:EmergencevForcing.MillValley,CA:SociologyPress.
Goodman,R.(1997).TheStrengthsandDifficultiesQuestionnaire:aresearchnote.Journalofchildpsychologyandpsychiatry,38(5),581-586.GoodmanR,RenfrewD,MullickM(2000)PredictingtypeofpsychiatricdisorderfromStrengthsandDifficultiesQuestionnaire(SDQ)scoresinchildmentalhealthclinicsinLondonandDhaka.EuropeanChildandAdolescentPsychiatry,9,129-134.GovernmentofChina.(2013).Announcementonthecareandeducationforruralschoolagedleft-behindchildrenbyMinistryofEducationandfourotheragencies.Retrievedfrom:http://www.gov.cn/zwgk/2013-01/10/content_2309058.htmGraham,E.,&Jordan,L.P.(2011).MigrantParentsandthePsychologicalWell-BeingofLeft-BehindChildreninSoutheastAsia.JournalofMarriageandFamily,73(4),763-787.Graham,E.,Jordan,L.,Yeoh,B.,Lam,T.,&Asis,M.(2012).Transnationalfamiliesandthefamilynexus:perspectivesofIndonesianandFilipinochildrenleftbehindbymigrantparent(s).EnvironmentandPlanningA.44(4)793–815Gray,M.,Misson,S.,&Hayes,A.(2005).Youngchildrenandtheirgrandparents.FamilyMatters,(72),10.Guo,Z.G.,Zhang,E.L.,Gu,B.C.,Wang,F.(2003)DiversityofChina’sFertilityPolicy.PopulationResearch.27(5),1-10.Han,J.,Ye,Y.,Bian,H.&Gao,W.(2012).PsychologicalresiliencefeaturesofShaanximigrantchildrenanditsinfluenceontheirsenseofsecurity.JournalofNorthwestUniversity(NaturalScienceEdition)42(4),613-619.
310
He,J.Ma,H.,&Xu,L.(2011).Investigationondepressionstatusamongleft-behindchildreninruralareasofAnhuiProvince,2008.PreventiveMedicineTribune,4,008.He,B.,Fan,J.,Liu,N.,Li,H.,Wang,Y.,Williams,J.,&Wong,K.(2012).Depressionriskof‘left-behindchildren’inruralChina.Psychiatryresearch,200(2),306-312.Hill,R.(1949).Familiesunderstress.NewYork:Harper&Brothers.Hill,R.(1958).Genericfeaturesoffamiliesunderstress.SocialCasework,49,139-150.Hoang,L.A.,Lam,T.,Yeoh,B.S.,&Graham,E.(2015).Transnationalmigration,changingcarearrangementsandleft-behindchildren'sresponsesinSouth-eastAsia.Children'sGeographies,13(3),263-277.Hoang,L.A.,&Yeoh,B.S.(2015).Children'sagencyanditscontradictionsinthecontextoftransnationallabormigrationfromVietnam.GlobalNetworks,15(2),180-197.Hopkins,J.(1990).Theobservedinfantofattachmenttheory.JournalofPsychotherapy,6,460–470.Hou,K.,Liu,Y.,Qu,Z.,Zhang,Y.&Jiang,S.(2015).Influenceofclassstructureonthementalhealthofleft-behindchildren:synthesizingorcontrastingeffects?PsychologicalDevelopmentandEducation,31(2),220-229.Hu,H.,Lu,S.,&Huang,C.C.(2014).Thepsychologicalandbehavioraloutcomesofmigrantandleft-behindchildreninChina.ChildrenandYouthServicesReview,46,1-10.Huang,Y.&Li,L.(2007).Mentalhealthstatusofdifferenttypesofleft-behindchildren.ChineseMentalHealthJournal,21(10),669.Huang,Y.,Zhong,X.N.,Li,Q.Y.,Xu,D.,Zhang,X.L.,Feng,C.,...&Deng,B.(2015).Health-relatedqualityoflifeoftherural-Chinaleft-behindchildrenoradolescentsandinfluentialfactors:across-sectionalstudy.Healthandqualityoflifeoutcomes,13(1),1.
311
Jaccard,J.,&Guilamo-Ramos,V.(2002).Analysisofvarianceframeworksinclinicalchildandadolescentpsychology:Issuesandrecommendations.JournalofClinicalChildandAdolescentPsychology,31(1),130-146. Janson,S.(2014).Childrenleftbehind.ActaPaediatrica,103(6),572-573.Jia,Z.,&Tian,W.(2010).Lonelinessofleft-behindchildren:across-sectionalsurveyinasampleofruralChina.Child:Care,HealthandDevelopment,36(6),812-817.Jia,Z.,Shi,L.,Cao,Y.,Delancey,J.,&Tian,W.(2010).Health-relatedqualityoflifeof“left-behindchildren”:across-sectionalsurveyinruralChina.QualityofLifeResearch,19(6),775-780.Jiang,Q.(2013).Emotionalandbehavioralproblemsamongleft-behindmiddleschoolchildreninFujianprovince.ChineseJournalofPublicHealth,29(12),1765-1768.Jiang,Y.,Sun,Y.,Yang,L.,Chen,M.,Lv,L.,Sun,Y…&Zhang,H.(2013).Analysisontheresilienceofleft-behindchildreninruralAnhuiProvinceanditsassociatedfactors.ChinaJournalofSchoolHealth,(3),365-367.Jordan,L.P.,&Graham,E.(2012).ResilienceandWell-BeingAmongChildrenofMigrantParentsinSouth-EastAsia.ChildDevelopment,83(5),1672-1688.KaihuaWomen’sFederation.(2013).Reportonsituationofleft-behindchildren.Unpublished;accessedbypersonalcorrespondence.King,R.(2012).Theoriesandtypologiesofmigration:anoverviewandaprimer.WillyBrandtseriesofworkingpapersininternationalmigrationandethnicrelations,3,12.Kobak,R.,&Madsen,S.(2008).DisruptionsinAttachmentBonds:ImplicationsforTheory,Research,andClinicalIntervention.InJ.Cassidy&P.R.Shaver(Eds.),Handbookofattachment:Theory,research,andclinicalapplications(2nded.,pp.23–48).NewYork:Guilford.Kou,J.,Du,Y.,Xia,L.(2007).FormulationofChildrenStrengthsandDifficultiesQuestionnaire(theeditionforstudents)forShanghainorm.ChinaJournalofHealthPsycholoy15:35–38
312
Lan,Y.L.,Li,Y.,Tang,X.J.,Zhang,Y.Y.,Tan,Y.N.,&Tan,J.Y.(2009).Personalityanddepressivesymptomsandtheirinfluentialfactorsinchildrenleft-behindinruralarea.ChineseJournalofPublicHealth,25,901-903.Lam,T.,Ee,M.,Anh,H.L.,&Yeoh,B.S.(2013).Securingabetterlivingenvironmentforleft-behindchildren:implicationsandchallengesforpolicies.AsianandPacificMigrationJournal,22(3),421-445.Lahaie,C.,Hayes,J.A.,Piper,T.M.,&Heymann,J.(2009).Workandfamilydividedacrossborders:TheimpactofparentalmigrationonMexicanchildrenintransnationalfamilies.Community,Work&Family,12(3),299-312.Lauby,J.,&Stark,O.(1988).Individualmigrationasafamilystrategy:YoungwomeninthePhilippines.Populationstudies,42(3),473-486.Lazarus,R.S.(1966).Psychologicalstressandthecopingprocess.NewYork:McGraw-Hill.Lazarus,R.S.(1976).Patternsofadjustment(3rded.).NewYork:McGraw-Hill.Lazarus,R.S.(1977).Cognitiveandcopingprocessinemotion.InA.Monat,ed.&R.Lazarus(Eds.),Stressandcoping.NewYork:ColumbiaUniversityPress.Laudy,J.,&Stark,O.(1988).Individualmigrationasafamilystrategy:YoungwomeninthePhilippines.PopulationStudies,42,473–486.Lee,E.S.(1966).Atheoryofmigration.Demography,3(1),47-57.Levitt,P.(1998).Socialremittances:Migrationdrivenlocal-levelformsofculturaldiffusion.Internationalmigrationreview,926-948.Li,G.Y.,Luo,T.M.,&Tao,F.B.(2012).Effectsofparentalworkingout-of-homeonhealthriskbehaviorandpsychologicalstatusamongleft-hehindchildren[J].ChineseJournalofPublicHealth,7,16-21.Li,G.Y.,&Tao,F.B.(2009).ZunyiMedicalCollege,ZhuhaiCampus(Zhuhai519041,China);Surveyonmentalhealthstatusandsuicidalideationamongruralchildrenleft-behind.ChineseJournalofPublicHealth,8.905-907.
313
Li,S.,Sato,H.,&Sicular,T.(Eds.).(2013).RisinginequalityinChina:Challengestoaharmonioussociety.CambridgeUniversityPress.Li,W.(2015).Kaihua,Zhejiang:FindinggoldintheNationalPark.ChinaEconomicWeekly,2015(35).Ling,H.,Fu,E.,&Zhang,J.R.(2015).Effectsofseparationageandseparationdurationamongleft-behindchildreninChina.SocialBehaviorandPersonality:aninternationaljournal,43(2),241-253.Liu,J.,Yan,C.,Liu,J.,Geng,C.,Li,M.,Wang,X…&Zeng,L.(2011).AStudyofPsychologicalStatusofLeoverChildreninXibaipoTownandRelativeInuenceFactors.ChineseJournalofEvidence-basedMedicine,11(5),489-492.Liu,J.,Yang,X.,Yang,Y.,Zhou,H.,LuW.,&Liu,Q.(2014).Multilevelmodelanalysisonchangesofpsychologicalhealthamongruralleft-behindchildreninjuniorhighschoolsinSichuan.ChineseJournalofSchoolHealth,35(11),1648-1650.Liu,L.J.,Sun,X.,Zhang,C.L.,Wang,Y.,&Guo,Q.(2010).AsurveyinruralChinaofparent-absencethroughmigrantworking:theimpactontheirchildren'sself-conceptandloneliness.BMCpublichealth,10(1),32.Liu,Z.(2009).LiushouChildreninaChineseVillage:ChildhoodApartfromParents.ChineseSociology&Anthropology,41(3),71-89.Liu,Z.,Li,X.,&Ge,X.(2009).Lefttooearly:theeffectsofageatseparationfromparentsonChineseruralchildren'ssymptomsofanxietyanddepression.AmericanJournalofPublicHealth,99(11),2049.Lu,W.,Hu,Y.,Zhang,D.,Peng,G.,Yang,S.,Xu,Y…&Han,J.(2014a).Left-behindcharacteristicsandtheeffectsofrelatedfactorsonpsychologicalhealthofruralleft-behindchildren.MaternalandChildHealthCareofChina29(21),3463-3466.Lu,Y.,Yu,Y.&Zhu,Y.(2014b).Mentalhealthstatusandfamilyfactorsinjunior-highschoolstudentsinruralGuizhou.ChineseJournalofSchoolHealth,35(9),1413-1415.Luca,A.M.(2007)‘Long-distancefamilies’,RomaniaNewsWatch:TransitionsOnline,Availableat:
314
http://www.romanianewswatch.com/2007/04/transitions-online-romania-long.html Luo,J.,Wang,L.G.,&Gao,W.B.(2012).Theinfluenceoftheabsenceoffathersandthetimingofseparationonanxietyandself-esteemofadolescents:across-sectionalsurvey.Child:Care,HealthandDevelopment,38(5),723-731.Massey,D.S.,Arango,J.,Hugo,G.,Kouaouci,A.,&Pellegrino,A.(1999).WorldsinMotion:UnderstandingInternationalMigrationattheEndoftheMillennium:UnderstandingInternationalMigrationattheEndoftheMillennium.ClarendonPress.McKenzie,D.,&Rapoport,H.(2006).Canmigrationreduceeducationalattainment?EvidencefromMexico.EvidencefromMexico(June1,2006).WorldBankPolicyResearchWorkingPaper,(3952).Main,M.(1996).Introductiontothespecialsectiononattachmentandpsychopathology:2.Overviewofthefieldofattachment.Journalofconsultingandclinicalpsychology,64(2),237.McCubbin,H.(1979).Integratingcopingbehaviorinfamilystresstheory.JournalofMarriageandtheFamily,41,237–244.McLanahan,S.andG.Sandefur(1994)GrowingUpwithaSingleParent.Cambridge,MA:HarvardUniversityPress.Moran-Taylor,M.J.(2008)‘Whenmothersandfathersmigratenorth:caretakers,children,andchild-rearinginGuatemala’,LatinAmericanPerspectives,35(4):79-95.Murphy,R.(2014).StudyandSchoolintheLivesofChildreninMigrantFamilies:AViewfromRuralJiangxi,China.DevelopmentandChange,45(1),29-51.Murray,J.,&Murray,L.(2010).Parentalincarceration,attachmentandchildpsychopathology.Attachment&humandevelopment,12(4),289-309.NationalBureauofStatistics(2008).Countyleveldatafromthe5thNationalCensus.Accessedfromhttp://www.stats.gov.cn/tjsj/ndsj/renkoupucha/2000fenxian/htm/table1.htm
315
NationalBureauofStatistics(2014).NationalStatisticalYearbook2014.Beijing.Nobles,J.(2013).Migrationandfatherabsence:ShiftingfamilystructureinMexico.Demography,50(4),1303-1314.Ochiai,E.(2009).CarediamondsandwelfareregimesinEastandSouth-EastAsiansocieties:Bridgingfamilyandwelfaresociology.InternationalJournalofJapaneseSociology,18(1),60-78.Pantea,M.C.(2012).Grandmothersasmaincaregiversinthecontextofparentalmigration:Bunicilecaîngrijitoareincontextulmigratieiparintilor.EuropeanJournalofSocialWork,15(1),63-80.Papalia,D.E.,Olds,S.W.,&Feldman,R.D.(2001).Humandevelopment(8thed.).NewYork:McGraw-Hill.Parreñas,S.R.(2005)ChildrenofGlobalMigration:TransnationalFamiliesandGenderedWoes.Stanford,CA:StanfordUniversityPress.Parreñas,S.R.(2008).Transnationalfathering:Genderedconflicts,distantdiscipliningandemotionalgaps.Journalofethnicandmigrationstudies,34(7),1057-1072.Patel,V.,Flisher,A.J.,Nikapota,A.,&Malhotra,S.(2008).Promotingchildandadolescentmentalhealthinlowandmiddleincomecountries.JournalofChildPsychologyandPsychiatry,49(3),313-334.Pawson,R.,&Tilley,N.(1997).RealisticEvaluationSage.LosAngeles,London,NewDelhi,Singapore.Piore,M.J.(1983).Labormarketsegmentation:towhatparadigmdoesitbelong?.TheAmericanEconomicReview,73(2),249-253.Pottinger,A.M.(2005).Children'sExperienceofLossbyParentalMigrationinInner-CityJamaica.AmericanJournalofOrthopsychiatry,75(4),485-496.Pribilsky,J.(2001).NerviosandModernChildhood:MigrationandShiftingContextsofChildLifeintheEcuadorianAndes.Childhood,8(2),251-273.
316
Prempeh,O.,Adjibolosoo,S.,&Mensah,J.(2004).Globalizationandthehumanfactor:Criticalinsights.Aldershot,England:Ashgate.Qiao,X.C.,&Huang,Y.H.(2013).FloatingPopulationsAcrossProvincesinChina-AnalysisBasedontheSixthCensus.Population&Development.19(1):13-28.Rapoport,H.,&Docquier,F.(2006).Theeconomicsofmigrants'remittances.Handbookoftheeconomicsofgiving,altruismandreciprocity,2,1135-1198.Chicago.Ren,Q.,&Treiman,D.J.(2016).TheconsequencesofparentallabormigrationinChinaforchildren'semotionalwellbeing.Socialscienceresearch,58,46-67.Roberts,K.(2002).FemaleLaborMigrantstoShanghai:Temporary“Floaters”orPotentialSettlers?1.InternationalMigrationReview,36(2),492-519.Robson,M.,Luo,J.,Peng,X.,Zong,R.,Yao,K.,Hu,R.,...&Zhu,M.(2008).Thestatusofcareandnutritionof774left-behindchildreninruralareasinChina.Publichealthreports,382-389.Rothbaum,F.,Rosen,K.,Ujiie,T.,&Uchida,N.(2002).FamilySystemsTheory,AttachmentTheory,andCulture.Familyprocess,41(3),328-350.Samers,M.(2010).Migration.London:RoutledgeSassen,S.(1988).TheMobilityofLaborandCapital.NewYork:CambridgeUniversityPress.Schaffer,H.R.,&Emerson,P.E.(1964).Thedevelopmentofsocialattachmentsininfancy.MonographsoftheSocietyforResearchinChildDevelopment,29(3,SerialNo.94),1–77.Schapiro,N.A.,Kools,S.M.,Weiss,S.J.,&Brindis,C.D.(2013).Separationandreunification:theexperiencesofadolescentslivingintransnationalfamilies.Currentproblemsinpediatricandadolescenthealthcare,43(3),48-68.Senaratne,B.C.V.,Perera,H.,&Fonseka,P.(2011).Mentalhealthstatusandriskfactorsformentalhealthproblemsinleft-behindchildrenofwomenmigrantworkersinSriLanka.CeylonMedicalJournal,56(4),153-158.
317
Shang,X.,&Katz,I.(2014).MissingElementsintheProtectionofChildren:ThreeCasesfromChina.JournalofSocialServiceResearch,40(4),545-559.Shen,M.,Yang,S.,Han,J.,Shi,J.,Yang,R.,Du,Y.,&Stallones,L.(2009).Non-fatalinjuryratesamongthe“left-behindchildren”ofruralChina.Injuryprevention,15(4),244-247.Shi,T.(1999).VillagecommitteeelectionsinChina:Institutionalisttacticsfordemocracy.WorldPolitics,51(03),385-412.Short,S.E.,Chen,F.,Entwisle,B.,&Fengying,Z.(2002).MaternalworkandchildcareinChina:amulti-methodanalysis.PopulationandDevelopmentReview,28(1),31-57.Short,S.E.,Fengying,Z.,Siyuan,X.,&Mingliang,Y.(2001).China'sone-childpolicyandthecareofchildren:Ananalysisofqualitativeandquantitativedata.SocialForces,79(3),913-943.Sjaastad,L.A.(1962).TheCostsandReturnsofHumanMigration,JournalofPoliticalEconomy,70(1):80-93.Senaratne,B.C.V.,Perera,H.,&Fonseka,P.(2011).Mentalhealthstatusandriskfactorsformentalhealthproblemsinleft-behindchildrenofwomenmigrantworkersinSriLanka.CeylonMedicalJournal,56(4),153-158.Smeekens,C.,Stroebe,M.S.,&Abakoumkin,G.(2012).TheimpactofmigratoryseparationfromparentsonthehealthofadolescentsinthePhilippines.SocialScience&Medicine,75(12),2250-2257.Sofranko,A.J.,&Idris,K.(1999).Useofoverseasmigrants’remittancestotheextendedfamilyforbusi-nessinvestment:Aresearchnote.RuralSociology,64,464–481.Sroufe,L.A.,Carlson,E.A.,Levy,A.K.,&Egeland,B.(1999).Implicationsofattachmenttheoryfordevelopmentalpsychopathology.DevelopmentalPsychopathology,11,1–13.Sroufe,L.A.,Egeland,B.,Carlson,E.,&Collins,W.A.(2005).Placingearlyattachmentexperiencesindevelopmentalcontext:theMinnesotalongitudinal
318
study.In:GrossmannKE,GrossmannK,WatersE,eds.AttachmentfromInfancytoAdulthood:TheMajorLongitudinalStudies.NewYork:TheGuilfordPress.p.49–70.Standing,G.(2001).Carework:Overcominginsecurityandneglect.In:Daly,M.,ed.CareWork:TheQuestforSecurity.Geneva:InternationalLaborOffice.Stark,O.,&Bloom,D.E.(1985).Theneweconomicsoflabormigration.TheamericanEconomicreview,75(2),173-178.Stark,O.,&Lucas,R.(1988).Migration,remittancesandthefamily.EconomicDevelopmentandCulturalChange,36,465–481Su,S.,Li,X.,Lin,D.,Xu,X.,&Zhu,M.(2012).Psychologicaladjustmentamongleft-behindchildreninruralChina:theroleofparentalmigrationandparent–childcommunication.Child:Care,HealthandDevelopment,39(2),162-170.
Strauss,A.andCorbin,J.(1998)BasicsofQualitativeResearch–TechniquesandProceduresforDevelopingGroundedTheory.(2nded).CA:Sage.
Strohschein,L.(2005).Parentaldivorceandchildmentalhealthtrajectories.JournalofMarriageandFamily,67(5),1286-1300.Tao,X.W.,Guan,H.Y.,Zhao,Y.R.,&Fan,Z.Y.(2013).Mentalhealthamongleft-behindpreschool-agedchildren:PreliminarysurveyofitsstatusandassociatedriskfactorsinruralChina.JournalofInternationalMedicalResearch,42(1),120-129. TheEconomist.(2015).China’sleft-behind:Littlematchchildren.(2015,Oct17).TheEconomist.Retrievedfromhttp://www.economist.com/news/briefing/21674712-children-bear-disproportionate-share-hidden-cost-chinas-growth-little-match-children.TheWorldBank(2016).MigrationandRemittancesFactbook2016.Washington,DC:TheWorldBankTodaro,M.P.(1969).Amodeloflabormigrationandurbanunemploymentinlessdevelopedcountries.TheAmericaneconomicreview,59(1),138-148.Tomșa,R.,&Jenaro,C.(2015).Childrenleftbehindinromania:anxietyandpredictorvariables.Psychologicalreports,116(2),485-512.
319
UNICEF(2006).Publicpolicyissues:MigrationandChildren.PanamaCity:UNICEFRegionalOfficeforLatinAmericaandtheCaribbean.UNICEF.(2007).Theimpactofinternationalmigration:ChildrenleftbehindinselectedcountriesofLatinAmericaandtheCaribbean.NewYork:UNICEFDivisionofPolicyandPlanning.WorkingPapersSeries. UNICEF.(2011).SituationAnalysisofVulnerable,ExcludedandDiscriminatedChildreninMoldova.Chisinau:UNICEFMoldova.UNICEF.(2014a).UNICEFAnnualReport2014China.Beijing:UNICEFChina.UNICEF.(2014b).WhatcensusdatacantellusaboutchildreninChina–FactsandFigures2013.Beijing:UNICEFChina.UnitedNations.(2013)Cross-nationalcomparisonsofinternalmigration:Anupdateonglobalpatternsandtrends.NewYork:UnitedNationsDepartmentofEconomicandSocialAffairs.PopulationDivisionWorkingPaper,2013/1.UnitedNations.(2014)WorldUrbanizationProspects:The2014Revision.NewYork:UnitedNationsDepartmentofEconomicandSocialAffairs,PopulationDivision.Valtolina,G.G.,&Colombo,C.(2012).Psychologicalwell-being,familyrelations,anddevelopmentalissuesofchildrenleftbehind.Psychologicalreports,111(3),905-928.Walker,A.J.(1985).Reconceptualizingfamilystress.JournalofMarriageandtheFamily,47,827-837.Wallerstein,J.S.,&Kelly,J.B.(2008).Survivingthebreakup:Howchildrenandparentscopewithdivorce.BasicBooks.Wang,T.Z.,Chen,M.C.,Sun,Y.H.,Yang,L.S.,Sun,Y.,Cao,Q.Q.,...&Han,T.W.(2011a).ResearchonChildren’sdepressionandtheinfluenceofleft-behindstatusinruralarea.ChineseJournalofSchoolHealth,32,1445-1447.Wang,D.(2006).China'surbanandruraloldagesecuritysystem:Challengesandoptions.China&WorldEconomy,14(1),102-116.
320
Wang,L.,Feng,Z.,Yang,G.,Yang,Y.,Dai,Q.,Hu,C.,Liu,K.,Guang,Y.,Zhang,R.,Xia,F.,&Zhao,M.(2015a).Theepidemiologicalcharacteristicsofdepressivesymptomsintheleft-behindchildrenandadolescentsofChongqinginChina.Journalofaffectivedisorders,177,36-41.Wang,X.,&Guo,Q.,(2010).Analysisofthefrequencyofnegativelifeeventsandtheinfluencingfactorsamongleft-behindchildreninamountainousruralareainGuizhouprovince.ChineseJournalofSocialMedicine27,351–353.Wang,X.L.,Hu,X.Y.,&Shen,J.L.(2011b).Affectionofleftchildren’sfriendshipqualityonlonelinessanddepression.ChineseJournalofClinicalPsychology,19(2),252-254.Wang,X.R.,Hui,E.C.M.,Choguill,C.,&Jia,S.H.(2015b).ThenewurbanizationpolicyinChina:Whichwayforward?.HabitatInternational,47,279-284.Wang,X.,Zhang,P.(2004)TheFormationalCauseandtheEducationalTacticsofThePhenomenonofJuniorMiddleSchoolStudentsinGradeTwo.JournalofChengduCollegeofEducation,18(8),68-70.Wang,W,Zhao,H.,Guo,Y.&Duan,Z.(2014).Studyonthementalhealthstatusandtheinfluencingfactorsofleft-behindchildreninruralareas.ModernPreventiveMedicine,41(6),994-996.Wen,M.(2008).Familystructureandchildren'shealthandbehavior:Datafromthe1999NationalSurveyofAmerica'sFamilies.JournalofFamilyIssues,29(11),1492-1519Wen,M.,&Lin,D.(2012).ChilddevelopmentinruralChina:childrenleftbehindbytheirmigrantparentsandchildrenofnonmigrantfamilies.Childdevelopment,83(1),120-136.Wen,M.,Su,S.,Li,X.,&Lin,D.(2015).PositiveyouthdevelopmentinruralChina:Theroleofparentalmigration.SocialScience&Medicine,132,261-269.Whitley,D.M.,Kelley,S.J.,&Sipe,T.A.(2001).Grandmothersraisinggrandchildren:Aretheyatincreasedriskofhealthproblems?.Health&SocialWork,26(2),105-114.
321
Wickramage,K.,Siriwardhana,C.,Vidanapathirana,P.,Weerawarna,S.,Jayasekara,B.,Pannala,G.,...&Sumathipala,A.(2015).Riskofmentalhealthandnutritionalproblemsforleft-behindchildrenofinternationallabormigrants.BMCpsychiatry,15(1),1.WorldBank(2015).WorldDevelopmentIndexDatabase.Washington,DC:WorldBank.Wu,F.(Ed.).(2007).China'semergingcities:themakingofnewurbanism.Routledge.Wu,Q.,Lu,D.,&Kang,M.(2015).SocialcapitalandthementalhealthofchildreninruralChinawithdifferentexperiencesofparentalmigration.SocialScience&Medicine,132,270-277.Xie,X.,Lu,S.,Huang,C.C.,Wang,Y.,&Pei,F.(2016).AdministrativeEfficiencyandDonationofFoundationsinChina.HumanServiceOrganizations:Management,Leadership&Governance,1-11.Yang,X.H.,Zhu,C.Y.,Yang,Q.S.,Li,Z.H.,&Xie,G.R.(2013).Left-homechildren’shopeanditsrelationshipwithmentalhealth.ChineseJournalofClinicalPsychology,21(3),505-507.Yang,Y.,Tao,F.,&Wan,Y.(2010).Depressivesymptomsandtheinfluencingfactorsamongleft-behindchildren.ChineseJournalofSchoolHealth,31(3),321-323.Ye,J.,&Lu,P.(2011).Differentiatedchildhoods:impactsofrurallabormigrationonleft-behindchildreninChina.TheJournalofpeasantstudies,38(2),355-377.Zhan,X.,Li,S.,Liu,C.,&Zhang,L.(2014).EffectofMigrationonChildren'sSelf-esteeminRuralChina.China&WorldEconomy,22(4),83-101.Zhang,J.,ZhouW.,&ShenL.(2008).,AnalysisonthecurrentsituationofmigrantworkersinZhejiangandthegovernment'sresponsibilities.ZhejiangStatistics,2008(6):p.7-9.Zhang,L.,Fu,W.,Wang,D.&Bao,Z.(2015).Aqualitativestudyontheperceptionofdiscriminationanditsinfluencesonbehavioralproblemsinmiddleschoolleft-behindchildren.ChineseSpecialEducation,7,(9)53-59.
322
Zhan,S.(2015).Fromlocalstatecorporatismtolandrevenueregime:urbanizationandtherecenttransitionofruralindustryinChina.JournalofAgrarianChange,15(3),413-432.Zhang,S.,Wang,L.F.,Sun,Y.H.,Chen,X.F.,Zhang,X.J.,Gao,R.,&Xing,X.Y.(2007).Epidemiologicalsurveyonthecurrentsituationofsocialanxietyofleftbehindchildreninprimaryschool.ModernPreventiveMedicine,34,441-443.Zhao,K.F.,Su,H.,He,L.,Wu,J.L.,Chen,M.C.,&Ye,D.Q.(2009).Self-conceptandmentalhealthstatusof‘stay-at-home’childreninruralChina.ActaPaediatrica,98(9),1483-1486.Zhao,H.,Luo,J.,&Li,Zuowei.(2006).ControlStudyofPersonalityandSelf-consciousnessinChildrenofCivilianWorkers.ChinaJournalofHealthPsychology,6,012633-634.Zhao,M.,Li,H.,Li,J.,Li,G.,Wang,C.,Nicholas,S.,&Meng,Q.(2012).Studyingontheinfluenceofmigrantparentsonmentalhealthoftheirleft-behindchildreninruralChina.TheChineseHealthServiceManagement,283(1),60-3.Zhao,X.,Chen,J.,Chen,M.C.,Lv,X.L.,Jiang,Y.H.,&Sun,Y.H.(2014).Left-behindchildreninruralChinaexperiencehigherlevelsofanxietyandpoorerlivingconditions.ActaPaediatrica,103(6),665-670.Zhao,Y.,Li,G.,&Zhang,H.(2015).MentalhealthstatusofmiddleschoolstudentsandassociatedfactorsinruralareasofShangshuiCounty.ChineseJournalofSchoolHealth,12,044.Zheng,Y.(2014).Currentstatusofmentalhealthofruralleft-behindchildrenandimprovementstrategies.EducationReview,(10),97-99.Zhou,B.(1991).Mentalhealthtest.EastChinaNormalUniversityPress,Shanghai.Zhou,W.J.,Gao,W.B.,Sun,X.Y.,&Luo,J.(2011).Psychologicalresiliencefeaturesofurbanmigrantchildrenandruralleft-behindchildreninSichuanprovinceofChina.JournalofPekingUniversity.Healthsciences,43(3),386-390.
323
Zimet,G.D.,Dahlem,N.W.,Zimet,S.G.,&Farley,G.K.(1988).Themultidimensionalscaleofperceivedsocialsupport.Journalofpersonalityassessment,52(1),30-41.ZhejiangOnline(2011).Kaihuahasreducedthenumberofadministrativevillagesto255withoptimizedplanningandstrengthenedcapacities.Retrievedfromhttp://site.zjol.com.cn/khnews/system/2011/01/08/013121196.shtmlZhu,S.(2016).Impactofparentalmigrationonthementalhealthofruralleft-behindchildren.JinanJournal(PhilosophyandSocialSciences),205(2)84-94.Zhu,Y.,Hu,J.,Yu,Y.,&Lu,Y.(2014).Mentalhealthandcopingstylesamongruralleft-behindchildren.ChineseJournalofSchoolHealth,35(11),1657-1659.Zou,Y.(2012).Qualitativestudyonhealthanddevelopmentalstatusofleft-behindchildreninnorthernGuizhoururalareas.ChineseJournalofSchoolHealth,33(7)837-840.
324
Annex:Quantitativesurveyquestionnaire
In order to better understand themental health anddevelopment of children in
Zhejiang, as researchers fromZhejiangUniversity,weare conducting this survey
and would like your cooperation. This questionnaire will be anonymous and
self-administered;itisonlyforourresearchandnoinformationwillbedisclosed.
Sopleaseanswerthequestionsaccordingtowhatyouconsidertobetrue,asthere
isnotarightorwronganswer.Thankyouforyourhelp.
ZhejiangChildDevelopmentQuestionnaire
(Forstudentsatorabove5thyearofprimaryschool)
Firstofall,pleasetellussomethingaboutyourselfandyourfamily.
1. Pleasetelluswhetheryouareagirloraboy. Girl Boy
2. Yourbirthday?________(mm/yyyy)
3. Whatisthenameoftheschoolyouattend?__________
4. Whichschoolyearareyouin?
5. What'sthelevelofyourfamily'seconomicconditionascomparedtoothersinyour
town?
①Verygood ②Good ③Fair ④Poor ⑤Verypoor
6. Doyouhavethefollowingitemsathome?
325
①airconditioner ②washingmachine ③television ④refrigerator
⑤computer/laptop ⑥motorcycle
7. Howmanysiblingsdoyouhave?______
Howmanyofthemareyoureldersister,youngersister,elderbrother,andyounger
brother,respectively?___________________________
8. Is your father living away from home for work (or business) now? ①yes,
currently ②yes,previously③no,never(ifno,gotoquestion11)
Ifyes,howoldwereyouwhenyourfatherlefthomeforwork(orbusiness)forthe
firsttime?______
Howlonghashebeenawayafteryouwereborn?______
Whichisthemostrecentplacehehasbeento?______
Howmanytimesdidhecomebackhomeduringthelastyear?______
Eachtimehecameback,howmanydaysonaveragedidhestayfor?______
9. Is your mother living away from home for work (or business) now? ①yes,
currently ②yes,previously③no,never(ifno,gotoquestion11)
Ifyes,howoldwereyouwhenyourmotherlefthomeforwork(orbusiness)forthe
firsttime?______
Howlonghasshebeenawayafteryouwereborn?______
Whichisthemostrecentplaceshehasbeento?______
Howmanytimesdidshecomebackhomeduringthelastyear?______
Eachtimeshecameback,howmanydaysonaveragedidshestayfor?______
10. (If both your parents live with you, skip this question) How many times have you
talkedwithyourmotherorfatherinthepastonemonthbyphoneoronQQ?______
326
11. Howmuchpocketmoneydoyougeteachmonth?______
12. Whoareyoulivingwithnow?(multiplechoices)
①yourfather’sfather②yourfather’smother③yourmother’sfather ④your
mother’smother ⑤yourfather ⑥yourmother ⑦otherrelatives
⑧neighbour ⑨nobody⑩otherpeople
Amongthem,whotakescareofyoumost?______
Isthatpersonmaleoffemale?______
Howoldishe/she?______
His/hereducationlevel?______
①primaryschool/neverwenttoschool ②juniormiddleschool ③highschool
④vocationalcollege ⑤universityorhigher
Wealsowanttoknowhowthingsaregoingforyouatschool.Pleasecompletethe
itemsaccordingtoyouractualsituationandfeelings.
13. Idon'twanttogotoschool
Nottrue Somewhattrue Certainlytrue
14. Ifindschoolinteresting/Ienjoyschoolactivities
Nottrue Somewhattrue Certainlytrue
15. Iworryaboutmyfuture
Nottrue Somewhattrue Certainlytrue
327
16. Iworryaboutgettingbadmarksorgrades
Nottrue Somewhattrue Certainlytrue
17. Itishardtopayattentioninclass
Nottrue Somewhattrue Certainlytrue
18. Ieasilyforgetthings
Nottrue Somewhattrue Certainlytrue
Nottrue Somewhattrue Certainlytrue
19. Ioftenanswerquestionsinclass.
Nottrue Somewhattrue Certainlytrue
20. Ifinishmyhomeworkontime
Nottrue Somewhattrue Certainlytrue
21. Ifinishmyhomeworkonmyown
Nottrue Somewhattrue Certainlytrue
22. Generallyhowareyoudoingwithyourstudyamongallstudentsinyourclass?
Top Middle Bottom
328
Wewould also like to knowwhether youdo things below to further understand
youandotherchildren.Yourparentsandteacherswilldefinitelyneverknowyour
answer.
23. Smoking:
Never sometimes less than 1 cigarette per week more than 1
cigaretteperweek
24. Drinkingalcohol
Never sometimes less than once a month more than once a
month
25. Skipclasses
Never Sometimes Manytimes
26. Cheatinexams
Never Sometimes Manytimes
27. Gotointernetbars
Never Sometimes Manytimes
28. Helpwithhousework
Never Sometimes Manytimes
29. Helpwithfarmworkorfamilybusiness
Never Sometimes Manytimes
30. Stealthings
Never Sometimes Manytimes
31. Lying
329
Never Sometimes Manytimes
32. Vandalism
Never Sometimes Manytimes
33. Bullyclassmatesorotherchildren
Never Sometimes Manytimes
34. Bulliedbyclassmatesorotherchildren
Never Sometimes Manytimes
35. Physicallyfightwithpeople
Never Sometimes Manytimes
Pleasecompletethefollowingitemsaccordingtoyourexperiencesduringthepast
sixmonths.Thereisnotanyrightorwronganswer,justtelluswhatyouconsider
tobetrue.
36. Itrytobenicetootherpeople.Icareabouttheirfeelings
Nottrue Somewhattrue Certainlytrue
37. Iamrestless,Icannotstaystillforlong
Nottrue Somewhattrue Certainlytrue
38. Igetalotofheadaches,stomach-achesorsickness
Nottrue Somewhattrue Certainlytrue
330
39. Iusuallysharewithothers(food,games,pensetc.)
Nottrue Somewhattrue Certainlytrue
40. Igetveryangryandoftenlosemytemper
Nottrue Somewhattrue Certainlytrue
41. Iamusuallyonmyown.Igenerallyplayaloneorkeeptomyself
Nottrue Somewhattrue Certainlytrue
42. IusuallydoasIamtold
Nottrue Somewhattrue Certainlytrue
43. Iworryalot
Nottrue Somewhattrue Certainlytrue
44. Iamhelpfulifsomeoneishurt,upsetorfeelingill
Nottrue Somewhattrue Certainlytrue
45. Iamconstantlyfidgetingorsquirming
Nottrue Somewhattrue Certainlytrue
46. Ihaveonegoodfriendormore
331
Nottrue Somewhattrue Certainlytrue
47. Ifightalot.IcanmakeotherpeopledowhatIwant
Nottrue Somewhattrue Certainlytrue
48. Iamoftenunhappy,down-heartedortearful
Nottrue Somewhattrue Certainlytrue
49. Otherpeoplemyagegenerallylikeme
Nottrue Somewhattrue Certainlytrue
50. Iameasilydistracted,Ifinditdifficulttoconcentrate
Nottrue Somewhattrue Certainlytrue
51. Iamnervousinnewsituations.Ieasilyloseconfidence
Nottrue Somewhattrue Certainlytrue
52. Iamkindtoyoungerchildren
Nottrue Somewhattrue Certainlytrue
53. Iamoftenaccusedoflyingorcheating
Nottrue Somewhattrue Certainlytrue
332
54. Otherchildrenoryoungpeoplepickonmeorbullyme
Nottrue Somewhattrue Certainlytrue
55. Ioftenvolunteertohelpothers(parents,teachers,children)
Nottrue Somewhattrue Certainlytrue
56. IthinkbeforeIdothings
Nottrue Somewhattrue Certainlytrue
57. Itakethingsthatarenotminefromhome,schoolorelsewhere
Nottrue Somewhattrue Certainlytrue
58. Igetonbetterwithadultsthanwithpeoplemyownage
Nottrue Somewhattrue Certainlytrue
59. Ihavemanyfears.Iameasilyscared
Nottrue Somewhattrue Certainlytrue
60. IfinishtheworkI'mdoing.Myattentionisgood
Nottrue Somewhattrue Certainlytrue
Pleasecompletethefollowing4itemsifyourbothyourparentsdonotlivewithyou
333
61. Imissmyparent(s)alot
Nottrue Somewhattrue Certainlytrue
62. Ifeelsadeachtimemyparentsleavehomeforwork
Nottrue Somewhattrue Certainlytrue
63. Ifeellonelyathome
Nottrue Somewhattrue Certainlytrue
64. Iwishmyparentswereathome
Nottrue Somewhattrue Certainlytrue
Whowouldhelpyouortalkwithyouinthefollowingsituations?
65. Whenyouwerehavingaproblemwithyourstudies
Parent Caregivers (if not living with parents) Teachers
Friends/Classmates
Others Nobody
66. Whenyouwereworriedordepressedaboutsomething
Parent Caregivers (if not living with parents) Teachers
Friends/Classmates
334
Others Nobody
67. Whenyouwerebeingteasedorbulliedbyanotherchild
Parent Caregivers (if not living with parents) Teachers
Friends/Classmates
Others Nobody
68. Whowouldyoushareyourhappiness(e.g.didgreatinexam)with?
Parent Caregivers (if not living with parents) Teachers
Friends/Classmates
Others Nobody
69. Whowouldyoutalktowhenyouareunhappy?
Parent Caregivers (if not living with parents) Teachers
Friends/Classmates
Others Nobody
70. Whowoulddisciplineyouortellyouoffwhenyoudidsomethingwrong?
Parent Caregivers (if not living with parents) Teachers
Friends/Classmates
Others Nobody
71. Whocaresforyoumost? ____________
335
72. Whodoyoulikemost? ____________
73. Whatdoyouwishtodowhenyougrowup? _______________
Please choose from options in the items below according to your thoughts and
feelingsduringthepasttwoweeks.
74. ①Irarelyfeelunhappy. ②Isometimesfeelunhappy ③Ioftenfeelunhappy
75. ①Ioftenwanttocry ②Isometimeswanttocry ③Irarelywanttocry
76. ①Iwouldliketobewithpeople ②Ioftendon'tliketobewithpeople
③Idon'tliketobewithpeopleatall
77. ①Irarelyfeellonely ②Isometimesfeellonely ③Ioftenfeellonely
78. ①Noneoftheadultsreallyloveme ②I'mnotsureifanyadultloveme
③I'msuresomeadultloveme
79. ①Iwanttocommitsuicide ②Ithinkaboutsuicide,butwon'tdoit.
③Ineverthoughtaboutsuicide.
Pleasetellusaboutyourfriends.
80. Duringthepastyear,didanyofyourfriends
1) drinkfrequently? ①Yes ②No
2) smoke? ① Yes ②
No
3) physicallyfightwithpeople? ①Yes ②No
4) gotointernetbarfrequently? ①Yes ②No
336
5) steal? ①Yes ②No
6) oftenskipclasses? ①Yes ②No
7) oftencheatinexams ①Yes ②No
8) spendmoneylavishly? ①Yes ②No
9) oftendisobeyteachersandparents? ①Yes ②No
10) hangoutwithbadguys?①Yes ②No
11) verballybullyotherchildren?(e.g.curseormockothers)①Yes ②No
12) physicallybullyotherchildren?(e.g.hitothers) ①Yes ②No
Wewould like toknow ifyousuffer fromviolence fromyourparents.During the
pastyear,didyourparents:
81. Screamedatyouveryloudandaggressively,saidmeanthingsorcursedyou?
Manytimes Sometimes Never Notinthepastyearbutthis
hashappened
82. Threatenedtoleaveyouorabandonyou?
Manytimes Sometimes Never Not in the past year but
thishashappened
83. Hitorslappedonyourface?
Manytimes Sometimes Never Notinthepastyear
84. Beatenorphysicallyhurtyoubyhand?
Manytimes Sometimes Never Notinthepastyear
337
85. Beatenorphysicallyhurtyouwithimplementslikesticksandwhips?
Manytimes Sometimes Never Notinthepastyear
86. Locked you out of the home, made you stand or do exercise for a long time as
punishment?
Manytimes Sometimes Never Not in the past year but
thishashappened
87. Orotherkindsofpunishmenttoyou?_____________________________________
Wewouldliketoknowifyousufferfromviolencefromyourcaregivers.Duringthe
pastyear,didyourcaregivers:
88. Screamedatyouveryloudandaggressively,saidmeanthingsorcursedyou?
Manytimes Sometimes Never Notinthepastyearbutthis
hashappened
89. Threatenedtoleaveyouorabandonyou?
Manytimes Sometimes Never Not in the past year but
thishashappened
338
90. Hitorslappedonyourface?
Manytimes Sometimes Never Notinthepastyear
91. Beatenorphysicallyhurtyoubyhand?
Manytimes Sometimes Never Notinthepastyear
92. Beatenorphysicallyhurtyouwithimplementslikesticksandwhips?
Manytimes Sometimes Never Notinthepastyear
93. Locked you out of the home, made you stand or do exercise for a long time as
punishment?
Manytimes Sometimes Never Not in the past year but
thishashappened
94. Orotherkindsofpunishmenttoyou?_____________________________________
Yoursafetyisthetopconcernofyourparentsandteachers-pleaseletusknowif
youhavebeenhurtinthefollowingwaysduringthepast12months.
95. Mechanicalinjury(suchasbonefracture,sprain): ①Yes,____times; ②No
96. Animalbites: ①Yes,____times; ②No
97. Burn: ①Yes,____times; ②No
98. Drowning: ①Yes,____times; ②No
99. Trafficaccident: ①Yes,____times; ②No
100. Others
101. Have you tried to hurt yourself during the past year? (e.g. by cutting, scratching,
jumpingfromhighplaces,overdosing,swallowingindigestiblethings):
339
①Never ②Once ③Morethanonce
Thankssomuchforyourhelp!
Wasthisahardquestionnairetoanswer?
Yes No
Isthereanythingthatyoudidn’tunderstand?
Yes No
Isthereanythingelseyouwouldliketosayaboutwhathappenedtoyouorabout
fillinginthequestionnaire?
Top Related