Digital and Social Media for SBCC March 2019 · Digital and Social Media for Social and Behaviour Change Communication (SBCC) 4 SKAM, a Norwegian program dealing with teen girls’

DigitalandSocialMediaforSocialandBehaviourChangeCommunication

LindaRaftreeforiMediaAssociates

March2019

© March 2019, iMedia Associates and Gates Foundation. This report provides an overview of the uses of digital and social media in social and behaviour change communication initiatives. It is licensed under a creative commons attribution-noncommercial-sharealike license.

Creative Commons: Attribution-NonCommercial-ShareAlike Please cite this work as follows: Produced by iMedia Associates. Author: Linda Raftree. Editor: Nicola Harford. Cover Photo: Emrys Schoemaker

Digital and Social Media for Social and Behaviour Change Communication (SBCC) 2


LindaRaftreeforiMediaAssociates

ExecutiveSummaryAsaccesstomobilephonesandtheInternethasexpandedoverthepastdecade,sotoohave

effortstoincorporatenewmediachannels,digitalplatformsandsocialmediaintosocialandbehaviourchangecommunication(SBCC)efforts.IntheGlobalSouth,accessandusearestillnot

synonymous,however.Althoughyoungpeopleinlow-incomesettingsindeedaccessmobilephones,theirsisnotanexperienceof“alwayson,alwaysconnected.”Formost,theonlineexperienceismobile-ratherthancomputer-based.Whilethereisnosuchthingas

homogeneousmobileorInternetaccess,manyyoungpeopleinlower-incomecontextsshareandborrowphonesfromfriendsandrelativesratherthanowningaphone.Thesephonesmaybelow-endphones,second-handphones,orpiratedphonesthatdonotofferconsistent

reliabilityandwhosefeaturesarebehindthemarket.Despitethechallenges,youngpeopletheworldoverdofindwaystonavigatefinancialand

socialconstraintssothattheycanparticipateinadolescentandteensociallifeandyouthculture.Education-entertainmentprogrammingcantapintotheseurgesandcapitalizeonmulti-modal,transmediaapproachestoreachandengagewithyoungpeople.Thekeyisto

understandyouth’sinterestsandneeds,theirsocioeconomiccontexts,andtheirmediaandtechnologyhabitsandpreferencesandtohaveastrongunderstandingofsocialandbehaviourchangetheoriesandpractices.

Thispaperexploresthewaysinwhichdigitalandsocialmediachannelscanbeusedsuccessfully,appropriately,andresponsiblytodrivebehaviourchangeinyoungmenand

womenfromlowersocioeconomicgroups.Weexaminedpeer-reviewedliterature,greyliterature,NGOresearchandevaluationreports,industryframeworkandguidancedocuments,andarticlesandblogpostsinordertoexplorerelevantSBCCtheories,models,frameworksand

approaches;touncoverlearningfromcurrentandpastdigitalSBCCinitiatives;toidentifygoodpracticeandcommonpitfallsindigitalSBCCdesign,tolayoutgoodpracticesfordigitalsafeguarding;andtoprovideafewexamplesofhowmonitoringandevaluationandlearningcan

bedesignedforSBCCinitiatives.

SBCCtheories,models,frameworksandapproachesMostSBCCframeworkshaveemergedfromthehealthsector,andexistingtheoriesareofteninadequateformobileordigitalSBCCduetotheirinteractiveandadaptivenature.DigitalSBCCinterventionsshouldbegroundedintheory,andtheymaybemoreapplicableiftheyareco-


designedwithusersandalignedwiththeirmediausagepreferencesandtheirinformationseekinghabitsandbehaviours.

Theories,models,frameworksandapproachesthathavesomerelevancefordigitalSBCCinclude:

• Trans-theoretical(StagesofChange)Model• SocialCognitiveTheory• BehaviouralEconomicsandNudgeTheory

• Mentalmodels• EcologicalModelsofBehaviourChange• SocialNormTheories

• NarrativePersuasionTheories• DiffusionofInnovationTheory

Despitelendingthemselvestonewwaysofreachingandengagingwithtargetgroups,someofthesetheoriesraiseethicalissuesduetotheirpotentialformisuseormanipulation,andpromptquestionssuchas‘whodecideswhatisdesirablebehaviour?’

WhatdocurrentandpastdigitalSBCCinitiativestellus?Severalstudiesconductedoverthepastdecadehavefoundthatteensandyouthareopento

textmessagesforhealthpromotion,andtheyfindthemappealingandculturallyacceptable.BenefitsofdigitalandsocialmediaforSBCCinclude:

• increasedinteractionwithothers;

• moreavailable,shared,andtailoredinformation;• increasedaccessibilityandwideningaccesstohealthinformation;• peer/social/emotionalsupport;

• publichealthsurveillance;and• potentialtoinfluencehealthpolicy.

Behaviourchangeinterventionsappeartobemosteffectivewhenpeopleareengagedonmultiplelevels.Reminderscanbesuccessfulforsimplerbehaviours,andthereisvariedevidenceonwhethervirtualbehaviourchangeeffortsareequallyassuccessfulasofflineones,orif

blendedapproachesthatcombineonlineandofflineapproachesworkbestandwhen.Usersarelargelyunawareofdataprivacyandconfidentiality,datasecurityandthedangersofdisclosingpersonalhealthinformationonlineoroverviainsecuremeanssuchastextmessage.Privacyis

particularlychallenginginenvironmentswheremobilesaresharedorborrowed.Someentertainment-educationexamplestoexplorefurtherinclude16andPregnant,which

mayhaveinfluencedtherateofteenpregnanciesintheUS;EastLosHigh,whichwassuccessfulatreachingAmericanLatina/oyouthwithsexualandreproductivehealthmessages;theVoicesforChangebroadprogramthataimedtochangesocialnormsaroundgenderinNigeria;and


SKAM,aNorwegianprogramdealingwithteengirls’day-to-daychallengeswhichbecameaculthitamongglobalteens.

GoodpracticeandcommonpitfallsindigitalSBCCdesignAnumberofgoodpracticesarehighlightedinpublishedliteratureandindocumentationfrompractitionersandorganizations,including:

Groundeffortsintheoryandfindwaystounderstandandrespondtothecomplexityof

underlyingdeterminantsofbehaviour.

Understandthetargetpopulationandthesocialcontext.Successfulapproachesarecontext-

specificandrootedinanunderstandingofbehaviour,itssocial,environmental,economicandmotivationaldeterminants,andobstaclestochange.Programsshouldaddresswidereconomicandculturalbarrierstobehaviourchange.

Understandanddesignforuserhabits,preferences,andinterests.Thisincludesthedevices,channels,mediasources,influencers,datapackages,networkstrength,andotherelementsthat

aredirectlyrelatedtowhetherandhowpeopleaccessandusedigitaldevices,servicesandplatforms.

Watchoutforcostanddatalimitations.Low-incomeuserstendtohavefrugalaccesshabits.SBCCprogramsshouldcarefullyconsiderhowtominimizedataconsumptionbydesigningspecificallyforthosewhohavelimitedminutesortextmessagesallowed.

Carefullythinkaboutlanguageandliteracy.Textmessagesforhealthinformation,motivatingindividuals,andencouragingself-managementispopular,howeverthesedonotworkwhere

thereislowliteracy.Languagecanalsobeapoliticizingfactor,andlanguagechoicesshouldbemadecarefullytoavoidcreatingconflictorexclusion.

Determinetherightfrequencyofmessagingorengagement.ThoughSMS-basedreminderstriggerindividualstothinkabouttopicstheymightnototherwiseconsciouslyconsider,oronlyconsideratcertaintimes,ifmessagesareseenas‘nagging’theywillnotbeeffective.

Buildtrustandfindtherightvoiceandmessenger(s).Tomaximizelikelihoodofbehaviourchange,therecipientneedstobelievethesenderofthemessageisacrediblesourceof

information.Findingtrustedmessengersandinfluencersandtherighttoneisaknowngoodpracticefrompre-Internetdays,anditholdstruefordigitalSBCC.

Preparewithsignpostingandensurecapacitytorespondtodemandgenerated.SBCCprogramsshouldplanforhowtheywillrespondtoanydemandcreatedbyincreasedawareness.Reliable,trustworthy,vettedonlineandofflinesupportsystemsandservicesshould

beinplacetowhichuserscanbesignposted.


Budgetformoderationofcontentandcomments.BothFacebookandWhatsApphavebeen

usedtospreadfalseandmisleadinginformationthatmanybelievehasinfluencedelectionprocesses,fueledhateandviolence,andledtowiderdivisionsontopicslikeraceandgender.Goodmoderationcanhelptostemfalseormisleadinginformationfromspreadingonsocial

channels.

Don’tbefooledbyvanitymetrics.Vanitymetricsarethosethatfocusonlargenumbersof

reachorengagement:forexample,pageviews,clicks,orlikes,yetsaylittleaboutthequalityofengagementorthewiderimpactonbehaviourchange.

DigitalSafeguardinggoodpracticesTheumbrellaof“DigitalSafeguarding”includesa)safeguardingwhenpeopleusedigitaldevicesorengagewithdigitalservicesorplatforms,b)protectingdataanddataprivacy,andc)securing

data.Womenandgirlsareespeciallyvulnerabletoonlineviolence,harassmentandabuse,aswellasLGBTQidentifyingpersonsandlowersocietalstrata,suchastheDalit,insomecases.

Therearemultiplesafeguardingareastoconsiderwhenengagingadolescentsindigitalinitiatives,including:

• bigpictureethicalandriskreview(isembarkingonthiseffortethical?);

• safeguarding,informedconsent,privacyandsecurityofdatacollectedduringformativeorongoingresearchactivities;

• buildingprivacyanddatasecurityintoplatform,productoroutreachdesign;

• ensuringthatthebenefitsofcollectingpersonalorsensitivedataoutweightheprivacyrisks;

• followingapplicablecountryandgloballegalframeworksrelatedtodataprivacyand

transmissionacrossborders;• protectingandsecuringdatacollectedorgeneratedthroughouttheengagementprocess;• ensuringclarityonhowandwheredataflowsandreviewinghowpartnerswillmanage

dataprivatelyandsecurely;and• considerationsaroundtheonlineexperienceitselfsuchassafeandappropriatecontent,

commentsandmoderation,andprotectionfromonlineabuse,bullying,harassment,

hatespeech,violence,scamsorgrooming.

Monitoring,evaluationandlearningfordigitalSBCCTheevidencebaseondigitalSBCCisstillweakanditisrecommendedthatfuturestudiesadopt

rigorousdesignsandendpoints.Creativeapproachestomonitoring,evaluationandlearning(MEL)ofdigitalSBCCincludethosefoundinthefollowingprograms:16andPregnant,EastLosHigh,VoicesforChange,SpringsterandHNI’s3-2-1.Somedigitaltoolsandemergingmethods,

includinginteractivevoiceresponse(IVR),Internetmessagingplatforms,andbigdatacanenableinteractive,engagingand/orinnovativeevaluationprocesses.


ExecutiveSummary .....................................................................................................2

Acronymsandabbreviations........................................................................................7

Glossaryofterms.........................................................................................................8

Introduction.................................................................................................................9

1.Mobileaccessanduse...........................................................................................10

2.Relevanttheories,models,frameworksandapproaches.......................................11

3.WhatdocurrentandpastdigitalSBCCinitiativestellus? ......................................18SMSforhealthbehaviourchange .................................................................................................................... 19Socialmedia,multi-mediaandtransmedia ................................................................................................. 2016andPregnant.......................................................................................................................................................21EastLosHigh .............................................................................................................................................................22VoicesforChange(V4C) .......................................................................................................................................22SKAM .............................................................................................................................................................................25

Virtualversusin-personbehaviourchange ................................................................................................ 264.GoodpracticeandcommonpitfallsindigitalSBCCdesign ......................................28

5.DigitalSafeguardinggoodpractices........................................................................34Genderdimensionsofonlineharassment,violenceandabuse.......................................................... 34Emergingprivacyregulations ........................................................................................................................... 35DigitalSafeguardingandResponsibleDatainthecivilsocietyspace ............................................. 36

6.Monitoring,evaluationandlearningfordigitalSBCC..............................................3716andPregnant ...................................................................................................................................................... 38EastLosHigh ............................................................................................................................................................ 38VoicesforChange(V4C) ...................................................................................................................................... 39Springster................................................................................................................................................................... 40HNI3-2-1Program................................................................................................................................................. 41ICTtoolsandtechnologyapproachesforMELofSBCCinterventions ............................................ 42Socialmediacommentsanalysisforsocialnormsresearch .................................................................42IVRforMonitoringandEvaluatingMediaInterventions ......................................................................42UsingInternetMessagingplatformsforMEL .............................................................................................43Bigdataandevaluation........................................................................................................................................44

7.Conclusion .............................................................................................................45

References.................................................................................................................46


AcronymsandabbreviationsAIDA: Awareness,Interest,Desire,Actionframework

ASRH: AdolescentSexualReproductiveHealth

BCC: BehaviourChangeCommunication

GDPR: GeneralDataProtectionRegulation

ICT: InformationandCommunicationTechnology

(I)NGO: (International)Non-GovernmentalOrganization

IVR: InteractiveVoiceResponse

MEL: Monitoring,EvaluationandLearning

MERL: Monitoring,Evaluation,ResearchandLearning

SBCC: SocialandBehaviourChangeCommunication

SMS: ShortMessageService

STI: SexuallyTransmittedInfection

ToC: TheoryofChange

TTM: Trans-theoreticalModelofBehaviourChange

V4C: VoicesforChange


GlossaryoftermsBasicphone:amobilephonethatcanonlycallortext,doesnothaveInternet(alsoknownasa

featurephone)

Chatbot:amenu-basedtextserviceswithautomatedreplies

Choicearchitecture:theorderinwhichchoicesareprovidedonamenuorsurvey

Datapackages:pre-paiddataor‘airtime’thatispurchasedforusingamobile

Discoverability:levelofeasewithwhichaplatformorserviceisdiscovered/found

GroundTruthing:furtherinvestigationofdatagatheredremotely,oftenviadigital

InternetMessaging:platformslikeWhatsApporFacebookMessenger

InteractiveVoiceResponse(IVR):avoice-andmenu-basedserviceforsharinginformation

mHealth:mobilehealthinterventions

Mobileinternet:Internetaccessedviamobilephone,ratherthanbroadbandorwifi

Smartphone:amobilephonethatcanaccesstheinternetandloadapplications

SMS:shortmessageservice,inotherwords,atextmessage

Transmedia:anarrativeorprojectusingmultipleplatformsandformatsincludingcurrentdigital

technologies

USSD:aformoftextmessagethatusesstarsandhashtags;commoninmobilefinance



LindaRaftreeforiMedia

IntroductionAsaccesstomobilephonesandtheInternethaveexpandedoverthepastdecade,sotoohaveeffortstoincorporatenewmediachannels,digitalplatformsandsocialmediaintosocialand

behaviourchangecommunicationefforts.Inthe‘GlobalSouth,’accessanduseisstillnotthesamething.Youngpeopledoaccessmobilephones,however,unliketheircounterpartsintheGlobalNorth,theirsisnotanexperienceof“alwayson,alwaysconnected”andformost,the

onlineexperienceisfullymobilephone-based.WhilethereisnosuchthingashomogeneousmobileorInternetaccess,itiscommonforyoungpeopleinlower-incomecontextstoshareandborrowphonesfromfriendsandrelativesratherthantoowntheirownphone.Thesephones

maybelow-endphones,second-handphones,orpiratedphonesthatdonotofferconsistentreliabilityandwhosefeaturesareseveralyearsbehindthemarket.Girlsinsomecontextsenjoylowerlevelsofaccessandusethanboys,andyouthwithlessdisposableincomealsoexperience

lowerlevelsofaccessanduse.Youngpeopletheworldover;however,findwaystonavigatefinancialandsocialconstraintsso

thattheycanparticipateinadolescentandteensociallifeandyouthculture.Education-entertainmentprogrammingcantapintothesetrendsandcapitalizeonmulti-modal,transmediaapproachestoreachingandengagingwithyoungpeople.Thekeyistounderstand

youth’sinterestsandneeds,theirsocio-economiccontexts,andtheirmediaandtechnologyhabitsandpreferencesalongwithastrongunderstandingofsocialandbehaviourchangetheoriesandpractices.

Thispaperexploresthewaysinwhichcurrentandemergingdigitalandsocialmediachannelscanbeusedsuccessfully,appropriately,andresponsiblytodrivebehaviourchangeinyoung

menandwomenfromlowersocioeconomicgroups.Weexaminedpeer-reviewedliterature,greyliterature,NGOresearchandevaluationreports,industryframeworkandguidancedocuments,andarticlesandblogpostsinordertoexplorerelevantsocialandbehaviourchange

communication(SBCC)theories,models,frameworksandapproaches;touncoverlearningfromcurrentandpastdigitalSBCCinitiatives;toidentifygoodpracticeandcommonpitfallsindigitalSBCCdesign,tolayoutgoodpracticesfordigitalsafeguarding;andtoprovideafewexamplesof

howmonitoringandevaluationandlearningcanbedesignedorSBCCinitiatives.


1.MobileaccessanduseA2018PewResearchCenterstudyshowsariseinthenumberofpeoplewhousetheInternet,asmartphoneorsocialmediainemergingeconomies/developingeconomies.Overthepastfive

years,globalInternetusehasrisenfrom42to64percent,smartphoneusefrom24to42percentandonlinesocialnetworkingfrom34to53percent.Internetandsocialmediausecontinuetolaginsomedevelopingeconomies,however.InIndiainparticular,35percentof

peopleage18-36reportedoccasionallyusingtheInternetorowningasmartphone(asopposedtojust13percentofthoseover37yearsold).53percentofthosewithahighereducationlevelreportedInternetand/orsmartphoneuseversus11percentofthoseinlowereducation

brackets.Overall,justone-in-fiveadultsinIndiausesocialnetworkingplatformssuchasFacebook,andagenderdivideexists,with28percentofadultmalesusingsocialnetworkingplatformsascomparedto11percentofwomen.Sub-SaharanAfricaisalsobehind,asoneofthe

leastconnectedpartsoftheglobe.Developednationuse,ontheotherhand,hasplateauedoverthepastfewyears(Poushter,BishopandChwe2018).

A2018VodafoneFoundationandGirlEffectstudythatincludedoverthreethousandgirlsintwenty-fivecountriesemphasizes,“Thereisnosuchthingashomogenousaccess.”Inparticular,themobilegendergapishighlycontextdependent,withanumberofintersectingfactorsand

implications.Sincegirlsoftengainaccessbyborrowingand/orsharingfriendsandrelatives’phones,theymaybelesslikelytohavetheirownaccountsorenjoyprivacyofusage,sotheytendtousephonesmorecautiously.Whengirlscanaccessphones,theyusethemasmuchas

possible.Overhalfofthegirlsincludedinthestudysaidtheyuseaphoneeverydayormultipletimesaday,anda32%saidtheyuseaphoneatleastonceaweek.Thestudyhighlightsthatwhilebothboysandgirlsfacecostbarriers,girlsalsohavesafetyconcernsrelatedtomobile

use,andsodotheirparents(VodafoneFoundationandGirlEffect2018).Thedistinctionbetweenaccessanduseiscritical(Banaji,Livingstone,etal.2018)andshould

beconsideredwhendesigningdigitalorsocialmediaoutreachbecausephoneownershipandusagecanbetransient.IntheGlobalNorththereisnomeaningfuldifferencebetweenmobileaccessanduse,butintheGlobalSouth,thisisnotthecase.Youngpeopleoftenborrowrather

thanownmobiles.Youthmayselltheirphonesiftheyneedcash,andmobilephoneborrowship1isinfluencedbyfactorssuchasfamilystatus(e.g.fosterstatus),householdresourcesand

livelihoodpatterns,andintergenerationalrelationships(Porter,etal.2012)(Banaji2017).Youngpeopleusedifferenttacticsusedtoovercomeconstraintsthataffecttheirmediause--rationingphoneusetosavebatterylife,limitingphonefunctionstosavedatacosts,findingfreeorcheap

gamesorapplications,deletinggamestofreeupstoragespace,andavoidingcommercialtacticstoeatintoairtime(WaltonandPallitt2012).Theserealitiesandbehavioursneedtobeconsideredwhendesigningdigitalsocialandbehaviourchangecommunications.

1Theterm“borrowship”iscommonlyusedinthemobilespacetodistinguishbetweenthosewhoareowners(mobilephoneownership)andthosewhoshareorborrowmobilephones(mobilephoneborrowship).


2.Relevanttheories,models,frameworksandapproachesMostSBCCframeworksandevidencehaveemergedfromthehealthsector.Reviewsofmobilehealthbehaviourinterventionshaveconcludedthattheywouldbenefitfromgreaterapplication

ofbehaviouraltheories,yetexistingtheoriesareofteninadequateforinformingmobileSBCC,whichismoreinteractiveandadaptive(Riley,Rivera,etal.n.d.).Evenwheninterventionsareinitiallyinformedbytheory,theydonotalwaysreflectatheory-basedapproachthroughout

theirimplementation.Additionally,availabletheoriesdonotalwaysmeshwiththedynamicnatureofmobiledevicesandsocialmedia(Shane-Simpson,etal.2017)(Painter,HynesandGlanz2008)(Riley,Rivera,etal.2011).Thefieldisstilllearninghowtomanagethecomplexmix

ofingredientsinamobileeffort.Inotherwords,thosedesigningmobileSBCCinitiativesmustconsiderthecapabilitiesandaffordancesofaparticulardevice(basicphone?smartphone?);thenatureofthechannel(SMS?Call?InternetMessaging?);whatcanbecarriedonthechannel

(Text?Voice?Video?);andtheinterests,habits,incomes,needsandcharacteristicsofthetargetpopulation(HealthCommunicationCapacityCollaborative(HC3)2013).

InaliteraturereviewconductedforGirlEffect,thebehaviourchangeresearchgroupNewKnowledgeexploredpossibilitiesandlimitationsrelatedtoknowledge,attitude,andpractice(behaviour)changethatcanbesupportedondigitalplatforms.Theyfoundthatdigital

interventionsthatlinktheoryandpracticetochangenormsareuncommonandthatexistingeffortsarestillinearlystages.Theyrecommendaholisticapproachthatgoesbeyondafocusonjustknowledgeandconfidenceandthatalsofindswaystoaddresspotentialbarriersto

behaviourchange.Thereisalsoconsistentagreementthatdigitalbehaviourchangeinterventionsdesignedforadolescentsshouldbeco-designedwithyoungpeople,funandengaging,andalignedwithhowtheyusemedia/socialmediaandtheirinformationseeking

habitsandbehaviours.Theliteraturethatfollowsdigitalbehaviourchangeinterventionsoveranextendedperiodoftimeislimited,andmuchoftheliteratureiscontext-specific.Thismakesitdifficulttocompareacrossmultipleinterventionsduetothevariationfoundamongdigital

platforms.Forthisreason,questionsremainaboutsustainabilityanddurationofchangeandabouthowmuchchangeneedstohappenoverhowlongforittobedurableforthelong-term(Shane-Simpson,etal.2017).

Belowwelistsomeofthetheories,models,frameworksandapproachesthataremost

relevantfordigitalsocialandbehaviourchangecommunicationwork:TheTrans-theoretical(StagesofChange)Model(TTM)featuresascaleformeasuringan

individual’sself-efficacy(confidenceinone’sownabilitytoachieveadesiredactionorchange)andoutlinesfivestagesofchange:pre-contemplation(nointentiontotakeactionwithinthenextsixmonths),contemplation(anintenttotakeactionwithinthenextsixmonths),

preparation(anintenttotakeactionwithinthenext30daysandhastakensomebehaviouralstepsinthisdirection),action(changedovertbehaviourforlessthansixmonths),and


maintenance(changedovertbehaviourformorethansixmonths).Thesubsequentstage,characterizedbynotemptationtorelapseandcompleteconfidenceinone’sabilitytomaintain

thechangedbehaviourpattern,istheterminationphase.WhenitcomestoTTM,interventionswithhighlevelsofutilityandimpacthavethemost

potentialforchange.Itmaybeusefultoestablishbothautilitycriterionthatcandemonstrateefficacyinchangingagivenbehaviour,andanimpactcriterionthatcandemonstrateinterventioneffectivenessintermsofmeasurablechangeamongthetargetpopulation

(Prochaska,WrightandVelicer2008).TTMdoesnotconsidersocialcontextanditsinfluenceonindividuals;ratheritassumesindividualsmakelogicaldecisionsateachstage.Additionally,theTTMmustaddresseachstageseparatelytobemosteffective.TTMhasbeenappliedtotailored

healthbehaviourchangeinterventionssuchasdietarychange(Barnett,etal.2016).MaxSt.JohnoffersaformulaforusingTMMwithsocialmediaSBCC(StJohn2011).Itincludes

phasesofinsight,interventionandevaluation.Attheinsightphase,herecommendsdoingresearchandtalkingwithuserstounderstand:Whoistheaudience?Wheredotheygoandwhatdotheydo?Wheredotheyhangout?Whataretheirattitudesandlanguage?Whodo

theylistento?Whoaretheirinfluencers?Attheinterventionphase,herecommendstyinggamingandgametheoryintoTMM.Duringthe

pre-contemplation(“I’munawarethatIhaveaproblem”)asoftintroductionoftheissuesisintroducedviagameplay.Atthecontemplationphase(“Ineedtodosomethingaboutthis”)use

educationthroughcommunityengagementviaaFacebookpage.Atthepreparationphase(“IknowwhatI’mgoingtochange”)providerecipesandothertakeawaystomakechangeeasy.Attheactionphase(“I’verecentlychangedmybehaviour”)encouragetheaudiencetoself-report

throughcomments.Andduringthemaintenancephase(“Ihaven’trelapsed”)uselong-termcontactthroughFacebooktohelpmaintainbehaviours.

Duringtheevaluationphase,oneshouldlookforcontributionandindicatorsthatdemonstratesuccesssincechangeishardtoattribute.Possiblemetricsinclude:awareness(exposuretomessaging,gameplays),contemplation(engagementthroughapage,likesandcomments),and

preparation(uptakeoftools,downloadsofrecipes).Tomeasuresocialreturnoninvestment,onlineandofflinequalitativeandquantitativesurveyscanbeused.Socialmediacanbeusedtomeasuresocialcapitalintermsofcommunitymake-up(sizeandsegmentationbybehaviour),

intensityofuse(regularityanddepthofengagement)andnetworkdata(numberandstrengthofconnections)(StJohn2011).

SocialCognitiveTheory,originallycalledSocialLearningTheorythisisanon-staticandnon-linearmodelshowinghowbeliefsoperatetogetherwithgoals,outcomeexpectations,andperceivedenvironmentalimpedimentsandfacilitatorsintheregulationofmotivation,

behaviour,andthestateofwellbeing.Itpositsthatbehaviouralchangeisembeddedwithinthe


perceptionofaperson’sbeliefthattheycanaccomplishagoal(Bandura2002)(Barnett,etal.2016).

SocialCognitivetheorycanbeapproachedonsocialmediainthefollowingway.Forreciprocaldeterminism(what’stellingmetochange?)oneshouldfindtherightpartners,thebestwaysto

drivepeople/traffictoasite,andtherightsocialmediachannels.Forbehaviouralcapability(amIabletomakethischange?)it’simportanttoprovidetools,advice,trainingandinformation.Self-efficacy(canIreallydothis?)canbeseenasbuildingconfidencethroughtheabilityto

createnewonlinenetworks.Forobservationallearning(iftheycandoit,socanI),it’simportanttopromoterealisticrolemodelsthatproveit’spossible.Finally,toreinforce(howcanIcelebratemyachievement)feedbackmechanismscanprovideencouragement.Henotesit’s

importanttouserolemodelswhoarecredibleexperts,celebrities,individualstoriesofachievement,wisdomofthecrowds(weightofnumbers),andwisdomoffriends(peopleyouknowandtrustinreallife)intheprocess(StJohn2011).

SocialCognitiveTheoryhasstrongtheoreticallinkstobehaviouraleconomics,asBandurawasoneoftheearlytheoristsaroundidentity,whichhasgrowntoholdanimportantspacein

behaviouraleconomictheory.TheWorldBank’sWorld2015DevelopmentReporton‘MindandSociety’isevidenceofthegrowingroleofbehavioureconomicsindevelopment,theimportanceofbehaviourchangeindevelopment,andtheimportanceofidentityinboth(WorldBankGroup

2015).BehaviouralEconomicsandNudgeTheoryemergedin2008fromtheeconomicsfield.Itproposesthat“higherorderpositivepursuitsdrivechangemoreproductivelythan

contemplatingtheeliminationofriskorthreat.”Thistheorypositsthatpeopleemploysimplifyingmodelstoconsideropportunitiesthatusegroupingorschemaswhichidentifygeneralizedmodels,pastpractice,andrecommendations.A‘nudge’directsapersontowarda

specificchoicebychangingthedefaultoption,description,anchor,orreferencepoint.Accordingtothistheory,socialchangecanbeachievedusingacommercialmodelofchoice

directionwhilepreservingaperson’sindividualfreedomofchoice(ThalerandSunstein2008)(Thaler,SunsteinandBalz2014).

Thestructureandmechanismsofsocialmediahaveprovenidealfor‘nudging’becausesocialmediasitescollectandanalyzethepersonaldataoftheirusers(oftenunbeknownsttothem).Thisdata(inthecaseofmassivelypopulationsiteslikeFacebookandon-linedatingplatforms

forexample)canrevealauser’sattitudes,opinions,purchases,habits,frequentedlocations,politicalleanings,sexualorientation,religionandmore.Evenwhenonesitedoesnotholdahugeamountofdataaboutanindividualperson,becauseindividualsaretrackedacrossmultiple

platforms,thecombineddataallowsthemtobetargetedand,insomecases,manipulatedbymaliciousornefariousactors.


MentalModelscanbehelpfulforsituatingtheculturaldeterminantsofbehaviourandthedecision-makingprocess.Peopledrawonconcepts,categories,identities,prototypes,

stereotypes,causalnarratives,andworldviewsfromtheircommunitieswhentheyframeandrespondtoexperiences.Thesementalframeworkshelpmakesenseofhowtheworldworksandone’stheirroleintheit.Mentalmodelsdifferfromsocialnormsbecausetheydonotneedtobe

enforcedbydirectsocialpressureanddonotnecessarilyfocusonspecificbehaviours.Peoplecanholdontomultiple,harmful,and/orevencontradictorymentalmodels,andtheonetheydrawondependsonhowagivencontexttriggersaspecificwayofunderstandingtheworld.

Exposingtargetgroupstoexperiencesthatbroadentheirmentalmodelsmaytriggerachangeinbeliefsaboutwhatapersoniscapableof(WorldBankGroup2015)(Shane-Simpson,etal.2017).

Ethicalchallengeshavebeenraised,however,withregardtonudgingandmentalmodels.Thesecanbeseenaspaternalisticapproacheswhichassumethatdevelopmentagencies,

governments,ortheprivatesectorshouldbenudgingpeopletowardsparticularbehaviours.NewKnowledgehighlightsthatnudgetheoryisseenbymanyaspaternalisticinitsbeliefthatpeoplewanttofollowtheirdesiresbutarecontenttolimittheirchoicestopre-established

optionsandthatpeopleareeasilyswayedtowardssolutionsthatothershaveprioritizedforthem.Theynotetheundueinfluenceofthosewhodetermineandsettheoptionsandhowthoseoptionsarestructuredtoachieveparticularendsandperceived‘choice’(Shane-Simpson,

etal.2017).

ExampleslikethatofCambridgeAnalytica’sinfluencingoftheUSelectionsandBrexitintheUK,andtheaccusationsofRussianmeddlinginglobaldemocraciesusingsocialmediatacticsthatwereperfectedintheUkrainecangivepauseforthinkingabouttheactionsofanykindof

behaviourchangecommunicationcampaignthatisnotfullytransparent.Choicearchitecturecanbeusedtooppressandmanipulatebyencouragingabehaviourbeforeapersonmightwillinglychooseitontheirown,inotherwords,movingsomeonefrompre-contemplationto

actionwithouttheirhavingexercisedagencyorself-efficacyinthatdomain.Whenuserprofilingisconductedbasedondataandpeoplearetargetedwithnudgesorsocialmediacontentthatpushesthemtochangewithouttheirknowledge,thelackoftransparencycanbeunsettling

(WorldBankGroup2015).TheEU’sGeneralDataProtectionLegislation(GDPR)hasspecificlegalorientationaboutwhatisacceptableprofilingonsocialmediaandotherdigitalplatforms,andstricterguidancewhenthatprofilingisonchildrenundertheageof13(andinsomeEU

countriestheageof16)(EuropeanUnion2016).EcologicalModelsofBehaviourChangerecognizemultiplelevelsofinfluence(intrapersonal,interpersonal,organizational,community,andpolicylevel)onindividuals’behavioursandthus

recommendstrategicinterventionswhichaddressthesemultiplelevelsastheyinfluenceindividualswithintheirspecificcontext.Ifanylayerofthecontextincludessubstantialbarrierstobehaviourchange,anindividualmaybelesslikelytomakethatchange,evenifmotivatedand

regardlessofwhethertheyhavethenecessaryskillstomakethechange.


Ecologicalmodelsaremosteffectivewheninformingbehaviouralinterventionsandtargeting

veryspecificbehaviours(e.g.,condomuse),butitcanbedifficultforresearchersandpractitionerstomakesenseofhowbroaderlevelsofinfluenceinteractwithlevelsmoreproximatetotheindividual(Sallis,OwenandFisher2015).

SocialNormTheories.Thisgroupoftheoriesconsidernormstobethebeliefsheldaboutwhichbehavioursareappropriateinaspecificgroup.Socialnormsestablishbehaviourexpectations

andtheseinturnaffectaperson’sdecisions,eitherbyshutteringpossibilitiesoropeningopportunities.AsNewKnowledgepointout,normsdonotinfluenceactionsinastaticwaybecauseanindividual’schoicescanreflectsubtleordeepchangesinbeliefsandattitudesthat

occurbecauseofsocialinteractions.Theseshiftingnormativeperceptionsinfluencebehaviouralchange,andsinceadolescentsarehighlyinfluencedbytheirpeers,itisimportanttoconsidersocialnormsandtheirroleinbehaviourchange.

A2011studycitedbyNewKnowledgeassessedhowofflineperceptionsofpeernormscouldinfluencewhetheradolescentsparticipateinriskysexualbehavioursonline.Thefindings

indicatethatbothinjunctiveanddescriptivepeernormspredictedwhetheranadolescentengagedinriskyonlinesexualbehaviour.Descriptivepeernormsweremorelikelytopredicttheriskybehaviourovertime(Baumgartner,ValkenburgandPeter2011).Adolescentsmay,then,

bemoreinfluencedbyhowtheybelievetheirpeersactuallybehavethanbytheirpeers’approvalofaparticularbehaviour.NewKnowledgerecommendsthatdigitalinterventions

targetingadolescentsconsideropportunitiestoillustratehowpeersbehaveinwaysthatchallengeordisruptsystemsandsocialnorms(Shane-Simpson,etal.2017).NewKnowledgegoesontonoteresearchshowingthatsmallgroupsareoftenincubatorsofsocialchangeand

thattheformationofsmall,safegroupsmightbeagoodstrategyforencouragingcollectiveidentitydevelopmentforyoungpeoplethatstartsinasmallgroupandthenexpandsbeyondit,thuscreatingasafespaceforindividualbehaviourchangeandadoptionofnewnorms(Fine

andHarrington2004)(Shane-Simpson,etal.2017).NarrativePersuasionTheoriesexplainthefactorsandprocessesthatfacilitatechangesin

audiencemembers’knowledge,attitudes,andbehaviours.Theyarecommonlyusedinedutainmentprogramming.Elementsofnarrative-basedentertainmentcanhelpanaudiencesuspenddisbelief,inturnreducingresistancetocounter-arguments(SlaterandRounder2002)

(Moyer-Guse2008).Edutainmentprogramsoftenusesocialcognitivetheoryalongwithdirectexposuretomediamodelsandindirectsociallearningthroughinterpersonaldiscussions(Bandura,A.2004).Narrativesresonateespeciallywellwhentheaudienceperceivesthe

storiestoberealisticandrelateswiththecharactersbecauseofsimilaritiesorbecauseoftheirdesirableattributes(Moyer-Guse2008)(Murphy,etal.2011).Additionally,narrativesbuiltwithfamiliarculturalmarkersareespeciallyeffectivewhentargetingminoritypopulations

(WangandSinghal2016)(LarkeyandHecht2010)(Lee,FawcettandDeMarco2015).


Twometa-analysesdemonstratedadvantagesofusingnarrativesinhealthinterventions,findingthatnarrativescanproduceasizablysignificantimpactoncombinedchangesinattitudes,

intention,andbehaviour,evenwhentheyhavearelativelysmalleffectonindividualoutcomes.Narrativesprovokeaffectiveaudienceresponse,andthosedeliveredthroughaudioandvideohavebeenfoundtobemoreeffectivethanthosedeliveredthroughprint.(Shen,SheerandLi

2015)(Zebregs,etal.2015).NarrativepersuasiontheorieswereusedforacclaimedEastLosHighprogramdevelopmentforcharacterbuildingandcontentproductionandtheyalsoguidedevaluationoftheprogramsintermsofkeymeasuresandhypotheses(WangandSinghal2016).

DiffusionofInnovationsTheory(Rogers2003)explainshowindividualsandorganizationsadoptinnovations—suchasideas,products,orservices—overtimeandexamineshowadopters

communicateandpromoteinnovations.Successfulinnovationstypicallyspreadgraduallyfromafewearlyadopterstothelargerpopulation.Thetheorydistinguishesbetween5differenttypesofadopters:innovators,earlyadopters,earlymajority,latemajority,laggards).Itincludesfive

factorsthatdetermineadoptionorrejectionofanewbehaviour:perceivedcomplexity,compatibility,observability,trialabilityandrelativeadvantage.Thoseusingthistheoryarerecommendedtoidentifytypeofadopterstoreach,developastrategy(knowledge,persuasion,

decision,implementation,confirmation),anddesignknowledgemanagementproductsandservicestoreachthem.

Right:Thediffusionprocess,adaptedby(Ohkubo,Limaye

andHarlan2015)from(Rogers2003).

Asthegraphshows,acurveplottingthecumulativepercentageofapopulation

adoptinganinnovationovertimetypicallyhasanS-shape.Someinnovationshavea

rapidrateofadoption(Innovation1)andothershaveaslowerrate(Innovation3).Critiquesofthetheoryincludeconcernsthatitdoesnotconsiderindividualresourcesorsocialsupportforbehaviourchange(Barnett,etal.2016).

BrandandCommercialFrameworks.InrecentyearssuchframeworkshavesometimesbeenadoptedforSBCC.OnesimplemodelusedbythemarketresearchcompanyTNSiscalledthe

AIDA:Awareness,Interest,Desire,Actionmodel,whereawarenessreferstocreatingawarenessofabrand,product,orservice;interestreferstogettingpeopleinterestedinknowingmoreaboutit;desirereferstodemonstratingthebrandhasbenefitstotheconsumer,sothatpeople

wanttoownit;andactionreferstopromptingpeopletotakethefinalstepandcommittopurchasingtheproduct(TNS2016).


TheNikeFoundationuseda“brandedmedia”approachforitsSBCCwork,notingthatbrands

havebeeninspiringpeopletochangebehaviourfordecades.Theideaisthatbrandsarecriticaltoolsforachievingsustainedchangeatscalebecausetheyserveasintegratorsanduniteawiderangeofactors.Brandsdevelopintangiblebenefits,understandingandmotivationsandareable

topersuadepeoplewithpromisesthatdrawthemintoproductsorservicesthataremeaningfulandthatinspirepeopletoact.Brandsalsocreatecommunitiesaroundthemthatpeoplewanttobeapartof,anduseinfluencersthatpeoplewanttobelike(NikeFoundation2013).

Anothercommercialmodelisthe‘customerjourney’usedfordiscoveryanduseofdigitalproductsandservices.Inthismodel,fivestepsareconsidered:non-aware(usershavepotential

toaccesstheservicebutarenotawareofit);aware(usershavebecomeawareoftheservice);cursory(usershavenotaccessedthecontentbuthavehadasuperficialinteractionwiththeservice);occasional(usershaveengagedwiththeservicealimitednumberoftimes);andrepeat

(userswhohaveusedtheservicesseveraltimesandaccessedcontentrepeatedly)(Wills,MorainandBaudeau2016).Thecustomerjourneymodelisusefultounderstandifandhowusersarriveto,interactwith,and/ordropoffofdifferentpartsofadigitalplatform,serviceor

site.Thisinturncanhelpdesignerstoimprovethetechnicalside(userinterface),contentoruserexperiencetoensurethatusersareabletoeasilynavigatethroughtheplatformandsothattheystaytherelongenoughtointeractwiththecontent(anecessarystepalongtheroadto

achievingimpact).

SocialMediaand‘Transmedia’ApproacheshavebecomeincreasinglycommonasmobileandInternetusehasgrown.Intransmediastorytelling,narrativeelementsarecreativelycoordinatedacrossdifferentmediaplatformstobuildastoryworld,engageabroaderaudience

andprovidethemanenrichingexperiencethatgoesbeyondpureentertainment.Becauseyouthareincreasinglycomfortablenavigatingmultipledigitalplatforms,transmediaedutainmentholdspromiseasahealthpromotionandeducationinterventionaltool(Singhal,

WangandRogers2013).StartingwithanunderlyingSBCCtheory,thoseusingtransmediaapproachescancreateamulti-prongedapproachbychoosingamediachannelandmessagingcombinationthatismostapplicabletothedesiredtypeofbehaviourchange.Thismustbe

basedonathoroughstudyofthemediahabitsandusesofthetargetaudience,however,ormessageswillbeflowingonchannelsthatarenotaccessibletothecoreaudienceorthatarenotconducivetothemessage(WangandSinghal2016).Certainmediachannels,forexample,

appeartoworkbetterforone-waycommunication,othersforinteractiveengagement,andothersforcomplexlearning(HealthCommunicationCapacityCollaborative(HC3)2013).

TheglobalhumanrightsorganizationBreakthroughimplementedwhattheyreferredtoasa‘360degree’campaignasearlyas2008.BellBajao,astheprogramwascalled,wasaimedatendingdomesticviolenceinIndia.Itisanearlyexampleofhowamulti-pronged,multi-level

campaignwasusedtoencouragesocialandbehaviourchangeatvariouslevelsoftheecologicalmodel.Thecampaign,developedbythebrandingcompanyOgilvy-Mather,workedthrougha


blogandsocialmediaaswellasthroughtraditionalmediachannelslikeTV,print,radioandavideovan.Coupledwiththe‘aircover’(astheyreferredtothemassmediaefforts),

Breakthroughalsouseda‘groundcover’approachthroughcommunity-basedprogrammingaimedatwomen’srightandpreventionofdomesticviolenceandengagementofmenandboysto‘ringthebell’againstdomesticviolence.Thecampaignengagedsurvivorsofdomestic

violencethemselves,malemembersofsociety(asprimaryagentsofchangeamongpeersandnetworks),andwidersocietythroughbroadmulti-mediaeducationandawarenessprograms(Chakraborty2010).

TheBBCMediaActionAnanyaprograminIndiausedasimilar‘360degree’strategytotacklethehighlevelsofmaternalandchildmortalityinBiharstate.Localcommunityhealthworkerswere

equippedwith‘MobileKunji’adeckofcardswithadviceaboutmaternalandchildhealth.Eachcardsportsauniquecodethatthecommunityhealthworkerusedtoaccessanaudiomessageonherphonethatwasvoicedbythecharacter‘Dr.Anita’.Thesamecharacterappearedagainin

aradioshowcalledKhirkiMehendiwali(MehendiOpensAWindow)whichairedthreetimesaweekandreinforcedmanyofthehealthmessagesthroughdramaandcomedy.Inaddition,amobileinformationservicecalledKilkariallowedfamiliestoreceiverecordedmessagesfromthe

sameDr.Anita.Lastly,thecampaignfeaturedatelevisedsocialadvertisementcalledEkTeenDo(One,Three,Two)whichstressedtheimportanceofleavingathree-yeargapbetweenpregnancies(Swarup2012)(Bachan2016).

3.WhatdocurrentandpastdigitalSBCCinitiativestellus?Althoughaccesstosmartphonesisincreasing,asnotedinSection1above,thoseinlower

incomebracketsorwhoareinotherwaysmarginalizedorrestricted(duetolocation,genderorculture)maynothaveaccesstotheInternetortoasmartphone.Forthisreason,wecoveravailableliteratureonprogramsusinglow-end,basicmobilephones(e.g.,thosethatcanonly

callandtext)aswellasthatoneffortsdesignedforamorerobust,Internet-enabledapproach,andmulti-mediaor‘transmedia’approaches.Muchoftheearlierliteratureonlow-incomecountriesisfocusedonmHealthandtheuseofSMS,whereastheearlyliteratureonInternet-

enabledapproachestendstofocusonmiddle-andhigh-incomecountries/contexts.AsaccesstotheInternet,especiallymobileInternet,hasexpandedtomoreyouthinlow-and

middle-incomecountries,theliteraturehasfollowedsuit.However,therearestillstrongbarrierstoaccessanduseofsmartphonesandtheInternetformany,perhapsmost,ofthepoorestandmostvulnerableyouth,especiallyyouthinruralareasandgirlslivinginrestrictive

homesorsociallycontrolledenvironments.ThechallengeforMTVNishedhwillbetoexplorewaysthatbasicdigitalchannelscanbecoupledwithmoretraditionalchannelssuchastelevisionandradio,andtofuture-proofitsinterventionsbytestingresponsiblewaysofusingemerging

channelslikeWhatsApp,FacebookLite,orchatbotstoenhanceengagementandimpact.


Someofthemainchallengesindevelopingandevaluatingdigitalinterventionsforbehaviourchangeinhealthandhealthcareweresummarizedinarecentarticle,including:therapidpace

withwhichtechnologychangesandtheneedtocontinuallyupdateandadapt;engagementwithdigitalinterventionsoftenbeingtoolimitedtosupportbehaviourchange-ortheengagementismultidimensionalandcannotbeevaluatedsimplybydigitalmeans;engagementmaybe

unequalandreinforcedisparitiesorinequalities;thereisatheoreticallackofclarityaroundthemechanismsthroughwhichdigitalbehaviourchangeinterventionshaveaneffect;itisdifficulttospecifymeaningfulcontrolsorcomparators;thecomplexmulti-componentnatureofdigital

interventionsrequiresaniterativedesignandtestingcycle;cost-effectiveevaluationmodelsandtechniquesarelacking;fundingmechanismsarenotalignedwithdigitalcycles;therearecompetingcommercialandethicaldemandsondataownershipandintellectualproperty;

qualityandethicalstandardsaregenerallyunclearorundefined(Michie,etal.2017).

SMSforhealthbehaviourchangeSeveralstudiesoverthepastdecadehavefoundthatteensandyouthareopentotext

messagingforhealthpromotion.InonestudyonwhetherSMSmessageswithreproductivehealthinformationwereacceptabletoAustralianyouth,thesurveyedyouthsaidmessagesprovidednewinformation,areminderofexistinginformationandreducedtheirapprehension

aroundtestingforsexuallytransmittedinfections(STIs).Arandomizedcontrolledtrialfrom2006-2007foundthosereceivingSMSandemailmessagesaboutsexualhealthimprovedtheirknowledge.Inthesamestudy,femaleswhoreceivedmessagesweremorelikelytogettested

forSTIsthanthosewhodidnotreceiveanymessages(Lim,etal.2011)(GoldandLim2010).Preventioninformationviatextmessagewasfoundtobeeffectivewithyoungpeopleaged10

to24inSanFrancisco(theSEXINFOproject,whichtargetedat-riskAfricanAmericanadolescentswithsexualhealthtextmessages)aswellasintheDemocraticRepublicofCongo(theLigneVertehotline,whichprovidedconfidentialinformationonfamilyplanningandcontraceptives

alongwithreferralstoclinics)(TothandPSIDRCongo2008).TexttoChangetrialedanSMSbehaviourchangeapproach,includinganSMS-basedmultiple-choicequiz,inanefforttosynergizecommunication,linkwithexistingpreventioncampaignsandofferservicesfor

HIV/AIDSinUganda.Findingsindicatedthatitiscriticaltohaveanextensivemarketingcampaign(includingradio,billboards,andnewspapers),anintroductorymessagewithanexplanationoftheprogramandhowanonymityisguaranteed,ashorterprogramdurationso

thatpeoplewillnotloseinterest,andvarioustechnicalimprovements(TexttoChange2008)(Mecheal,etal.2010).(IppolitiandL'Engle2017).

AglobalbriefonmHealthforadolescentsexualreproductivehealth(ASRH)byIppolitialsofoundthatthatASRHinformationprovidedviamobileisappealing,feasible,andculturallyacceptabletoyouth;andthatitcanpositivelyinfluenceyouthSRHoutcomesbyimproving

knowledge,reducingsexualriskbehaviour,andincreasingutilizationofhealthservices.mHealthprogramsalsoofferedcost-savingsascomparedtoothertypesofscalingmodels.Examples


reviewedsuggestthatSMScanenableinnovativehealthmessaging,includingdirecttext/SMSmessages,role-modelstorylines,psychosocialsupport,andcounseling.Inadifferent

approach,inEthiopia,MarieStopesInternationalsentelectronicvouchers(e-vouchers)directlytoauser’sphone.Thevoucherscouldberedeemedatanaffiliatedhealthfacilityforfreecounselingsessionsoncontraceptivemethodsaswellasmedicalabortionandpost-abortion

careservices.Intermsofage,olderadolescents(ages15andabove)tendedtobethemainusersofthesemHealthinterventions,withlowerprogramuptakeamongyoungeradolescents.(Ippoliti2017).

FundamentalandcommonchallengeswithmHealthforASRHincludeaccess,security,infrastructure,functionalityandlanguage.Theseneedtobeaddressedindesignandpilot

stages.Contentshouldreflectboththeprimarylanguagesspokenandthevernacularwithwhichyouthfeelmostaccustomed.BecausebasicSMSislimitedtotext,itcaninhibitrobustcontentprovisionandfallshortinengagingusersintwo-waycommunication.Poornetworkconnectivity

isamajorissueaffectingcontinuity,reach,andqualityofSMS-basedinterventions.Maintainingthesecurityandconfidentialityofuserinformationanddataisamajorchallengeformobileanddigitalinterventions.Mobilephonesareoftensharedamongfamilymembersinlower-

incomecontexts,reducingprivacy.Thisisespeciallyimportanttoconsiderwhenaddressingsensitiveortaboohealthtopics(N.Ippoliti2017).

Socialmedia,multi-mediaandtransmediaA2013reviewlookedat98researchstudiesonsocialmediaforhealthcommunication.Themethodologicalqualityofthestudieswasdeemedaslow-mostwereexploratoryanddescriptiveratherthanfocusedonmethodology.Sixkeybenefitsofsocialmediawerefound:

increasedinteractionwithothers;moreavailable,shared,andtailoredinformation;increasedaccessibilityandwideningaccesstohealthinformation;peer/social/emotionalsupport;publichealthsurveillance;andpotentialtoinfluencehealthpolicy.Limitationsincludedthose

relatedtoquality,lackofreliability,confidentiality,andprivacy.Usersarelargelyunawareofdataprivacyandconfidentiality,datasecurityandthedangersofdisclosingpersonalhealthinformationonlineoroverviainsecuremeanssuchastextmessage(Moorhead,etal.2013).

A2014studyonsocialmediaandtransmedia’sinfluenceonchildhealthcovered10studies,mostofwhichwerefocusedonmultiple-mediaplatformsthatwerenotstrictlytransmediaor

socialmediaperse.Therewerelimitedfindingsthatcouldpointtomultimediaplatformsonhealthybehavioursdirectlyinfluencingchildsurvival.However,theauthorsdidfindstudiesthatshowedapositiveimpactonchangesinattitudes,awareness,andknowledge.Widerfindings

includedthatbehaviourchangeinterventionsweremosteffectivewhenpeoplewereengagedonmultiplelevels;inotherwords,wheninitiativesincluded“behavioural(performance-basedtechniques);social(interpersonalandsmallmedia);sensory(massmedia,materials);cognitive

(problemsolving)”.Itcouldnotbedeterminedfromtheexistingstudiestheextenttowhichonetechniquewasmoreeffectivethananother(Higgs,etal.2014).


CasestoexplorefurtherFourrecententertainment-educationinitiativesaimedatyoungpeoplethatincorporatedsocial

mediaanddigitalengagementforSBCCare:MTV’s16andPregnant,EastLosHigh,VoicesforChange’sPurple,andtheNorwegianBroadcastingCompany’sSKAM.TheseareworthexaminingfurtherbecausealthoughtheyarenotaperfectfitforMTVNishedh,theyoffervariousinsights

onusingSBCCtheoryinthedevelopmentofEEforyouthwithagenderfocus,incorporatingdigitalaspectsintomulti-mediaandtransmediaapproaches,andgatheringfeedbackandmeasuringresultsthroughdigitalchannels.

16andPregnantThisprogramfollowedthelivesofpregnantteenagersintheUnitedStatesinthefinalmonthsof

theirpregnancyandtheirfirstexperiencesofmotherhood.A2015studymatchedVitalStatisticsbirthdatatoNielsentelevisionratingsdatatodeterminewhetherexposuretotheshowhadanimpactonteenchildbearingrates.Resultsimpliedthattheshowledtoa4.3percentreduction

inteenbirthsinthe18monthsfollowingitsinitialairing.Theauthorsbelievethiscanaccountfor24percentoftheoveralldeclineinteenbirthsintheUSduringthatperiod.TheysupplementedtheirfindingswithdatafromGoogleTrendsandTwitter,whichsuggestedthat

theshowledtoincreasedinterestincontraceptiveuseandabortion(KearneyandLevine2015).Theshowgarneredmixedreactions,withcriticsbelievingthattheshowglamorizedteen

pregnancybymakingthewomenfeaturedintorealitystars.AstudyfortheNationalCampaigntoPreventTeenandUnplannedPregnancy(Albert2010)reportedthat82percentofteenswhowatchedtheshowsaidithelpedteensbetterunderstandthechallengesofpregnancyand

parenthoodwhereasonly17percentsaiditglamorizedteenpregnancy.

Researcher,JenniferAubrey,questionedtheseearlierfindings,notingthat“conclusionsarebasedoncorrelationalresultsofaggregatedata–therewasabiggerdropinteenpregnancyinareaswhereteenagerswerewatchingmoreMTVprogramming,notonlythe16andPregnant

series…Theyarelookingatdataattheaggregatelevelbutmakingassumptionsaboutwhatishappeningattheindividuallevel,sayingthat16andPregnantiseffectivesexeducationforteens"(Harwood2014).Aubrey’sownstudysuggeststhatthosewhoidentifiedtheleastwith

themothersontheshowhadthemostnegativeattitudesuponviewingit,whereasthosereportingthehighestlevelofaffinityandhomophilyhadmorepositiveattitudestowardteenpregnancy.Theseresearchersalsofoundthatteenswhothoughttheyweresimilartothe

pregnantteensweremorelikelytohavealowerperceptionoftheirownriskforpregnancy,higheracceptanceofmythsaboutteenpregnancyandamorefavorableattitudewithregardtoteenpregnancy,andthatteenswhofelta‘friendshiprelationship’withtheteenmomsalsohad

adecreasedbehaviouralintentiontoavoidpregnancy(Behm-Morawitz,etal.2017).


EastLosHighThiswasatransmediaeducation-entertainmentprogramaimedatyoungLatina/oAmericans.Theprogramembeddededucationalmessagesinentertainmentnarrativesacrossdigital

platformstopromotesexualandreproductivehealth.EastLosHighwasdevelopedtoreach,engageandinfluenceyoungLatina/oAmericansonthetopicofadolescentpregnancyandbettersexualreproductivehealth.OneinthreeLatinaadolescentsintheUSbecomepregnant

beforetheyare20yearsold(1.5timesthenationalaverage),yetthereareveryfeweffectivehealthinterventionsaimedatLatina/oyouthintheUS.(WangandSinghal2016).

EastLosHighaimedtocreateaculturallysensitiveintervention.ItwasdesignedtosubvertthestereotypesofLatina/ocharactersasgardeners,maidsandgangmembers.TheplotcoveredrealisticissuesthatyoungLatinas/osfaceandofferedmultipleentrypointsforLatina/oyouth,

whoavidlyconsumedramaanddigitalentertainment,throughtheirpreferredplatform(Jenkins2006)(Castro2013)(Lopez,Gonzalez-BarreraandPatten2013).Theteamtookastrategicapproachbasedonexistingmediahabits.Latinas/osare40percentmorelikelythanthe

generalpopulationtowatchtelevisionandvideosonlineoronasmartphone,and3timesmorelikelytocheck(viasocialmedia)whatprogramstheirfriendsarewatching(Razzetti2012).

EastLosHighwasavailableontheHuluweb-streamingsite,andattheendofeachepisode,viewerswereencouragedtovisitthewebsitewheretheycouldaccessadditionaltransmedianarrativeextensions,forexample,anextendedscenewhereacharactertalkedwithahealth

counseloraboutcondomuse,orasetofvideologsfromoneofthemaincharacters.Fromanyoftheextensionmaterials,viewerscouldvisitaresourcespage,findlocalhealthclinicsbasedontheirlocation,orclickotherlinksformoreinformation.Thetransmediaextensionswere

promotedonFacebook,TwitterandInstagram.Engagementwiththeshowandothermaterialsweretrackedthroughwebanalytics,viewersurveys,socialnetworkanalysisandcontentanalysistounderstandsocialdynamics,messageframingandusergeneratedcontentonEast

LosHigh’ssocialmediasites(WangandSinghal2016).

VoicesforChange(V4C)V4CwasaverybroadprogramgearedtowardsstrengtheningtheenablingenvironmentforgenderequalityinNigeria.Itappliedsocialnormstheoryatscaleandaimedtoaddressstructuralbarrierstogenderequality,inparticular,discriminatoryandharmfulattitudes,

behavioursandsocialnorms.V4Csoughttochangethreenormativeareas:women’svoiceandleadership,women’sroleindecisionmaking,andviolenceagainstwomenandgirls.Theprogramtargetedyoungwomenandmenaged16-25yearsoldandoperatedinfourstatesin

Nigeriaforsomeactivitiesandforothersitcoveredthewholecountry.

Believingthatchangeneededtohappenatscale(e.g.,atindividuallevelandalsowithinwidersociety),theprogramworkedatmultiplelevels:individual(providinggirlsandyoungwomenwiththeskills,knowledgeandconfidencetochallengediscriminatorysocialnormsandcreate

changeintheircolleges,homes,workplacesandcommunities);community(engagingmenand


boys,religiousandtraditionalleaders,andnetworksofwomenandgirlstocreateacriticalmassofsupportforgenderequality,acceleratingchangeandshiftingnegativenorms);andsocial-

structurallevel(changingdiscriminatorylaws,creatingbetterpolicies,directingassetstowardswomenandgirls,sendingamessageaboutchangedsocialnormsthroughpoliticalandlegalstructures)(Fraser,EkeobaandFeather2017).

V4CwasunderpinnedwithaTheoryofChange(ToC)expressingalong-term(20year)visionforgenderequality.Threedomains(Self,SocietyandFormalInstitutions)werethecoresitesat

whichchangewassought.Thedomainofthe‘Self’isthepersonaltransformationfoundationfordiffusingchangesmorewidely.The‘Society’domainisaimedatchangingsocialnormsatscaleandlayingfoundationsforagenderequalitymovement.The‘FormalInstitutions’domainis

wherelegalandpolicychangescansupportandcreatepopulation-widesocialchanges.TheToCenvisionschangestowardsgenderequalityineachdomainthatworktogethertoamplifyandstrengthenthebroaderenablingenvironmentforgenderequality(MilwardandNelson2017).

V4C’sprogramstrategyworkedattwolevels–‘scale’(whichtargetedyoungpeoplesociety-wideinthefourstateswithabrandedcommunicationscampaign),and‘focusedintensive’

(whichcreatedintensivelearningopportunitiesinbothface-to-faceandonlinespacesforkeyindividualsselectedfortheirpotentialtoamplifyandpromotethechangeoncetheyhadgonethroughtheirownpersonaltransformation).Differentinterventionsweredesignedforeach

levelandthesedrewontwomodelsofchange:a‘scalemodel,’whichcentralizedanideaofthediffusionofnewattitudesandbehavioursconcerninggenderequality,andsocialnormschange

asthecatalystforthis;anda‘stagesofchange’modelwhereindividualsgothroughdifferentstepsofPre-Contemplation;ContemplationorAwarenessRaising;Planningthechange/Persuasion;Action;andMaintenance,andwhichcentralizedanideaofself-transformationas

thefoundation(VoicesforChange2017).Theselectionofparticipantsforintensiveprogrammingwasbasedonthediffusionmodel(Who

wouldbebestplacedtodrivediffusion?)Thestagesofchangemodeldrovethecontentandimmediateobjectivesfortheintensiveinputs.Abrandedcommunicationscampaign,‘Purple’,workedalongsidethesetostimulateinitialstagesofchangeamongawiderpopulation,and

preparethatpopulationforthediffusioneffectscarriedforwardbykeyinfluencers,whilealsocreatingasenseofsociety-wideapprovalforthechanges,aspartofthesocialnormsapproach.Thestrategyusedawiderangeofcommunicationschannelswiththeaimofachievingalayered

experienceinwhichindividualswereaddressedwithsimilarmessagesthroughdifferentmedia/interventions.Someareaswere‘saturated’withmultiplemessages,sothatachangingpopulationwouldbesupported,andchangesreinforced,fromdifferentsources.Inshort,the

mediacampaignaimedtocreatetheconditionsfordiffusion.ThePurplebrandaimedtospreadtheideaswidelysothatdiffusionbykeyinfluencerswouldbeamplifiedandacceleratedatscalebydiffusionvianewsocialnorming.Lastly,legalandpolicychangebuttressedtheseareas,and

evidencegenerationaimedtostimulatediscussionanddebate.V4Cusedarangeofquantitativeandqualitativetoolstotrackresults,createanevidencebaseforstrategicmid-program


adjustments,andlayafoundationforunderstandingoutcomes(VoicesforChange2017).

AstudytodeterminebrandawarenessofPurpleshowedthatitwashighestinpost-secondaryinstitutionswherePurpleactivitiesweretakingplace.Awarenessamongthoseoutsidepost-secondaryinstitutions,whosecontactwithPurplewasthroughmediasourcesalone,was

superficialandunderstandingofthemessagewasvague.AwarenessofthePurplewebsitewasverylow,at2%ofthetotalsample,whichwasproblematicgiventhecentralityoftheonlinecampaigninthestrategy.TherewasalackofclarityaroundthePurple“CalltoAction.”Young

peopledidnotknowwhatanextsteptowardsactionwouldbeforthem.Thoughthesiteofferedexamplesofactions(talkingtoafriend,encouragingarelativetospeakup)thesiteneededtobemuchcleareraboutwhatpeopleweresupposedtodoafterexposuretothe

campaign(TNS2016).Anendlinereport(DennyandHughes2017)ontheV4Cprogramfoundthatlargescalechanges

inyoungpeople’sgenderattitudesandbehaviourstookplaceinthefourtargetstatesoverthe2014-2017period;howevertheexactcontributionV4Cmadetoimprovedattitudesandbehaviourswasdifficulttoascertain.ThereiscompellingevidencethatthePurplebrandwas

animportant(andlikelyunderestimated)driverofchangeinalmostallareas.V4Cactionsintheselfandsocietydomainshadapositiveeffect,irrespectiveofculturalcontextortheoriginalstartingpointinastate.PhysicalSafeSpaceswerefoundtobeparticularlyeffective,especially

giventheirabilitytomobilizeyoungpeopletodiffusenewattitudesandbehaviourstoothers.

Radiowasoneofthemosteffectivemethodsofpromotingpositivegenderattitudesandbehaviour,especiallyoutsideofmedia-saturatedLagos,andradiowasthemostsuccessfulinreachingpoorergroups.V4Cintentionallytookforwardamulti-mediacampaigntosaturate

youngpeoplewithconsistentmessagesfromdifferentsources,givingasenseofsocietywidesupportfornewnorms.Theprogram,however,hadlimitedeffectisinchangingyoungpeople’sattitudesandbehaviourstoviolenceagainstwomenandgirls.Reasonsforthiswerenotclear

butcouldinvolvepeoplebeingmoreawareofwhatconstitutesviolenceandthereforereportingviolencewherepreviouslytheywouldnot,oritcouldbethatfactorsotherthansocialnormswerestronginfluencingfactors.Onlinespaceswerenotasimpactfulashoped.Theyrequire

carefulmoderationtobeeffective,andevenwhenthatishappening,onlineusersarelikelytobeaccessingotherwebsitesorsocialmediaplatformsthatarenotregulatedinthesameway,meaningthattheyarealsoexposedtocounterproductiveviewsandthiscanunderminethe

resultsachieved.KeylearningsfromV4Cinclude:thatbrandcanmoveyoungpeopletochangeattitudesand

behaviours,andpotentiallytakeaction.ThePurplebrandalsoprovidedthe“glue”forprograminterventionsandcreatedalayeredeffectthatwasinfusedthroughoutallcommunicationsanddigitalcontent.Thediffusionmodelcanbeausefulasawaytoframebehaviouradoptions,and

knowledge,skillsandself-confidenceareimportantforsupportingdiffusionofnewbehaviours.Workingwithmenandboyswascentraltoachievingchangeatscale.Intensive


programeffectsseemedtobestrongestamongmenastheywereassociatedwithdramaticallyhigherratesofchangeandpositivebeliefscomparedtomeninthegeneralpopulation,

suggestingthatcarefullyconstructededucationaboutgenderinsafe,supportiveenvironmentshastransformativeeffectsparticularlyamongmen.Mensometimesrespondedtomessagingmorestronglythanwomen,notableinearly-stagechangedimensionssuchascontemplation,

andthissuggeststhatdifferencesinthe‘startingpoint’ofpeopleengagingwithPurpleinfluencedthescaleofchange.Inchallengingcontexts(forexample,religiouslyconservativeones),gettingmessaging‘right’canunlockanevengreaterscopeforchangethaninless

challengingcontexts.(DennyandHughes2017).TheV4Cendlinefoundthatdigitalapproachessignificantlyhelpedtogetissuesofgender

equalityintopublicdebate,addingvisibilityandreachtointensiveprogramming.Thedepthofengagementwiththecontentwasacriticalcomponent,withimplicationsbothforhowthesetoolsareusedaswellasformeasurementoftheirimpact.Inmoderatedspaces,attitudes

andbehaviourschanged–likelybecausepeoplehadtheopportunitytoreflectanddebateonline.Socialmediauseingeneral,however,iscorrelatedwithlesspositiveindividualchangerelatedtoviolenceagainstwomenandgirls,possiblyduetootherharmfulgendermessagesand

imageryaccessedonline.Togetherthesefindingssuggesttheneedforclosemoderationofon-lineplatformsandtheneedforcombiningdigitalapproacheswithothermoreintensivecommunicationsapproaches,includingsafespacesandmoderatedradiodiscussionstoachieve

moreconsistentchangesinattitudesandpractices.Anotherpointrelatedtodigitalisthatusingavarietyofmediatoolswasimportantinrespondingtodifferenttargetingobjectives.Online

spaceshadsomebiastowardsabettereducatedpopulation–andamoresociallyconnectedone.Radiohadthewidestreach,andwasalsoeffectiveatreachinglower-educated,poorerpartsofthepopulation(DennyandHughes2017).

SKAMSKAMwasacharacter-drivenonlinedramaseriesproducedbytheNorwegianBroadcastingCompany(NRK)thatairedinNorwayandgarneredateenfollowingintherestoftheworld.The

SKAMstorylinewasplaced‘soclosetorealitythatitcouldbetrue’(Nyborg2012).Thestorywasrevealeddailyandin‘realtime’onablogusingvideoclips,chatmessagesandpicturesand

theaudiencewasencouragedtocommentandengage.SKAMtookitsleadfromdigitalandtransmediastorytelling,integratingmultipletextstocreateanarrativewhere‘eachmediumdoeswhatitdoesbest’(Jenkins2006)andinvitingtheaudiencetobepartoftheprocess.SKAM

builtonanearlieronlinedramaseriesthattriedtotackleseriousissuesandtogiveteengirlsauthenticexamplesofhowtocopewithlife.TheNorwegianBroadcastingCompanypickeduptheideaanddecidedin2014todevelopasimilarconceptforanaudienceof16-year-old

Norwegiangirls.Thetighteraudiencetargetservedtomorespecificallytailorcontent(Nyborg2012)(Fladalen2016).

TheteamdidextensiveaudienceresearchwithteensalloverNorway,includinginterviews,reportsandstatistics,socialmediascanning,andschoolvisitswiththegoalofunderstanding


Norwegianteensandidentifyinghowanewonlinedramaconceptcouldservetheirneeds(Magnus2017).Thepointwastocreatecontentthatwascatchy,relevantandauthenticfor

teens,saidtheSKAMproducer;‘Ithadtohavetruthandhonestyabouttheirownculture,somethingtheyhadn’tseenanywherebeforeandcouldn’tgetanywhereelse.Theyhadtorelatetoitandidentifywithitmorethananyotherseries’(Pickard2016).Theresultingshow

wasanonlinedramaseriesfollowingacastofteenagersataHighSchoolinOsloastheynavigatedlifebothathomeandschool.Elementsfromtelevisiondrama,sitcomandsoapoperawerecombinedtotellthestories(Fladalen2016).

SKAMwasproducedandpublishedacrossmultiplemediaplatformsselectedtofitNorwegianteens’naturalmediahabits.ThemainplacewastheSKAMwebsite(skam.p3.no)whichhad

videoclips,chatmessagesandpicturespublishedonadailybasisduringeachseason.ThekeycharactersalsohadInstagramaccounts.Thesewereusedasintermediariesbetweenthefictionaluniverseandtheaudience,allowinga‘fictionalslideintoreality’,whichwasthe

signaturefeatureoftheseries(Magnus2016).OnFridays,videoclipswerecombinedintoafullepisodethatwasmadeavailableonNRK’smobileappandtheiryouthchannel.SKAMworkedbothasanonlinedramatoldacrossmultipleplatformsandastraditionaltelevision.All

essentialinformationwentintovideoclips,andotherplatformsfleshedoutthestorylinesandcharacterstokeepthemrelevantandinsyncwithcurrentevents’(Furevold-Boland2016).

Thestorylinewaspublished“inrealtime”sothatithadacontinuitythatwasdifferentfromweeklytelevisionshowsorfrombinge-watching(popularonNetflix).Thestoryblendedinto

viewers’reallives,whichenhancedthereality-fictionblurring,andtheschedulingwasalsodrivenbythestoryandcharacters.Forexample,inoneepisode,acharactersentanotheratextmessageat17:32,themessagewasuploadedtotheSKAMsiteatthesametime(Magnus2017)

Thisalsomeansthatthepublishingwasirregularandunpredictable,makingtheaudiencecheckbackmoreoftenfornewsorupdates.Usersengagedtodebatethestorylineinmultipleplaces,butofmostinteresttotheproducerswasthecommentssectionwherediscussionstook

placeaftereveryupdate.TheSKAMteamusedtheaudiencecommentstoadjustthestoryline(Magnus2017).FansalsodiscussedSKAMonsocialmedia.Fanactivityrangedfromseriousdiscussionsofkeythemestospeculation,selfies,fanfiction,art,memesandgifs.Fan-generated

contenthelpedfillingapsbetweentheupdatesandallowedSKAMtobeanongoingandimmersiveexperience(Atkinson2016)(SchankeSundet2017).(Nyborg2012).

Virtualversusin-personbehaviourchangeAnunpublishedreportforGirlEffectfromtheBehaviouralInsightsTeam(BIT),agroupthatappliesbehaviouralscienceandnudgetheorytosocialandgovernmentalproblems,exploreda

seriesofquestionsrelatedtotheeffectivenessofvirtualreferencegroupsforbehaviourchange.Theyfoundthatthereislimitedevidenceonwhetheronlinegroupsforbehaviourchangework,butthattheevidencethatdoesexistsuggeststhatonlinegroupscanbeeffective,though

engagementandretentionmayneedtobeaddressed.Offlinegroups,peerinfluencersand


buddiesdohavepowertoinfluencebehaviour,thoughthismaybemorelikelyforbehavioursthatarevisible,andit’snotentirelyclearhowthismighttransfertotheonlineenvironment

(TheBehavioralInsightsTeam2018).BIT’sfindingsincludethatalthoughparticipantsdon’tneedtoknowoneanothertomakean

onlinegroupeffective,behaviourchangeinterventionsmaybemoreeffectiveifindividualsinthegrouparemoredemographicallysimilar(e.g.,arepartofthesamein-group)(TheBehavioralInsightsTeam2018).Thisfindingissupportedbytheearlierexampleinthispaperof

EastLosHigh’sresonancewithLatina/os,however,itwouldbeimportanttoconsidermessaging–inthecaseof16andPregnant,theyoungpeoplewhomostrelatedtotheteenagemomstendedtoembracethenon-desiredbehaviours.BIT’sreviewoftheliteraturefoundfourmeta-

analysesthatdirectlycompareddigitalandin-personinterventionsforbehaviourchange.Thesecoveredonlyhealthinterventionsandonlinelearning,butbasedonthoselimitedstudies,BITsuggeststhatbothinpersonanddigitalinterventionscanbeimpactful,andblended

interventionsareverypromising.Thisalignswithearliersuggestionsinthispaperthatmulti-mediaandtransmediaapproachesaremoreeffectivethansingleprongedones.

AstudyinChilecomparedin-personpeersupportandSMSremindersandshowedthatdigitalreminderscanbejustaseffectiveasin-personones.Inthisstudy,savingsincreasedamongmembersofgroupswhometonceaweekandhadtopubliclyannouncetheirsavings

commitment.Inafollowupstudy,textmessagesaboutsavers’progressandtheaggregatebehaviouroftheirpeerswasaseffectiveassavingsgroupsandaddingina‘savingsbuddy’for

face-to-facemeetingshadnoadditionaleffect.Theresearchershypothesizedthataccountabilityandreminderswereimportantwaystoencouragecertainbehaviours(Kast,MeierandPomeranz2012).BITconcludesthatwhenonlineandin-personmethodsaredirectly

compared,successisprobablyduetooverallprogramandnotattributabletoadistributionchannel(BehavioralInsightsTeam2018).Thisfindingalignswithawidebodyofresearchandpracticeregardingtechnologyinterventionsacrosssectorsandcountries–inotherwords,a

singletechnologyinterventionisusuallynotenoughtoclaimdirectattribution,andtherearenormallymultiplefactorsintheexternalenvironmentaswellasinprogramdesignthatinfluencehoweffectiveaparticularchannelisintermsofbehaviourchange.

BITpositsthatsocialmediaallowsforpubliccommitments,andthatthesecanbeusefulforholdingpeopleaccountabletoeachother.Inaddition,theynotethatbehaviourandsocial

normsoftenchangebecauseofiterativesocialreferencing.Becausedigitalcommunicationmakesremindersandquickcommunicationeasy,remindershaveproventobeanimportantaspectofbehaviourchangeprograms(includingwhentheyareincorporatedalongsideother

interventions).BITrecommendsthattheyshouldbeincorporatedoftenintoonlinegroups.Remindersaremosteffectivewhentheyprovidefeedbackonperformance,remindapersonoftheirgoals,highlightpositivesocialnorms,oraresignedbyapersonthatrecipientsknow

andtrust(e.g.ajobcentercoach,friendorteacher)(BehavioralInsightsTeam2018).


OtheraspectscoveredbytheBITliteraturereviewincludetheimportanceofsupportingintrovertsorpeoplewhomaybeshyortimid.Digitalgroups,theynote,mightallowintroverts

toengagemoreactivelythantheywouldinperson.Theyalsonotethatpeoplearemorelikelytoengagewhenitiseasy.Movingcertainactivitiesonlinemaycutoutsignificantfrictioncostsandbarriersandallowpeopletoengagefromtheplacesandatthetimesthataremost

convenienttothem.Lastly,theysuggestthatpeoplemightbemorehonestabouttheirownnegativebehavioursinananonymoussetting,astheyfeellessjudgment.Thiscanbackfirehowever,ifanonymityleadstolesssocialrestraintandmorenegativeengagement(e.g.

bullying,hatespeechtrollingorharassment)onanonlinesite(BehavioralInsightsTeam2018).

4.GoodpracticeandcommonpitfallsindigitalSBCCdesignDesignofdigitalSBCCprogramsmusttakeintoconsiderationanumberofaspects,including:

users’mediaandtechnologyhabits,engagingandeffectivemessaging,andappropriateandaccessiblechannelsformessages.Goodpracticesaresummarizedbelow.

GroundeffortsintheorySBCCprogramsshouldbegroundedinappropriate,relevantandupdatedtheory.OfteninSBCCprogramming,thereislimitedornoattentiontobehaviourchangetheoryorarelianceon

outdatedorsimplistictheory).Behaviouraldeterminantsneedtoexpandbeyondtheindividualleveltoconsidercontext,environmentandinter-personaldeterminants.Implementersshouldfindwaystounderstandandrespondtothecomplexityofunderlyingdeterminantsof

behaviour(Barnett,etal.2016).Understandthetargetpopulationandthesocialcontext(socialnorms/gendernorms)

Itiscriticaltounderstandthetargetpopulation(includingtheirwillingnesstochange).Moresuccessfulapproachesarecontext-specificandrootedinanunderstandingofcurrentbehaviour,itssocial,environmental,economicandmotivationaldeterminants,andpotentialobstaclesto

change.Interventionsthataremoredeeplyembeddedincontextmaybeperceivedasmorerelevantandthusmaybemoreeffectiveintriggeringbehaviourchange.Knowledgeandawarenessareusuallyinsufficientforchangetooccur,andprogramsneedtoaddresswider

barrierstobehaviourchangesuchaseconomicandculturalfactors(Barnett,etal.2016).Inaprogramencouragingvoluntarytestingandcounselinguptake,forexample,contributionsofmobilehealthmessagingwerefoundtobecounterbalancedbythepersistenceofblaming

attitudesandemergenceofnewsourcesofstigmaassociatedwithanti-retroviralprovision(Mecheal,etal.2010).

Researchconductedby2CVandGirlEffectwithadolescentgirlsinseveralcountrieswasaimedatbetterunderstandinggirls’informationneeds,theresearcherscreatedasetof24cardswith

topicsonthemandgirlsworkedcollaborativelyinsmallgroupstosortthecardsintopilesof“I


amnotinterestedinthis/don’twantthisinformation”“IaminterestedinthisbutIalreadygetit”and“IaminterestedinthisandIdon’tknowwheretofindit.”Thegirlsthendiscussedthe

cardsandprioritizedthemintotheirtopinterests.Theresearcherswerealsoabletoaskprobingquestionsinanaimtodeterminewhetheranytopicsinthe“I’mnotinterested”pileweretooembarrassingtotalkaboutinpublicbutmightbesecretlyofinteresttothegirlsinthetarget

group.Additionally,a“secretenvelope”waspassedaroundwheregirlscouldaskquestionsthattheyweretooshytoaskinfrontofothergirls,andthisinformationwasusefulforresearcherstothendesigncontentinwaysthatwouldgivegirlstheinformationtheyneededinwaysthat

wereadaptedtotheirlevelsofcomfort(GirlEffectand2CV2017).Understandanddesignforuserhabits,preferences,andinterests

Anuancedunderstandingofmediaandtechnologyaccessanduseiscritical.Thisincludesthedevices,channels,mediasources,influencers,datapackages,networkstrength,andotherelementsthataredirectlyrelatedtowhetherandhowpeopleaccessandusedigitaldevices,

servicesandplatforms.ReviewofafreeinformationhotlineinMadagascar,forexample,exploredusers’channelpreferencesandtheprosandconsofaccessingthehotlineviaIVR,SMSandUSSD.Mostuserswantedtolistentomessages,notreadthem,makingIVRmore

appropriateforthepopulationwhohadlowliteracylevels.However,someuserswantedtostoretheinformationreceivedontheirphones,andSMSwaspreferableintheircase.‘Discoverability’wasbetterwithSMSmessagesascomparedtoIVR,however,meaningmore

usersfoundoutabouttheservicebySMS(Wills,MorainandBaudeau2016).

ThosedesigningandimplementingSBCCprogramswithdigitalormobilecomponentsshouldalsobecarefulaboutmakingassumptionsbasedonhigh-leveldata.‘Ground-truthing’needstotakeplacesothatanuancedunderstandingguidesdesign.AnexampleisaprograminSierra

Leonewhereimplementersassumedthatthemoremarginalizedthecommunity,thelowerthemobilephoneaccessforadolescentgirls.SincethecommunityhadnomobilecoverageorradiosignalandwasstrictlyMuslim,itwasassumedthegirlswouldbehardtoreach,howevergirls

hadalotofaccesstomobilebecauseparentswouldsendgirlsseveralmilesoutofthecommunitywiththefamilymobiletocheckmessagesandcallfamilymembers,andgirlswouldthenhaveplentyoftimealonewiththephone(Bachan2016)

2CVandGirlEffect‘sresearch(mentionedabove)foundthatgirls’hadarichandvariedhabitswhichvariedbytheirlocationwithinaparticularcountry(e.g.,urbanorrural).Relatively

wealthiergirlsinsomecountrieshadgreateraccesstomobilesthantheirout-of-schoolcounterparts,whereasinothercountriesgirlswhowerewealthierwereusuallyinschoolanditwasconsideredinappropriateforthemtohavephones,howevergirlswhowereoutofschool

tendedtohavejobsaswellasphones(GirlEffectand2CV2017).Watchoutforcostanddatalimitations

Lowincomeuserstendtohaveveryfrugalaccesshabits,andSBCCprogramsaimedatthesepopulationsshouldcarefullyconsiderhowtominimizedataconsumption.Thismeansthat


programsneedtobedesignedforthosewhohavelimitedminutesortextmessagesallowed.Eveniftheyhavesmartphones,theymaybemostlyofflineandrarelyconnected.Somepopular

smartphonescomeequippedwithapplicationslikeWhatsAppandFacebookLiteorfreeFacebookpre-installed,andsomedatapackagescomewithacertainnumberofminutesforthesespecificplatforms.Becauseofthis,usersmayspendmostoftheirtimeinsideofthose

platforms,makingitdifficulttoreachthemelsewhere(deLanerolle,WaltonandSchoon2017).ThinkcarefullyaboutlanguageandliteracymHealthinterventionsfordiseasepreventionoftenusetextmessagesforhealthinformation,motivatingindividuals,andencouragingself-management,howeverilliteracycanbeanissueif

basingoutreachonSMS(Mecheal,etal.2010).Languagecanalsobeapoliticizingfactor.WhentheV4CprogramwasbeingimplementedinNigeria,thequestioncameupofwhethertouseEnglish,Hausaorotherlocallanguages.Researchwithstakeholders,includinganonline

survey,WhatsAppsurveyswithnon-Hausaspeakers,reviewoftrafficandengagementdatawasconducted.Inaddition,asocialmediadiscussionwaslaunchedaroundwhetheraHausawebsitewasappropriateforNigeriaorifitwouldsparkpolitical/socialissuesorfeelingsofpreferences

orexclusion.HausaspeakersstronglysupportedtranslatingthewebsiteanditsadvertisingintoHausa.However,BrandAmbassadorsandfacilitatorsfeltthatlocalizationshouldnotbetakenlightlyasnon-Hausaspeakerperceptionofthebrandneededtobeconsideredalso.Intheend,

itwasdecidedtokeepwebsiteinEnglish,buttoaddkeyphrasesaboutunityinYoruba,andIgbo.Thebrandwasalsoharnessedtosendamessageofunity.Somelocaladvertisingwasdoneinlocallanguages,withcarefulattentiontowhichlanguagewasusedinareasthathadmixed

Hausa-non-Hausaspeakers.Additionally,promotionalandwebimagesbeganusinga50-50splitofmaleandfemaleandHausa/non-Hausapersons(Every1Mobile2016).DeterminetherightfrequencyofmessagingorengagementThoughSMS-basedreminderstriggerindividualstothinkabouttopicstheymightnototherwiseconsciouslyconsider,oronlyconsideratcertaintimes,suchasduringacrisis,itischallengingtoprovideremindersthatindividualswanttoreceivewithoutbeingseenasrepetitive,annoying

or'nagging'whichwillmakethembelesslikelytoa)payattentiontothemessageandb)continuetosubscribetoreceivethemessages.Thetimingofmessagebroadcastshouldbedesignedtobebothacceptabletoparticipantsandrelevanttotheinformationinthemessage

(GoldandLim2010).TailormessagestotheaudienceandtheindividualuserDigitalinterventionshaveadvantagesovertraditionalSBCCmodesbecausetheycanbe

designedflexiblytoincludeinteractivecomponents,opportunitiesforfeedback,andconvenientlytimedmessagingtoengagedifferentusersintargetedways.Tailoredintheonlinecontextmeansthatindividualizedmessagesaredesignedusingapre-assessmentofkey

individualdifferencevariablesorcharacteristicslinkedtothebehaviourchangemodelthatisbeingused.Tailoredmessagesaremorerelevanttousersandtendtocommandtheirattentionbetterandpersuadethemmoreeffectively.Tailoredmessagingalsousersmorelikelytoprocess


andhaveapositiveperceptionofeducationalmaterials(Lustria,etal.2013).Interventionsshouldalsobetailoredtothetargetaudience,forexample,theirdigitalpreferencesand

habits(KordaandItani2013).Levelsofengagementandusetendtoincreasewheninterventioniterativelymodifyaplatformtoreflectemergingtrendswithadolescents(Baños,etal.2017).Interventionswithmultiple,complimentary,focal-themedmessagesthatencourageusersto

engagewiththeplatformandinteractwithotherusersontheplatformarerecommended(KordaandItani2013)(Shane-Simpson,etal.2017).Usemulti-mediaandtransmediaapproachesMulti-mediaandtransmediaapproachesmayhavebetterresultsformorecomplexbehaviours.InthecaseofEastLosHigh,forexample,transmediayieldedsignificantlybetteroutcomesthandidotherconditionsovertime.Thoughcondompromotionmessageswerethe

sameacrossallexperimentalconditions,whentheywereaddressedontheshowandalsoinanextendedvideoclipdialoguewithamaincharactertheinterventionsawastronguptickininterestandclicks.(WangandSinghal2016).

Buildtrustandfindtherightvoiceandmessenger(s)Tomaximizelikelihoodofbehaviourchange,inadditiontomessagesbeingnovelorengaging,therecipientneedstobelievethesenderofthemessageisacrediblesourceofinformation

(GoldandLim2010).Findingtrustedmessengersandinfluencersisaknowngoodpracticefrompre-Internetdays,anditholdstrueinthedigitalsphereofSBCC.It’simportanttoconductresearchandtestingtodeterminewhoisthebestmessengertocarryaparticularmessagetoa

specificaudience,andtodeveloptherighttoneofvoiceorpersonalityforamessageorconversation.ThiscouldbeanyonefromatrustedINGOorNGO,aTelecomscompany,agovernmentagency;acelebrity,afictionalcharacterdevelopedspecificallyforSBCCpurposes,

tofriendsandfamilies.PayattentiontoframingofmessagesMessageframingseemstosignificantlyinfluencemessageeffectiveness.Positivelyand

negativelyframedmessagesaroundhygienehaveledtoasignificantimprovementinhygienepractices,whereasneutralmessagesledtoincreasedawarenessbuthardlyanychangeinbehaviours(Barnett,etal.2016).

PreparewithsignpostingandensurecapacitytorespondtodemandgeneratedSBCCprogramsshouldbepreparedtorespondtoanydemandthatiscreatedbyincreasedawarenessandhavereliable,trustworthyandvettedonlineandofflinesupportsystems/

servicestowhichuserscanbesignposted.AcommunityandradioeducationcampaignopenedupnewchannelsforreportingchildabuseinBenin,forexample.Despiteearlyengagementwithlocalchildprotectionservices,oncereportsstartedcominginlocalgovernmentdidn’thave

capacitytorespondtothecalls.Insomecases,childvictimsofabusethathadbeenreportedthroughtheprogramsSMSservicewereremovedfromabusivefamiliesandplacedwithlocalpoliceofficers,leavingthemopentootherpotentialformsofabuse(Bachan2016).


DevelopengagingcontentBuzzfeed2hasaneffectiveengagementmethodthatinvolvespackaginginformationinconcise

andestheticallyappealingformats.Goodpracticesforcreatingengagingcontentincludeattractingattentionwithheadlinesofeightwords--thesuggestednumberofwordsforthehighestclick-rate.Buzzfeedusesa“Bite-Snack-Meal”modelwherethebiteistheheadline,the

snackisthethumbnailimageandshortdescriptionofthearticle,andthemealisthefullarticle.Buzzfeedalsopersonalizescontent,basedonuserpersonasandcharacteristicsgeneratedfromclick-dataresearchthatsuggestthatcertainindividualswouldchoosecertainarticles.Buzzfeed

alsoallowsitsreaderstocategorizeandorganizearticlesinwaysthataremeaningfultothem.Itfeaturestimelycontentandtrendingtopicsthatareconstantlyupdated.Lastly,Buzzfeedcontentispre-packagedtobeeasilyshareable.Thismeansthatreaderssharemoreoftenand

becomepromotersoftheBuzzfeedbrand.(Manzo2015)(Shane-Simpson,etal.2017).Itispossibletoengageoff-lineusersthroughlow-endmobilesaswell.GramVaani’sMobile

Vaaniisdescribedasasocialmediaplatformforruralareas.ItrunsonanIVRsystemthatallowspeopletocallintoleaveamessageabouttheircommunityortolistentomessagesleftbyothers.TheMobileVaaninetworkcoversseveralstatesinIndiaandithasprovedtobeas

effectiveasconventionalmediaand,insomecases,moreeffective,sinceinmanyruralhouseholdsthemobilephoneistheonlyelectronicpossession.MainKuchBhiKarSaktiHoon(EmpoweringWomen.AcceleratingChange)wasacampaignthatusedamixofinter-personal,

groupandmassmediachannelswithanemphasisonincorporatingparticipantvoices.AnaudioversionoftheMainKuchBhiKarSaktiHoontelevisionseries(whichfocusesonayoungwoman’sabilitytobreakfreefromhertraditionalsocietyandbecomeaprofessional)was

created.Eachaudioepisodewasfollowedbyasetofquestionstoinitiateandencouragediscussiononthatparticularepisode.UseropinionsweresolicitedviaGramVaani,andadiversesetoflistenersfromvariousregions,classesandethnicitiesengagedindiscussions.Atfifty-two

percent,contributionsfromwomenintheGramVaanicampaignwerereportedlyhigherthaninmainstreammediacampaigns,whichhoverbetweentwenty-fiveandthirtypercent.Overall,

therewereoverathousandcontributionstothecampaigndiscussion,andalmost29,000peopleaccessedthecontent(GramVaani2015)(GramVaanin/d).

CGNetSwaraisasimilarsystem,designedforsharingnewsinareaswherethereislittlemainstreammediacoverage.Thevoice-basedportalisfreelyaccessibleviamobilephone,andanyonecanreportandlistentostoriesoflocalinterest.Reportedstoriesaremoderatedand/or

annotatedoreditedbyjournalistspriortobecomingpubliclyavailableforplaybackonlineandovermobile(CGNetSwaran/d).

2BuzzFeed,Inc.isanAmericanInternetmedia,newsandentertainmentcompanythatiswell-knownforitsdigitalsavvyandfocusonviralcontent.Buzzfeedwasinitiallyfamousforonlinequizzes,listsandpopculturearticles;howeverovertime,thecompanygrewintoaglobalmediaandtechnologycompanyofferingpolitical,do-it-yourself,animalandbusinesscontent.


ConsiderusingInternetMessagingPlatformsand/orChatbotsAlthoughmessagingplatformsonlyworkonsmartphones,theirusehasgrowndramaticallyinrecentyears.By2018,3.6billionofpeople(almosthalftheworld’spopulation)wasexpected

tobeusingthem.Theshifttomessagingappswasdrivenbyeconomics–theyalloweduserstosendtextmessagesatlow(orno)cost.Assmartphone,Wi-Fiandmobile-internetaccesshasgrown,messagingappshaveevolvedintomulti-mediahubsthatofferusersnewwaysofsharing

informationandimproveddatasecurity.A2018studynotesthatpeoplearemorelikelytoopenmessagesthatcomeinviamessagingapplicationsthantheyaretoopenSMSmessages(e.g.,theyhaveahigh“openrate”).Inmid-2016,WhatsApprolledoutencryptedmessagesasthe

default,andothermessagingappshavefollowed.Thismeansthatthecontentofmessagessentontheseapplicationsisnotaccessiblebyanyoneotherthanthesenderandtheuser.(ICRC;TheEngineRoom;BlockParty;2017)(EchoMobileandDigitalImpactAlliance2018).

AlongsidetheexpansionofInternetMessagingplatformsisanincreaseintheuseofChatbots.Mostmessagingplatformssupportbottechnology,thoughthecomplexityanddegreeof

artificialintelligenceofthebotofferingsdiffer.Mostbotsperformsimplefunctionslikeaspromptinguserstopresssuggestedresponsebuttonsorreplyingtousers’questionswithshortanswers.Basicbotsarefairlyinexpensivetodeployandhaveno“mindoftheirown”that

couldtakeconversationsintounpredictableplaces.Anumberofanglesneedtobeconsideredifconsideringusingabot,souserandcontextresearchareacriticalpointwhendesigningorincorporatingachatbotintoaprogram.Aspectsofinclusionneedthoroughexplorationaswell

(EchoMobileandDigitalImpactAlliance2018).BudgetformoderationofcontentandcommentsBothFacebookandWhatsApphavebeenusedtospreadfalseandmisleadinginformationthat

manybelievehasinfluencedelectionprocesses,fueledhateandviolence,andledtowiderdivisionsontopicslikeraceandgender.MisinformationspreadonWhatsAppgroupsinBrazilisreportedtohavesealedthewinofextremistJairBolsonaroinBrazil’s2018Presidential

electionsandWhatsApp-spreadmisinformationhasignitedviolentattacksandevenmurderforpeopleinIndia,Mexico,andelsewhere.(Kayyali2018).Tocounterthespreadoffakenews,inJanuaryof2019WhatsAppdecreasedfromtwentytofivethenumberofcontactsorgroup

chatstowhichausercanforwardamessage.AheadofnationalelectionsinIndia,however,thischangedidnotappeartobehavingastrongimpact.IndianpoliticalpartiesreportedlypouredmoneyintothecreationofhundredsofthousandsofWhatsAppgroupstospreadpolitical

messages,whichaccordingtoresearchers,frequentlycontainanddisseminatefalseinformationandhatefulrhetoric,especiallythatfueledbyHindunationalists(Perrigo2019).

Astheeffectsofmisinformationanddisinformationhavebecomemoreapparentglobally,platformcompaniesarebeingcalledontomakedecisionsaboutspeech/freespeechandwhat

contentisallowedontheirplatforms(Caplan2018).Womenandgirlsinparticularalsofaceharassmentonline,especiallywhentheyspeakoutongenderdiscriminationorgoagainstsocialandculturalnorms(Vlahakis2018)andforthisreason,platformsanddigitalinitiatives


mustensurethattheyhavesufficientresourcingformoderationthatcanmanagecommunitiesandwatchforanymisuseoftheplatformorharmfulcontentorcomments.

Don’tbefooledbyvanitymetricsVanitymetricsarethosethatfocusonlargenumbersofreachorengagement,forexample,pageviews,clicks,orlikesasopposedtoqualityofengagementwithcontentorimpactofthe

content/platform/service.Numberscanbedeceivinginmanyways,forexample,thenumberofcallsreceivedtoanIVRlinemightbequitelarge,butifthenetworkisweakandthemajorityofcallsdrop,thenthenumberofcallsreceivedisnotverymeaningfulintermsofimpactbecause

veryfewpeoplewillhavelistenedtoarecording.Ifpeoplearriveatawebsitepageandleaveimmediately,chancesaretheyarrivedatthewrongpageanddidnothaveinterestinthecontentthere,andthisshouldnotbeconsideredasignofhighuptakeorinterest.Identifying

solidmetricsandunderstandingfromwherenumbersemanateiscriticaltodesigningimpactmeasurement.Inadditiontonumbersandanalytics,itisimportanttodeterminewhetheronlineknowledgeorstatedintenttranslatesintoofflinebehaviour.

5.DigitalSafeguardinggoodpracticesThe‘DigitalSafeguarding’umbrellacoversa)safeguardingwhenpeople,inthiscase,children

andadolescents,usedigitaldevicesorengagewithdigitalservicesorplatforms;b)dataprivacyandprotection;andc)datasecurity(Raftree2018a).Therearemultiplesafeguardingareastoconsiderwhenengagingadolescentsindigitalinitiatives,including:

• bigpictureethicalandriskreview(isembarkingonthiseffortethical?);• safeguarding,informedconsent,privacyandsecurityofdatacollectedduringformative

orongoingresearchactivities;

• buildingprivacyanddatasecurityintoplatform,productoroutreachdesign;• ensuringthatthebenefitsofcollectingpersonalorsensitivedataoutweightheprivacy

risks;

• followingapplicablecountryandgloballegalframeworksrelatedtodataprivacyandtransmissionacrossborders;

• protectingandsecuringanydatacollectedorgeneratedthroughouttheprocess

• ensuringaclearhandleonhowandwheredataflowsandreviewinghowpartnerswillmanagedataprivatelyandsecurely;

• andconsiderationsaroundtheonlineexperienceitselfsuchassafeandappropriate

content,commentsandmoderation,andprotectionfromonlineabuse,bullying,harassment,hatespeech,violence,scamsorgrooming(Raftree2018).

Genderdimensionsofonlineharassment,violenceandabuseOnlineviolenceandabusecanmanifestassexual,psychologicaloremotional.Onlinesituations

canalsoleadtophysicalharm,violenceandabuseoffline.(EuropeanInstituteforGenderEquality2017).Therearemajorgapsindataontheprevalenceandimpactofonlineviolence


andabuseagainstwomenandgirls,includinggapsininformationabouttheexperiencesofwomeninAfricaandAsia(VanderWilk2018).Researchthatdoesexistsuggeststhatwomen

aremorelikelytoexperienceonlineviolencethanmenandtheyfacemoreseriousconsequences(Powell2018).AmnestyInternationalfoundthatalmostaquarterofwomensurveyedacrosseightcountriessaidtheyhadexperiencedonlineabuseorharassmentatleast

once(AmnestyInternational2017).Inadditiontohighlevelsofharassmentandabuse,womenexperiencemoresevereformsofonlineviolencethanmen,includingphysicalthreats,harassmentoverasustainedperiod,sexualharassmentandstalking(PewResearchCentre

2018).Thenatureofthedigitalspace,witheasyamplificationandscale,enablesharmcausedto

survivorstoalsobeamplified.Violenceandabuseagainstwomenarecarriedouttodayatgreaterdistance,speedandrates.Meanwhile,anonymityandencryptionprotectperpetratorsfrombeingknowntobothsurvivorsandauthorities(Aziz2017).Themostcommonsocialmedia

platformhighlightedinonestudy–bothforexperiencingandwitnessingabuse–wasFacebook.OthersocialmediaplatformswhereresearchparticipantsexperiencedwereTwitter,InstagramandYouTube.Otherplaceswhereabusehappensincludedontheinternetin

general,viamobilephones,andonmessagingservicessuchasWhatsAppandemail.Sexualorientation,genderidentity,raceandageoftenexacerbatethetypeofonlineviolenceandabusethatwomenface(Vlahakis2018).Forexample,LoomNepalreportedthat100percentof

transgenderrespondentstotheirstudyreportedhavingbeensubjectedtoviolenceonlineand84percentofrespondentswhoidentifiedasgay,lesbian,bisexual,orothernon-heterosexual

orientationhadbeensubjectedtoonlineviolence.Adisproportionatelyhighnumberofpeoplereportingonlineviolenceinthestudyidentifiedaslesbian,gay,bisexualortrans(LGBT)orDalit.Onlineharassmentandabuseoftenleadtoself-censorshipandthesilencingofwomen

(LOOMNepal2017).

A2018reportongirls’perceptionsandexperienceswithmobilephonesshowsthatgirlsacross

differentcountriesareacutelyawareof‘safety’asanissuewhenthinkingaboutmobileaccessanduse.Whilegirlswerepositiveabouttheroleofphonessupportingtheirsafety,theyalsopointedoutthattheyandtheirparentshadanumberofsafetyconcernsandfears,forexample,

thatphonescouldbeagatewayto‘badpeople,’namely,boysandmen,orothergirlswhocouldbeabadinfluence.Girlsandboysalsosaidthatmobilescanleadthemtodangerousinformationandimagerysuchpornographyand‘badmessages’thatupsetthemand‘corrupt’

theirmoralbeing.Otherfearsthatgirlsspecificallymentionedwereexposureto‘sugardaddies’;watchingpornography,which‘tempts’girlsintotryingsex;beingtrickedintomeetingmenandboysforsex,orevenprovidingtheopportunityforthemtobeputunderaspellandseduced.

(VodafoneFoundationandGirlEffect2018).

EmergingprivacyregulationsTheEuropeanUnionpassednewdataprivacyregulationsinMayof2018calledtheGeneralDataPrivacyRegulation(GDPR).Thisoverarchingframeworkhassetthebarforarights-based


approachtodataprivacythatplacesagreateramountofcontrolintothehandsof‘datasubjects’whosedataiscollectedandusedby‘datacontrollers’and‘dataprocessers.’Though

theGDPRdoesnotapplytoallcountries,itdoesrequirecompanieswishingtotargetEUcitizenswithoffersandservicestocomply.ThelegislationalsocoversEU-basedentitiesthatarecollectingdatainothercountries.RightsandprinciplescontainedintheGDPRinclude

conceptssuchaspersonaldata;sensitivedata;legalbasesfordatacollection;legitimatepurposefordatacollectionanduse;informedconsent;dataminimization;andrightstorectification,erasure,informationabouthowpersonaldataisbeingused,andrightstorefuseor

restrictprofilingonthebasesofdata(EuropeanUnion2016).TheGDPRmandatesparticularattentiontodataprivacyandethicswhenworkingwithchildren,includingthestipulationsaroundparentalconsentandspecialprivacyconsiderationsforthedataofchildren(UK

InformationCommissioner'sOffice2018).ManydevelopmentorganizationsaremakingeffortstoadoptthephilosophiesoftheGDPRduetothebeliefthatallpeopledeservethesamedataprivacyrights,andthattheseareespeciallyimportantforvulnerablegroupswhomayface

evengreaterriskifdataprivacyisbreachedortheirdataisabusedoraccidentallylostorleaked(Porter,etal.2018).

Indiaisinthemidstofchangestoitsprivacylaws,withproposedrulesthatwouldamendSection79ofthecountry’sITAct,theprimarylawononlinecommerceandcybercrime.Iftheseamendmentsareapproved,platformslikeFacebookandTwitterwouldberequiredtocensor

contentthattheIndiangovernmentconsidersinappropriate.ThismovecouldalsoaffecthowcontentisservedoutsideIndianborders.Theamendedlawwouldalsorequiresocialmedia

companiestoofferupusermessagesatgovernmentrequest.Thisisachallengeforend-to-endencryptedserviceslikeWhatsApp,whichhasspecificallyencryptedmessagesinanefforttoassureusersthattheirpersonalmessagesaresafefromcorporateorgovernmenteyes(Kelly

2019).MTVNishedhwillneedtoreviewitslegalobligationswithregardtoprivacylawsfortheEU,US,India,andanyothercountrywheredatamightbecollected,transmittedorprocessedbypartnersorsub-contractors.

DigitalSafeguardingandResponsibleDatainthecivilsocietyspaceDigitalsafeguardingandresponsibledatamanagementhaveemergedaskeyconcernsandareas

offocusforthoseimplementingprogramsthatincludedigitaldata,digitalplatformsordigitalcontent.Whiledigitalplatformsthemselvestrafficinuserdata,developmentpractitionersandhumanrightsactivistshavebecomeincreasinglyvocalabouttheneedforgreaterattentionto

howdatafromvulnerableorat-riskpopulationsiscollected,transmitted,used,storedandsharedandhowonlinespacescanbeprotectedfrommanipulation,hatespeechandabuse.

TheconceptofResponsibleDataemergedsomefiveyearsagooutofeffortsbyTheEngineRoomandotherinterestedpartners.Responsibledataisdefinedas“Thedutytoensurepeople’srightstoconsent,privacy,securityandownershiparoundtheinformationprocesses

ofcollection,analysis,storage,presentationandreuseofdata,whilerespectingthevaluesoftransparencyandopenness”(ResponsibleDataForum2014).


Alistoftenrecommendationstoenhancedataprivacywasdevelopedforthedesignphaseof

theGirlEffectMobileplatformtohelpensuregirls’safeguardingandtoaddresstechnological,cultural,legal,andcontextualrisksaswellastoensurethatreputationalrisks,trustworthiness,onlineabuseanddataprivacyandsecuritymeasuresweremitigated.Theseincludedaspects

likeassessingrisktogirlsatvariouslevels,includingtechnology,culture,legal,contextual,trust/abuse/reputation,dataandprivacybreachesanddataethics;understandingdigitalprivacyandsafetyfromgirls’perspectiveineachcontextbeforedesign;workingwithgirlgroupsto

identifyprivacy,securityandreputationalrisks,todesignanddevelopofsafetyandsecurityprotocols;toflagnewrisksthatariseduringimplementation,andtoadaptanditeratearoundsafeguardingapproacheswhennewconcernsarise.Anotherrecommendationwastoassignor

consultwithadigitalsecurityexpertineachpilotefforttoconductaprivacyimpactassessmentandrecommendcontextuallyspecificapproachesanddesignelementsthatarerelevanttoeachcontextthatcanlaterbefilteredupintoaglobalsafeguardingapproach.Theguidance

recommendedfollowinglegalguidelinesthatapplyinthevariouscountriesofoperationorreachandestablishingorganizationalpoliciesandstandardsondatacollectionandsecurity,includingPrivacybyDesignstandards.Trainingforstaffonpotentialrisksandorganizational

liabilities,accountabilityprocesses,responsibilitiesandconsequencesforprivacybreacheswerealsorecommended,aswellastrainingforlocalpartnersandallies(Raftree2016).

Anumberofotherguidingdocumentsexisttosupportpractitionerswhoarethinkingthroughtheseareas.Asetofrelevantpolicyandguidanceforthedevelopmentandsocialimpactfield

wereassembledonthe“ResponsibleDataHackPad”byagroupofResponsibleDataenthusiastsasaresourcefortheMERLTechconferencein2016andsubsequentlyturnedintoalivingdocumentthatiscontinuallyupdatedasnewresearchandguidancearise(MERLTech

Participants2019).

6.Monitoring,evaluationandlearningfordigitalSBCCAsnotedearlierinthispaper,theevidencebaseondigitalSBCCissomewhatweak.Higgsetal.,

whenreferringtothemHealth,socialandtransmediaapproaches,andbehaviourchangeevidencebase,recommendthatfuturestudiesadoptrigorousdesignsandendpoints.Theevidencebaseneedstobesystematicallybuilttoavoidwastingresourcesandenable

comparisonsacrossstudies,modalities,andoutcomes.Additionally,dataanalysisandreportingonthesetypesofprogramsshouldsystematicallyarticulatebasicimplementationinformation,suchasthetypeofmediaplatformused,thecontentpresented,mediadistribution

andreach,consumerexposure,andotherspecificsrelatedtothemediainterface.Additionally,theyidentifyaneedforqualitativestudiesthatprovidearicherunderstandingabouthowculturalfactorsshapetheadoptionandsuccessofnewtechnologies.Theyrecommendthat

qualitativestudiesexaminehowbehavioursandsocialnormsinteractandhowsocialnormsshiftinimportantwaysduetotheinfusionofnewtechnologiesinthecommunity,and


differencesinroledefinitionsbroughtaboutbecauseofnewtechnologies(Higgs,etal.2014).BelowarehighlightedsomecasesthatmayprovideideasforhowtodesigntheMELforMTV

Nishedh.Additionally,somenewertoolsthatcouldbeincorporatedintoMELarecovered.

16andPregnantTheevaluationteammatchedVitalStatisticsbirthdatatoNielsentelevisionratingsdatato

determinewhetherexposuretotheshowhadanimpactonteenchildbearing.Theyimplementedaninstrumentalvariables(IV)strategyusinglocalareaMTVratingsdatafromapre-periodtopredictlocalarea16andPregnantratings,andalsointroducedeventstudy

methods,usingthespecifictimingoftheshow’sintroductiontoidentifyacausaleffect.ThosefindingsweresupplementedwithareviewofdatafromGoogleTrendsandTwitterthatsuggestedthattheshowledtoincreasedinterestincontraceptiveuseandabortion,ascaptured

byinternetsearchandtweetingbehaviour(KearneyandLevine2015).Asnotedearlierinthispaper,theseresultswerequestionedbyresearcherJenniferAubreywhoconcludedthattheywerebasedonfaultyanalysisofcorrelationalresultsfromaggregatedata,andthatinfactthe

showhadnuancedimpacts.Girlswhoidentifiedwiththemaincharactershadmorepositiveattitudesaboutteenpregnancythangirlswhocouldnoteasilyrelatetothem(Behm-Morawitz,etal.2017).

EastLosHighEvaluatorsusedamixedmethodsapproachincludinganalyticstrackingtoassessaudiencereach,whichmonitoredWebtraffictotheEastLosHighWebsiteandtoNGOpartners’Web

sitesandwidgets;aviewersurveytoassessnarrativeengagementandintendedoutcomes;andalaboratoryexperimentwithnon-viewersofEastLosHightocomparetheeffectoftransmediaedutainmentwithotherformsofnarrativepresentation.Complementarymethodsincluded

socialnetworkanalysisandcontentanalysistounderstandthesocialdynamics,messageframing,anduser-generatedcontentonEastLosHigh’ssocialmediapresenceandparticipantobservationandin-depthinterviewswithyoungLatinocouplestorevealEastLosHigh’s

influenceontheirsexualdecision-making.Morespecifically,theteamcollectedanonymousunobtrusivetrackingdatathroughGoogle

AnalyticstocaptureEastLosHighaudiencereachduringthepre-programpublicity,thepremiereofseason1,andafterseason1ended.Theteamrecordedthenumberofvisitors,numberofpageviews,averagedurationofpageviews,andgeographicallocationofvisitors.

Usingageographicalinformationsystemsoftware,theygeneratedspatiotemporaldynamicvisualizationsofthegeographicaldiffusionofEastLosHighwebsitevisitorsovertime.NGOpartnersPlannedParenthoodandStayTeen.orgindependentlymonitoredandsharedthe

statisticsoftheirWebtrafficandhealthservicewidgetsontheEastLosHighwebsite.AnonlinesurveywasembeddedintotheEastLosHighwebsiteandpromotedonsocialmedia

withcustomincentives.Participantsansweredclosed-andopen-endedquestionsabouttheir


impressionsoftheprogram,engagementwithitsnarrativeelements,interpersonaldiscussionsabouttheshow,andknowledge,attitudes,andbehaviouralintentionsrelatedtosexualand

reproductivehealth.Theevaluationteamderivedmeasuresofaudienceexperiencefromestablishedscalesoftransportation,identification,andnarrativeengagement.

Theteamusedapartialfactorialdesigntotesttheeffectofdifferentstorytellingformatsontheshow’stargetaudience.Thecontentwasselectedaccordingtotheprogram’sobjectives,andthemessageswereidenticalacrossallexperimentalconditions.Repeatedmeasuresofnarrative

engagement,knowledge(ofcorrectuseofcondoms,birthcontrolpills,andemergencycontraception),attitudes(onsexeducation,women’srighttochooseabortion,theimportanceofHIVtesting,theimportanceofsexcommunication),andbehaviour(usedprotectionduring

lastinstanceofsexualintercourse,engagedinsexcommunication)occurredatbaseline,posttestanda2-weekfollow-up(WangandSinghal2016).

VoicesforChange(V4C)ATheoryofChange(ToC)wasdevelopedasaniterativetoolguidingthedesignandevolutionoftheV4Cprogramandtohelporganizeanddevelopthinkingthroughouttheprogramteamaboutwhatwasexpectedtohappen,andwhyspecificinterventionswereexpectedtohavethe

desiredeffects.TheToCrepresentsadevelopmentofthesocio-ecologicalmodelofchangeusedinothersocialsciencecontextsanddepictsthreedomainsofchangein‘Self’,‘Society’and‘FormalInstitutions’whereV4Cdelivereddifferenttypesofintervention.

• Self:Youngwomen,youngmenandanumberof‘keyinfluencers’wereexpectedto

acquireknowledgeandskillsaboutgenderequalitythroughprogramcoursesandother

activities–virtualandphysical–whichusedtransformationalmethodologiestoinspiredeeppersonalchangeandcommitment.

• Society:A‘risingtide’populationofgeneralmembersofsocietynotdirectlytargetedby

theprogramwasexpectedtoundergochangeinknowledge,attitudesandpracticesinareaskeytogenderequality(e.g.women’sleadership,women’sdecision-makingandVAWG).Thechangewouldhappenthroughsocialnormschangestimulatedbyamedia

campaignandothermediatoolstogeneratemessages,aswellasimages/depictionsofnewsocialnorms.‘Keyinfluencers’wereexpectedtocontributetothissocialnormschangebydiscussinggenderequalityamongtheirvariousconstituents/listenersand

puttingforwardalternativenorms.

• Institutions:V4Cactivitiestargetedchangesinthelaw,policychange,changesin

governmentplanningandbudgetingprocesses,andchangesinwomen’spoliticalleadershipbroughtaboutpartlybyworkingoncommitmentswithinpoliticalparties.

Thepersonaltransformationinthe‘Self’domainisrootedintheStagesofChangemodelofchange.Theactivityinthe‘Society’domaindrawsona‘diffusion’modelofchangewhichexplainshowchangeatindividuallevelsspreadsthroughsociety(MilwardandNelson2017).


SpringsterThisisamobile-firstglobalplatformthatdigitallyconnectsmarginalizedandvulnerablegirlsaroundtheworld.Itfeaturescontentdesignedforgirlsandcreatedbygirls,aimingtoput

essential,tailoredinformationdirectlyintotheirhands,andtohelpgirlsfindmeaningandstrengthineachother’sexperiences.GirlEffectcommissioned2CVtohelpthemunderstandrealSouthAfricanSpringsterusers’localcontextsandchallenges,userjourneystotheservice,

overalluserexperience,andgirls’viewsonexistinglocalcontent.TheresearchalsoexploredoveralluserengagementandSpringsterimpact–includingchangingknowledge,attitudesand

practices–andwhetherthisvariedfordifferenttypesofusers.BecauseSpringstertargetusersareprimarilybasedinurbanareas,thisresearchtookplaceintheJohannesburgarea,whichhadthemostresponsestotherecruitmentsurvey.GirlswererecruitedfromtheSpringstersiteby

samplingregisteredusersviaanonlinesurveywhichappearedatthetopofusers’platformpage.Thesampleframewaslimitedgiventhenumberofgirlswithinreasonablegeographicrangesforresearchpurposesandwhethertheparticipantcouldbecontacted.Thesample

include14to18-year-oldusersacrossarangeofdemographicsandusageexperiences–includingthoseintheearlystagesofplatformexploration(1-2visits)anddedicatedlong-termusers(6+visits).Despitenotexplicitlysamplingforvulnerability,allusersincludedinthis

researchmet1ormoreoftheSpringstertargetvulnerabilitycriteria–withmostparticipantsmeeting2+criteria.

Theapproachbeganwithdepthinterviewswiththe‘seedrespondent’–theSpringsteruser.Firstresearchersconducteddeep-diveexplorationofthegirluser’scontext,pathwaystoSpringsteruse,engagementanduserexperience,andimpactonknowledge,viewsand

behaviour.Wherepossible,theythenconducteddepthorpairedinterviewswithfriendsandfamilyfromthegirl’swidersocialnetwork–includingfriendsand/orfamilymembers.Theseinterviewstriangulateddataaroundthegirl’sambitions,challengesandsupportneeds,and

behaviourorattitudechanges,andviewsonSpringster.Intotal,theteamconducted20in-depthhomeinterviews,10indepthfamilyorgatekeeperinterviews,and15friendshipinterviews.

Additionally,girls’digitalrepertoireswereresearched,includingthehandsetstheyuseandhowtheyuseandaccessdata,thecostsandkindsofmobileanddataplans,theplatformsandapps

theyuse,informationtheyseekfromtheInternet,how/howoftentheyaccessInternet,howtheyfindtheSpringsterplatform,whattheydoonit,theirexpectationsoftheplatform,usepatterns,emotional/intellectualengagementwiththeplatform,theirperceptionsof‘whothe

platformisfor’,theuserinterfaceandhowtheynavigateit,andaspectsrelatedtothesite’scontent,tone,voice,andinteractivity.

Lastlyevaluatorslookedathowgirlsshareinformationandknowledgegainedfromthesite,theinfluenceorimpactofthesiteonusers(bothhighlyengagedusersand“lurkers”)andmeasuredearlyimpactsinconfidence,knowledge,feelinglessalone,andrelationshipswithfamilyand

friends(JacobsandUpton2018).


Inadditiontothefocusedresearchhighlightedabove,GirlEffectisusingarangeofmethodsto

betterunderstandhowusers’engagementandparticipationrelatetochangesingirls’livesbeforeandaftertheyreadcontentonSpringsterandengageinconversationswithotherusers.Usingdifferentdatasources,suchasGoogleAnalytics,commentanalysis,onlinesurveysand

socialmediaanalytics,theteamiscombiningbigdatawithtraditionalapproachestolearnmoreabouttheintersectionofdigitalbehaviour,participationincommunityspacesandengagementwithcontentandhowparticipationinthisonlinesafespaceinfluenceschange.

(AbreuLopesandBailur2018).

HNI3-2-1ProgramThisprogramofferedmobilephoneusersinMadagascaraccesstoinformationonanumberoftopicsandthemes,includinggenderandgender-basedviolencethroughIVRline,SMSorUSSD

(whichissimilartoSMS,butuserscanhaveaback-and-forthconversation).Midwaythroughtheproject,auserjourneyevaluationwasconductedaswellasabusinessmodelassessmenttoseeiftheprogramcouldfindawaytobecomesustainablewithoutsupplementalfunding.

Corequestionsaskedduringtheuserjourneyevaluationincluded:

• Wherearethebottlenecksinthecustomerjourney?

• Whatarethedriversofrepeatusage?Whatisthemosteffectivecontent?• Howcanwedeterminewhichusersaremorevaluableforamobileoperator?• Whatisuserwillingnesstopayfortheservice?

• Whatgenderisusingtheservice?Whatcontentismosteffectivebygender?• Howdomarketing/promotionalactivitiesimpact3-2-1?

Businessmodelquestionsincluded:• Howcanwetacklethebottleneckintheuserjourney?• Whatisuserwillingnesstopay?IsthereagoodB2Cmodel?

• HowcanHNIbetterusetheirdatatosupporttheircoststructure?• HowcanHNIstrengthentieswithallkeypartnerswhoarecriticalforbusinessmodel?

Thethreechannels(IVR,SMS,USSD)weremonitoredtodeterminetheiruptakeandeffectiveness,wherecustomersweregettingstuckintheuserjourney,atwhatpointtheybecomeengagedandrepeatusers(orwhytheydidn’t),userinterfacetesting,thevalueof

differenttypesofcustomers(cursoryandrepeatcustomers).Customersweresegmentedbytypeofuserandwheretheywereintheuserjourney.

Byfocusingonthelisteningbehaviourofusersanddetermininghowmanyare‘engaged’(thosewholistenedtothemajorityofamessage)theteamcouldassesshowmanyusersinthebaseusedtheserviceinawaythatcouldcreaterealvalue.Ifbehaviourchangewastohappen,itwas

apre-conditionthattheuserhavelistenedtoamessageinfull,thereforeonly‘engagedusers’couldbeexpectedtoexhibittherelevantchangesinbehaviour.


Usingdataanalyticstounderstandtrendsandengagementacrosscontentareas,theteamcould

assesswhichcontentareasgainthemosthits(initialdemand)andthentoassesslevelsofattritionversusthoseuserswhohadtrulyengagedwiththecontent.Usertestimonialswerelayeredovertheanalyticstocompareandunderstand(groundtruththenumbers).Basedonthe

data,ahypothesisaroundwhatiscreatinganybarrierstousecanbecreated,andthedesignoftheservicecouldbecarefullyadjusted,andeachchangeindesignoftheplatform/servicecouldbetestedtoseeifuserengagementimproves(Wills,MorainandBaudeau2016).Asnoted

aboveinthispaper,thistypeofmonitoringandassessmentisaimedatunderstandingandimprovingelementsofplatformorsitedesign,notevaluatingoverallimpactofaninitiative.

ICTtoolsandtechnologyapproachesforMELofSBCCinterventions

SocialmediacommentsanalysisforsocialnormsresearchOnestudysampled2872obesity-relevantcommentsfromthreeyearsofinterestfromamulti-topiconlinemessageboardandconductedaninductivethematicanalysistoidentifykeythemes.Thecommentsanalysisallowedthemtobetterunderstandobesitystigma,dominant

representationsofobesepersons,experiencesofexplicitstigma,thesocialandpsychologicalrepercussionsofovertstigmaandnormsregardingtheperceptionofobesebodies.Theresults

highlighttherichnessofdataavailableviasocialmediainteractionsasawindowintonaturally-occurringdiscourse(DeBrun,etal.2014).

IVRforMonitoringandEvaluatingMediaInterventionsApilotstudywasimplementedinRwandawithVOTOMobiletodeterminewhethermobilephonetechnologiescanamplifythereachofradioprogramsandstreamlinemonitoringand

evaluationofradioprograms.IVRwasofinterestbecauseitofferedthepossibilityofcollectinglargequantitiesofqualitativeandquantitativedataatlowcostbecausemobilephoneuserscouldbeaskedautomatedquestionsthroughtheirphone,thusavoidingface-to-faceinterviews.

IVRwasalsoseentoofferbenefitsbeyondSMSdataasitcouldcollectvoicedata.ThestudyusedIVRbothtodistributecontentandtocollectdatafromlisteners,throughasurveythatlistenersheardattheendofaradioprogram.

Collectingdataonmediacontentviathemobilephonesurveywasfoundtohavedisadvantages.Thesurveywassetupsothatonlythosewhohadlistenedtothecontentwould

receivethesurvey,andthismadeitimpossibletocomparewiththosewhodidnothearthecontent.Surveyquestionsneededtobesimpleandbriefsothatrecipientswouldunderstandthemandbeabletoanswerthem,andlistenersmustbewillingtostayonthelinethroughthe

endofthesurvey.Fortheseandotherreasons,thestudyconcludedthatIVRisunlikelytoofferafullreplacement

formorerigorousevaluationmethods.IVRdoesofferseveralM&Ebenefitshowever,inthatitcanallowprojectimplementersmonitorhowlistenersareexperiencingprograms,trackmobilephonereach(thenumberoflistenerswholistenthroughtheirmobiledevice),gauge


knowledge(whetherthemessagesprovidedbytheprogramareproperlyunderstood),andobtainfeedbackonprogramcontent(engagementwithprogram,enjoyment,desiretolisten

inthefuture,identificationwith/likabilityofcharacters,open-endedfeedback,etc.).Allofthisinformationcanbeusedthroughoutaprojecttoimprovedesignandunderstandimpact.

AdditionalfindingsincludethatdespitethepromisingattributesofIVR,thetechnicallimitationsofthetestedIVRsystemandtheresultingcostofgatheringthedataweretoogreattowarrantarecommendationoftheplatformforconductingjointquantitativeandqualitative

M&E,atleastfororganizationswithoutin-houseexpertiseincomputerprogramming.FindingssuggestthatIVRtechnologyisnotsufficientlydevelopedtosupportM&Eeffortsthatcombinequantitativeandqualitativefeedback.IVRcould,however,beusedasapartof

formativeevaluationandperiodicmonitoringtogetasenseofhowaprojectisdoingandwheretweaksmightbemade(Kraidy2016).

AnotherstudyofVOTOMobilewasaimedatanalyzingwhethermobilephone-basedsurveysarefeasibleandcost-effectiveforgatheringstatisticallysignificantinformationinfourlow-incomecountries(Afghanistan,Ethiopia,Mozambique,andZimbabwe).Inthiscase,resultsshowedthat

thecostofatypicalhouseholdsurveycouldbereducedbymorethan90percentand,withcare,mobilesurveyscanbeafeasibletooltoreplacemoretime-consumingandexpensiveapproaches.Lessonsfromtheexerciseincludedthatunlessthereisspecialattentiontodetail,

mobilesurveyresponseratescanbelowerthanexpectedandtheyoftenexcludethevoiceofpoorandvulnerablegroups.Inclusioncanbeimprovedbyusingvoicemessaginginsteadof

SMS,offeringachoiceoflanguages,creatingaretrypatterntoensurereceipt,providingacallbackchanneltocapturecallsthathavebeenmissed,usinglongerintroductionstosurveystoprovidemorecontext,choosingafemalevoice,andpayingcloseattentiontothetimeofdayof

calls(LeoandMorello2015).

UsingInternetMessagingplatformsforMELIn2016,aheadofSouthAfrica’smunicipalelectionstheAfrica’sVoicesFoundationpartnered

withLivityAfricatoevaluatetheimpactofacampaigntoencourageyoungpeopletovoteandhighlightissuesthatmatteredtothem.Theyusedonlinesurveysofyoungpeople,conductedvia

emailandthroughWhatsAppandFacebookMessenger,andsocialmediaposts.WhatsAppandMessengerwereselectedbecauseoftheirpopularityamongyouth.ItwasfeltthatWhatsAppgroupsencouragedconversationsthatwouldyieldparticularlyusefulfeedbackbecausethedata

thatcanbegatheredisrich,authentic,andprovidesinsightsintosocio-culturalbeliefsandbehaviours.Therewereconcerns,however,aboutprivacywhenusingbothFacebookMessengerandWhatsApp.Althoughtheteamsoughtconsentandstoredthedatasecurely,

theydidnotfeelconfidentabouthowthedatamightbeusedbytheplatforms.Thiswasofparticularconcernbecauseyouthwereaskedforpersonalinformationlikevotinganddemographics.Theorganizationshavedecidedthattheywouldnotusetheseplatformsinthe

futureunlessprivacyrisksarefullyunderstoodbeforestarting(ICRC;TheEngineRoom;BlockParty;2017).


Anotherchallengewithusingmessagingappsisthepotentialforbias.Relyingondata

gatheredsolelythroughoneapprisksintroducingstatisticalbiasintoanalysesofaparticularissue.Biasesarerelatedtodifferentialaccesstotheinternetandmobilephoneswhichaffectsrepresentationofdifferentpopulationsandpotentiallyreinforcinginequalities.Thistypeofbias

couldalsoexacerbateblindspots:forexample,reportscollectedthroughmessaging-appdatamightgivetheimpressionthataneventorsituationwasconcentratedinaparticularpartofacountry,withouttakingaccountofnetworkoutagesorlimitedinternetaccessinotherpartsof

thecountry.Additionally,womenmaynotbeabletoaccessorsubmitdataoveramessagingappwithouttheapprovalofamanintheirhousehold.(ThoughitshouldbenotedthatthisistrueformanyICTsandfeedbackinitiatives)(ICRC;TheEngineRoom;BlockParty;2017).

BigdataandevaluationWhenbehaviourchangeprogramsarecarriedoutonmultipleplatformsthatgeneratedata,itis

possibletoemployBigDatamethodstoconductresearch,monitorandinsomecasesevaluatethem.GlobalPulse,forexample,useddigitaldatatounderstandconversationsaboutcontraceptionandteenagepregnancyamongUgandanyouth,andtoanalyzetheirperceptions

towardsdifferenttypesofcontraception.DataforthestudywasextractedfrompublicFacebookpostsandUNICEF’sU-reportplatform,anSMS-basedpollingsystemforUgandanyouth.Theanalysiswasvisualizedinaninteractivedashboardthatofferedsnapshotsofthe

conversationsaboutdifferentmethodsofcontraceptionovertime.Italsovisualizedpopulardiscussiontopicsassociatedwithcontraceptionandteenpregnancy.Thestudyshowedthatthereispotentialfordigitaldatatohelptrackattitudesandperceptionsaboutreproductive

healthandrightsinrealtime,whichcouldsupportmonitoringandevaluationofdevelopmentprograms.(UNGlobalPulse2014)(AbreuLopesandBailur2018)

Inanotherexample,WellToldStoryconductedsentimentanalysisonmillionsofdigitalmediaexchangesamongfans.Theanalysisproducedinsightsonthepowerofcollectivediscussionstostimulatemassivenormativechangesaroundcontraception.Alongitudinalpanelstudylinked

thenormativeshiftstobehaviourchangeatscale.WellToldStoryisaresearchandmediacompanyworkingonsocialandbehaviourchangeinEastAfrica.Theagencyproducesafree,

data-driven,multimedia(comic,radio,events,digitalmedia)brandcalledShujaaz.AsacomponentoftheShujaazmedia,guided,dynamicdigitalconversationsamongShujaazfansonFacebook,WhatsAppandviaSMScreatearichdatasetthathelpsinformlearningandstrategy

(WellToldStory2017)(AbreuLopesandBailur2018).AlthoughthereareemergingexamplesofhowbigdatacouldbeusedtoevaluateSBCC,there

areplentyoflimitationsincludingconcernsaboutitsuse.CarelessinterpretationofbigdatamightleadtodisproportionaterepresentationofthosewhoareabletoaccessandusetheInternetandleavebigdatatrails.Useofbigdatarequiresadequatenormativeframeworks,

protectionofprivacy,andrisksandharmsmitigationforindividualsandgroupsofindividuals.Accesstobigdatahasproventobechallenging,anditrequirestechnicalknowledge,skillsand


budgettoretrieve,handleandstoreit(AbreuLopesandBailur2018).

7.ConclusionTheevidencebaserelatedtodigitalsocialandbehaviourchangecommunicationsisgrowingandbecomingmorerobust,yetthereisstillaneedformoretheory-basedapproachesandfor

on-goingresearchandevidencebuildingtobetterunderstandthecomplexnuancesofthisspace.DigitaltoolsandsocialmediaareanaturalfitforSBCCbecauserichandmulti-layeredapproachestendtohaveagreaterimpactonbehaviourchange.

Atthesametime,theprocessofcreatingandfinessingadigitalSBCCapproachisacomplexonethatdependsonfactorsattheindividual,family,community,societylevels,aswellasmedia

contextandaccesstodigitaldevices,channels,platforms-andskillstonavigatethem.DigitalSBCCpractitionersmustbeawareofthepotentialpaternalisticandmanipulativesidesof

digitalbehaviourchangeapproachesbothforethicalreasonsandbecausenewandemerginglegislationprovidesgreaterrightstodatasubjectsintheareaofdataprivacy,protection,securityandprotectionfromdigitalprofiling.

Anumberofgoodpracticeshaveemergedfromtheliteratureandfromknowngoodpracticeinthefieldsofbehaviourchange,digitaldevelopment,communicationfordevelopment

approaches,andcommercialtacticsforusingdigitaltoengagedifferentpopulationsineffectiveways.Thesemustbeframedinstrongapproachestodigitalsafe-guardingtoensurethatethics,dataprivacy,datasecurity,andoverallchildprotectionaspectsaremaintained.

SeveralcreativeapproachestousingdigitalandsocialmediaforSBCChaveemergedoverthepastfiveyears.Theyworkthroughanumberofchannelsandplatformstocreatetransmedia

approachesthatarecementedinyouthinterests,contexts,needsandmedia/digitalhabits.CloserreviewoftheseeffortsoffersmuchlearningthatcanbeappliedtofutureSBCCeffortsthatutilizesocialmediaanddigitaltoolsandplatforms.

Aninterestingchallengeishowtocreatelivelyandengagingtransmediaapproachesthatrelyon

lower-endtoolsandthatengageuserswhohavelittletonoInternetaccess.Thetrickwillbetotakeideasfromhigh-powereddigitalapproachesandadaptthemtolow-techenvironmentsinwaysthatarehighlyengagingforyoungpeople.Inaddition,thereisaneedtodesignMERL

approachesthattakeadvantageoftransmediatouchpointswhereparticipantopinionsandotherdatacanbecollectedandusedtoinformandshapethecourseofon-goingactivities.

ThispaperaimstostimulateideasonhowtoaddressthesechallengessothatdigitalSBCCinitiativescanbeimplementedandevaluatedmoreeffectivelyandwithgreaterimpact.


References

2CV.GirlEffectMobileFormativeResearch.London:GirlEffect,2014.

AbreuLopes,Claudia,andSavitaBailur.Genderequalityandbigdata.NewYork,NY:GlobalPulseandUNWomen,2018.

Ajzen,I.,andM.Fishbein.Understandingattitudesandpredictingsocialbehaviour.EnglewoodCliffs,NJ:Prentice-Hall,1980.

Albert,Bill.WithOneVoice2010:America’sAdultsandTeensSoundOffaboutTeenPregnancy.Washington,DC:TheNationalCampaigntoPreventTeenandUnplannedPregnancy,2010.

AmnestyInternational.Amnestyrevealsalarmingimpactofonlineabuseagainstwomen.November2017.https://www.amnesty.org/en/latest/news/2017/11/amnesty-reveals-alarming-impact-of-online-abuse-against-women/.

Armitage,C.J.,andM.Conner."Efficacyofthetheoryofplannedbehaviour:Ameta-analyticreview."BritishJournalofSocialPsychology40,no.4(2001):471-499.

Atkinson,S."WhyNorway’sMostHypedShow‘Skam’Isn’t‘Skins’ButYouShouldWatchitAnyway."Highsnobiety,December22,2016.

Aziz,Z."Duediligenceandaccountabilityforonlineviolenceagainstwomen."2017.

Bachan,Keshet.Lessonsandgoodpracticeforgirls'mobileandsocialnetworks.InternalReport,GirlEffect(Unpublished),2016.

Bailey,J.,etal."Sexualhealthpromotionforyoungpeopledeliveredviadigitalmedia:ascopingreview."PublicHealthResearch3,no.13(2015).

Banaji,Shakuntala.ChildrenandmediainIndia:Narrativesofclass,agencyandsocialchange.London:Routledge,2017.

Banaji,Shakuntala,SoniaLivingstone,AnulekhaNandi,andMariyaStoilova."Instrumentalisingthedigital:adolescents’engagementwithICTsinlow-andmiddle-incomecountries."DevelopmentinPractice28,no.3(April2018):432-443.

Bandura,A."Socialcognitivetheoryforpersonalandsocialchangebyenablingmedia."InEntertainment–EducationandSocialChange:History,Research,andPractice,editedbyA.Singhal,M.J.Cody,E.M.RogersandM.Sabido,75-96.Mahwah,NJ:Erlbaum,2004.

Bandura,A."Socialcognitivetheoryinculturalcontext."AppliedPsychology51,no.2(2002):269-290.

Baños,R.M.,E.Etchemendy,A.Mira,G.Riva,A.Gaggioli,andC.Botella."Onlinepositiveinterventionstopromotewell-beingandresilienceintheadolescentpopulation:Anarrativereview."FrontiersinPsychiatry8(2017):1-9.

Barnett,Inka,NigelScott,SimonBatchelor,andLawrenceHaddad.Dial‘N’fornutrition?Alandscapeanalysisofwhatweknowaboutm-nutrition,m-agricultureandm-development.WorkingPaper,InstituteofDevelopmentStudies,2016.

Baumgartner,S.E.,P.M.Valkenburg,andJ.Peter."Theinfluenceofdescriptiveandinjunctivepeernormsonadolescents'riskysexualonlinebehaviour."Cyberpsychology,BehaviourandSocialNetworking14,no.2(2011):753-758.

BehaviouralInsightsTeam.Reviewofevidenceononlinevsin-persongroup-basedinterventionsforbehaviouralchange.Unpublished,2018.

Behm-Morawitz,Elizabeth,JenniferStevensAubrey,HillaryPennell,andKyungBoKim."ExaminingtheeffectsofMTV’s16andPregnantonadolescentgirls’sexualhealth:Theimplicationsofcharacteraffinity,


pregnancyriskfactors,andhealthliteracyonmessageeffectiveness."HealthCommunication,November2017.

Caplan,Robyn.Contentorcontextmoderation:Artisanal,community-reliantandindustrialapproaches.NewYork:Data&Society,2018.

Castro,T."EastLosHighonHuluisfirstEnglishlanguageshowwithallLatinocast."HuffingtonPost,June6,2013.

CGNetSwara.AboutCGNetSwara.n/d.http://cgnetswara.org/about.html.

Chakraborty,Swati.BellBajaoend-linesurveyondomesticviolenceandHIV/AIDS.Breakthrough,2010.

Cugelman,B.,M.Thelwall,andP.Dawes."Onlineinterventionsforsocialmarketinghealthbehaviourchangecampaigns:Ameta-analysisofpsychologicalarchitecturesandadherencefactors."JournalofMedicalInternetResearch13,no.1(2011).

DeBrun,A,MMcCarthy,KMcKenzie,andAMcGloin."Weightstigmaandnarrativeresistanceevidentinonlinediscussionsofobesity."Appetite,no.72(2014):73-81.

deLanerolle,Indra,MarionWalton,andAletteSchoon.Izolo:mobilediariesofthelessconnected.ReesarchReport,Brighton:MakingAllVoicesCount:IDS,2017.

Denny,Elaine,andClaireHughes.Attitudes,practicesandsocialnormssurvey:EndlineReport.Voices4Change,2017.

EchoMobileandDigitalImpactAlliance.Messagingappsfordevelopment.WhitePaper,DigitialImpactAlliance,2018.

Elliot,Mark,ElaineMackey,KieronO'Hara,andCarolineTudor.Theanonymisationdecision-makingframework.UKAN-UniversityofManchester,2016.

EuropeanInstituteforGenderEquality."Cyberviolenceagainstwomenandgirls."2017.

EuropeanUnion."GeneralDataProtectionRegulation."2016.

Every1Mobile.LocalisingPurple?VoicesforChange,2016.

Fine,G.A.,andB.Harrington."Tinypublics:Smallgroupsandcivilsociety."SociologicalTheory22(2004):341-356.

Fladalen,J.I."-Nerveni“Skam”skalværesterkogrelevant."RushprintNo4,April2016.

Fraser,Erika,PatieneEkeoba,andJoFeather.YoungpeopleandchangeinNigeria:Opportunitiesandbarriers.Abuja,Nigeria:VoicesforChange,2017.

Furevold-Boland,M.,interviewbyV.SchankeSundet.(December13,2016).

GirlEffectand2CV.FormativeResearchforSpringster.London:Unpublished,2017.

Gold,Judy,andMeganLim."What'sinamessage?DeliveringsexualhealthpromotiontoyoungpeopleinAustraliaviatextmessaging."BioMedPublicHealth(BilMedCentralLtc.),2010.

GramVaani.Behaviourchangecommunicationthroughserialsaccessibleonmobilephones.June3,2015.http://gramvaani.org.

—.HowMobileVaaniworks.n/d.http://gramvaani.org.

Harwood,L.UAresearcherfindsnegativeimpactofMTVshowSixteenandPregnant.Mar25,2014.http://www.arizona.edu(accessedSep16,2018).

HealthCommunicationCapacityCollaborative(HC3).UsinghealthcommunicationframeworksinICTinterventionstoguideRMNCHdemand.JohnsHopkinsUniversityCenterforCommunicationPrograms(Unpublished),2013.


Higgs,Elizabeth,etal."UnderstandingtheroleofmHealthandothermediainterventionsforbehaviourchangetoenhancechildsurvivalanddevelopmentinlow-andmiddle-incomecountries:Anevidencereview."JournalofHealthCommunication(Routledge)19(2014):164-189.

Hutchinson,Paul,etal."EvaluationoftheMTVShugafamilyplanningdramainNigeria."TulaneUniversity,2018.

ICRC;TheEngineRoom;BlockParty;."Humanitarianfuturesformessagingapps."2017.

Ippoliti,N.,andK.L'Engle."Meetusonthephone:mobilephoneprogramsforadolescentsexualandreproductivehealthinlow-to-middleincomecountries."ReproductiveHealth,2017.

Ippoliti,Nicole.MobilePhoneProgramsForAdolescentAndYouthSexualAndReproductiveHealthInLow-AndMiddle-IncomeCountries.ResearchBrief,USAIDandKnowledgeforHealth,Washington,DC:USAIDandKnowledgeforHealth,2017.

Jacobs,Lani,andJoUpton.SpringsterUFOs,SouthAfricadebrief.2CVandGirlEffect,Unpublished,2018.

Jenkins,H.ConvergenceCulture.WhereOldandNewMediaCollide.NewYork:Routledge,2006.

Kast,F.,S.Meier,andD.Pomeranz.Under-saversanonymous:Evidenceonself-helpgroupsandpeerpressureassavingscommitmentdevice.Cambridge:NationalBureauofEconomicResearch,2012.

Kayyali,Dia.HarmreductionforWhatsApp.November13,2018.https://blog.witness.org/2018/11/harm-reduction-whatsapp/(accessedNovember17,2018).

—.What'sup,WhatsApp?November13,2018.https://blog.witness.org/2018/11/whats-up-whatsapp(accessedNovember18,2018).

Kearney,Melissa,andPhillipLevine.Mediainfluencesonsocialoutcomes:TheimpactofMTV's16andPregnantonteenchildbearing.Workingpaper,NationalBureauofEconomicResearch,Cambridge:NationalBureauofEconomicResearch,2015.

Kelly,M.Indiawantssocialmediaplatformstoremovecontentitdeems'unlawful'.January20,2019.https://www.theverge.com/2019/1/20/18188587/india-facebook-twitter-tumblr-remove-content-unlawful-speech-freedom-internet-foundation.

Korda,H,andZItani."Harnessingsocialmediaforhealthpromotionandbehaviourchange.."HealthPromotionPractice14,no.1(2013):15-23.

Kraidy,Marwan.TheuseofIVRtosupportmonitoringandevaluationofmediainterventions:AcasestudyoftheVOTOsysteminRwanda.CenterforAdvancedResearchinGlobalCommunicationandAnnenbergSchoolforCommunication,UniversityofPennsylvania,2016.

Krebs,P.,J.O.Prochaska,andJ.S.Rossi."Ameta-analysisofcomputer-tailoredinterventionsforhealthbehaviourchange."JournalofPreventiveMedicine51,no.3(2010):214-221.

Larkey,L.K.,andM.Hecht."Amodelofeffectsofnarrativeasculture-centrichealthpromotion."JHealthCommunications15,no.2(2010):114-135.

Lee,H.,J.Fawcett,andR.DeMarco."Storytelling/narrativetheorytoaddresshealthcommunicationwithminoritypopulations."AppliedNarrativeResearch30,no.154(2015):58-60.

Leo,B.,andR.Morello.Domobilephonesurveysworkinpoorcountries?CenterforGlobalDevelopment,2015.

Lim,M.,J.Hocking,C.K.Aitken,L.Jordan,andC.Fairely."Arandomisedcontrolledtrialoftextandemailmessagingforsexualhealthpromotiontoyoungpeople."JournalofEpidemiologyandCommunityHealth66,no.1(2011):69-74.


Limaye,RupaliJ.,andAnneBallardSara.Assessingtheeffectsofastorytellinginitiative:Resultsofthefamilyplanningvoicesevaluationphase1.EvaluationReport,Baltimore,MD:JohnsHopkinsCenterforCommunicationPrograms,2017.

LOOMNepal."Nepal1.Theinternetandsexualexpression:ExploringtheuseoftheinternetamonggenderequalityandsexualrightsadvoctesinNepal."InEROTICSSouthAsiaexploratoryresearch:Sex,rightsandtheinternet,byAssociationofProgressiveCommunications.AssociationofProgressiveCommunications,2017.

Lopez,M.H.,A.Gonzalez-Barrera,andE.Patten.Closingthedigitaldivide:Latinosandtechnologyadoption.PewHispanic,2013.

Lustria,M.,S.Noar,J.Cortese,S.VanStee,R.Glueckauf,andJ.Lee."Ameta-analysisofweb-delivered,tailoredhealthbehviorchangeinterventions."JournalofHealthCommunications18,no.9(2013):1039-1069.

Magnus,M.,interviewbyV.SchankeSundet.Inpersoninterview,onlineproducerofSKAMOslo12,(January12,2017).

Magnus,M."SKAM–nårfiksjonogvirkelighetmøtes."NordicomInformation38,no.2(2016):31-38.

Manzo,C."5lessonslibrarywebsitescanlearnfromBuzzfeed."Weave:JournalofLibraryUserExperience1,no.3(2015).

Mecheal,Patricia,etal."Barriersandgapsaffectingmhealthinlowandmiddleincomecountries."PolicyWhitePaper,mHealthAllianceandTheEarthInstituteatColumbiaUniversity,Washington,DC,2010.

MERLTechParticipants."ResponsibleDataHackpad."ResponsibleDataHackpad.2019.

Michie,S.,L.Yardley,R.West,K.Patrick,andF.Greaves."Developingandevaluatingdigitalinterventionstopromotebehaviourchangeinhealthandhealthcare:Recommendationsresultingfromaninternationalworkshop."JournalofMedicalInternetResearch19,no.6(2017).

Milward,Kirsty,andKelsyNelson.Howchangehappens:Alighttouchstudyonthetheoryofchange.Voices4Change,2017.

Moorhead,SAnne,DianeE.Hazlett,LauraHarrison,JenniferK.Carroll,AntheaIrwin,andCiskaHoving."Anewdimensionofhealthcare:Systematicreviewoftheuses,benefits,andlimitationsofsocialmediaforhealthcommunication."JournalofMedicalInternetResearch15,no.4(2013).

Moyer-Guse,E."Towardatheoryofentertainmentpersuasion:explainingthepersuasiveeffectsofentertainment–educationmessages."CommonTheory18,no.3(2008):407-425.

Murphy,S.T.,L.B.Frank,M.B.Moran,andP.Patnoe-Woodley."Involved,transported,oremotional?Exploringthedeterminantsofchangeinknowledge,attitudes,andbehaviourinentertainment–education."JCommun61,no.3(2011):407-431.

NikeFoundation.Scale,engagement+inspiration:Howbrandandintegratestrategiescandrivemassivesocialchange.Unpublished,2013.

Nyborg,I.M."MIA–etstortogstilledramapånett."NRKBeta,May3,2012.

Ohkubo,Saori,RupaliLimaye,andSarahHarlan.K4Healththeoryprimer:Diffusionofinnovation.JohnsHopkinsUniversity,2015.

Painter,J.,M.Hynes,andK.Glanz."Theuseoftheoryinhealthbehaviourresearchfrom2000-2005:Asystematicreview."AnnalsofBehaviouralMedicine35,no.3(2008):358-362.

Perrigo,B.HowvolunteersforIndia'srulingpartyareusingWhatsApptofuelfakenewsaheadofelections.January25,2019.http://time.com/5512032/whatsapp-india-election-2019/.

PewResearchCentre.Onlineharassment.PewResearchCenter,2018.


Pickard,M."NorwayfeelsShame."DramaQuarterly(Drama),August2016.

Porter,G.,etal."Youth,mobilityandmobilephonesinAfrica:Findingsfromathree-countrystudy."InformationTechnologyforDevelopment18,no.2(2012):145-162.

Porter,R.,L.Raftree,A.Blanchfield,andJ.Urbanowicz."BeyondGDPR:Thewhatswhysandhowsofdataprotection."InteractionForum.2018.

Poushter,Jacob,CaldwellBishop,andHanyuChwe.Socialmediausecontinuestoriseindevelopingcountriesbutplateausacrossdevelopedones.PewResearchCenter,2018.

Powell,HenryN."Technology-facilitatedsexualviolence:Aliteraturerevuewofempiracalresearch."Trauma,Violence&Abuse19,no.2(2018):195-208.

Prochaska,J.,J.Wright,andW.Velicer."Evaluatingtheoriesofhealthbehaviourchange:AhierarchyofcriteriaappliedtotheTranstheoreticalModel.."AppliedPsychology:Aninternationalreview57,no.4(2008):561-588.

Raftree,Linda.Digitalsafeguardingtipsandguidance.Guidelines,GirlEffect,GirlEffect,2018.

Raftree,Linda.DigitalSafeguarding:Modernizingourpolicyandpractice.London:GirlEffect,2018a.

Raftree,Linda.Girlsafeguardingonmobileandsocialnetworks.InternalReport,GirlEffect(Unpublished),2016.

Razzetti,G.Latinosandmediausagerevisited.www.clickz.com,2012.

ResponsibleDataForum.ResponsibleDataForum.2014.https://responsibledata.io/.

Riley,W.T.,D.Rivera,A.Atienza,W.Nilsen,S.M.Allison,andR.Mermelstein."Healthbehaviourmodelsintheageofmobileinterventions:Areourtheoriesuptothetask?"TranslationalBehaviouralMedicine1(2011):53-71.

Riley,W.T.,D.Rivera,A.Atienza,W.Nilsen,S.M.Allison,andR.Mermelstein."Healthbehaviourmodelsintheageofmobileinterventions:Areourtheoriesuptothetask?"TranslationalBehaviouralMedicine1:53-71.

Rogers,EverettM.DiffusionofInnovations.5thEdition.Simon&Schuster,2003.

Rosenstock,I."HistoricalOriginaloftheHealthBeliefModel."HealthEducationMonographs2,no.4(1974).

Sallis,J.F.,N.Owen,andE.Fisher."Ecologicalmodelsofhealthbehaviour."InHealthbehaviour:Theory,resarchandpractice,byK.Glanz,B.K.RimgerandK.Viswanath.SanFrancisco,CA:Jossey-Bass,2015.

SchankeSundet,Vilde."From'secret'onlineteendramatointernationalcultphenomenon:TheglobalexpansionofSKAManditspublicservicemission."(forthcoming),2017.

Shand,Tim.GenderdimensionsofprogrammingformenandboysasFPusers.InstituteforReproductiveHealth,GeorgetownUniversity,2017.

Shane-Simpson,Christina,JenniferDixon,TawnyaFaySwitzer,JoannaLaursenBrucker,andJohnFraser.GirlEffectliteraturereview:Digitaltoolsforbehaviourchange.NewKnowledge,NewYork:NewKnowledge,2017.

Shen,F.,V.C.Sheer,andR.Li."Impactofnarrativesonpersuasioninhealthcommunication:ameta-analysis."JournalofAdvertising44,no.2(2015):105-113.

Sheridan,Jonah.Howtoassessavendor'sdatasecurity?ElectronicFrontierFoundation,2018.

Singhal,A.,H.Wang,andE.M.Rogers."Theentertainment-educationcommunicationstrategyincommunicationcampaigns."InPublicCommunicationCampaigns,byC.edsRiceR.E.andAtkins,321-334.BeverleyHills,CA:Sage,2013.


Slater,M.D.,andD.Rounder."Entertainment–educationandelaborationlikelihood:understandingtheprocessingofnarrativepersuasion."CommonTheory12,no.2(2002):173-191.

StJohn,Max.Socialmediaandbehaviourchange:planninganddoing.NixonMcInnes,2011.

Sullivan,TaraM.,MollyStrachan,andBarbaraK.Timmons.Guidetomonitoringandevaluatinghealthinformationproductsandservices.USAID,2007.

Swarup,S.Thecircleoflife:A360-degreeapproachtocommunication.August2012.

TexttoChange.ExpandingHIVpreventionanduptakeofHIVVoluntaryCounselingandTesting(VCT)throughinteractiveSMSmessaginginMbarara,Uganda.2008.

Thaler,R.H.,andC.R.Sunstein.Nudge:Improvingdecisionsabouthelth,wealthandhappiness.NewHaven,CT:YaleUniversityPress,2008.

Thaler,R.H.,C.Sunstein,andJ.Balz."Choicearchitecture."InThebehaviouralfoundationsofpublicpolicy,byE.(ed)Shafir,428-439.Princeton,NJ:PrincetonUniversityPress,2014.

TheBehaviouralInsightsTeam.Efficacyofvirtualreferencegroupsforbehaviouralchange.UnpublishedworkforGirlEffect,2018.

TNS."VoicesforChangeResearchSummary."2016.

Toth,C,andPSIDRCongo.CellphonehotlinespreadsfamilyplanninginformationinDRCongo.USAIDfromtheAmerianPeople,2008.

TrustLaw,mHealthAlliance,Baker&McKenzie,MSD.Patientprivacyinamobileworld:Aframeworktoaddressprivacylawissuesinmobilehealth.TrustLaw,mHealthAlliance,Baker&McKenzie,MSD,2013.

UKInformationCommissioner'sOffice."ChildrenandtheGDPRguidance."2018.

UNGlobalPulse.Analyzingattitudestowardscontraception&teenagepregnancyusingsocialdata.UNGlobalPulse,2014.

UNHumanRightsCouncil.ReportoftheSpecialRapporteuronthepromotionandprotectionoftherighttofreedomofopinionandexpression,DavidKaye.UNHumanRIghtsCouncil,2015.

VanderWilk,A.Cyberviolenceandhatespeechonlineagainstwomen.EuropeanParliament'sPolicyDepartmentforCitizen'sRightsandConstitutionalAffairs,2018.

Vlahakis,Maria.Breakingthesilence:Endingonlineviolenceandabuseagainstwomen'srightsactivists.WomankindWorldwide,2018.

VodafoneFoundationandGirlEffect.Realgirls,reallives,connected:Aglobalstudyofgirls'accessandusageofmobile,toldthrough3000voices.London:VodafoneFoundationandGirlEffect,2018.

VoicesforChange.Strengtheningtheenablingenvironmentforwomen'sempowermentinNigeria:AchievementsandlearningfromVoicesforChange.VoicesforChange,VoicesforChange,2017.

Walton,M.,andN.Pallitt."GrandtheftSouthAfrica:Games,literacyandinequalityinconsumerchildhoods."LanguageandEducation26,no.4(2012):347-361.

Wang,Hua,andArvindSinghal."EastLosHigh:TransmediaedutainmenttopromotethesexualandreproductivehealthofyoungLatina/oAmericans."AmericanJournalofPublicHealth106,no.6(2016):1002-1010.

WellToldStory.Canweseenewnormsform?Newtechmeanswecannotonlywatchtheprocessbutalsounderstandandinfluenceit.WellToldStory,2017.

Whitton,Alexix,etal."Breakingopentheblackbox:Isolatingthemostpotentfeaturesofawebandmobilephone-basedinterventionfordepression,anxiety,andstress."JMIRMentalHealth2,no.1(Jan-Mar2015).


Wills,Adam,GillesMorain,andRodolpheBaudeau.M4DImpactServiceEvaluation:Airtel&HNI's3-2-1Madagascarservice.GSMA,2016.

WorldBankGroup.Worlddevelopmentreport:Mind,socityandbehaviour.Washington,DC:TheWorldBank,2015.

WorldHealthOrganizationandInternationalInitiativeforImpactEvaluation.Anevidencemapofsocial,behaviouralandcommunityengagementinterventionsforreproductive,maternal,newbornandchildhealth.Geneva:WorldHealthOrganization,2017.

Zebregs,S.,B.vandenPutte,P.Neijens,andA.deGraaf."Thedifferentialimpactofstatisticalandnarrativeevidenceonbeliefs,attitude,andintention:ameta-analysis.."HealthCommunications30,no.3(2015):282-289.

Digital and Social Media for SBCC March 2019 · Digital and Social Media for Social and Behaviour Change Communication (SBCC) 4 SKAM, a Norwegian program dealing with teen girls’

Documents