Teens and Technology 2013

MARCH13,2013
Teens and Technology 2013

Smartphone adoption among teens has increased substantially and mobile access to the
internet is pervasive. One in four teens are cell-mostly internet users, who say they mostly
go online using their phone.

MaryMadden
SeniorResearcher,PewInternetProject
AmandaLenhart
SeniorResearcher,DirectorofTeensandTechnology
Initiatives,PewInternetProject
MaeveDuggan
ResearchAssistant,PewInternetProject
SandraCortesi
Fellow,DirectoroftheYouthandMediaProject,
BerkmanCenterforInternet&Society
UrsGasser
ExecutiveDirector,BerkmanCenterforInternet&
Society
http://www.pewinternet.org/Reports/2013/TeensandTech.aspx
FOR FURTHER INFORMATION, CONTACT:
Pew Research Centers Internet & American Life Project

1615 L St., N.W., Suite 700
Washington, D.C. 20036

Media Inquiries:
202.419.4500
REPORTEMBARGOEDUNTIL12:01AMEASTERNONMARCH13,2013
pewi nt er net . or g 2
SummaryofFindings
SmartphoneadoptionamongAmericanteenshasincreasedsubstantiallyandmobileaccesstotheinternetis
pervasive.Oneinfourteensarecellmostlyinternetusers,whosaytheymostlygoonlineusingtheirphone
andnotusingsomeotherdevicesuchasadesktoporlaptopcomputer.Theseareamongthenewfindings
fromanationallyrepresentativesurveyof802teensages1217andtheirparentswhichshowsthat:
- 78%ofteensnowhaveacellphone,andalmosthalf(47%)ofthoseownsmartphones.Thattranslates
into37%ofallteenswhohavesmartphones,upfromjust23%in2011.
- Oneinfourteens(23%)haveatabletcomputer,alevelcomparabletothegeneraladultpopulation.
- Nineinten(93%)teenshaveacomputerorhaveaccesstooneathome.Seveninten(71%)teenswith
homecomputeraccesssaythelaptopordesktoptheyusemostoftenisonetheysharewithother
familymembers.
MobileaccesstotheinternetiscommonamongAmericanteens,andthecellphonehasbecomeanespecially
importantaccesspointforcertaingroups:
- Aboutthreeinfour(74%)teensages1217saytheyaccesstheinternetoncellphones,tablets,and
othermobiledevicesatleastoccasionally.
- Oneinfourteensarecellmostlyinternetusersfarmorethanthe15%ofadultswhoarecell
mostly.Amongteensmartphoneowners,halfarecellmostly.
- Oldergirlsareespeciallylikelytobecellmostlyinternetusers;34%ofteengirlsages1417saythey
mostlygoonlineusingtheircellphone,comparedwith24%ofteenboysages1417.Thisisnotable
sinceboysandgirlsareequallylikelytobesmartphoneowners.
- Amongolderteengirlswhoaresmartphoneowners,55%saytheyusetheinternetmostlyfromtheir
phone.
Inoverallinternetuse,youthages1217whoarelivinginlowerincomeandlowereducationhouseholdsare
stillsomewhatlesslikelytousetheinternetinanycapacitymobileorwired.However,thosewhofallinto
lowersocioeconomicgroupsarejustaslikelyandinsomecasesmorelikelythanthoselivinginhigherincome
andmorehighlyeducatedhouseholdstousetheircellphoneasaprimarypointofaccess.
About the survey
Thesefindingsarebasedonanationallyrepresentativephonesurveyof802parentsandtheir802teensages
1217.ItwasconductedbetweenJuly26andSeptember30,2012.InterviewswereconductedinEnglishand
Spanishandonlandlineandcellphones.Themarginoferrorforthefullsampleis4.5percentagepoints.This
reportisthesecondinaseriesofreportsissuedincollaborationwiththeBerkmanCenterforInternet&
SocietyatHarvard.Thefirstrelease,Teens,ParentsandOnlinePrivacy,waspublishedinNovember2012
andisavailableat:http://www.pewinternet.org/Reports/2012/TeensandPrivacy.aspx
MainFindings
Fully95%ofteensareonline,apercentagethathasbeenconsistentsince2006.Yet,thenatureofteens
internetusehastransformeddramaticallyduringthattimefromstationaryconnectionstiedtodesktopsin
thehometoalwaysonconnectionsthatmovewiththemthroughouttheday.Inmanyways,teensrepresent
theleadingedgeofmobileconnectivity,andthepatternsoftheirtechnologyuseoftensignalfuturechangesin
theadultpopulation.Teensarejustaslikelytohaveacellphoneastheyaretohaveadesktoporlaptop
computer.Andincreasinglythesephonesareaffordingteensalwayson,mobileaccesstotheinternetin
somecases,servingastheirprimarypointofaccess.Smartphoneownershipamongteenshasgrown
substantiallysince2011;37%ofAmericanyouthages1217nowhaveasmartphone,upfrom23%in2011.
Tabletsarealsotakinghold,asclosetooneinfourteenssaytheyhaveoneofthesedevices.Takentogether,
teenshavemorewaysthanevertostayconnectedthroughoutthedayandnight.

Internet use over time by teens and adults
%withineachagegroupwhogoonline
Source:ThePewResearchCentersInternet&AmericanLifeProjectsurveys.Allteendatacomesfromseparatesurveys
ofteensandtheirparents.Methodologicalinfoforeachsurveyisavailableat:http://pewinternet.org/Data
Tools/DownloadData
95%
95%
89%
77%
52%
0%
20%
40%
60%
80%
100%
Nov04 Nov06 Nov07 Feb08 Sept09 July11 Sept12
1217 1829 3049 5064 65+
Three in four teens access the internet on cell phones, tablets, and other mobile
devices.
Aboutthreeinfour(74%)teensages1217aremobileinternetuserswhosaytheyaccesstheinterneton
cellphones,tablets,andothermobiledevicesatleastoccasionally.
1
Bycomparison,55%ofadultsaremobile
internetusers.
2
However,thisgapisdrivenprimarilybyadultsages65andolder,manyofwhomarenotusing
theinternetinanycapacity,letaloneonamobiledevice.Adultsundertheageof50,ontheotherhand,are
justaslikelyasteenstobemobileinternetusers;74%ofadultsages1849accesstheinternetonacellphone,
tablet,orothermobiledevice.
Teen Internet Access Demographics
%ofteensineachdemographicgroup
Accessthe
Internet
MobileAccessto
Internet(phone,
tablet,etc)
AccessInternet
MostlyonCell
Phone
Allteens,ages1217(n=802) 95% 74% 25%
TeenGender
a Boys(n=405) 97
b
72 20
b Girls(n=397) 93 76 29
a
AgeofTeen
a 1213(n=246) 93 71 16
b 1417(n=556) 96 76 29
a
TeenGenderandAge
a Boys,1213(n=122) 96 70 12
b Boys,1417(n=283) 97
c
73 24
a
c Girls,1213(n=124) 91 72 20
d Girls,1417(n=273) 95 78 34
abc
ParentRace/ethnicity
a White,NonHispanic(n=542) 98
bc
77
c
24
b Black,NonHispanic(n=122) 92 74 33
c Hispanic(n=92) 88 63 21
ParentEducation
a LessThanHighSchool/Highschoolgrad(n=244) 91 71 27
b SomeCollege(n=192) 99
a
71 24
c College+(n=363) 99
a
81
ab
22
ParentHouseholdIncome
a Lessthan$30,000/yr(n=154) 89 66 30
c
b $30,000$49,999(n=155) 94 72 25
c
c $50,000$74,999(n=110) 99
a
79 14
1
Thesurveyalsoincludedseparatequestionsaboutteenmobileappusagethatwillbereleasedinaforthcomingreport.
2
AccordingtoourSeptember2012TrackingSurvey:http://www.pewinternet.org/SharedContent/Data
Sets/2012/September2012HealthTracking%28prelim%29.aspx
d $75,000+(n=335) 99
ab
79
a
24
c
Urbanity
a Urban(n=278) 94 74 27
b Suburban(n=410) 96 72 24
c Rural(n=101) 99
a
79 21
Source:PewInternetTeensandPrivacyManagementSurvey,July26September30,2012.N=802parentsofteens
ages1217and802teensages1217.Marginoferroris+/4.5percentagepoints.
Note:Columnsmarkedwithasuperscriptletter(
a
)oranotherletterindicateastatisticallysignificantdifference
betweenthatrowandtherowdesignatedbythatsuperscriptletter.Statisticalsignificanceisdeterminedinsidethe
specificsectioncoveringeachdemographictrait.
93% of teens have a computer or have access to one.
Eightoutoftenteenshaveadesktoporlaptopcomputer.Amongthe20%ofteenswhodonothavetheirown
computer,twothirds(67%)haveaccesstoonetheycanuseathome.Takentogether,thismeansthat93%of
teenshaveacomputeroraccesstoone.
Whiteteensaremorelikelythanblackteenstohaveadesktoporlaptopcomputer.Some81%ofwhiteteens,
comparedwith64%ofblackteens,ownacomputer.Olderteensages1417aremorelikelytohavea
computerthanyoungerteensages1213(83%vs.72%).Teenslivinginsuburbanareasaremorelikelytohave
acomputerwhencomparedwithurbanteens(84%vs.75%).
71% of teen computer users say the computer they use most often is shared with
family members.
Fewteenshaveexclusiveaccesstothecomputertheyusemostoften.Some71%ofteenswhohavea
computeroraccesstooneathomesaythatthecomputertheyusemostoftenisonetheysharewiththeir
parents,siblings,orothermembersoftheirfamily.Thevastmajorityofyoungerteenssharethecomputer
theyusemostoften;81%ofteencomputerusersages1213relyonsharedcomputerscomparedwith66%of
olderteencomputerusersages1417.Byage17,onlyhalf(51%)ofteencomputeruserssaythecomputer
theyusemostoftenisonetheysharewithfamily.Teencomputeruserslivinginruralareasaremorelikely
thansuburbanteenstosaythecomputertheyusemostoftenisasharedcomputer(80%vs.67%).
23% of teens have a tablet computer.
Oneinfourteens(23%)saytheyhaveatabletcomputer,alevelcomparabletothegeneraladultpopulation
(25%ofAmericanadultshaveatabletcomputer).Muchlikeadults,familyeducationandincomelevelsare
strongindicatorsofteentabletownership.Almostathirdofteenswhoseparentshaveatleastsomecollege
educationorhavehouseholdincomesover$75,000ayearowntablets.Olderandyoungerteensareequally
likelytohavetablets.
Teen Computer and Tablet Ownership Demographics

OwnaComputer OwnaTablet
Allteens,ages1217(n=802) 80% 23%
TeenGender
a Boys(n=405) 77 20
b Girls(n=397) 82 27
a
AgeofTeen
a 1213(n=246) 72 26
b 1417(n=556) 83
a
22
TeenGenderandAge
a Boys,1213(n=122) 66 23
b Boys,1417(n=283) 82
a
18
c Girls,1213(n=124) 79 28
d Girls,1417(n=273) 84
a
27
b
25
b Black,NonHispanic(n=122) 64 19
c Hispanic(n=92) 79 21
ParentEducation
a LessthanHighSchool/Highschoolgrad(n=244) 77 16
b SomeCollege(n=192) 78 30
a
c College+(n=363) 85 29
a
a Lessthan$30,000/yr(n=154) 73 15
b $30,000$49,999(n=155) 82 19
c $50,000$74,999(n=110) 84 27
d $75,000+(n=335) 81 31
ab
Urbanity
a Urban(n=278) 75 25
b Suburban(n=410) 84
a
23
c Rural(n=101) 72 23
a
37% of all teens have smartphones, up from 23% in 2011.
Cellphoneownershipamongteenshasbeenstablesince2011,butsmartphoneownershiphasgrown
considerably.Some78%ofteensnowhaveacellphone,andalmosthalf(47%)ofthosesaytheyhave
smartphones.Thattranslatesinto37%ofallteenswhohavesmartphones,upfromjust23%in2011.By
comparison,45%ofalladultshaveasmartphone.
3
Teenswithhighlyeducatedparentsorwithparentsinthehighestincomebracketarealsomorelikelytohave
cellphones.However,parentincomelevelsdonotmapasneatlywithsmartphoneownershipamongteens.
Forinstance,teenslivinginthelowestearninghouseholds(under$30,000peryear)arejustaslikelyasthose
livinginthehighestearninghouseholds($75,000ormore)toownsmartphones(39%vs.43%).Olderteensare
morelikelythanyoungerteenstohavecellphonesandthosephonesaremorelikelytobesmartphones.Rural
teensaresignificantlylesslikelytohaveasmartphonethanurbanorsuburbanteens.
Source:PewInternetTeensandPrivacyManagementSurvey,July26September30,
2012.N=802parentsofteensages1217and802teensages1217.Marginoferroris
+/4.5percentagepoints.
One in four teens are cell-mostly internet users far more than the 15% of
adults who are cell-mostly. Among teen smartphone owners, half are cell-mostly.
Whilemanyteenshaveavarietyofinternetconnecteddevicesintheirlives,thecellphonehasbecomethe
primarymeansbywhich25%ofthoseages1217accesstheinternet.Amongteenswhoaremobileinternet
users,thatnumberrisestooneinthree(33%).Amongteensmartphoneowners,50%saytheyusetheinternet
mostlyviatheircellphone.
Whilecellmostlyinternetusetrackswithsomeofthesamedemographictrendsassmartphoneownership
(suchasage),genderstandsoutasanespeciallyimportantindicator.Althoughteengirlsandboysareequally
likelytohavesmartphonesandareequallylikelytousesomekindofmobileaccesstotheinternet,girlsare
significantlymorelikelythanboystosaytheyaccesstheinternetmostlyusingtheircellphone(29%vs.20%).
Olderteengirlsrepresenttheleadingedgeofcellmostlyinternetuse;34%ofthemsaythatmostoftheir
3
Forourlatestadultfiguresonsmartphoneusageandtrendsinsmartphoneadoptionovertime,see:
http://www.pewinternet.org/StaticPages/TrendData%28Adults%29/DeviceOwnership.aspx
77
78
23
37
0
10
20
30
40
50
60
70
80
90
Jul11 Sep12
Teen Cell Phone and Smartphone Ownership
%ofallteensages1217
Cellphone
Smartphone
internetusehappensontheircellphone.Amongolderteengirlswhoaresmartphoneowners,55%saythey
usetheinternetmostlyfromtheirphone.
Inoverallinternetuse,youthages1217whoarelivinginlowerincomeandlowereducationhouseholdsare
stillsomewhatlesslikelytousetheinternetinanycapacitymobileorwired.However,thosewhofallinto
lowersocioeconomicgroupsarejustaslikelyandinsomecasesmorelikelythanthoselivinginhigherincome
andmorehighlyeducatedhouseholdstousetheircellphoneasaprimarypointofaccess.
A2010PewInternetProjectreportonteensmobilephoneusefoundthatamongteencellphoneowners,
21%ofthosewhodidnotgoonlineoruseemailthroughaconventionalcomputerinsteadusedtheirphone
handsettogoonline.Atthetime,teencellphoneownersinthelowestincomehouseholdsweremorelikely
thanthoselivinginhigherincomehouseholdstousetheirphonetogoonline.
4
Thesefindingsaresimilartotrendswehaveobservedwithadults:Thoseadultswithanannualhousehold
incomeoflessthan$50,000peryearandthosewhohavenotgraduatedcollegearemorelikelythanthose
withhigherlevelsofincomeandeducationtousetheirphonesformostoftheironlinebrowsing.Whenasked
forthemainreasonwhytheymostlygoonlinefromamobilephone,adultcellmostlyusersprimarilycited
convenience,butoneintenpointedtowardsalackofotheraccessoptionsasthemainreasonwhythey
primarilyusetheirphonetogoonline.
5
(Tablebelow)
4
See:http://www.pewinternet.org/Reports/2010/TeensandMobilePhones/Chapter2/Part4.aspx
5
Formoredetailonadultsandcellmostlyinternetuse,seeCellInternetUse2012.Availableat:
http://www.pewinternet.org/Reports/2012/CellInternetUse2012.aspx
Teen Cell Phone and Smartphone Ownership Demographics

OwnaCellPhone(anykind) OwnaSmartphone
Allteens,ages1217(n=802) 78% 37%
TeenGender
a Boys(n=405) 77 36
b Girls(n=397) 78 38
AgeofTeen
a 1213(n=246) 68 23
b 1417(n=556) 83
a
44
a
TeenGenderandAge
a Boys,1213(n=122) 65 20
b Boys,1417(n=283) 83
ac
43
ac
c Girls,1213(n=124) 71 26
d Girls,1417(n=273) 82
a
44
ac
c
35
b Black,NonHispanic(n=122) 72 40
c Hispanic(n=92) 64 43
ParentEducation
a LessThanHighSchool/Highschoolgrad(n=244) 71 35
b SomeCollege(n=192) 79 35
c College+(n=363) 87
ab
41
a Lessthan$30,000/yr(n=154) 69 39
b
b $30,000$49,999(n=155) 74 24
c $50,000$74,999(n=110) 81 38
d $75,000+(n=335) 86
ab
43
b
Urbanity
a Urban(n=278) 76 42
c
b Suburban(n=410) 81 39
c
c Rural(n=101) 73 19
a
Questions
Teens and Pri vacy Management Survey 2012
Final Topline 10/9/2012
DataforJuly26September30,2012
Princeton Survey Research Associates International for
the Pew Research Centers Internet & American Life Project

Sample: n= 802 parents of 12-17 year olds and 802 teens ages 12-17
Interviewing dates: 07.26.2012 09.30.2012

Margin of error is plus or minus 4.5 percentage points for results based on total parents [n=802]
Margin of error is plus or minus 4.5 percentage points for results based on total teens [n=802]
Margin of error is plus or minus 4.6 percentage points for results based on total teens [n=781]
Margin of error is plus or minus 4.6 percentage points for results based on teen internet users [n=778]
Margin of error is plus or minus 5.1 percentage points for results based on teen SNS or Twitter users [n=632]
Margin of error is plus or minus 5.3 percentage points for results based on teens with a Facebook account [n=588]
Margin of error is plus or minus 9.4 percentage points for results based on teens with a Twitter account [n=180]

TEEN INTERVIEW
KI NTUSE Do you use the internet or email, at least occasionally?
KI NTMOB Do you access the internet on a cell phone, tablet or other mobile device, at least
occasionally?
6

USESINTERNET
DOESNOTUSE
INTERNET
CurrentTeens 95 5
July2011 95 5
Sept2009
i
93 7
Feb2008
ii
93 7
Nov2007 94 6
Nov2006 93 7
Nov2004 87 13
6
Trendquestionpriorto2006wasDoyouevergoonlinetoaccesstheInternetorWorldWideWebortosendand
receiveemail?TrendquestionfromNov2006thruSept2009wasDoyouusetheinternet,atleastoccasionally?/Do
yousendorreceiveemail,atleastoccasionally?TrendquestioninJuly2011wasDoyouusetheinternet,atleast
occasionally,forexampleoneitheracomputeroracellphone?/Doyousendorreceiveemail,atleastoccasionally?
K2 As I read the following list of items, please tell me if you happen to have each one, or not. Do
you have...[INSERT ITEMS IN ORDER]?
YES NO DON T KNOW REFUSED

a. Acellphone...oranAndroid,iPhoneorother
devicethatisalsoacellphone
7

CurrentTeens 78 22 0 0
July2011 77 23 0 0
September2009 75 25 0 0
February2008 71 29 0
November2007 71 29 0
b. Adesktoporlaptopcomputer
8
July2011 74 26 0 0
September2009 69 31 0 0
February2008 60 40 0
c. AtabletcomputerlikeaniPad,Samsung
Galaxy,MotorolaXoom,orKindleFire

K2a_1 Some cell phones are called smartphones because of certain features they have. Is your cell
phone a smartphone, such as an iPhone or Android, or are you not sure?
9

Basedonteencellphoneowners
CURRENT
TEENS
JULY
2011
%
47 Yes,smartphone 30
49 No,notasmartphone 56
4 Notsure/Dontknow 14
7
Priorto2009,trendwordingwasAcellphone.ItemwordinginSeptember2009andJuly2011was:Acellphone...or
aBlackberry,iPhoneorotherdevicethatisalsoacellphone
8
InNovember2004andNovember2006,desktopcomputerandlaptopcomputerwereaskedasseparateitems.
Resultsshownherehavebeenrecalculatedtocombinethetwoitems.
9
July2011wordingwas:Isthatasmartphoneornot...orareyounotsure?
0 Refused 0
[n=637]
[n=642]

K2b_1 Is there a computer that you can use at home?
Basedonteenswhodonothaveacomputer
CURRENT
TEENS
JULY
2011
% 67 Yes 63
33 No 37
0 Dontknow 0
0 Refused 0
[n=161] [n=175]
K3 Thinking about the computer you use most often, do your parents, siblings or other members
of your family share that computer with you?
Basedonteenswhohaveacomputerorhaveaccesstoone[N=764]

CURRENT
TEENS
%
71 Yes
29 No
0 Dontknow
0 Refused

Methodology
2012 Teens and Pri vacy Management Survey
PreparedbyPrincetonSurveyResearchAssociatesInternational
forthePewResearchCentersInternetandAmericanLifeProject
October2012
SUMMARY

The2012TeensandPrivacyManagementSurveysponsoredbythePewResearchCentersInternetand
AmericanLifeProjectobtainedtelephoneinterviewswithanationallyrepresentativesampleof802teensaged
12to17yearsoldandtheirparentslivingintheUnitedStates.ThesurveywasconductedbyPrincetonSurvey
ResearchAssociatesInternational.TheinterviewsweredoneinEnglishandSpanishbyPrincetonDataSource,
LLCfromJuly26toSeptember30,2012.Statisticalresultsareweightedtocorrectknowndemographic
discrepancies.Themarginofsamplingerrorforthecompletesetofweighteddatais4.5percentagepoints.
Detailsonthedesign,executionandanalysisofthesurveyarediscussedbelow.
DESIGNANDDATACOLLECTIONPROCEDURES
SampleDesign
Acombinationoflandlineandcellularrandomdigitdial(RDD)sampleswasusedtorepresentallteensand
theirparentsintheUnitedStateswhohaveaccesstoeitheralandlineorcellulartelephone.Bothsamples
wereprovidedbySurveySamplingInternational,LLC(SSI)accordingtoPSRAIspecifications.
BothsamplesweredisproportionatelystratifiedtoincreasetheincidenceofAfricanAmericansandLatinos.
Thesamestratificationschemewasusedforbothsampleframesandwasbasedontheestimatedincidenceof
minoritygroupsatthecountylevel.AllcountiesintheUnitedStatesweredividedintotenstratabasedonthe
estimatedproportionofAfricanAmericanandLatinopopulations.Stratawithhigherminoritydensitieswere
oversampledrelativetostratawithlowerdensities.Phonenumbersweredrawnwithequalprobabilities
withinstrata.Thedisproportionatesampledesignwasaccountedforintheweighting.
TosupplementthefreshRDDsample,interviewswerealsocompletedamongasampleofparentswho
recentlyparticipatedinthePSRAIWeeklyOmnibussurvey.Table1showsabreakdownofthenumberof
interviewscompletedbysamplesegment.
ContactProcedures
InterviewswereconductedfromJuly26toSeptember30,2012.Asmanyas7attemptsweremadetocontact
andinterviewaparentateverysampledlandlinetelephonenumberandasmanyasfiveattemptsweremade
tocontactandinterviewaparentateverysampledcellnumber.Aftertheparentinterview,anadditional10
callsweremadetointerviewaneligibleteen.Samplewasreleasedforinterviewinginreplicates,whichare
representativesubsamplesofthelargersample.Usingreplicatestocontrolthereleaseofsampleensuresthat
completecallproceduresarefollowedfortheentiresample.Callswerestaggeredovertimesofdayanddays
oftheweektomaximizethechanceofmakingcontactwithpotentialrespondents.Eachtelephonenumber
receivedatleastonedaytimecallinanattempttocompleteaninterview.
Contactprocedureswereslightlydifferentforthelandlineandcellsamples.Forthelandlinesamples,
interviewersfirstdeterminedifthehouseholdhadany12to17yearoldresidents.Householdswithnoteens
werescreenedoutasineligible.Ineligiblehouseholds,interviewersfirstconductedashortparentinterview
witheitherthefather/maleguardianormother/femaleguardian.Theshortparentinterviewaskedsomebasic
householddemographicquestionsaswellasquestionsaboutaparticularteeninthehousehold(selectedat
randomifmorethanoneteenlivedinthehouse.)
Forthecellphonesamples,interviewsfirstmadesurethatrespondentswereinasafeplacetotalkandthat
theywerespeakingwithanadult.Callsmadetominorswerescreenedoutasineligible.Ifthepersonwasnot
inasafeplacetotalkacallbackwasscheduled.Interviewersthenaskedifany12to17yearoldslivedintheir
household.Caseswherenoteenslivedinthehouseholdwerescreenedoutasineligible.Iftherewasanage
eligibleteeninthehousehold,theinterviewersaskedifthepersononthecellphonewasaparentofthechild.
Thosewhowereparentswentontocompletetheparentinterview.Thosewhowerenotparentswere
screenedoutasineligible.
Forallsamples,aftertheparentinterviewwascompleteaninterviewwascompletedwiththetargetchild.
Datawaskeptonlyifthechildinterviewwascompleted.
Table1.SampleSegments
Segment #ofints.
FreshRDDlandline 267
FreshRDDcell 134
Callbacklandline 265
Callbackcell 136
WEIGHTINGANDANALYSIS
Weightingisgenerallyusedinsurveyanalysistocompensateforpatternsofnonresponseand
disproportionatesampledesignsthatmightbiassurveyestimates.Thissamplewasweightedinthreestages.
ThefirststageofweightingcorrectedforthedisproportionateRDDsampledesigns.Foreachstratumthe
variableWT1wascomputedastheratioofthesizeofthesampleframeinthestratumdividedbytheamount
ofsampleorderedinthestratum.Forthecallbacksamples,theweightsfromtheoriginalsurveyswasbrought
inandusedasWT1.
Thesecondstageofweightinginvolvedcorrectingfordifferentprobabilitiesofselectionbasedon
respondentsphoneusepatterns.Respondentswhohavebothalandlineandacellphonehaveagreater
chanceofbeingsampledthanrespondentswithaccesstoonlyonekindofphone.Tocorrectforthiswe
computedavariablecalledPUA(PhoneUseAdjustment).ThePUAwascomputedusingthefollowingformula
wheren1isthenumberofrespondentshavingonlyonekindofphone(landlineorcell,butnotboth)andn2is
thenumberofrespondentshavebothalandlineandacellphone.
WT1andPUAwerethenmultipliedtogethertouseasaninputweight(WT2)forpoststratificationraking
Theinterviewedsamplewasrakedtomatchnationalparametersforbothparentandchilddemographics.The
parentdemographicsusedforweightingwere:sex;age;education;race;Hispanicorigin;numberof1217year
oldsinhousehold;numberofadultsinthehousehold;phoneuseandregion(U.S.Censusdefinitions).The
childdemographicsusedforweightingweregenderandage.Theparameterscamefromaspecialanalysisof
theCensusBureaus2011AnnualSocialandEconomicSupplement(ASEC)thatincludedallhouseholdsinthe
UnitedStates.ThephoneuseparameterwasderivedfromrecentPSRAIsurveydata.
RakingwasaccomplishedusingSampleBalancing,aspecialiterativesampleweightingprogramthat
simultaneouslybalancesthedistributionsofallvariablesusingastatisticaltechniquecalledtheDeming
Algorithm.Weightsweretrimmedtopreventindividualinterviewsfromhavingtoomuchinfluenceonthefinal
results.Theuseoftheseweightsinstatisticalanalysisensuresthatthedemographiccharacteristicsofthe
samplecloselyapproximatethedemographiccharacteristicsofthenationalpopulation.Table2compares
weightedandunweightedsampledistributionstopopulationparameters.

Table2:SampleDemographics
Parameter Unweighted Weighted
CensusRegion
Northeast 17.8 13.8 17.1
Midwest 22.2 21.2 21.0
South 36.0 36.9 36.8
West 24.0 28.1 25.1

Parent'sSex
Male 43.3 35.5 41.2
Female 56.7 64.5 58.8

Parent'sAge
LT35 10.3 6.5 9.9
3539 18.1 12.7 17.7
4044 25.6 21.4 24.6
4549 24.4 24.2 25.0
5054 14.6 21.1 15.0
55+ 7.1 14.2 7.8

Parent'sEducation
LessthanHSgrad. 12.7 6.4 11.7
HSgrad. 33.5 24.2 31.8
Somecollege 23.3 24.0 24.2
Collegegrad. 30.5 45.4 32.2

Parent'sRace/Ethnicity
White~Hispanic 63.0 68.0 63.3
Black~Hispanic 11.2 15.3 12.0
Hispanic,nativeborn 6.7 4.5 6.4
Hispanic,foreignborn 12.5 7.0 11.8
Other~Hispanic 6.5 5.1 6.6

Parent'sPhoneUse
Landlineonly 7.8 6.7 8.0
DualUsers 59.8 78.4 62.4
CellPhoneonly 33.1 14.8 29.6

#of1217KidsinHH
One 70.2 64.5 69.0
Two 25.2 27.4 25.9
Three+ 4.6 8.1 5.1

#ofadultsinHH
One 10.5 13.0 11.5
Two 58.6 58.6 57.7
Three+ 30.9 28.4 30.8
(Continued)
Table2:SampleDemographics(continued)
Parameter Unweighted Weighted
Kid'sSex
Male 51.3 50.5 51.0
Female 48.7 49.5 49.0

Kid'sAge
12 16.7 14.1 15.6
13 16.7 16.6 17.1
14 16.7 15.6 16.0
15 16.7 16.8 17.3
16 16.7 19.3 17.4
17 16.7 17.6 16.6

EffectsofSampleDesignonStatisticalInference
Postdatacollectionstatisticaladjustmentsrequireanalysisproceduresthatreflectdeparturesfromsimple
randomsampling.PSRAIcalculatestheeffectsofthesedesignfeaturessothatanappropriateadjustmentcan
beincorporatedintotestsofstatisticalsignificancewhenusingthesedata.Thesocalled"designeffect"ordeff
representsthelossinstatisticalefficiencythatresultsfromsystematicnonresponse.Thetotalsampledesign
effectforthissurveyis1.69.
PSRAIcalculatesthecompositedesigneffectforasampleofsizen,witheachcasehavingaweight,w
i
as:

In a wide range of situations, the adjusted standard error of a statistic should be calculated by
multiplyingtheusualformulabythesquarerootofthedesigneffect(deff).Thus,theformulaforcomputing
the95%confidenceintervalaroundapercentageis:
2
1
1
2
|
.
|
\
|
=
=
=
n
i
i
n
i
i
w
w n
deff
formula1
|
|
.
|
\
|

n
p p
deff p

) 1 (
96 . 1
formula2
where pisthesampleestimateandnistheunweightednumberofsamplecasesinthegroupbeing
considered.
Thesurveysmarginoferroristhelargest95%confidenceintervalforanyestimatedproportionbased
onthetotalsampletheonearound50%.Forexample,themarginoferrorfortheentiresampleis4.5
percentagepoints.Thismeansthatin95outevery100samplesdrawnusingthesamemethodology,
estimatedproportionsbasedontheentiresamplewillbenomorethan4.5percentagepointsawayfromtheir
truevaluesinthepopulation.Itisimportanttorememberthatsamplingfluctuationsareonlyonepossible
sourceoferrorinasurveyestimate.Othersources,suchasrespondentselectionbias,questionnairewording
andreportinginaccuracy,maycontributeadditionalerrorofgreaterorlessermagnitude.

Response Rate
Table3reportsthedispositionofallsampledcallbacktelephonenumberseverdialed.Theresponserateis
calculatedaccordingtoAmericanAssociationofPublicOpinionResearchstandards.
Table 3:Sample Disposition

Landline
Fresh
RDD
Cell
Fresh
RDD
LL
Callback
Cell
Callback
267 134 265 136 I=Completes
17 9 9 10 R=Refusal known to be eligible
11197 14226 501 448 UO
R
=Refusal eligibility status unknown

4733 8666 56 63 NC=Non contact known working number
211 108 2 3 O=Other
54721 17757 126 98 OF=Business/computer/not working/child's cell phone
4960 1043 10 1
UHUO
NC
=Non-contact - unknown household/unknown
other
3383 3475 89 101 SO=Screenout

0.31 0.61 0.88 0.89
e1=(I+R+UO
R
+NC+O+SO)/(I+R+UO
R
+NC+O+SO+OF) -
Assumed working rate of non-contacts
0.08 0.04 0.75 0.59
e2=(I+R)/(I+R+SO) - Assumed eligibility of unscreened
contacts

16.1% 12.4% 37.7% 30.2%
AAPOR
RR3=I/[I+R+[e2*(UOR+NC+O)]+[e1*e2*UHUO
NC
]]
i
September 2009 trends based on the Parents and Teens Cell Phone Use Survey conducted June 26 September 24, 2009
[n=800 parents of 12-17 year-olds, n=746 internet teens ages 12-17 and 54 offline teens ages 12-17].
ii
February 2008 trends based on the Gaming & Civic Engagement Survey of Teens/Parents conducted November 1, 2007
February 5, 2008 [n=1,102 parents of 12-17 year-olds, n=1,033 internet teens ages 12-17 and 69 offline teens ages 12-17].

Teens and Technology 2013 - Pew Research Center

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Teens and Technology 2013 - Pew Research Center

Uploaded by

Copyright:

Available Formats

MARCH13,2013

FOR FURTHER INFORMATION, CONTACT:

Pew Research Centers Internet & American Life Project

Teen Computer and Tablet Ownership Demographics

Teen Cell Phone and Smartphone Ownership Demographics

YES NO DON T KNOW REFUSED

Table 3:Sample Disposition

You might also like