Ngày tải lên :
12/02/2014, 23:20
... thepopulation.Thenumberofwomen
sherepresentsinthepopulationiscalled
her‘‘samplingweight.’’Sampling
weightsmayvaryconsiderablyfromthis
averagevaluedependingonthe
respondent’srace,theresponseratefor
similarwomen,andotherfactors.As
withanysamplesurvey,theestimatesin
thisreportaresubjecttosampling
variability.SignificancetestsonNSFG
datashouldbedonetakingthesampling
designintoaccount.
Nonsamplingerrorswereminimized
bystringentquality-controlprocedures
thatincludedthoroughinterviewer
training,checkingtheconsistencyof
answersduringandaftertheinterview,
imputingmissingdata,andadjustingthe
samplingweightsfornonresponseand
undercoveragetomatchnationaltotals.
Estimatesofsamplingerrorsandother
statisticalaspectsofthesurveyare
describedinmoredetailinanother
separatereport(13).
Thisreportshowsfindingsby
characteristicsofthewomaninterviewed,
includingherage,maritalstatus,
education,parity,householdincome
dividedbythepovertylevel,andraceand
Hispanicorigin.Ithasbeenshownthat
blackandHispanicwomenhavemarkedly
lowerlevelsofincome,education,and
accesstohealthcareandhealthinsurance,
thanwhitewomen(14).Theseandother
factors,ratherthanraceororiginperse,
probablyaccountfordifferencesinthe
behaviorsandoutcomesstudiedinthis
reportamongwhite,black,andHispanic
women(15).
TableBshowsafactorthatshould
beconsideredininterpretingtrendsin
pregnancy-relatedbehaviorintheUnited
States:thechangingagecompositionof
thereproductive-agepopulation.In
1982,therewere54.1millionwomenof
reproductiveageintheUnitedStates;in
1988,57.9million;andin1995,60.2
million(16).Thelargebabyboom
cohort,bornbetween1946and1964,
was18–34yearsofagein1982,24–42
yearsofagein1988,and31–49years
ofagein1995.Theselargebirthcohorts
werepreceded(upto1945)and
followed(1965–80)bysmallercohorts.
Whiletheoverallnumberofwomen
15–44yearsofageroseby6million,or
11percentbetween1982and1995,the
numberofteenagewomendroppedby
about6percent,thenumberofwomen
20–24yearsofagedroppedby
15percent,andthenumberofwomen
25–29droppedby6percent(tableB).In
contrast,thenumberofwomen30–44
yearsofageincreasedsharply—for
example,thenumberofwomen40–44
yearsofageincreasedby59percent
between1982and1995.Also,women
30–44yearsofageaccountedfor
54percentofwomen15–44yearsofage
in1995comparedwith44percentin
1982.Thesedifferencesinage
compositionmayberelevantwhenever
timetrendsamongwomen15–44years
ofagearebeingdiscussed.
Publicusefilesbasedonthe1995
NSFGareavailableoncomputertape.
TheywillalsobeavailableonCompact
DiscRead-OnlyMemory(CD-ROM).
Questionsaboutthecostandavailability
ofthecomputertapesshouldbedirected
totheNationalTechnicalInformation
Service(NTIS),5285PortRoyalRoad,
Springfield,VA22161,703–487-4650,
or1–800-553-NTIS.Questionsregarding
theCD-ROMfilesshouldbedirectedto
NCHS’DataDisseminationBranchat
301–436-8500.
Results
T
ables1–17containmeasuresof
pregnancyandbirthintheUnited
States.
ChildrenEverBornandTotal
BirthsExpected
In1995,women15–44yearsof
ageintheUnitedStateshadhadan
averageof1.2birthsperwoman
(table1).Thiscompareswith1.2in
1988and1.3in1982(17).In1995,
women15–44yearsofageexpectedto
finishtheirchildbearingwithan
averageof2.2childrenperwoman
(table1)comparedwith2.2in1988
and2.4in1982(17).
Theproportionwhoreportthatthey
haveneverbeenpregnantwasmarkedly
higherforcollegegraduatesthanfor
thosewhodidnotcompletehighschool
(table3).Thissamepatternbyeducation
isalsoseenwhendataforlivebirthsare
examined(tables4–5):about49percent
ofwomen22–44yearsofagewhohad
graduatedfromcollegehadhadnolive
birthsasofthedateofinterview
comparedwithjust8percentofwomen
22–44yearsofagewithoutahigh
schooldiploma(table4).Withinrace
andHispanicorigingroups,thepattern
wasthesame:collegegraduateshad
markedlyhigherpercentschildlessthan
womenwithlesseducation(table5).
Table6showsacomparison
betweenlivebirthsreportedinthe
NSFGandlivebirthsregisteredonbirth
certificatesintheyears1991–94.In
eachindividualcalendaryearandfor
thesumoftheyears1991–94,the
NSFGestimateofthenumberofbirths
isveryclosetothebirthcertificatetotal
anddiffersfromitbylessthanthe
NSFG’ssamplingerror.TheNSFG
estimateisalsoverycloseforwhite
women.TheNSFGestimateforblack
womenisslightlylower,andthe
estimateforotherracessomewhat
higherthanthebirthcertificatedata.A
discussionofthisdifferenceisgivenin
thedefinitionof‘‘RaceandHispanic
origin’’inthe‘‘DefinitionsofTerms.’’
Overall,andbycharacteristicsother
thanrace,however,table6showsthat
TableB.Numberofwomen,byage:UnitedStates,1982,1988,and1995
Age¬ ... thepopulation.Thenumberofwomen
sherepresentsinthepopulationiscalled
her‘‘samplingweight.’’Sampling
weightsmayvaryconsiderablyfromthis
averagevaluedependingonthe
respondent’srace,theresponseratefor
similarwomen,andotherfactors.As
withanysamplesurvey,theestimatesin
thisreportaresubjecttosampling
variability.SignificancetestsonNSFG
datashouldbedonetakingthesampling
designintoaccount.
Nonsamplingerrorswereminimized
bystringentquality-controlprocedures
thatincludedthoroughinterviewer
training,checkingtheconsistencyof
answersduringandaftertheinterview,
imputingmissingdata,andadjustingthe
samplingweightsfornonresponseand
undercoveragetomatchnationaltotals.
Estimatesofsamplingerrorsandother
statisticalaspectsofthesurveyare
describedinmoredetailinanother
separatereport(13).
Thisreportshowsfindingsby
characteristicsofthewomaninterviewed,
includingherage,maritalstatus,
education,parity,householdincome
dividedbythepovertylevel,andraceand
Hispanicorigin.Ithasbeenshownthat
blackandHispanicwomenhavemarkedly
lowerlevelsofincome,education,and
accesstohealthcareandhealthinsurance,
thanwhitewomen(14).Theseandother
factors,ratherthanraceororiginperse,
probablyaccountfordifferencesinthe
behaviorsandoutcomesstudiedinthis
reportamongwhite,black,andHispanic
women(15).
TableBshowsafactorthatshould
beconsideredininterpretingtrendsin
pregnancy-relatedbehaviorintheUnited
States:thechangingagecompositionof
thereproductive-agepopulation.In
1982,therewere54.1millionwomenof
reproductiveageintheUnitedStates;in
1988,57.9million;andin1995,60.2
million(16).Thelargebabyboom
cohort,bornbetween1946and1964,
was18–34yearsofagein1982,24–42
yearsofagein1988,and31–49years
ofagein1995.Theselargebirthcohorts
werepreceded(upto1945)and
followed(1965–80)bysmallercohorts.
Whiletheoverallnumberofwomen
15–44yearsofageroseby6million,or
11percentbetween1982and1995,the
numberofteenagewomendroppedby
about6percent,thenumberofwomen
20–24yearsofagedroppedby
15percent,andthenumberofwomen
25–29droppedby6percent(tableB).In
contrast,thenumberofwomen30–44
yearsofageincreasedsharply—for
example,thenumberofwomen40–44
yearsofageincreasedby59percent
between1982and1995.Also,women
30–44yearsofageaccountedfor
54percentofwomen15–44yearsofage
in1995comparedwith44percentin
1982.Thesedifferencesinage
compositionmayberelevantwhenever
timetrendsamongwomen15–44years
ofagearebeingdiscussed.
Publicusefilesbasedonthe1995
NSFGareavailableoncomputertape.
TheywillalsobeavailableonCompact
DiscRead-OnlyMemory(CD-ROM).
Questionsaboutthecostandavailability
ofthecomputertapesshouldbedirected
totheNationalTechnicalInformation
Service(NTIS),5285PortRoyalRoad,
Springfield,VA22161,703–487-4650,
or1–800-553-NTIS.Questionsregarding
theCD-ROMfilesshouldbedirectedto
NCHS’DataDisseminationBranchat
301–436-8500.
Results
T
ables1–17containmeasuresof
pregnancyandbirthintheUnited
States.
ChildrenEverBornandTotal
BirthsExpected
In1995,women15–44yearsof
ageintheUnitedStateshadhadan
averageof1.2birthsperwoman
(table1).Thiscompareswith1.2in
1988and1.3in1982(17).In1995,
women15–44yearsofageexpectedto
finishtheirchildbearingwithan
averageof2.2childrenperwoman
(table1)comparedwith2.2in1988
and2.4in1982(17).
Theproportionwhoreportthatthey
haveneverbeenpregnantwasmarkedly
higherforcollegegraduatesthanfor
thosewhodidnotcompletehighschool
(table3).Thissamepatternbyeducation
isalsoseenwhendataforlivebirthsare
examined(tables4–5):about49percent
ofwomen22–44yearsofagewhohad
graduatedfromcollegehadhadnolive
birthsasofthedateofinterview
comparedwithjust8percentofwomen
22–44yearsofagewithoutahigh
schooldiploma(table4).Withinrace
andHispanicorigingroups,thepattern
wasthesame:collegegraduateshad
markedlyhigherpercentschildlessthan
womenwithlesseducation(table5).
Table6showsacomparison
betweenlivebirthsreportedinthe
NSFGandlivebirthsregisteredonbirth
certificatesintheyears1991–94.In
eachindividualcalendaryearandfor
thesumoftheyears1991–94,the
NSFGestimateofthenumberofbirths
isveryclosetothebirthcertificatetotal
anddiffersfromitbylessthanthe
NSFG’ssamplingerror.TheNSFG
estimateisalsoverycloseforwhite
women.TheNSFGestimateforblack
womenisslightlylower,andthe
estimateforotherracessomewhat
higherthanthebirthcertificatedata.A
discussionofthisdifferenceisgivenin
thedefinitionof‘‘RaceandHispanic
origin’’inthe‘‘DefinitionsofTerms.’’
Overall,andbycharacteristicsother
thanrace,however,table6showsthat
TableB.Numberofwomen,byage:UnitedStates,1982,1988,and1995
Age¬ ... 1995
National Survey of Family Growth.
Vital Health Stat Series 2 (in
preparation).
14. U.S. Bureau of the Census. Statistical
abstract of the United States, 1996.
Washington: U.S. Department of
Commerce....