search

ajdamico / asdfree

analyze survey data for free
http://asdfree.com/
GNU General Public License v3.0
612 stars 449 forks source link

pns `dom` table does not get used in the import script #71

Closed ajdamico closed 8 years ago

ajdamico commented 8 years ago

djalma, i made some changes to the main script to get started--

https://github.com/ajdamico/asdfree/commit/e7d9b6eb5985b5203b7b40d3ccb06142bca2888d

--but the dom table never gets used, and i believe it should be merged onto the pes table before the survey design gets created? that way, it is a rectangular file (like the censo demografico gets dom and pes merged). what merge fields should be used? so that i can merge on the household-level information

using the current version of the import file, the intersecting fields are--

> intersect(names(dom),names(pes))
 [1] "v0001"     "v0024"     "upa_pns"   "v0006_pns" "upa"       "v0028"     "v0029"     "v00281"    "v00291"   "v00282"    "v00292"    "v00283"    "v00293"

thanks!

DjalmaPessoa commented 8 years ago

Hi, I'm out of town. I'll be back on wednesday. Have a wonderful 2016!

Em 2 de jan de 2016, às 18:19, Anthony Damico notifications@github.com escreveu:

djalma, i made some changes to the main script to get started--

https://githubcom/ajdamico/asdfree/commit/e7d9b6eb5985b5203b7b40d3ccb06142bca2888d

--but the dom table never gets used, and i believe it should be merged onto the pes table before the survey design gets created? that way, it is a rectangular file (like the censo demografico gets dom and pes merged) what merge fields should be used? so that i can merge on the household-level information

using the current version of the import file, the intersecting fields are--

intersect(names(dom),names(pes)) [1] "v0001" "v0024" "upa_pns" "v0006_pns" "upa" "v0028" "v0029" "v00281" "v00291" "v00282" "v00292" "v00283" "v00293" thanks!

— Reply to this email directly or view it on GitHub.

DjalmaPessoa commented 8 years ago

To merge "dom" and "pes" please use the variables:

V0001 (State), V0024 (Stratum), UPA_PNS (PSU) and V0006_PNS (household order number in the PNS).

Note that for people it is needed to work with two different designs:

  1. people with age >18 answering the long questionnaire. It is defined in the the data subset:

    !is.na(M001)&M001=="1"

  2. For people in the whole population.

Please define them separately and poststratify each one

In the script I sent before there are examples of estimates defined for the the two cases.

To estimate a household characteristic please use the

design:

dom_des <- subset (pes_des, C001== "01")

where "pes_des" is the design for the whole population of people (2).

This defines the data relative to the head of the household (one for each household), eliminating household data duplication for people living in the same household in the merged file.

On Sun, Jan 3, 2016 at 11:05 AM, Djalma Pessoa pessoad@gmail.com wrote:

Hi, I'm out of town. I'll be back on wednesday. Have a wonderful 2016!

Em 2 de jan de 2016, às 18:19, Anthony Damico notifications@github.com escreveu:

djalma, i made some changes to the main script to get started--

https://githubcom/ajdamico/asdfree/commit/e7d9b6eb5985b5203b7b40d3ccb06142bca2888d

--but the dom table never gets used, and i believe it should be merged onto the pes table before the survey design gets created? that way, it is a rectangular file (like the censo demografico gets dom and pes merged) what merge fields should be used? so that i can merge on the household-level information

using the current version of the import file, the intersecting fields are--

intersect(names(dom),names(pes)) [1] "v0001" "v0024" "upa_pns" "v0006_pns" "upa" "v0028" "v0029" "v00281" "v00291" "v00282" "v00292" "v00283" "v00293"

thanks!

— Reply to this email directly or view it on GitHub https://github.com/ajdamico/asdfree/issues/71.

DjalmaPessoa commented 8 years ago

new script with changes after merging dom and pes.

On Thu, Jan 7, 2016 at 5:12 PM, Djalma Pessoa pessoad@gmail.com wrote:

To merge "dom" and "pes" please use the variables:

V0001 (State), V0024 (Stratum), UPA_PNS (PSU) and V0006_PNS (household order number in the PNS).

Note that for people it is needed to work with two different designs:

  1. people with age >18 answering the long questionnaire. It is defined in the the data subset:

    !is.na(M001)&M001=="1"

  2. For people in the whole population.

Please define them separately and poststratify each one

In the script I sent before there are examples of estimates defined for the the two cases.

To estimate a household characteristic please use the

design:

dom_des <- subset (pes_des, C001== "01")

where "pes_des" is the design for the whole population of people (2).

This defines the data relative to the head of the household (one for each household), eliminating household data duplication for people living in the same household in the merged file.

On Sun, Jan 3, 2016 at 11:05 AM, Djalma Pessoa pessoad@gmail.com wrote:

Hi, I'm out of town. I'll be back on wednesday. Have a wonderful 2016!

Em 2 de jan de 2016, às 18:19, Anthony Damico notifications@github.com escreveu:

djalma, i made some changes to the main script to get started--

https://githubcom/ajdamico/asdfree/commit/e7d9b6eb5985b5203b7b40d3ccb06142bca2888d

--but the dom table never gets used, and i believe it should be merged onto the pes table before the survey design gets created? that way, it is a rectangular file (like the censo demografico gets dom and pes merged) what merge fields should be used? so that i can merge on the household-level information

using the current version of the import file, the intersecting fields are--

intersect(names(dom),names(pes)) [1] "v0001" "v0024" "upa_pns" "v0006_pns" "upa" "v0028" "v0029" "v00281" "v00291" "v00282" "v00292" "v00283" "v00293"

thanks!

— Reply to this email directly or view it on GitHub https://github.com/ajdamico/asdfree/issues/71.

setwd( "C:/Djalma/PNS/" )

load("dom.rda") load("pes.rda")

merge the microdata for household and people:

DomPes <- merge(dom,pes, by.x=c("V0001","V0024", "UPA_PNS", "V0006_PNS"), by.y= c("V0001","V0024", "UPA_PNS", "V0006_PNS") )

save(DomPes,file="DomPes.rda")

##################################################################

design object for people answering the long questionaire

############################################################# pes_sel<-subset(DomPes,!is.na(M001)&M001=="1")

people with self evaluated health good or very good

pes_sel<-transform(pes_sel,saude.b.mb=ifelse(N001%in%c("1","2"),1,0))

urban and rural

pes_sel<-transform(pes_sel,situ=substring(V0024,7,7)) pes_sel$situ<-factor(pes_sel$situ,labels=c("urbano","rural"))

sexo

pes_sel$C006<-factor(pes_sel$C006,labels=c("masculino","feminino"))

UF

estado.names <- c( "Rondônia" , "Acre" , "Amazonas" , "Roraima" , "Pará" , "Amapá" , "Tocantins" , "Maranhão" , "Piauí" , "Ceará" , "Rio Grande do Norte" , "Paraíba" , "Pernambuco" , "Alagoas" , "Sergipe" , "Bahia" , "Minas Gerais" , "Espírito Santo" , "Rio de Janeiro" , "São Paulo" , "Paraná" , "Santa Catarina" , "Rio Grande do Sul" , "Mato Grosso do Sul" , "Mato Grosso" , "Goiás" , "Distrito Federal" ) pes_sel$UF<-factor(pes_sel$V0001,labels=estado.names)

Region

region.names<-c("Norte","Nordeste","Sudeste","Sul","Centro-Oeste") pes_sel$Region<-substring(pes_sel$V0001,1,1) pes_sel$Region<-factor(pes_sel$Region,labels=region.names) table(pes_sel$Region)

pes_sel$P04101<- as.numeric(pes_sel$P04101) pes_sel$P04102<- as.numeric(pes_sel$P04102) pes_sel$P04301<- as.numeric(pes_sel$P04301) pes_sel$P04302<- as.numeric(pes_sel$P04302)

pes_sel<-transform(pes_sel,tempo_desl_trab=ifelse(is.na(P04101),0,P04101 * 60 + P04102)) pes_sel<-transform(pes_sel,tempo_desl_athab =ifelse(is.na(P04301),0,P04301 * 60 + P04302)) pes_sel<-transform(pes_sel,tempo_desl = tempo_desl_trab + tempo_desl_athab) pes_sel<-transform(pes_sel,atfi04=ifelse(tempo_desl >= 30,1,0))

variable 1

pes_sel$one<-1

survey design

library(survey) pes.sel.des<-svydesign(ids=~UPA_PNS,strata=~V0024,data=pes_sel,weights=~V0029.y,nest=TRUE)

post-stratify design

post.pop<-unique(subset(pes_sel,select=c(V00293.y,V00292.y))) names(post.pop)<-c("V00293.y","Freq")

pes.sel.des.pos<-postStratify(pes.sel.des,~V00293.y,post.pop)

examples:

############################################################

compute estimates of table tabela 5.1.1.1 in

ftp://ftp.ibge.gov.br/PNS/2013/pns2013.pdf

#################################################################

Brasil

saudbr<-svymean(~saude.b.mb, design=pes.sel.des.pos) c(round(100_coef(saudbr),1),round(100_confint(saudbr)[1,],1))

by sex

saudbrsex<-svyby(~saude.b.mb,~C006, design= pes.sel.des.pos, vartype="ci", level = 0.95, svymean) cbind(saudbrsex[,1],round(100*saudbrsex[,2:4],1))

Situation

by situation (Rural and urban)

saudsitu<-svyby(~saude.b.mb,~situ, design= pes.sel.des.pos, vartype="ci", level = 0.95, svymean) cbind(saudsitu[,1],round(100*saudsitu[,2:4],1))

situation x sex

saudsitusex<-svyby(~saude.b.mb,~situ+C006, design= pes.sel.des.pos, vartype="ci", level = 0.95, svymean) cbind(saudsitusex[,1:2],round(100*saudsitusex[,3:5],1))

by UF

sauduf<-svyby(~saude.b.mb,~UF, design= pes.sel.des.pos, vartype="ci", level = 0.95, svymean) cbind(sauduf[,1],round(100*sauduf[,2:4],1))

UF x SEX

saudufsex<-svyby(~saude.b.mb,~UF+C006, design= pes.sel.des.pos, vartype="ci", level = 0.95, svymean) cbind(saudufsex[,1:2],round(100*saudufsex[,3:5],1))

###########################################################################################

Estimativas da Tabela 3.4.1.1 em

ftp://ftp.ibge.gov.br/PNS/2013/pns2013.pdf

####################################################

Compare Estimates against ,

% of ind. above 18yearsold that practice active travel for > 30minutes

Brasil

atfibr<-svymean(~atfi04, design=pes.sel.des.pos) svymean(~atfi04, design=pes.sel.des.pos,vartype="ci", level = 0.95) c(round(100_coef(atfibr),1),round(100_confint(atfibr)[1,],1))

by situation (Urban vs Rural)

atfisitu<-svyby(~atfi04, ~situ, design= pes.sel.des.pos, vartype="ci", level = 0.95, svymean,na.rm=T) cbind(atfisitu[,1],round(100*atfisitu[,2:4],1))

by Region

atfireg<-svyby(~atfi04, ~Region, design= pes.sel.des.pos, vartype="ci", level = 0.95, svymean) cbind(atfireg[,1],round(100*atfireg[,2:4],1))

by Race

atfirace<-svyby(~atfi04, ~C009, design= pes.sel.des.pos, vartype="ci", level = 0.95, svymean) cbind(atfirace[-6,1],round(100*atfirace[-6,2:4],1))

by UF

atfiuf<-svyby(~atfi04,~UF, design= pes.sel.des.pos, vartype="ci", level = 0.95, svymean) cbind(atfiuf[,1],round(100*atfiuf[,2:4],1))

UF x SEX

atfiufsex<-svyby(~atfi04,~UF+C006, design= pes.sel.des.pos, vartype="ci", level = 0.95, svymean) cbind(atfiufsex[,1:2],round(100*atfiufsex[,3:5],1))

###############################################################

Examples for the whole population

match with Gráfico 8 in http://biblioteca.ibge.gov.br/visualizacao/livros/liv94074.pdf

############################################################

Create variables

idade categorizada

DomPes$C008<-as.numeric(DomPes$C008) DomPes$age.cat<-cut(DomPes$C008,c(0,18,30,40,60,150),right=FALSE,include.lowest=TRUE)

raça:

DomPes$raca<-DomPes$C009 DomPes$raca[DomPes$raca%in%c("3","5","9")]<-NA DomPes$raca<-factor(DomPes$raca,labels=c("Branca","Preta","Parda"))

Educ

DomPes$educ<-DomPes$VDD004 DomPes$educ[DomPes$educ==" "]<-NA DomPes$educ[DomPes$educ%in%c("1","2")]<-"1" DomPes$educ[DomPes$educ%in%c("3","4")]<-"2" DomPes$educ[DomPes$educ%in%c("5","6")]<-"3" DomPes$educ<-factor(DomPes$educ,labels=c("SinstFundi","FundcMedi","MedcSupi","Supc"))

DomPes$Region<-substring(DomPes$V0001,1,1) DomPes$Region<-factor(DomPes$Region,labels=region.names)

DomPes$C001<- as.numeric(DomPes$C001)

Question answered by all people

design object for all people

pes.all.des<-svydesign(ids=~UPA_PNS,strata=~V0024,data=DomPes,weights=~V0028.y,nest=TRUE)

post-stratification

post.pop.all<-subset(DomPes,select=c(V00283.y,V00282.y)) post.pop.all<-unique(post.pop.all) names(post.pop.all)<-c("V00283.y","Freq") pes.all.des.pos<-postStratify(pes.all.des,~V00283.y,post.pop.all)

Examples:

Variable VDD004:

Higher schooling level ( age>=5)

1 No instruction

2 incomplete elementary or equivalent

3 complete elementary or equivalent

4 incomplete middle or equivalent

5 complete middle or equivalent

6 incomplete university or equivalent

7 Graduation

svymean(~VDD004,pes.all.des.pos)

people with health insurance

svymean(~I001,pes.all.des.pos,na.rm=TRUE)

by sex

plansex<-svyby(~I001,~C006, design=pes.all.des.pos,vartype="ci", level = 0.95, svymean,na.rm=TRUE) plansex.res<-data.frame(Sex=c("Male","Female"),round(100*plansex[,c(2,4,6)],1))

bar plot

library(ggplot2) ggplot(plansex.res,aes(x=Sex,y=I0011,fill=Sex))+ geom_bar( stat="identity", width=0.9) + geom_errorbar(aes(ymin=ci_l.I0011, ymax=ci_u.I0011))+ xlab("Sex")+ylab("% Health Insurance")+ ggtitle("Proportion of people having health insurance by sex")+ theme_bw()

by age class

Estimates

planage<-svyby(~I001,~age.cat, design=pes.all.des.pos,vartype="ci", level = 0.95, svymean,na.rm=TRUE) planage.res<-data.frame(age.cat=planage[,1],round(100*planage[,c(2,4,6)],1))

bar plot

ggplot(planage.res,aes(x=age.cat,y=I0011,fill=age.cat))+ geom_bar( stat="identity") + geom_errorbar(aes(ymin=ci_l.I0011, ymax=ci_u.I0011))+ xlab("Age")+ylab("% Health Insurance")+ ggtitle("Proportion of people having health insurance by age")+ theme_bw()

by education class

planeduc<-svyby(~I001,~educ, design=pes.all.des.pos,vartype="ci", level = 0.95, svymean,na.rm=TRUE) planeduc.res<-data.frame(educ=planeduc[,1],round(100*planeduc[,c(2,4,6)],1)) planeduc.res

bar plot

ggplot(planeduc.res,aes(x=educ,y=I0011,fill=educ))+ geom_bar( stat="identity") + geom_errorbar(aes(ymin=ci_l.I0011, ymax=ci_u.I0011))+ xlab("Education")+ylab("% health insurance")+ ggtitle("Proportion of people having health insurance by education")+ theme_bw()

by race

planraca<-svyby(~I001,~raca, design=pes.all.des.pos,vartype="ci", level = 0.95, svymean,na.rm=TRUE) planraca.res<-data.frame(race=planraca[,1],round(100*planraca[,c(2,4,6)],1)) planraca.res

bar plot

ggplot(planraca.res,aes(x=race,y=I0011,fill=race))+ geom_bar( stat="identity") + geom_errorbar(aes(ymin=ci_l.I0011, ymax=ci_u.I0011))+ xlab("Education")+ylab("% health insurance")+ ggtitle("Proportion of people having health insurance by race")+ theme_bw()

Example 1

library(survey)

design object for people answering the long questionaire

pes_sel<-subset(pes,!is.na(M001)&M001=="1")

registers with missing weights

pes.peso0<-subset(pes_sel,is.na(V0029))

discard regiters with missing weights. Not needed, these lines were discarded by IBGE

pes_sel0<-subset(pes_sel,!is.na(V0029))

people with self evalueted health good or very good

pes_sel0<-transform(pes_sel0,saude.b.mb=ifelse(N001%in%c("1","2"),1,0))

urban and rural

pes_sel0<-transform(pes_sel0,situ=substring(V0024,7,7)) pes_sel0$situ<-factor(pes_sel0$situ,labels=c("Urbano","Rural"))

sex

pes_sel0$C006<-factor(pes_sel0$C006,labels=c("Masculino","Feminino"))

States

estado.names <- c( "Rondônia" , "Acre" , "Amazonas" , "Roraima" , "Pará" , "Amapá" , "Tocantins" , "Maranhão" , "Piauí" , "Ceará" , "Rio Grande do Norte" , "Paraíba" , "Pernambuco" , "Alagoas" , "Sergipe" , "Bahia" , "Minas Gerais" , "Espírito Santo" , "Rio de Janeiro" , "São Paulo" , "Paraná" , "Santa Catarina" , "Rio Grande do Sul" , "Mato Grosso do Sul" , "Mato Grosso" , "Goiás" , "Distrito Federal" ) pes_sel0$V0001<-factor(pes_sel0$V0001,labels=estado.names)

this is a separate sample just for a group of people not a domain

survey design

library(survey) pes.sel.des<-svydesign(ids=~UPA_PNS,strata=~V0024,data=pes_sel0,weights=~V0029,nest=TRUE)

design post-stratification

post.pop<-unique(subset(pes_sel0,select=c(V00293,V00292))) names(post.pop)<-c("V00293","Freq")

pes.sel.des.pos<-postStratify(pes.sel.des,~V00293,post.pop)

####################Example 1#############################################

Brasil

tab5.1.1.1.BR<-svymean(~saude.b.mb,pes.sel.des.pos)

round(100_c(coef(tab5.1.1.1.BR),coef(tab5.1.1.1.BR)-2_SE(tab5.1.1.1.BR), coef(tab5.1.1.1.BR)+2*SE(tab5.1.1.1.BR)),1)

create a funtion to organize the results of the svyby function

tab5111<-function(svyby.obj,nomevar){ LI<-coef(svyby.obj)-2_SE(svyby.obj) LS<-coef(svyby.obj)+2_SE(svyby.obj) result<-data.frame(svyby.obj[,1:length(nomevar)],saude.b.mb=round(100_coef(svyby.obj),1), LI=round(100_LI,1),LS=round(100*LS,1)) names(result)[1:length(nomevar)]<-nomevar rownames(result)<-NULL result }

sex

tab5.1.1.1_BR_SEXO<-svyby(~saude.b.mb,~C006,pes.sel.des.pos,svymean) tab5111(tab5.1.1.1_BR_SEXO,"sexo")

situation (urban, rural)

tab5.1.1.1_SITU<-svyby(~saude.b.mb,~situ,pes.sel.des.pos,svymean) tab5111(tab5.1.1.1_SITU,"situação")

situação e sexo

tab5.1.1.1_SITU_SEXO<-svyby(~saude.b.mb,~situ+C006,pes.sel.des.pos,svymean) tab5111(tab5.1.1.1_SITU_SEXO,c("situação","sexo"))

UF

tab5.1.1.1_UF<-svyby(~saude.b.mb,~V0001,pes.sel.des.pos,svymean) tab5111(tab5.1.1.1_UF,"UF")

UF X sexo

tab5.1.1.1_UF_SEXO<-svyby(~saude.b.mb,~V0001+C006,pes.sel.des.pos,svymean) tab5111(tab5.1.1.1_UF_SEXO,c("UF","sexo")) ###########################End of Example 1 ########################################################3

Example 2:

Question answered by all people

design object for all people

pes.all.des<-svydesign(ids=~UPA_PNS,strata=~V0024,data=pes,weights=~V0028,nest=TRUE)

post-stratification

post.pop.all<-subset(pes,select=c(V00283,V00282)) post.pop.all<-unique(post.pop.all) names(post.pop.all)<-c("V00283","Freq") pes.all.des.pos<-postStratify(pes.sel.des,~V00283,post.pop.all)

Higher schooling level ( age>=5)

svymean(~VDD004,pes.all.des.pos)

x<-subset(pes.all.des.pos,C008>=5) svymean(~VDD004,x)

this result was not published yet.

dom01<-subset(dom,V0015=="01")

design object for people answering the long questionaire

pes_sel<-subset(pes,!is.na(M001)&M001=="1")

registers with missing weights

pes.peso0<-subset(pes_sel,is.na(V0029))

discard regiters with missing weights. These registers were discard by IBGE

pes_sel0<-subset(pes_sel,!is.na(V0029))

people with self evalueted health good or very good

pes_sel0<-transform(pes_sel0,saude.b.mb=ifelse(N001%in%c("1","2"),1,0))

urban and rural

pes_sel0<-transform(pes_sel0,situ=substring(V0024,7,7))

survey design

library(survey) pes.sel.des<-svydesign(ids=~UPA_PNS,strata=~V0024,data=pes_sel0,weights=~V0029,nest=TRUE)

pós-estratificação do desenho

post.pop<-unique(subset(pes_sel0,select=c(V00293,V00292))) names(post.pop)<-c("V00293","Freq")

pes.sel.des.pos<-postStratify(pes.sel.des,~V00293,post.pop)

calcula estimativas da tabela 5.1.1.1

Brasil

tab5.1.1.1.BR<-svymean(~saude.b.mb,pes.sel.des.pos) class(tab5.1.1.1.BR) round(100_c(coef(tab5.1.1.1.BR),coef(tab5.1.1.1.BR)-2_SE(tab5.1.1.1.BR), coef(tab5.1.1.1.BR)+2*SE(tab5.1.1.1.BR)),1)

Brasil por sexo

tab5.1.1.1_BR_SEXO<-svyby(~saude.b.mb,~C006,pes.sel.des.pos,svymean) tab5.1.1.1_BR_SEXO<-transform(tab5.1.1.1_BR_SEXO,LI=saude.b.mb-2_se,LS=saude.b.mb+2_se)

masculino

cbind(tab5.1.1.1_BR_SEXO[1,1],round(100*tab5.1.1.1_BR_SEXO[1,c(2,4,5)],1))

feminino

cbind(tab5.1.1.1_BR_SEXO[2,1],round(100*tab5.1.1.1_BR_SEXO[2,c(2,4,5)],1))

Situação

Rural e urbana

tab5.1.1.1_SITU<-svyby(~saude.b.mb,~situ,pes.sel.des.pos,svymean) tab5.1.1.1_SITU<-transform(tab5.1.1.1_SITU,LI=saude.b.mb-2_se,LS=saude.b.mb+2_se) cbind(tab5.1.1.1_SITU[,1],round(100*tab5.1.1.1_SITU[,c(2,4,5)],1))

Rural e urbana por sexo

tab5.1.1.1_SITU_SEXO<-svyby(~saude.b.mb,~situ+C006,pes.sel.des.pos,svymean)

tab5.1.1.1_SITU_SEXO<-transform(tab5.1.1.1_SITU_SEXO,LI=saude.b.mb-2_se,LS=saude.b.mb+2_se) tab5.1.1.1_SITU_MASC<-subset(tab5.1.1.1_SITU_SEXO,C006==1) tab5.1.1.1_SITU_FEM<-subset(tab5.1.1.1_SITU_SEXO,C006==2)

SITU, sexo masculino

cbind(tab5.1.1.1_SITU_MASC[,1],round(100*tab5.1.1.1_SITU_MASC[,c(3,5,6)],1))

SITU, sexo feminino

cbind(tab5.1.1.1_SITU_FEM[,1],round(100*tab5.1.1.1_SITU_FEM[,c(3,5,6)],1))

UF

tab5.1.1.1_UF<-svyby(~saude.b.mb,~V0001,pes.sel.des.pos,svymean)

tab5.1.1.1_UF<-transform(tab5.1.1.1_UF,LI=saude.b.mb-2_se,LS=saude.b.mb+2_se) cbind(tab5.1.1.1_UF$V0001,round(100*tab5.1.1.1_UF[,c(2,4,5)],1))

UF x SEXO

tab5.1.1.1_UF_SEXO<-svyby(~saude.b.mb,~V0001+C006,pes.sel.des.pos,svymean)

tab5.1.1.1_UF_SEXO<-transform(tab5.1.1.1_UF_SEXO,LI=saude.b.mb-2_se,LS=saude.b.mb+2_se)

UF, Sexo masculino

tab5.1.1.1_UF_MASC<-subset(tab5.1.1.1_UF_SEXO,C006==1) cbind(tab5.1.1.1_UF_MASC[,1],round(100*tab5.1.1.1_UF_MASC[,c(3,5,6)],1))

UF, Sexo feminino

tab5.1.1.1_UF_FEM<-subset(tab5.1.1.1_UF_SEXO,C006==2) cbind(tab5.1.1.1_UF_FEM[,1],round(100*tab5.1.1.1_UF_FEM[,c(3,5,6)],1))

#################################################################

examples of household characteristics estimation

#######################################################

household design

dom_des <- subset(pes.all.des.pos,C004=="01" )

householder density for Brasil

svymean(~~C001,dom_des )

householder density by region

svyby(~C001,~Region, dom_des,svymean)

ajdamico commented 8 years ago

dom table now incorporated