Pooled DHS data dictionary

Getting started

First we load the required packages.

easypackages::libraries(
  # Data i/o
  "here",                 # relative file path
  "rio",                  # file import-export
  
  # Data manipulation
  "janitor",              # data cleaning fns
  "haven",                # stata, sas, spss data io
  "labelled",             # var labelling
  "readxl",               # excel sheets
  # "scales",               # to change formats and units
  "skimr",                # quick data summary
  "broom",                # view model results
  
  # Analysis output
  "gt",
  # "modelsummary",          # output summary tables
  "gtsummary",            # output summary tables
  "flextable",            # creating tables from objects
  "officer",              # editing in office docs
  
  # R graph related packages
  "ggstats",
  "RColorBrewer",
  # "scales",
  "patchwork",
  
  # Misc packages
  "tidyverse",            # Data manipulation iron man
  "tictoc"                # Code timing
)

Next we turn off scientific notations.

options(scipen = 999)

Next we set the default gtsummary print engine for tables.

theme_gtsummary_printer(print_engine = "flextable")

Now we set the flextable output defaults.

set_flextable_defaults(
  font.size = 11,
  text.align = "left",
  big.mark = "",
  background.color = "white",
  table.layout = "autofit",
  theme_fun = theme_vanilla
)

Document introduction

We document the variable labels and other contents of all variables across all the pooled Demographic and Health Survey (DHS) birth history datasets of South Asian countries here. This will be primary document for viewing what variables will be used for analysis in PhD research. This document will guide us on the variable availability in the pooled datasets of the various South Asian countries.

This document contains variable details for the following Afghanistan DHS surveys:

# Creating the table of pooled surveys to be used for analysis
sasiadhs_tbl <- tibble(
  ccode = c(1:2),
  country = c("Bangladesh", "Nepal"),
  n_sample = c(nrow(bdbr_pl), nrow(npbr_pl)),
  n_variables = c(ncol(bdbr_pl), ncol(npbr_pl))
)

# Convert and view as flextable
sasiadhs_tbl |> 
  mutate(n_sample = prettyNum(n_sample, big.mark = ",")) |> 
  # convert to flextable object
  qflextable() |> 
  align(align = "left", part = "all")

Table 1: South Asian DHS datasets and their sample size to be used for pooling

ccode	country	n_sample	n_variables
1	Bangladesh	349,962	120
2	Nepal	161,330	124

Here we will create basic data dictionaries of each datasets using the labelled package. In our earlier efforts we realized that the method for generation data dictionaries for each dataset are similar. Hence, we will create a function using the look_for() command to generate the data dictionaries and view them.

Bangladesh DHS pooled data dictionary

Here we check the variable details of the Bangladesh birth history recode (BR) pooled dataset.

svy_name	variable	label	col_type	missing
bdbr	ctr_name	Country name	chr	0
bdbr	ctr_code	Country code	dbl	0
bdbr	year	Survey year	dbl	0
bdbr	uniqueid	Unique birth ID	chr	0
bdbr	commid	Community ID	chr	0
bdbr	hhid	Household ID	chr	0
bdbr	motherid	Mother ID	chr	0
bdbr	caseid	case identification	chr	0
bdbr	bidx	birth column number	dbl	0
bdbr	v000	country code and phase	chr	0
bdbr	v001	cluster number	dbl	0
bdbr	v002	household number	dbl	0
bdbr	v003	respondent's line number	dbl	0
bdbr	v004	ultimate area unit	dbl	0
bdbr	v005	sample weight	dbl	0
bdbr	svy_unf	Un-normalization factor	dbl	0
bdbr	svy_wwgt6	Women's weight un-normalized	dbl	0
bdbr	svy_psu	primary sampling unit	chr	0
bdbr	svy_strata	sample strata	chr	0
bdbr	v006	month of interview	dbl	0
bdbr	v007	year of interview	dbl+lbl	0
bdbr	v008	date of interview (cmc)	dbl	0
bdbr	v009	respondent's month of birth	dbl	0
bdbr	v010	respondent's year of birth	dbl	0
bdbr	v011	date of birth (cmc)	dbl	0
bdbr	v012	current age - respondent	dbl	0
bdbr	v021	primary sampling unit	dbl	0
bdbr	v022	sample strata	dbl+lbl	0
bdbr	v025	type of place of residence	dbl+lbl	0
bdbr	v106	highest educational level	dbl+lbl	0
bdbr	v107	highest year of education	dbl+lbl	124078
bdbr	v151	sex of household head	dbl+lbl	0
bdbr	v152	age of household head	dbl+lbl	13
bdbr	v201	total children ever born	dbl	0
bdbr	v209	births in past year	dbl+lbl	0
bdbr	bord	birth order number	dbl	0
bdbr	b0	child is twin	dbl+lbl	0
bdbr	b1	month of birth	dbl	0
bdbr	b2	year of birth	dbl+lbl	0
bdbr	b3	date of birth (cmc)	dbl	0
bdbr	b4	sex of child	dbl+lbl	0
bdbr	b5	child is alive	dbl+lbl	0
bdbr	b6	age at death	dbl+lbl	312326
bdbr	b7	age at death (months-imputed)	dbl	312306
bdbr	b8	current age of child	dbl	37656
bdbr	b9	child lives with whom	dbl+lbl	37656
bdbr	b10	completeness of information	dbl+lbl	0
bdbr	b11	preceding birth interval	dbl	119918
bdbr	b12	succeeding birth interval	dbl	121849
bdbr	b13	flag for age at death	dbl+lbl	312306
bdbr	str_v001		chr	0
bdbr	str_v002		chr	0
bdbr	str_v003		chr	0
bdbr	str_bord		chr	0
bdbr	str_v021		chr	0
bdbr	str_v022		chr	0
bdbr	cons	constant == 1	dbl	0
bdbr	agecon	Age at conception of child (cmc)	dbl	0
bdbr	agemort	Age at death of child (cmc)	dbl	312306
bdbr	lag_b5	Prev child survival status	dbl+lbl	119918
bdbr	lag_agemort	Prev child age at mortality	dbl	315583
bdbr	sibsurv_all	Death scarring	fct	119918
bdbr	sibsurv_con	Death scarring	fct	119918
bdbr	sibsurv_3c	Death scarring	fct	119918
bdbr	sibsurv_all_nmv	Death scarring	fct	0
bdbr	sibsurv_con_nmv	Death scarring	fct	0
bdbr	sibsurv_3c_nmv	Death scarring	fct	0
bdbr	binterval_3c	Preceding birth interval (in months)	fct	119918
bdbr	binterval_3c_opp	Preceding birth interval (in months)	fct	119918
bdbr	binterval_6c	Preceding birth interval (in months)	fct	119918
bdbr	binterval_6c_opp	Preceding birth interval (in months)	fct	119918
bdbr	binterval_3c_nmv	Preceding birth interval (in months)	fct	0
bdbr	binterval_3c_nmv_opp	Preceding birth interval (in months)	fct	0
bdbr	binterval_6c_nmv	Preceding birth interval (in months)	fct	0
bdbr	binterval_6c_nmv_opp	Preceding birth interval (in months)	fct	0
bdbr	bord_c	Birth order	fct	0
bdbr	sex_mf	Sex of child	fct	0
bdbr	sex_fm	Sex of child	fct	0
bdbr	medu	Mother's education	fct	0
bdbr	medu_opp	Mother's education	fct	0
bdbr	macb	Mother's age at child birth (in years)	dbl	0
bdbr	macb_c	Mother's age at child birth (in years)	fct	0
bdbr	macb_c_opp	Mother's age at child birth (in years)	fct	0
bdbr	marital	Mother's marital status	fct	0
bdbr	pedu	Father's education	fct	0
bdbr	pedu_opp	Father's education	fct	0
bdbr	por	Place of residence	fct	0
bdbr	por_opp	Place of residence	fct	0
bdbr	head_sex_mf	Sex of household head	fct	0
bdbr	head_sex_fm	Sex of household head	fct	0
bdbr	head_age	Age of household head	fct	0
bdbr	head_age_opp	Age of household head	fct	0
bdbr	season	Season during birth	fct	0
bdbr	religion	Religion of the household	fct	26
bdbr	yoi_v007	Year of interview	dbl+lbl	0
bdbr	yob_v010	Mother's year of birth	dbl	0
bdbr	yob_b2	Child's year of birth	dbl	0
bdbr	cyob10y	Birth cohort of index child	fct	0
bdbr	cyob10y_opp	Birth cohort of index child	fct	0
bdbr	cyob5y	Birth cohort of index child	fct	0
bdbr	cyob5y_opp	Birth cohort of index child	fct	0
bdbr	myob10y	Mother's year of birth	fct	0
bdbr	myob10y_opp	Mother's year of birth	fct	0
bdbr	wi_score	Wealth index score	dbl	0
bdbr	wi_qt	Wealth index quintile	fct	0
bdbr	wi_qt_opp	Wealth index quintile	fct	0
bdbr	hhsize	Household size	int	0
bdbr	hhsize_dejure	Num of de jure residents	dbl	0
bdbr	hhsize_defacto	Num of de facto residents	dbl	0
bdbr	hhsize_c	Household size	fct	0
bdbr	hhstruc	Household structure	fct	0
bdbr	hhstruc_opp	Household structure	fct	0
bdbr	hhgen_num	Generations in household	dbl	0
bdbr	hhgen_3c	Generations in household	fct	0
bdbr	hhgen_2c	Generations in household	fct	0
bdbr	hhgen_2c_opp	Generations in household	fct	0
bdbr	infantd_01	Infant mortality	dbl	0
bdbr	infantd	Infant mortality	fct	0
bdbr	under5d_01	Under-five mortality	dbl	0
bdbr	under5d	Under-five mortality	fct	0

Nepal DHS pooled data dictionary

Here we check the variable details of the Nepal birth history recode (BR) pooled dataset.

svy_name	variable	label	col_type	missing
npbr	ctr_name	Country name	chr	0
npbr	ctr_code	Country code	dbl	0
npbr	year	Survey year	dbl	0
npbr	uniqueid	Unique birth ID	chr	0
npbr	commid	Community ID	chr	0
npbr	hhid	Household ID	chr	0
npbr	motherid	Mother ID	chr	0
npbr	caseid	case identification	chr	0
npbr	bidx	birth column number	dbl	0
npbr	v000	country code and phase	chr	0
npbr	v001	cluster number	dbl	0
npbr	v002	household number	dbl	0
npbr	v003	respondent's line number	dbl	0
npbr	v004	ultimate area unit	dbl	0
npbr	v005	sample weight	dbl	0
npbr	svy_unf	Un-normalization factor	dbl	0
npbr	svy_wwgt6	Women's weight un-normalized	dbl	0
npbr	svy_psu	primary sampling unit	chr	0
npbr	svy_strata	sample strata	chr	0
npbr	v006	month of interview	dbl+lbl	0
npbr	v007	year of interview	dbl	0
npbr	v008	date of interview (cmc)	dbl	0
npbr	v009	respondent's month of birth	dbl+lbl	0
npbr	v010	respondent's year of birth	dbl+lbl	0
npbr	v011	date of birth (cmc)	dbl	0
npbr	v012	current age - respondent	dbl	0
npbr	v021	primary sampling unit	dbl	0
npbr	v022	sample strata for sampling errors	dbl+lbl	0
npbr	v025	type of place of residence	dbl+lbl	0
npbr	v106	highest educational level	dbl+lbl	0
npbr	v107	highest year of education	dbl+lbl	92582
npbr	v151	sex of household head	dbl+lbl	0
npbr	v152	age of household head	dbl+lbl	0
npbr	v201	total children ever born	dbl	0
npbr	v209	births in past year	dbl+lbl	0
npbr	bord	birth order number	dbl	0
npbr	b0	child is twin	dbl+lbl	0
npbr	b1	month of birth	dbl+lbl	0
npbr	b2	year of birth	dbl	0
npbr	b3	date of birth (cmc)	dbl	0
npbr	b4	sex of child	dbl+lbl	0
npbr	b5	child is alive	dbl+lbl	0
npbr	b6	age at death	dbl+lbl	142535
npbr	b7	age at death (months-imputed)	dbl	142533
npbr	b8	current age of child	dbl	18797
npbr	b9	who child lives with	dbl+lbl	18797
npbr	b10	completeness of information	dbl+lbl	0
npbr	b11	preceding birth interval	dbl	50836
npbr	b12	succeeding birth interval	dbl	51315
npbr	b13	flag for age at death	dbl+lbl	142533
npbr	str_v001		chr	0
npbr	str_v002		chr	0
npbr	str_v003		chr	0
npbr	str_bord		chr	0
npbr	str_v021		chr	0
npbr	str_v022		chr	0
npbr	cons	constant == 1	dbl	0
npbr	agecon	Age at conception of child (cmc)	dbl	0
npbr	agemort	Age at death of child (cmc)	dbl	142533
npbr	lag_b5	Prev child survival status	dbl+lbl	50836
npbr	lag_agemort	Prev child age at mortality	dbl	144071
npbr	sibsurv_all	Death scarring	fct	50836
npbr	sibsurv_con	Death scarring	fct	50836
npbr	sibsurv_3c	Death scarring	fct	50836
npbr	sibsurv_all_nmv	Death scarring	fct	0
npbr	sibsurv_con_nmv	Death scarring	fct	0
npbr	sibsurv_3c_nmv	Death scarring	fct	0
npbr	binterval_3c	Preceding birth interval (in months)	fct	50836
npbr	binterval_3c_opp	Preceding birth interval (in months)	fct	50836
npbr	binterval_6c	Preceding birth interval (in months)	fct	50836
npbr	binterval_6c_opp	Preceding birth interval (in months)	fct	50836
npbr	binterval_3c_nmv	Preceding birth interval (in months)	fct	0
npbr	binterval_3c_nmv_opp	Preceding birth interval (in months)	fct	0
npbr	binterval_6c_nmv	Preceding birth interval (in months)	fct	0
npbr	binterval_6c_nmv_opp	Preceding birth interval (in months)	fct	0
npbr	bord_c	Birth order	fct	0
npbr	sex_mf	Sex of child	fct	0
npbr	sex_fm	Sex of child	fct	0
npbr	medu	Mother's education	fct	0
npbr	medu_opp	Mother's education	fct	0
npbr	macb	Mother's age at child birth (in years)	dbl	0
npbr	macb_c	Mother's age at child birth (in years)	fct	0
npbr	macb_c_opp	Mother's age at child birth (in years)	fct	0
npbr	marital	Mother's marital status	fct	0
npbr	pedu	Father's education	fct	0
npbr	pedu_opp	Father's education	fct	0
npbr	por	Place of residence	fct	0
npbr	por_opp	Place of residence	fct	0
npbr	head_sex	Sex of household head	fct	0
npbr	head_sex_opp	Sex of household head	fct	0
npbr	head_age	Age of household head	fct	0
npbr	head_age_opp	Age of household head	fct	0
npbr	birth_month	Month of birth	dbl	0
npbr	season	Season during birth	fct	0
npbr	season_wrong	Season during birth	fct	0
npbr	religion	Religion of the household	fct	0
npbr	nat_lang	Native language of the household	fct	0
npbr	yoi_v007	Year of interview	dbl	0
npbr	yob_v010	Mother's year of birth	dbl	0
npbr	yob_b2	Child's year of birth	dbl	0
npbr	cyob10y	Birth cohort of index child	fct	0
npbr	cyob10y_opp	Birth cohort of index child	fct	0
npbr	cyob5y	Birth cohort of index child	fct	0
npbr	cyob5y_opp	Birth cohort of index child	fct	0
npbr	myob10y	Mother's year of birth	fct	0
npbr	myob10y_opp	Mother's year of birth	fct	0
npbr	ecoreg	Ecological region	fct	0
npbr	wi_score	Wealth index score	dbl	0
npbr	wi_qt	Wealth index quintile	fct	0
npbr	wi_qt_opp	Wealth index quintile	fct	0
npbr	hhsize	Household size	int	0
npbr	hhsize_dejure	Num of de jure residents	dbl	0
npbr	hhsize_defacto	Num of de facto residents	dbl	0
npbr	hhsize_c	Household size	fct	0
npbr	hhstruc	Household structure	fct	0
npbr	hhstruc_opp	Household structure	fct	0
npbr	hhgen_num	Generations in household	dbl	0
npbr	hhgen_3c	Generations in household	fct	0
npbr	hhgen_2c	Generations in household	fct	0
npbr	hhgen_2c_opp	Generations in household	fct	0
npbr	infantd_01	Infant mortality	dbl	0
npbr	infantd	Infant mortality	fct	0
npbr	under5d_01	Under-five mortality	dbl	0
npbr	under5d	Under-five mortality	fct	0