The DHS Program User Forum      
Discussions regarding The DHS Program data and results
Home » Topics » Reproductive Health » ANC coverage for BDHS-2011
ANC coverage for BDHS-2011 [message #11893] Sun, 26 February 2017 13:22 Go to next message
Mamun is currently offline  Mamun
Messages: 2
Registered: February 2017
Location: Bangladesh
Member
I tried to estimate anc coverage for BDHS-2011 given in the table-9.3 page-126 using IR file. I used the following stata code

recode m14_1 (0=0 "none") (1=1) (2=2 "2") (3=3 "3") (4/20=3 "4+") (98=4 "dk"), gen(anc)
ta anc [iw=v005/1000000]

But I could not match the values. I also used the following codes using KR file

recode m14 (0=0 "none") (1=1) (2=2 "2") (3=3 "3") (4/20=3 "4+") (98=4 "dk"), gen(anc)
ta anc [iw=v005/1000000]

can you help to match these results and give an explanation

Thanks in advance
Re: ANC coverage for BDHS-2011 [message #11906 is a reply to message #11893] Tue, 28 February 2017 09:50 Go to previous messageGo to next message
Mlue
Messages: 26
Registered: February 2017
Location: Cape Town
Member
Please use code below, and use the births recode dataset...

Notice that there are some slight differences in the proportions with the DHS reports, but I don't believe that this matters...

*********


*** === BANGLADESH 2011 _ ANTENATAL VISITS === ***

*==============================================================================*

// USE THE BIRTHS RECODE - BDHS 2011

********************************************************************************
**

** WEIGHT VARIABLE
gen weight = v005/1000000

********************************************************************************

** SURVEY SET
gen psu =    v021
gen strata = v022
svyset psu [pw = weight], strata(strata)

********************************************************************************

** BIRTHS IN THE THREE YEARS PRECEDING THE SURVEY

keep if v238 > 0 & v238 !=. 

********************************************************************************

/*
	You were close to the results in the DHS report, but you did not limit
	your analysis to the three years preceding the survey, as in the report, 
	and you had made some mistakes with your recode [you had coded both these as "=4" (4/20=4 "4+") (98/99=5 "dk")]...
	
	The recode below should give you the exact figures as in the report

*/

recode m14 (0=0 "none") (1=1 "1") (2=2 "2") (3=3 "3") ///
(4/20=4 "4+") (98/99=5 "dk"), gen(anc)
ta anc v025 [iw=v005/1000000]
ta anc v025 [iw=v005/1000000], col nofreq

*** SURVEY TAB
/* This will give you the same results as in the report */

svy: tab anc v025, count format(%4.0f)
svy: tab anc v025, percent format(%4.1f) col

**============================================================================**

[Updated on: Tue, 28 February 2017 09:56]

Report message to a moderator

Re: ANC coverage for BDHS-2011 [message #11909 is a reply to message #11906] Tue, 28 February 2017 13:23 Go to previous messageGo to next message
Mamun is currently offline  Mamun
Messages: 2
Registered: February 2017
Location: Bangladesh
Member
Thank Dubile very much.
Mamun
Re: ANC coverage for BDHS-2011 [message #12130 is a reply to message #11906] Thu, 30 March 2017 21:30 Go to previous messageGo to next message
dhs110 is currently offline  dhs110
Messages: 30
Registered: October 2014
Location: korea
Member
dear Dubile
I want to re-code the variable M14_1 from Pakistan data set...PDHS 2013-13
but i don't want to replicate the table i want only two (dichotomous variable) like this:
0=no visit
1=at least 4 visit
Regards,
dhs110


dhs110
Re: ANC coverage for BDHS-2011 [message #12144 is a reply to message #12130] Sat, 01 April 2017 11:09 Go to previous messageGo to next message
Mlue
Messages: 26
Registered: February 2017
Location: Cape Town
Member
/* What about those who had one to three antenatal visits?

I would keep those with less than four antenatal visits, and have 0-3 = "0", and 4-20="1" when using the example above... Please remember that the maximum number of visits is not always "20", hence you need to check the variable for your given country before deriving an indicator out of it... */

** Example: On Stata

recode m14 (0/3=0 "None or <4") (4/20=1 "4+") (98/99=5 "dk"), gen(anc)
ta anc v025 [iw=v005/1000000]
ta anc v025 [iw=v005/1000000], col nofreq

** If you're not interested in do not know or missing (98/99):

recode m14 (0/3=0 "None or <4") (4/20=1 "4+") (98/99=.), gen(anc)
ta anc v025 [iw=v005/1000000]
ta anc v025 [iw=v005/1000000], col nofreq
//now your universe (i.e. total) will be less than that in the report because you've excluded missing/do not know

[Updated on: Sat, 01 April 2017 17:04]

Report message to a moderator

Re: ANC coverage for BDHS-2011 [message #12146 is a reply to message #12144] Sun, 02 April 2017 05:24 Go to previous message
dhs110 is currently offline  dhs110
Messages: 30
Registered: October 2014
Location: korea
Member
thank you.stay blessed

dhs110
Previous Topic: sas commands for delivery assistance by health professional
Next Topic: Replicating Table 9.7: Timing of first postnatal checkup for the mother
Goto Forum:
  


Current Time: Mon Dec 11 06:09:38 Eastern Standard Time 2017