The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Data » Weighting data » Pooled Cross sections
Re: Pooled Cross sections [message #10080 is a reply to message #10008] Thu, 23 June 2016 10:22 Go to previous messageGo to previous message
cbdolan is currently offline  cbdolan
Messages: 17
Registered: March 2013
Location: Williamsburg, VA
Member
Thanks for the detailed follow up and paper link. Both were helpful.

I think I have something wrong with the way I constructed the pooled weights based on the results of my descriptives. The N's shouldn't be this dissimilar. (Note: I previously merged the files with the spatial files so I'm using ADM1_CODE as the province level variable).

I annotated the code to clarify the steps. Please let me know if I've missed a step.


 tab ADM1_CODE[iweight=NW]

       ADM1_CODE |      Freq.     Percent        Cum.
-----------------+-----------------------------------
        Bandundu |329,552,409       16.16       16.16
       Bas-Congo | 90101292.7        4.42       20.58
        Equateur |264,708,100       12.98       33.56
Kasai-Occidental |173,724,497        8.52       42.07
  Kasai-Oriental |231,513,280       11.35       53.42
         Katanga |215,915,193       10.59       64.01
        Kinshasa |182,595,002        8.95       72.96
         Maniema | 72641458.3        3.56       76.53
       Nord-Kivu |135,628,336        6.65       83.18
       Orientale |189,207,489        9.28       92.45
        Sud-Kivu |153,913,256        7.55      100.00
-----------------+-----------------------------------
           Total | 2.0395e+09      100.00

. tab ADM1_CODE

       ADM1_CODE |      Freq.     Percent        Cum.
-----------------+-----------------------------------
        Bandundu |    272,736       12.63       12.63
       Bas-Congo |    118,666        5.49       18.12
        Equateur |    300,174       13.90       32.02
Kasai-Occidental |    195,252        9.04       41.06
  Kasai-Oriental |    234,033       10.84       51.90
         Katanga |    259,660       12.02       63.92
        Kinshasa |    156,412        7.24       71.17
         Maniema |    132,302        6.13       77.29
       Nord-Kivu |    140,773        6.52       83.81
       Orientale |    203,672        9.43       93.24
        Sud-Kivu |    145,920        6.76      100.00
-----------------+-----------------------------------
           Total |  2,159,600      100.00



I did the following to set up the pooled weights:

use "Y:\4_DHS_BirthRecode\CDBR61FL.dta"
*Original weight in DHS : v005 (which should preferably be divided by 1000000)
generate n_v005=(v005/1000000)
*note this is the population of 15-49 in DRC (2013) from United Nations, Department of Economic and Social Affairs, Population Division (2015). World Population Prospects: The 2015 Revision, custom data acquired via website.
generate P1549=16167000
*note this is the sample size from the individual recode file of women 15-49 interviewed
generate n1549=18827
*Country specific weight :CSW= P1549/n1549 (population aged 15-49 in the country / sample size of )
generate CSW=(P1549/n1549)
*New weight
generate NW=n_v005*CSW 
file Y:\4_DHS_BirthRecode\n_CDBR61FL.dta saved
clear

use "Y:\4_DHS_BirthRecode\CDBR50FL.dta"
*Original weight in DHS : v005 (which should preferably be divided by 1000000)
generate n_v005=(v005/1000000)
*note this is the population of 15-49 in DRC (2013) from United Nations, Department of Economic and Social Affairs, Population Division (2015). World Population Prospects: The 2015 Revision, custom data acquired via website.
generate P1549=13201000
*note this is the sample size from the individual recode file of women 15-49 interviewed
generate n1549=9995
*Country specific weight :CSW= P1549/n1549 (population aged 15-49 in the country / sample size of )
generate CSW=(P1549/n1549)
*New weight
generate NW=n_v005*CSW 
file "Y:\4_DHS_BirthRecode\n_CDBR50FL.dta saved
clear

use "Y:\4_DHS_BirthRecode\n_CDBR61FL.dta"
append using "Y:\4_DHS_BirthRecode\n_CDBR50FL.dta"


*generate weight: see code at top
*make unique strata values by region/urban-rural )
egen stratum=group(ADM1_CODE v025)
*tell stata the weight (using pweights for robust standard errors, cluster (psu), and strata 
svyset [pw=NW],psu(v021)strata(stratum)
*prefix regrss with "svy:stata will now know how to weight your data and compute the right standard errors */

 
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Previous Topic: Antenatal visit with skilled providers (Ethiopian DHS 2000, 2005, 2011 AND 2016))
Next Topic: Using R to weight data for domestic violence
Goto Forum:
  


Current Time: Fri Apr 26 20:45:10 Coordinated Universal Time 2024