The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Topics » General » append and merge data of 40 data set (append and merge HR MR & IR of sub-saharan countries)
append and merge data of 40 data set [message #22905] Wed, 02 June 2021 04:14 Go to next message
JaneQuan is currently offline  JaneQuan
Messages: 11
Registered: June 2021
Member
Hi! It's Jane.

I am working on Sub-saharan data of 40 datasets(some countries have one-year datasets, and others have more than one-year dataset). and I need:
a) HV244, HV245, AGLANDAMT2, AGLANDWHO, PLOTSOWNO variables from HR;
b) MV714 (CURRWORKMN), MV717 (WKCURRJOBMN), MV745B (MANOWNLAND), MV745D (MANLANDDEED), MANLANDSELL, MV739 (DECMANEARN), MV743A (DECMANHCARE), MV743B (DECBIGHHMN), MV744A (DVAGOOUTMN) from MR;
c) V717 (WKCURRJOB), V714 (CURRWORK),V745A (FEMOWNHOUSE), V745B (FEMOWNLAND),V745D (FEMLANDDEED),FEMLANDSELL (FEMLANDSELL),INHERITLANDHOUSE, V739 (DECFEMEARN), V746 (WKEARNSMORE),V743F (DECHUSEARN), V743A (DECFEMHCARE),V743B (DECBIGHH),V743D (DECFAMVISIT),V744C (DVAARGUE) from IR.

I have downloaded all the STATA data, and I am going to pool them together. my steps will firstly append all the HR, MR, and IR data respectively, and then merge MR and IR onto the HR dataset.
but I found observations among HR, IR, and MR are all different within the same country. how should I merge them? --the household ID and caseid are slightly different since the caseid includes both household id and respondent id.

by the way, all other data are collected from the Standard DHS survey, but Senegal's from 2014-2016 is the continuous DHS Survey. and Senegal Continuous DHS in 2014 and 2016 has two datasets in HR, MR, and IR respectively, such as
Senegal: 2014 "SNHR6RFL/SNHR70FL" "SNIR6RFL/SNIR70FL" "SNMR6RFL/SNMR70FL";
Senegal: 2016 "SNHR7IFL/SNHRG0FL" "SNIR7IFL/SNIRG0FL" "SNMR7IFL/SNMRG0FL"
Which one should I use?

Any hint will be appreciated. thanks!

[Updated on: Wed, 02 June 2021 04:17]

Report message to a moderator

Re: append and merge data of 40 data set [message #22908 is a reply to message #22905] Wed, 02 June 2021 09:58 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3214
Registered: February 2013
Senior Member
The SN70 files include just the 2014 round. The SN6R files include the combined 2012-13 and 2014 rounds. You should use SN6R for estimates for subnational regions. Men were included in the 2014 round but not the 2012-13 round. The SN71 files include just the 2016 round. SNG0 includes the combined 2015 and 2016 rounds.
Re: append and merge data of 40 data set [message #22910 is a reply to message #22908] Wed, 02 June 2021 11:37 Go to previous messageGo to next message
JaneQuan is currently offline  JaneQuan
Messages: 11
Registered: June 2021
Member
Thank you, Bridgette.

I have three more questions:
1) I saw some survey years of DHS data are 2015-2016, and some are 2015, what's the difference? 2015-2016 means two-year data set while 2015 means only a one-year dataset?
2) you said "SNG0 includes the combined 2015 and 2016 rounds while SN71 files include just the 2016 round", but there is a file for the year 2015, which is SN7H, so SNG0=SN71+SN7H, right?
3) "Men were included in the 2014 round but not the 2012-13 round"- this refers to the MR file, right?

Thanks:)
Re: append and merge data of 40 data set [message #22911 is a reply to message #22910] Wed, 02 June 2021 11:51 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3214
Registered: February 2013
Senior Member
1) I saw some survey years of DHS data are 2015-2016, and some are 2015, what's the difference? 2015-2016 means two-year data set while 2015 means only a one-year dataset?
Which DHS surveys are you referring to? Generally speaking, if a survey is titled 2015-16, it indicates that the year of fieldwork spanned both years.

2) you said "SNG0 includes the combined 2015 and 2016 rounds while SN71 files include just the 2016 round", but there is a file for the year 2015, which is SN7H, so SNG0=SN71+SN7H, right?
SNG0 contains the data for both the 2015 and 2016 rounds.

3) "Men were included in the 2014 round but not the 2012-13 round"- this refers to the MR file, right?
Men refers to what you will find in the MR data.
Re: append and merge data of 40 data set [message #22913 is a reply to message #22911] Wed, 02 June 2021 21:13 Go to previous message
JaneQuan is currently offline  JaneQuan
Messages: 11
Registered: June 2021
Member
Hi Bridgette,

I really appreciated your explanations on Senegal's data, They have helped me tremendously.

Thank you.
Previous Topic: Calculation of child marriage rates and v511 from PR dataset
Next Topic: pooling countries to run fixed effect
Goto Forum:
  


Current Time: Sun Dec 22 16:47:54 Coordinated Universal Time 2024