The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Topics » Service Provision Assessment (SPA) » 2019 Senegal SPA dataset merging issue
2019 Senegal SPA dataset merging issue [message #27824] Fri, 06 October 2023 15:38 Go to next message
kgeorg7 is currently offline  kgeorg7
Messages: 8
Registered: August 2022
Member
Hi there,

In an attempt to obtain the 'urban/rural variable' at the facility level, for the 2019-Senegal-SPA-Provider dataset, I tried merging it with the 2019-Senegal-SPA-Facility dataset, using the facility identifier variable (facil). The 'facil' variable uniquely identifies facilities across the datasets, allowing to merge them. However, all the rural observations have missing values for the 2019-Senegal-SPA-Provider variables. Thus, it appears that only urban facility providers were included in the 2019-Senegal-SPA-Provider dataset, and I was wondering if this was by design? This is particularly confusing given that within the 2012-Senegal-SPA-Provider-Recode dataset, both, urban and rural observations are present.

I'm using a simple Stata code, as seen below:
use "SPA\2019\SNFP8ADTSP\SNFP8AFLSP.dta", clear
merge m:1 facil using "SPA\2019\SNFC8ADTSP\SNFC8AFLSP.dta"

Any help would be much appreciated!

Thank you in advance,
Kevin
Re: 2019 Senegal SPA dataset merging issue [message #27840 is a reply to message #27824] Mon, 09 October 2023 16:51 Go to previous message
SaraDHS is currently offline  SaraDHS
Messages: 44
Registered: December 2020
Member
Hi Kevin,
I think the main issue is you are using the wrong file. The provider dataset is not SNFP8AFLSP.dta, this is the Family Planning client dataset. The provider dataset is SNPV8AFLSP.dta. If you merge the facility and provider dataset, there are providers who work in rural areas.

If you are actually interested in the Family Planning client information, the reason there is all missing data for those rural facilities is that they are from cases de sante, which only had a facility inventory and provider interview, and did not include FP client observation/exits. If you look in the FP client dataset, even urban cases de sante are missing client level data.

Best,
Sara


Sara Riese, PhD Senior Demographic and Health Researcher, DHS Program
Previous Topic: KEN SPA 2004-2005 RAW Variables
Next Topic: How to create Facility Recode dataset from Facility Raw for BDHS 2017 SPA
Goto Forum:
  


Current Time: Sun Apr 28 11:39:04 Coordinated Universal Time 2024