The DHS Program User Forum: Kenya

Home » Countries » Kenya » Clustering MIS

Show: Today's Messages :: Show Polls :: Message Navigator

Clustering MIS [message #22720]

Mon, 26 April 2021 17:51

toyosi
Messages: 3
Registered: January 2021

Member

Hi, I am trying to cluster my standard errors but I am getting all insignificant results. I am trying to see the effect maternal education has on malaria. Is there anything you would advice, I am using the Kenyan 2015 MIS. I am using the IR and PR files which I have already merged

I already tried
generate wt = v005/1000000
svyset (pweight=wt), psu (v021) strata (v022)

[Updated on: Tue, 27 April 2021 00:30]

Report message to a moderator

Re: Clustering MIS [message #22728 is a reply to message #22720]

Tue, 27 April 2021 16:06

Bridgette-DHS
Messages: 3208
Registered: February 2013

Senior Member

Following is a response from DHS Research & Data Analysis Director, Tom Pullum:

In general, and in the Kenya 2015 MIS, the PSU (sample cluster) is given by v001 or v021 in the IR file and hv001 or hv021 in the PR file. These two variables can almost always be used interchangeably. What do you mean by "didn't work"? Are you putting together an svyset command? What kind of error are you getting?

Report message to a moderator

Previous Topic:	MIS 2015
Next Topic:	Missing Values for 2014 Kenya DHS variable V302

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Wed Dec 11 18:47:52 Coordinated Universal Time 2024