The DHS Program User Forum: Mortality » Lack of clustering variable

Home » Topics » Mortality » Lack of clustering variable

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

Lack of clustering variable [message #15698]

Sat, 01 September 2018 08:04

zealma
Messages: 2
Registered: September 2018
Location: United States

Member

Hello:

I am using the DHS survey for the Palestinian Territories (2004). The cluster variable has been removed due to "identifiability" concerns by the gov. How can I account for this in my results. Would a multi-level model work (one that includes area nested within province)? What would I do with the weights?

To not, I am using cross sections over time (2004,2006,2010,2014).

Report message to a moderator

Re: Lack of clustering variable [message #15753 is a reply to message #15698]

Mon, 10 September 2018 15:06

Liz-DHS
Messages: 1516
Registered: February 2013

Senior Member

A response from Senior Sampler, Dr. Ruilin Ren:
Quote:

we do not have extra information outside the survey data file. The current data structure in DHS may not support multilevel modeling if PSU weight is required.

Report message to a moderator

Previous Topic:	Perinatal Mortality (still birth) calculation SPSS syntax
Next Topic:	SAS MACRO COMMAND FOR DHS_U5M

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Thu Oct 31 21:30:30 Coordinated Universal Time 2024