The DHS Program User Forum: Weighting data

Home » Data » Weighting data » Sampling

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

Sampling [message #30341]

Sat, 09 November 2024 10:39

Hejie Wang
Messages: 14
Registered: July 2024

Member

I have combined many surveys from many countries into a file, I want to create a complex sample object in R using svydesign function, what should I set the PSU and the stratification variable to respectively?

In addition, I have observed that different surveys in the same country, under the same PSU (v021), sometimes seem to come from different regions (v024 and v025), such as BDKR7RFL and BDKR41FL. Why is this? Is it because the cluster number (same country) is independent each time it is surveyed? Or is it because BDKR41FL is older?

I want to explore the effects of air pollution on children's anemia through the mixed effects model. Is it correct that I take variables v000 and v021 as random terms? In addition, I want to know whether I should set my random terms as random intercept terms or random coefficient terms or adjust them at the same time?

Report message to a moderator

[Message index]

		Sampling By: Hejie Wang on Sat, 09 November 2024 10:39
		Re: Sampling By: Bridgette-DHS on Mon, 11 November 2024 14:07

Previous Topic:	Multilevel weights level 2
Next Topic:	weighting data

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Wed Nov 27 03:42:42 Coordinated Universal Time 2024