The DHS Program User Forum      
Discussions regarding The DHS Program data and results
Home » Countries » Tanzania » Svy command
Svy command [message #4369] Fri, 15 May 2015 04:09 Go to next message
erimgina@yahoo.com is currently offline  erimgina@yahoo.com
Messages: 3
Registered: April 2015
Location: China
Member
Dear all
I am planning to use DHS 2010 data to investigate the relationship between family planning use and increased weight among women using modern methods of family planning, I getting difficult to use svy command in stata, I have read the user mannual and written the command as follows

gen psu= v021
gen wght=v005/1000000
egen strata = group(v022 v023), label
svyset psu [pweight=wght], strata(strata)
svy: mean age

The results that I get is missing some information as shown in the table below, the same problem occur when I do regression analysis
svy: mean age
(running mean on estimation sample)

Survey: Mean estimation

Number of strata = 51 Number of obs = 10139
Number of PSUs = 475 Population size = 10139
Design df = 424

------------------------------------------------------------ --
| Linearized
| Mean Std. Err. [95% Conf. Interval]
-------------+---------------------------------------------- --
age | 28.57465 . . .
------------------------------------------------------------ --
Note: missing standard error because of stratum with single
sampling unit.

Please any help!
Re: Svy command [message #4379 is a reply to message #4369] Sat, 16 May 2015 17:09 Go to previous messageGo to next message
Reduced-For(u)m
Messages: 279
Registered: March 2013
Senior Member


There are a couple of ways to deal with that. Here is Trevor's comment from a previous thread:

http://userforum.dhsprogram.com/index.php?t=tree&th=1807 &goto=3738&S=343ce2e112b3f15f2b333eec712dc2cf#msg_37 38

"3) While using the singleunit(centered) parameter is one way around the problem of a stratum with a single unit, in general we recommend to regroup the strata with a similar strata and not use the singleunit parameter.... The difference in the results between the two approaches however will be tiny."
Re: Svy command [message #4380 is a reply to message #4379] Sun, 17 May 2015 03:33 Go to previous message
erimgina@yahoo.com is currently offline  erimgina@yahoo.com
Messages: 3
Registered: April 2015
Location: China
Member
Dear All
Thank you for quick response, and it is very helpful

Erick
Previous Topic: Variable for children education/ enrollment ratio/grade attainment
Next Topic: Child Mortality
Goto Forum:
  


Current Time: Sun Jan 21 13:27:42 Eastern Standard Time 2018