Weights for Individual Recode SAS analysis [message #11928] |
Tue, 07 March 2017 15:08 ![Go to next message Go to next message](/theme/default/images/down.png) |
kdiclem1
Messages: 8 Registered: March 2017 Location: New Haven, CT
|
Member |
|
|
I am doing an analysis that looks at reproductive health indicators and hunger score variables from the household data set. I am using Tanzania's DHS-VII 2015-16 Data. I have merged them, and now for the my analysis in SAS, I wanted to confirm that for my proc surveys, I use:
weight --> (V005 Women's individual sample weight (6 decimals) / 1,000,000)
cluster --> V021 Primary sampling unit
strata --> V023 Stratification used in sample design
I also want to ensure that by doing this I am properly weighting all "ever married women" as my sample.
Thank you!
[Updated on: Tue, 07 March 2017 17:52] Report message to a moderator
|
|
|
Re: Weights for Individual Recode SAS analysis [message #11933 is a reply to message #11928] |
Wed, 08 March 2017 18:41 ![Go to previous message Go to previous message](/theme/default/images/up.png) |
Trevor-DHS
Messages: 802 Registered: January 2013
|
Senior Member |
|
|
If your base dataset is the women's recode file then yes, you have the weight, cluster and strata correctly given. If your base dataset is the household members file then it might be the household equivalent of these (hv005, hv021 and hv023).
You mention all "ever married women". Unless you restrict the dataset then your analysis will be for all women. If you want just ever married women then you would need to restrict your analysis to ever married women using v502 > 0.
|
|
|