topic Weight statement - need confirmation of understanding in Statistical Procedures

Weight statement - need confirmation of understanding

HyunJee — Wed, 14 Sep 2011 13:38:58 GMT

I am looking through a previous analyst's code and they frequently used proc surveyfreq and proc surveylogistic with the weights option. I have not done this in some time and wanted to make sure my understanding of the weight statement was accurate.

When you specify the weight statement and variable, instead of SAS recognizing each variable with a default weight of 1, it uses the numeric value in the weight variable. When I looked at the data I am analyzing, weights ranged from 13.1 up to 51000 and about 12000 of the 30000 records are missing weights (FYI: I am using a dataset that the CDC provides).

So to my understanding, a record with a weight of 51000 compared to a record with a weight of 30000 will have more impact on the sample and statistical output.

I know this may seem an elementary question, but it has been a few years and I want to make sure I fully grasp the concept. Thank you for any help you can provide!

Weight statement - need confirmation of understanding

art297 — Wed, 14 Sep 2011 13:50:32 GMT

That is what the documentation states. Take a look at:

http://support.sas.com/documentation/cdl/en/statug/63347/HTML/default/viewer.htm#statug_surveyfreq_sect010.htm

Weight statement - need confirmation of understanding

DLing — Wed, 14 Sep 2011 13:50:34 GMT

Yes, you're right, the more weighty observations carries more, ahem, weight, in the analysis Weighted means, weighted variances, weighted parameter estimates, and so on...

Details on how its used in each PROC is in the PROC documentation.

Weight statement - need confirmation of understanding

HyunJee — Wed, 14 Sep 2011 15:13:37 GMT

thank you art297 and DLing! Just what I needed to know