Hi,
I'm looking into using the 'svy' commands and am seeking some advice on their use on survey data when the design information is not known.
Say the survey was a stratified, multi-stage design, but the data does not contain information on the stratification or clustering, but does include weights and (jackknife) replicate weights. Can svy still be used, and would it be statistically valid to do so?
Thanks in advance for any help.
Cheers,
Melanie Black
Data Access and Confidentiality Methodology Unit
Australian Bureau of Statistics
ph. +61 (02) 6252 7241
------------------------------------------------------------------------------------------------
Free publications and statistics available on www.abs.gov.au
