It sounds like it was a 2-stage sampling plan. If you want to specify unequal WOR, you will need the joint inclusion probabilities of the units selected in the first stage.
Alex
-----Original Message-----
From: SPSSX(r) Discussion [mailto:
[hidden email]] On Behalf Of LCF
Sent: Monday, November 05, 2007 3:02 PM
To:
[hidden email]
Subject: How to design an unequal WOR plan in SPSS?
Hi there:
I have a question about the statistical analysis of complex samples. I want
to analyze data from NLAES used a complex sampling design by SPSS. I have
had the final weight, inc_prob1 and inc_prob2. But I don't know how to put
the variables into the syntax by the method of unequal WOR (unequal
probability sampling without replacement). Besides, the total population in
the USA is more than 200,000,000 and the sample size in the data is about
42,000. Is it reasonable that analyze the data by the method of WR (sampling
with replacement)?I
Thanks for any suggestions,
LCF
=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD
=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD
Locked
