News from the SPSS Community

[Jon K Peck]

There is now a forum for IBM SPSS Data Collection on the SPSS Community site making a total of 12 forums.

The SPSSINC RAKE extension command has been updated to provide better treatment of input weights.  The final raked weight now varies within each cell in proportion to the input weight while still matching the marginal totals.

Jon Peck (no "h") aka Kim
Senior Software Engineer, IBM
[hidden email]
new phone: [phone]

[Robert Walker]

This is a great extension. Question: is there an option to display the effective sample size (i.e., before vs. after weighting)? After weighting, the effective sample size should be less, since we would be weighting “down” in every case, rather than making smaller groups represent more than they otherwise should.

 

Regards,

 

Bob Walker

Surveys & Forecasts, LLC

www.safllc.com

 

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Jon K Peck
Sent: Monday, April 23, 2012 5:55 PM
To: [hidden email]
Subject: News from the SPSS Community

 

There is now a forum for IBM SPSS Data Collection on the SPSS Community site making a total of 12 forums.

The SPSSINC RAKE extension command has been updated to provide better treatment of input weights.  The final raked weight now varies within each cell in proportion to the input weight while still matching the marginal totals.

Jon Peck (no "h") aka Kim
Senior Software Engineer, IBM
[hidden email]
new phone: [phone]

[Jon K Peck]

You can see in the output a table with the sample efficiency.  That should give you the equivalent information, i.e., 100% efficiency would mean that the effective size equals the actual size.

Jon Peck (no "h") aka Kim
Senior Software Engineer, IBM
[hidden email]
new phone: [phone]




From:        Bob Walker <[hidden email]>
To:        Jon K Peck/Chicago/IBM@IBMUS, "[hidden email]" <[hidden email]>
Date:        05/23/2012 01:54 PM
Subject:        RE: News from the SPSS Community

---

This is a great extension. Question: is there an option to display the effective sample size (i.e., before vs. after weighting)? After weighting, the effective sample size should be less, since we would be weighting “down” in every case, rather than making smaller groups represent more than they otherwise should.
 
Regards,
 
Bob Walker
Surveys & Forecasts, LLC
www.safllc.com
 
From: SPSSX(r) Discussion [[hidden email]] On Behalf Of Jon K Peck
Sent: Monday, April 23, 2012 5:55 PM
To: [hidden email]
Subject: News from the SPSS Community
 
There is now a forum for IBM SPSS Data Collection on the SPSS Community site making a total of 12 forums.

The SPSSINC RAKE extension command has been updated to provide better treatment of input weights.  The final raked weight now varies within each cell in proportion to the input weight while still matching the marginal totals.

Jon Peck (no "h") aka Kim
Senior Software Engineer, IBM
peck@...
new phone: [phone]