I am trying to run PSMatching on a multiply imputed dataset in SPSS 25 using PSMatching3.04. I was previously able to run PSMatching on an MI dataset in SPSS 20 with R2.12. Since the switch to SPSS 25, it will only run PSMatching on the first imputation, or brings over some iterations of the matched cases but not all. Is there any way to run this in SPSS 25 while keeping all iterations of the matched cases?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Based on this, and other comments that we have received, it seems that
SPSS 25 may have broken some features.
We will need to investigate this first, and issue a patch. At the
moment, the best bet would be to try to run this with an older version
of SPSS, if at all possible.
A cumbersome alternative would be to run the imputed datasets
manually, and then record, and combine the estimates by hand -
tedious, but possible.
I am trying to run PSMatching on a multiply imputed dataset in SPSS 25 using
PSMatching3.04. I was previously able to run PSMatching on an MI dataset in
SPSS 20 with R2.12. Since the switch to SPSS 25, it will only run PSMatching
on the first imputation, or brings over some iterations of the matched cases
but not all. Is there any way to run this in SPSS 25 while keeping all
iterations of the matched cases?
I am trying to run PSMatching on a multiply imputed dataset in SPSS 25 using PSMatching3.04. I was previously able to run PSMatching on an MI dataset in SPSS 20 with R2.12. Since the switch to SPSS 25, it will only run PSMatching on the first imputation, or brings over some iterations of the matched cases but not all. Is there any way to run this in SPSS 25 while keeping all iterations of the matched cases?
Based on this, and other comments that we have received, it seems that
SPSS 25 may have broken some features.
We will need to investigate this first, and issue a patch. At the
moment, the best bet would be to try to run this with an older version
of SPSS, if at all possible.
A cumbersome alternative would be to run the imputed datasets
manually, and then record, and combine the estimates by hand -
tedious, but possible.
On Tue, Feb 6, 2018 at 4:18 PM, Kari Roberts
klroberts718@users.sourceforge.net wrote:
Hi there - How would one do this by hand?