Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: To:

From   Sara Anderson <>
To   undisclosed-recipients:;
Subject   st: To:
Date   Tue, 21 Feb 2012 22:48:34 -0500

I am attempting to merge to imputed datasets each with the same number of
imputations and participants. When I attempt the merge, I get this error:
variable id does not uniquely identify observations in the using data

Using these commands:
use impute_diss, clear
mi merge 1:1 id using pscore_only_mi

I have tried re-creating the files, confirming number of id's etc. The only
reason I can think for the error is that in one imputation, all
participants have values of the variables for all imputations. For the
other (impute_diss), not all participants have an id because some
participants had complete data so no imputations were not necessary. But I
somehow thought that the mi merge command would figure that out.

Is the only way out to export the imputed files, recombine not as mi, then
to merge by the id and _mi_m? This seems cumbersome, but I'm about to try
it out.

*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index