MULTIPLE IMPUTATION ERROR MESSAGE

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

MULTIPLE IMPUTATION ERROR MESSAGE

Harry Kullmann
Dear list,

when using multiple imputation with SPSS 17 (17.0.0., August 2008) I get
the following message:

"Some missing values cannot be imputed because a factor in the model has a
value that does not appear in the date used to build the model."

Despite this message, however, the data set seems to be completely imputed.
At least I cannot find any missing data in the five imputed data sets
generated.

Still, I am afraid there is something wrong with my imputation results.
Can anybody point me to what to think of the error message given above?

The spss-syntax reads as follows:

DATASET DECLARE impu.
DATASET DECLARE impu_iterationprotocol.
MULTIPLE IMPUTATION [varlist]
  /IMPUTE METHOD=FCS
   MAXITER= 10
   NIMPUTATIONS=5
   SCALEMODEL=PMM
   INTERACTIONS=TWOWAY SINGULAR=1E-012
    MAXPCTMISSING=NONE MAXMODELPARAM=800
  /MISSINGSUMMARIES NONE
  /IMPUTATIONSUMMARIES MODELS DESCRIPTIVES
  /OUTFILE IMPUTATIONS=impu
   FCSITERATIONS=impu_iterationsprotokol.

Thanking you very much in advance

Harry

=====================
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
Reply | Threaded
Open this post in threaded view
|

Re: MULTIPLE IMPUTATION ERROR MESSAGE

Joost van Ginkel
Dear Harry,

I get the same message sometimes. To my experience this only happens
when not all variables present in the data take part in the imputation
model. Maybe (but I don't know if I'm right) PASW sees the resulting
imputed data file with some scores still being missing, and erroneously
concludes that something must have gone wrong since not all missing
values have been imputed. That's at least what I can think of. Does
anyone have any other ideas?

Joost van Ginkel


Joost R. van Ginkel, PhD
Leiden University
Faculty of Social and Behavioural Sciences
PO Box 9555
2300 RB Leiden
The Netherlands
Tel: +31-(0)71-527 3620
Fax: +31-(0)71-527 3945


-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Harry Kullmann
Sent: Friday, January 21, 2011 11:48 AM
To: [hidden email]
Subject: MULTIPLE IMPUTATION ERROR MESSAGE

Dear list,

when using multiple imputation with SPSS 17 (17.0.0., August 2008) I get
the following message:

"Some missing values cannot be imputed because a factor in the model has
a value that does not appear in the date used to build the model."

Despite this message, however, the data set seems to be completely
imputed.
At least I cannot find any missing data in the five imputed data sets
generated.

Still, I am afraid there is something wrong with my imputation results.
Can anybody point me to what to think of the error message given above?

The spss-syntax reads as follows:

DATASET DECLARE impu.
DATASET DECLARE impu_iterationprotocol.
MULTIPLE IMPUTATION [varlist]
  /IMPUTE METHOD=FCS
   MAXITER= 10
   NIMPUTATIONS=5
   SCALEMODEL=PMM
   INTERACTIONS=TWOWAY SINGULAR=1E-012
    MAXPCTMISSING=NONE MAXMODELPARAM=800
  /MISSINGSUMMARIES NONE
  /IMPUTATIONSUMMARIES MODELS DESCRIPTIVES
  /OUTFILE IMPUTATIONS=impu
   FCSITERATIONS=impu_iterationsprotokol.

Thanking you very much in advance

Harry

=====================
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

**********************************************************************
This email and any files transmitted with it are confidential and
intended solely for the use of the individual or entity to whom they
are addressed. If you have received this email in error please notify
the system manager.
**********************************************************************

=====================
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