|
Hi, I apologise in advance for my simple questions ... but I am
new in Structural equation modelling and AMOS. My data set has missing values, and as such I am going down
the path of data imputation. IN this regard I have two questions based on
two different approaches. 1.
Firstly, if I have an imputed data file (created in
SPSS using the missing values module), how do I then use this file in AMOS that
will provide me a pooled estimate using my 20 imputed data sets? Is this
even possible? [Note I have attempted this, specifying through spss the
split file using the imputation variable, but when I run my amos model, it
assumes this file is complete and my sample size is 20 times, instead of the
pooled estimate. ] a. Also
critical, my computer does not have the missing values module (imputation was
performed on another computer), but does recognise the file as an imputed file
– I’ve been told by SPSS that this should not be a problem
(response below). 2.
Otherwise, can anyone provide me a good reference on
how to use the data imputation built in with AMOS v17. I have tried using
this function, but after a long time, it just keeps bombing out. Any guidance and recommended further reading would be
appreciated. Regards, Paola. Hi
Paola, Once
you have imputed values using the Missing Values module, if you then save your
data file, you will be able to continue your analysis and SEM using your Grad
Pack of SPSS 17 and AMOS on that data file with the imputed values. It is
highly recommended that you save the imputed values in a new file in case you
do need to refer back to the original data file. I
hope this helps. Thanks,
Date: Tue, 19 May
2009 16:34:11 +0800 HI, Regards, Paola “Ours has become a time-poor society, fatigued by
non-physical demands and trying to compartmentalize daily living tasks.
It is small wonder that physical activity is discarded in this
environment” p126 (Steinbeck, 2001) P Please consider the
environment before printing this email. |
| Free forum by Nabble | Edit this page |
