>> SPM does not provide SEM modelling so I can't help you with any software
>> tips here other than to say try the student edition of LISREL (but I've
>> never used it !).
>>
>> Best wishes,
>>
>> Will.
SPM does not provide SEM, but Christian Buechel wrote a nice SEM
'toolbox' which has been used in SPM-based studies over the last 10
years. I would be happy to send you a copy. As it stands, it integrates
well with SPM99/2, but a bit of tweaking is needed to read data in from
SPM5 .mat files because of the change in file structures. However, the
core SEM model specification and optimisation is very straightforward.
It gives the same results as LISREL (of course it should, but it was
nice to confirm it a few years back) but it was more convenient to be
able to specify just the SPM.mat file and the x,y,z and radius
coordinates of model regions. You are also able to include just voxels
that exceeded a threshold F-test across effects of interest; and have
the choice of regional mean, peak-voxel or first eigenvariate as the
time series. One can also use it for single subject or multi-subject
analyses.
James
|