Re: ANOVA/MANOVA

[John Darrington]

The Anova capabilities of PSPP are currently rather limited. Oneway
anova is supported through the ONEWAY command.  The latest master
also has the command GLM which, when complete, will be able to perform
everything you are looking for. Unfortunately, at the moment, it can 
only do "latin squares" designs.  

Jason Stover is working on the last bit necessary to allow a full factorial
univariate anova.  Hopefully, that will be finished soon.  From there, I think 
it'll be straightforward to extend to the multivariate case.  I'm not sure 
what would be involved to get it to accept covariates, which would do your 
ANCOVA/MANCOVA requirements.  Maybe you would like to look into  that?

If you want to contribute code to PSPP, you should also take a look at the 
PSPP developer's guide (which is in the tarball), and at the GNU Coding 
Standards http://www.gnu.org/prep/standards

Thanks for your interest in PSPP.

John

On Sat, Sep 17, 2011 at 01:42:47PM -0500, Trever L Hallock wrote:
     Hello,
     
     I would love to help with PSPP, but am relatively inexperienced. I am a
     student in both Math and Computer Science, and although I feel confident in
     C, I am learning about ANOVA/MANOVA and ANCOVA/MANCOVA in my classes, which
     is what I would like to implement over this coming semester. I would also
     need to learn GSL while coding these. If you think my code could help, let
     me know if you have any more specifics on the struct that I should use to
     store the results of my routines besides what I will be able to figure out
     by browsing other back end examples PSPP uses. Is there anyone else working
     on this?
     

-- 
PGP Public key ID: 1024D/2DE827B3 
fingerprint = 8797 A26D 0854 2EAB 0285  A290 8A67 719C 2DE8 27B3
See http://pgp.mit.edu or any PGP keyserver for public key.

signature.asc
Description: Digital signature