Hello all,

I hope this isn't a silly question, but I'm only slowly learning GM 
analyses.  I've been working with 2D data in MorphoJ, and I've gotten to 
the point where all the recommendations are to pull the PC or Procrustes 
coordinate data out and analyze them in another stats program like SPSS.

My question is, 1) what numbers, and 2) what analyses?

I see people recommending doing a MANOVA in SPSS, but I don't understand on 
what 
numbers, or what that would tell me.

My project is 2D outline analysis with 7 landmarks and 6 equidistant 
semi-landmarks between each pair (43 total).  I have various categories 
(location, ware, time, etc.) that I want to compare similarities/variation 
of shape within and between groups, and try to determine if different 
shapes are notably linked to different areas or time periods.  Differences 
between communities, changes over time, concepts like that.  I've run PCA 
in MorphoJ and DFAs.  I did a regression based on centroid size that 
indicated that 18% of shape difference could be indicated by size.  (I also 
did a Partial Least Squares "Within a Configuration" of landmarks on 
different parts of the vessel, but I'm also not sure I understand what 
that's telling me.)  I've come to the end of what I understand MorphoJ can 
do for me, however.  Would anybody be willing to explain what statistical 
analyses in SPSS make sense?

Thank you,

- Kate Barvick

-- 
You received this message because you are subscribed to the Google Groups 
"Morphmet" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/morphmet2/5cdb2c13-15de-4268-95fb-11fbcb710df2n%40googlegroups.com.

Reply via email to