[Ross Peacock]

Dear Lee,

Thanks for the email prompt.

My experience with SSH and other ordination methods is biased towards community compositional data, not the taxonomic interpretation of morphometrics (although I did give KYST a run in 1986 on the UTAS mainframe using morphometric data.).

I was interested in Mike Palmers take on this issue (http://ordination.okstate.edu/overview.htm):

‘… the number of dimensions worth interpreting is usually very low.’ and ‘.. Third and higher axes can be constructed. The choice of ‘when to stop’ interpreting new axes is largely a matter of taste, the quantity and quality of the data, and the ability to interpret the results.’.

My experience suggests that looking at 2-3 dimensions is a pragmatic but reasonable start, and if the other interpretation statistics look problematic, think about the properties of the data and how you are treating it.

Lee I was interested in your comment about stress values ideally being <0.15, and putting aside how stress is measured, I have generally used a rule of thumb of 0.2, although for large data sets exceeding 1000 objects, this is often a real struggle.

I hesitate to say much more without a better understanding of the mathematics of ordination methods.

My vote is therefore for 2-3 dimensions.

Ross

[Ross Peacock]

Dear Lee,
Thanks for the reply. We regularly use the +/- option in GSTA to provide quick summaries of species in groups as the classifications are being evaluated. To create these were have returned to DOS PATN, using the LF95 version. This has led to further issues with transferring the files between versions, the archives read into v3.02 well, however there is more work preparing the v3.02 data files in a format to read in via DATN. A version which includes attribute constancy and fidelity would be very useful.
Ross

[Ross Peacock]

Dear Lee,

Here is a description of how I use the old bitmap version of the dendograms. They aren’t pretty, but they are functional.

In a current case I have an 80 group dendogram of GIS polygon types (997 objects, 114 attributes) derived from remote sensing of overstorey species, and we simply dump the *.den file into EXCEL and then add some explanatory variables next to the dendogram arms. It’s a really simply way to look at relationships in the hierarchy. If we need a publication quality dendogram then the PATNv3 version is far superior. I would like to keep doing the exploratory work with the old bitmap version as well if we can still export it in this format.

Thanks
Ross

[Ross Peacock]

Dear Lee,

Thanks for the feedback.

The task of filling down the species labels into the empty cells would be lengthy if done via EXCEL using copy or fill down commands. I could probably find a programming solution in S-PLUS but that is more handling. If the group statistics output could simply fill these cells for the user it would be much easier.

I was able to save the bmp version of the dendogram in the reduced width format but cannot replicate the versatility of the DOS PATN version which could be annotated and sorted in EXCEL rows once pasted and formatted to a non-proportional font. The bmp version simply gets inserted into EXCEL as a picture, the dendogram arms cannot be accessed.

Ross

[Author]

Posts