Would you use more dimensions in SSH Ordinations?

[bowmanmi]

To all PATN users,

We had previously (a long time ago!) used the DOS version of PATN for SSH ordinations & have recently started using the windows version. In the DOS version, the user could select more than 3 axes to reduce ‘stress’. With the taxonomic data we ordinate, we often require more than 3 axes to obtain a reasonable stess level. We were wondering if anyone else would benifit from being able to select more than 3 axes in SSH ordinations?

Thanks, Michelle

[lee]

Hi Michelle

A fair request. It was pragmatic that we implemented only up to three dimensions in PATN (for Windows) V1. The significance of achieving our 3d display environment probably blinded us from thinking further, at least at that time. There is however no logic in assuming that there could be only three factors controlling the variation in a dataset.

What do you do if stress is greater than 0.15? At the moment, probably reduce the noise or the complexity of the data in one of a number of ways. I would not publish an ordination result greater than 0.15 myself.

If there is user support to go to say 5d, my strategy would be to enable selection of any 3 of the 5 axes to be selected to display in the 3d plot. The PCC strategy follows easily enough. Listing and output of the coordinates represents no problem.

Over to others to discuss-

Lee

[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

[lee]

Thanks Michelle and Ross for a useful discussion. I have sat on my thoughts for a few weeks about this issue.

There is truth in all that has been said. I have certainly had situations where the 4th dimension in an ordination provided evidence for a genuine process that was generating variation in the data. Ross is right however in that caution is well justified. It is often too easy to interpret patterns that may be suspect, particularly in higher dimensions.

I have put the request for more than 3d on the ‘TO DO’ list, but not as priorty-1 (eg, I’d like to add a nearest-neighbour list function first as a number have requested it). I figure additional dimensions in SSH need not be selected, but can be there if desired.

Ross, on stress I do tend to use ~0.15 as a cut-off myself. If stress is higher than 0.15, I seek ways to reduce it closer to 0.1. I look back at the coding, transformations and standardisations, eliminate outlier objects and noisy variables. I’d would like to add an ability of determining which objects were the most difficult to fit in SSH. Another wishlist item!

Lee

[Author]

Posts