measurementScale Decision Tree

[Tim Bergsma]

Scott,

I really like what you're doing with the decision tree idea.  I hope
whatever you come up with is exportable!

It seems to me that some supervision of the automatic process will be
inevitable.  In answer to your question below, we could try to associate
probable measurementScale with each DictionaryUnit.  At one point during
development, I naively suggested hard-coding measurement scale as a
field for each dictionary unit.  That won't work.  Take meters, for
instance.  Length can't be negative, so on first blush, meters is
ratio.  But height above sea level can be measured in meters, and some
points on the globe are actually below sea level, which instantly gives
you an interval scale.  One would expect that mass could never be
negative, but on a poorly calibrated balance, it might be.  In some
situations, the researcher may wish to report the data as is, but
without claiming that an interval scale was intended.  The issue for
you, I think, is how common/rare these sorts of examples are.  I can't
speak to that issue yet.

regards,

Tim.


Scott Chapal wrote:

> 
>    - Do we really know that the following are true?
> 
> > > "...interval and ordinal scales are much less common than ratio and
> > >  nominal scales in ecological data"
> 
> > > Someone has pointed out choice of measurement scale on the Stevens
> > > typology depends in part on what you want to do with the data.
> > > So, for instance, measurement scale can't be predicted from, say,
> > > DictionaryUnit
> 
> Should we try to associate probable measuremenScale choice with each
> DictionaryUnit?  It seems to me that many of them are pretty obvious.