Charles,
 
But how have you stratified the 10,000 homesteads?  If we first researh
the history of each building and then develop a model for
ethnic, gender, economic group, and vocation, then we can
talk seriously about adequate sampling.  Even so, we will find
significant variation within those groups that merits
sub-sampling in order to better characterize the sub-group.
 
I have been thrashing about another thorny problem with
sampling in my work at the County of San Diego.  Dennis
O'Neil in a 1992 or 93 issue of American Antiquity challenged
the premise that a 5% sample is all we need to take on
prehistoric sites to call it salvage.  He picked a site
that had the 5% sample and then put his field class to work
to dig 23%, 30+%, 50%, and 63%.  At 30+%, he bgan to detect
significant evidence of a Paleo-Indian component, as well
as historic beads suggesting contact.  This bore real fruit
at 50%, but was just repeated at 63%.  We have to be very
cautious about sampling.
 
And finally, I would like to mention that USFS contracts
tend to be very limited in scoping what contractors can do
with the homesteads they do dig.  Many of our colleagues have
been doing this for so long that they do sampling rather
mechanically and have lost their spark of research.  This
means that unless you have zealous folks out there working
on those 10,000 homesteads, you will not get quality field
work, analysis, and interpretation.
 
The 10,000 homesteads merit serious thought and considera-
tion.  I would fight hard to preserve them in-place and
not waste them with salvage archaeology or dismissive
government policies that future generations will greatly
regret.
 
Ron May
[log in to unmask]