I'll try to answer Phil's question on the NSF retention of data policy.  I
don't know whether they are actually ready to issue a policy, but if you
aren't aware you should know that NSF went to OMB last year for a
determination as to whether or not technical data is included in Section
__.53(a) of A-110.  Section __.53(a) specifies that financial records,
supporting documents, statistical records, and all other records pertinent
to an award shall be retained by the institution.  In an April 11, 1994
letter to NSF, OMB stated that the term "all other records pertinent to an
award" encompasses all records that were produced in connection with the
award, including laboratory data and primary data.

So, that brings us to where we are now.  Many of us are developing policies
with respect to data retention, not data ownership.  Mostly, those policies
are specifying that the records are in the custodianship of the investigator
and that they have the responsibility to retain them.  Further, those
policies are also stating procedures for when an investigator leaves the
institution.

The biggest problem in developing these policies is whether or not to assert
ownership.  Most of us are sidestepping that issue for many reasons
including political considerations (imagine the flap raised if you tell
faculty you own their data) and even storage reasons (if you assert you own
the data, are you prepared to deal with the deluge of faculty who wil tell
you, you own it, you store it!)

I think we would all be wise to address this issue at our own institutions.
 Retention of data is not only an NSF concern, although they have been more
overt lately.  It has also come up with PHS and seems a paramount issue in
many misconduct cases.  In that respect, it is probably self-protective to
have a retention policy.

Just my two cents worth...

Jane Youngers
Office of Research and Project Administration
University of Rochester
xxxxxx@orpa.rochester.edu (new address)