i am forwarding this for mr sklebar. please respond to him, not me!
lmv
------------- Begin Forwarded Message -------------
X-MIMETrack: Serialize by Router on gscodenh01/SERVER/USGS/DOI(Release 5.0.3
|March 21, 2000) at 09/27/2000 08:30:06 AM
MIME-Version: 1.0
Date: Wed, 27 Sep 2000 09:29:23 -0500
From: Tom Sklebar <tom_sklebar@USGS.GOV>
Subject: GIS and statistics
To: FWIM-L@LISTSERV.VT.EDU
I am interested in the current "state of the art" for statistical analysis
of GIS outputs, particularly in the area of landscape ecology and
biological data. It appears from this author's viewpoint that there exists
a disjunction between the available statistical analysis techniques and
standard GIS outputs. For GIS users, the frustration comes from the lack
of apparent interest by software makers to provide even elementary
statistical methods (e.g. logistic regression) inside the software or even
if implemented, fails to provide sufficient information for use by a
statistician or evaluation by the statistically literate. For
statisticians, the frustration results from not enough information in the
standard reports and tables output from GIS software to allow advanced
statistical techniques and/or lack of understanding of the underlying GIS
processes. For both groups, the frustration comes from evaluating the
biological relevance of the analysis.
This discussion thread resulted in 6 general questions:
Is there a "common ground" between the statistical analysis techniques and
the available GIS outputs?
Could there be a "better" set of data output from a GIS that may be more
amenable to statistical analysis?
Are there available statistical analysis techniques, currently unutilized
or underutilized, that could be applied to available GIS outputs that would
be more biologically relevant?
>From a landscape ecology perspective, what is the current and future of
landscape fragementation analysis and theory, and corridor theory and
analysis?
What is the current "state of the art" in pattern recognition theory and
applications and could, or should, those techniques be applied to GIS?
>From a conference standpoint, when was the last conference that brought
together statisticians/biometricians with GIS and Remote Sensing
specialists to discuss the merging of the two disciplines?
Just wondering. Will sum a list of URL's, contributed authors, and
discussion if you send me a copy along with posting a copy to the listserv.
Please note also that this could be a discussion topic at The Wildlife
Society GIS, Remote Sensing and Telemetry Working Group meeting at the
Midwest Fish and Wildlife Conference in Minneapolis in December, 2000.
Tom Sklebar
USGS
Northern Prairie Wildlife Research Center
701-253-5554
tom_sklebar@usgs.gov
------------- End Forwarded Message -------------
-----------------------------------------------------------------------
This message was distributed by s-news@wubios.wustl.edu. To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message: unsubscribe s-news
|