An attempt to employ probabilistic subsampling in a wooded study area is reviewed. Errors committed in attempts to test hypotheses about statistical associations are identified, and more suitable procedures are suggested. The site-location potential of regional subsampling is considered. On theoretical grounds it is concluded that regional subsampling employing small subunits in areas of low site density is probably inefficient and biased when used as a search procedure.
Email your librarian or administrator to recommend adding this journal to your organisation's collection.