[Sigcr-l] Exhaustivity and specifity of indexing

[Fred Lerner]

I agree with Susanne Humphrey's observation that "sometimes the gist of the article suggests multiple levels of specificity for indexing".

I produce the PILOTS Database, an online index to the worldwide psychotrauma literature. We have developed a controlled vocabulary of about 1200 descriptors (the PILOTS Thesaurus), and have used it to index over 30,000 publications. We often encounter a book chapter or journal article which discusses a particular aspect of psychotrauma in general terms and then discusses in some detail a specific example; for example, a paper on traumatic bereavement in natural disasters whose conclusions are meant to apply to disasters generally, but which supports its argument by reporting research conducted in the aftermath of a particular hurricane or earthquake. Our practice in such a case is to apply the general descriptor "natural disasters" to reflect the intended applicability of the paper, the specific descriptor "hurricanes" (or "earthquakes" or whatever)  to reflect the type of disaster discussed in the research results, and an additional term (or terms) taken from a separate controlled vocabulary of names of individual disasters and other traumatic incidents (in this case, it might be "Hurricane Katrina (2005)".

My justification for doing this is simply that I feel that it best serves the needs of the clinicians and researchers whom the PILOTS Database is intended to serve. Whether it would make sense to index at multiple levels of specificity in larger, more general databases is a matter that I will leave to those more learned in such matters than I. But what's the point in developing a small, specialised database for a particular community if you can't bend the rules to suit the nature of the literature and/or the needs of the users?

(Anyone interested in learning more about the PILOTS Database can find detailed information, and free access, at our website.)

Fred Lerner, D.L.S.
Information Scientist
National Center for PTSD
VA Medical Center (116D)
White River Junction, Vermont 05009 USA
phone (802) 296 5132
fax (802) 296 5135
internet <fred.lerner at dartmouth.edu>
website <www.ncptsd.va.gov>