We did a couple of similar studies here at TCNJ. One looked at initial
article searching, and another was more broad looking at all searches. I
don't have the numbers for the broader look at our searches (someone
else did that study - and hopefully it will be coming to a peer-reviewed
journal near you soon!), but I can tell you the numbers Bennett posted
to the list were similar to what was in our investigations.
I can also tell you that many students were resilient as Mike Berger's
research (According to Karen Coyle) shows.
Really, if you ever do this type of study on your own catalog, you will
be dismayed, but you will learn a lot.
Edward
Karen Coyle said the following on 02/07/2008 7:57 PM:
> Bennett Ponsford wrote:
>
>> Subject headings keyword - 40.31% resulted in no hits Journal title
>> keyword - 37.41% resulted in no hits Expert keyword - 31.98% Title
>> keyword - 31.51% Journal title starts with - 28.42% Title starts with
>> - 25.26% Author keyword - 25.12% Keyword - 20.53%
>
> It seems obvious but also interesting that Keyword is the least likely
> to get you zero hits. That could well lead to a certain "user
> satisfaction" with that type of search.
>
>>
>> One question is then, what do people do when their search results in
>> no hits? From just browsing through the logs, I can say that some
>> people just keep retrying until they find something they like, but I
>> don't have numbers on that yet. Nor do I have numbers on how many
>> people just give up and move on.
>
> Mike Berger studied this in the MELVYL logs for his doctoral
> dissertation. He was able to identify "sessions" and to see how people
> compensated for zero results. Unfortunately, he never wrote up anything
> other than his dissertation, which is:
>
> Author Berger, Michael George.
> Title Information-seeking in the online bibliographic system
> : an
> exploratory study / by Michael George Berger.
> Publisher 1994.
> Description v, 216 leaves ; 28 cm.
> Note Thesis (Ph. D. in Library and Information Studies)--University
> of California, Berkeley, May 1994.
>
> As I recall from conversations with him, users turned out to be quite
> resilient, trying lots of different angles before succeeding or giving
> up (from the logs you couldn't really tell, but most users moved from
> zero results to non-zero results). His figures also showed that the
> number of initial zero results was surprisingly high. I know when I
> tried to discuss the pitfalls of keyword searching with a class of
> computer science students at Stanford some years ago, their response was
> (in reference to search engines): "But I always get something." In the
> general equation of something v. nothing, something does tend to be a
> happier result, if even wrong.
>
> --
> -----------------------------------
> Karen Coyle / Digital Library Consultant
> kcoyle@kcoyle.net http://www.kcoyle.net
> ph.: 510-540-7596 skype: kcoylenet
> fx.: 510-848-3913
> mo.: 510-435-8234
> ------------------------------------
--
Edward M. Corrado
http://www.tcnj.edu/~corrado/
Systems Librarian
The College of New Jersey
403E TCNJ Library
PO Box 7718 Ewing, NJ 08628-0718
Tel: 609.771.3337 Fax: 609.637.5177
Email: corrado_at_tcnj.edu
Received on Fri Feb 08 2008 - 08:45:37 EST