We moved the LCC facet based on input received from planning and focus
groups, both at NCSU and for the larger TRLN Endeca project. While this
facet consistently scored high in percentage of usage at NCSU, it was
always unclear to us whether this was due to interest or purely to
visibility. In local focus group sessions, it seemed as though students
understood what this facet was doing, but they often passed it over as a
precision tool anyway, even when presented with user tasks where it was
the obvious choice to achieve the outcome required. The TRLN UI
planning group, where NCSU was joined by colleagues from Duke, NC
Central, and UNC, felt that the head of the results display was too
valuable a piece of real estate to choose one facet over the others for
special treatment, so it was really there that the LCC move to the
facets column first happened.
Since this change is so recent, I don't know that we have any statistics
yet that might cause us to reverse these decisions. In some ways, I'm
not sure that I would trust any statistics until the change had had a
chance to settle in, as I suspect that there are plenty of cultural
reasons why there needs to be a passage of time before one tests changes
to public interfaces. Both library staff and library users have
expectations of catalogs that are conditioned to some extent by what
they are used to, which might explain some of the differences between
NCSU statistics and those of FCLA. Plus, facet use may also reflect the
mix of disciplines and their library participation rates at different
institutions, and the ways people in those disciplines approach a
particular search application. If your user population is skewed
towards the humanities, you might expect more author/title than subject
searching and language might also be an important facet. Towards the
social sciences, subject searching would rise, including use of
classification. Towards engineering and technology, where subject
terminology in LCC is not so current, you might expect more general
keyword ("keyword anywhere") searching, with subject facets used to
increase recall. And so on. NCSU is largely an STM campus, although we
also have humanities and social sciences programs which make heavy use
of the library. Somebody from McMaster or FCLA can speak to their
collections, but you might also remember that Phoenix Public Library
(http://www.phoenixpubliclibrary.org/) also uses Endeca and their facet
placement and, I'm sure, usage is quite different from any of the
academic libraries. In fact, from their main library page, your eyes
are most drawn to their browse facets, which I'm guessing, may well be
populated by a DCC (Dewey Decimal) hierarchy!
Charley
Steve Toub wrote:
> Thanks, Emily.
>
> I do have some additional questions:
>
> 1. What was the rationale for moving the LCC facet from the top of the
> page to the left with the other facets? And has the usage of that facet
> has increased or decreased significantly since that change?
>
> 2. Do you have a sense of whether facet usage at NCSU's implementation
> is fairly typical of Endeca implementations at other academic libraries
> (e.g., TRLN, McMaster)? The reason I ask is that I notice that the
> overall numbers on search vs. navigate vs. search, then facet at FCLA
> [3] seem to be significantly different than the equivalent numbers in
> the NCSU PowerPoint.
>
> --SET
>
> [3]
> http://www.fcla.edu/FCLAinfo/stats/endeca_stat/endeca_stat_summary_2008_03.html
>
>
>
>
> Emily Lynema wrote:
>> Steve,
>>
>> While we haven't calculated the stats out at NCSU recently, I do have
>> some carefully configured stats from Jan - April 2007 that could serve
>> as representative. These numbers should have done a more complete job of
>> eliminating bot activity (bots *love* facets, even when you try to
>> exclude them) than the numbers found in the presentation, although the
>> overall outcome is similar. The percentage here is percent of facet
>> usage.
>>
>> Topic: 28.5%
>> LCC: 24.3%
>> Format: 14.4%
>> Library: 9.2%
>> New: 6.4%
>> Genre: 5.1%
>> Author: 3.9%
>> Region: 2.9%
>> Language: 2.1%
>> Era: 1.8%
>> Availability: 1.2%
>>
>> These numbers were calculated at a time when all facets showed at least
>> 5 values by default (we've since closed some of the facets by default in
>> the interface). Obviously, the placement of the facets and whether or
>> not values are show probably impact usage. For the order the facets
>> where presented at the time these stats were taken, I'd use the chart in
>> the powerpoint; we've also since re-ordered our facets.
>>
>> We might have more data to share if you have other specific questions.
>>
>> -emily
>>
>>>
>>> ------------------------------
>>>
>>> Date: Tue, 8 Apr 2008 13:39:08 -0700
>>> From: Steve Toub <stoub_at_YAHOO.COM>
>>> Subject: Which facets are most used?
>>>
>>> Hi--
>>>
>>> For those of you who have facets in your catalog, what level of usage
>>> are you seeing for individual facets (e.g., format vs. topic vs.
>>> library
>>> collection)?
>>>
>>> I've poked around a bit and have found some numbers for NCSU [1] and
>>> the
>>> State Library of Tasmania [2] but am curious about other data points
>>> for
>>> both public libraries and academic.
>>>
>>>
>>> --SET
>>>
>>> [1] See slides 53-55 in
>>> http://www.lib.ncsu.edu/endeca/presentations/200710-ncla-pennell.ppt
>>>
>>> [2] See figure 5 in
>>> http://www.valaconf.org.au/vala2008/papers2008/25_Sokvitne_Final.pdf
>>>
>>>
>>> --
>>> Steve Toub
>>> Product Manager
>>> BiblioCommons
>>>
>>> ------------------------------
>>
>>
>> --
>> Emily Lynema
>> Systems Librarian for Digital Projects
>> Information Technology, NCSU Libraries
>> 919-513-8031
>> emily_lynema_at_ncsu.edu
>>
--
__________________________________ __________________________________
"""""""""""""""""""""""""""""""""" """"""""""""""""""""""""""""""""""
Charley Pennell mailto:cpennell_at_unity.ncsu.edu
Principal Cataloger for Metadata voice: (919)515-2743
Metadata and Cataloging Department fax: (919)515-7292
NCSU Libraries, Box 7111
North Carolina State University
Raleigh, NC 27695-7111
Adjunct Librarian, Memorial University of Newfoundland
World Wide Web: http://www.ibiblio.org/hillwilliam/chuckhome.html
__________________________________ __________________________________
"""""""""""""""""""""""""""""""""" """"""""""""""""""""""""""""""""""
Received on Wed Apr 09 2008 - 16:47:49 EDT