The Usability of Electronic Finding Aids during Searches for Known Items

1
The Usability of Electronic Finding Aids during
Searches for Known Items
Society of American Archivists Annual
Meeting August 22, 2003 Los Angeles, California

• Christopher J. Prom
• Assistant University Archivist
• University of Illinois at Urbana-Champaign

2
introduction

• development of descriptive standards
• materials centric process
• not inherently user centered
• but lays groundwork for easy reformatting
• how to best present archival data for ease of use?

3
study goals (the why)

• user interactions with finding aids
• looking for known collection
• look for folder on a given topic
• make inferences regarding best type of
  designsusability
• what is outside scope?
• manuscripts only, not corporate
• not a report on descriptive standards

4
methodology (the how), slide 2

• developed website
• survey
• 9 usability tasks with controlled searches
• collection-level interfaces
• 11 option (links)
• 3 option (search, subject browse, provenance)
• 2 option (search and alpha list)
• 1 option (alpha list)
• folder-level interfaces
• PDF (Adobe)
• non-searchable EAD w/ nav bar (Cookbook)
• HTML, top table on contents
• searchable EAD w/ two pop up windows
• administered both on and off site

5
(No Transcript)
6
(No Transcript)
7
methodology, slide 2

• during the test
• used control on info searched for
• system recorded answers/times
• on sitepost test interviews
• after the test
• coded data coding
• tabulated, queried and analyzed
• correlated times to survey results
• transcribed interviews

8
results, slide 1

• respondents
• 89 participants 35 on site, 54 off site
• on site mostly students (77), off site
  archivists librarians (51)
• 72 claim to have use archives last year, 75 w/
  electronic finding aids
• experienced computer users
• 69 self rate as highly experienced computer
  users or above (I can install software or
  hardware, develop web pages, use databases.)

9
results, 2

• collection tasks
• tasks w/ fewest search options most successful
• alpha lists worked well

10
results, 3

• folder-level tasks, plotted search times

on-site users only
11
survey/task correlations

• will not discuss in detailtable on handout
• speed in using electronic finding aids correlates
  more to computer experience than archival/library
  experience.

12
suggested design features, 1 (based on
observations, interviews and, comments)

• factors specific to finding aids
• need complete description and context
• is some info available on line for every
  collection?
• provide browse option alongside search boxes
• people use a search box if available
• . . . but actually find known items faster w/
  lists
• search boxes for single finding aids
• 60 try browsers find in page
• dont break ability to do this w/out a search box

13
suggested design features, 2

• general factors
• use standard technologies and formats
• PDF?
• no pop ups beware complex javascript
• use standard search algorithms
• keep interfaces simple
• clear labels and layout
• Where is the label? (e.g. box number)
• detailed description vs folder list

14
linear format
default visited color
(hints CTRL-F will work)
visual interest (could move to right)
moderate number of links
to series descriptions w/ hyperlinks to folder
list
15
conclusions?
this presentation available from a link
at http//web.library.uiuc.edu/ahx/workpap/
I would like to thank the Research and
Publication Committee of the University of
Illinois Library, which provided support for the
completion of this research. Christopher J.
Prom Assistant University Archivist