This handout is the text of information on the poster session, http://www.slideshare.net/NASIG/correcting-accidentals-13238333 Sanjeet Mann

1. See the poster here:
http://www.slideshare.net/NASIG/correcting-accidentals
Correcting Accidentals: Using an Availability Study to Identify and Resolve the
Sanjeet Mann
“Suspensions” Impeding Access to Electronic Resources
Electronic Resources / Arts Librarian
sanjeet_mann@redlands.edu
http://goo.gl/606us
Literature Summary
The availability study is a time-honored technique for evaluating how effectively patrons can obtain
desired items from a library. Buckland (1975) applied the method to book availability; Kantor (1976)
visualized periodical availability using a branching diagram; Ciliberti et al. (1998) investigated items
from an OPAC and CD-ROM indexes and categorized error types; Nisonger reviewed 46 recent
availability studies (2007) and adapted the technique to evaluate electronic resources (2009).
Wakimoto et al. (2006) and Trainor and Price (2010) have conducted similar studies focusing on link
resolver performance, one crucial contributor to electronic resource availability. Availability studies
can be “simulated” (conducted by a researcher) or “real” (conducted with the assistance of patrons).
Research Objectives
To better understand the challenges facing Armacost Library’s users by estimating their likelihood of
obtaining full text or experiencing an e-resource error, to understand why these errors occur, and to
prioritize our efforts to troubleshoot and correct them.
Method
I conducted a simulated availability study of e-resources
available to users of Armacost Library at the University of
Redlands. I generated an interdisciplinary sample of 400 items
by deriving 4 sample searches in each of 10 humanities, social
science and science disciplines from actual research topics
reported in our LibStats reference desk statistical software,
running the searches in A&I-only databases, and selecting the
first 10 results from each search.
In a Microsoft Excel spreadsheet (available at
http://goo.gl/606us), I noted bibliographic metadata, recorded all
inbound “source” and outbound “target” URLs, and indicated the
presence of local full text or an error.
I classified errors according to a six-point system covering proxy,
source database, knowledge base, link resolver, target database,
and ILL system errors. I calculated availability and error rates
by discipline and overall as percentages, and visualized the
results using a branching diagram.

2. Assumptions
A simulated availability study involves judgment calls concerning the terms to be searched, which
results are tested, and what constitutes an error:
Decision Rationale
Use derived searches to generate sample Maximize variety of item types
Simple keyword searches including specific words Observed student research behavior at University of
from research topics reported in LibStats Redlands
Test only first 10 results from each search Armacost Library usability test finding
ILL requests not considered an error “Just in time” item access is becoming the norm
Circular linking and browsing target are errors Users perceive failure (Trainor and Price 2010, p. 16)
List as erroneous if any target links give error Cannot anticipate which target link patrons will click
If multiple errors occur, choose the error type that Usability findings indicate that students rarely persist
occurs earliest to the patron after encountering the first e-resource error
Missing title or date in ILL field is an error Would cause Direct Request to fail or ILL staff to
delay processing the request for additional verification
Results
150 of 400 (38%) of the items tested produced an error; 104 of 400 (26%) items were locally available
online. About 1 in 3 items were only obtainable by ILL. Error types were distributed as follows:
Next Steps
Fix proxy and Serials Solutions errors
Contact vendors to assist with source, link
resolver, and target errors
Customize ILLIAD forms
Integrate study findings into library
instruction and web site redesign
Recruit student and faculty participants in
availability studies involving live patrons, to
account for usability errors as an access barrier.
Explore how library instruction can
empower patrons to mitigate availability errors.
Works Cited
Buckland, Michael (1975). Book Availability and the Library User. New York: Pergamon.
Ciliberti, Anne, et al. (1998). Empty Handed? A Material Availability Study and Transaction Log
Analysis Verification. Journal of Academic Librarianship 24 (4): 282-289.
Kantor, Paul (1976). Availability Analysis. Journal of the American Society for Information Science
27 (5): 311-320.
Nisonger, Thomas E. (2007) A Review and Analysis of Library Availability Studies.Library
Resources & Technical Services 51 (1): 30-49.
---. (2009). A Simulated Electronic Availability Study of Serial Articles Through a University Library
Web Page.College & Research Libraries 70 (5): 422-446.
Trainor, Cindi, and Jason Price (2010).Rethinking Library Linking: Breathing New Life into
OpenURL. Chicago: American Library Association.
Wakimoto, Jina Choi, et al. (2006). The Myths and Realities of SFX in Academic Libraries. Journal of
Academic Librarianship 32 (2): 127-136.

3. Acknowledgements
This presentation would not be possible without the generous support of Armacost Library and the
University of Redlands College of Arts Dean’s Office. Catherine Walker and Osman Trad of ITS
provided timely assistance with printing.