Page 105 - JOURNAL OF LIBRARY SCIENCE IN CHINA 2015 Vol. 41

104 Journal of Library Science in China, Vol. 7, 2015

of the databases acquired by the library. Or the library didn’t purchase certain back issue
databases of some vendors. As for the first probable reason, because a single library has
little influence on the continuity of vendors’ full-text databases, the authors suggest that, in
the process of group purchase, the library consortia and Steering Committee of Academic
Libraries (SCAL) should put pressure on this problem. While, usually, the continuity of the
full-text database is ensured by the contract signed between the library and the vendor. So,
in most instances, the second reason is the major concern of the unavailable issue situation.
Facing the complicated development of electronic collection, wise suggestions cannot be
concluded from a separate analysis of title unavailability or issue unavailability. Instead, an
optimal combination of full-text databases according to the patrons’ needs should be proposed.
The academic libraries can assemble the following elements: 1) a sense of the whole picture
of journal collection development, 2) strict quality control of full-text databases in regard to
the collection continuity, 3) precise analysis of the temporal coverage of full-text databases,
and 4) vendors’ quotation.
As for the print collection, only 0.11% of the sample citations failed to be accessed for the
reason of unavailable issue. This finding owes to the endeavor of print collection building in
collection continuity. The ACF of unavailable titles is three times of that of unavailable issues.
Aiming to improve the literature guarantee rate, the print journal, which has a good quality and
academic reputation, is still a priority if vendors’ quotation exceeds expectations. Meanwhile,
academic libraries have to consider the bundled subscription in acquiring new databases, and
cancel the overlapped subscriptions in time.
As for the OA resources, sample citation access failure occurs much more often on
unavailable titles than on unavailable issues, this is probably because: 1) many OA journals
have not entered the vision of researchers, 2) prejudice from the researchers, and 3) unbalanced
development of the OA journals between different disciplines, etc. To promote the use of OA
journals and accelerate the process of the OAI, academic libraries should take the responsibility
for driving and submitting the OA policy of the university and the SCAL while integrating and
popularizing OA journals.
For an in-depth study of the influence of two causes on access failure, Figure 4 and Figure 5
show the access failure statics caused by unavailable titles and unavailable issues separately. On
the whole, the unavailable-issue curve of E∪P∪OA performs a continuum between the OA and
electronic unavailable-issue curve, and the most part of it falls within the range of 0-10%. The
unavailable-issue curve of E∪P∪OA fluctuates obviously before 1990, while behaving a gradual
downtrend thereafter. In Figure 5, the unavailable-title curve of E∪P∪OA stands at around 10%
and runs under the extra curves. Unavailable issues have less influence than unavailable titles.
Because unavailable issues are tacit and have a subtle overlapping ratio among the three resources,
they are hard to supplement. So the emphasis and difficulty lie in the accurate detection and the
pertinent remediation of unavailable issues. Owing to the overlap between the three resources and