THE LIBRARYS ROLE IN THE INSTITUTIONAL REPOSITORY

The Librarys Role in the Institutional Repository

Colleen Sanders
Emporia State University
July 27, 2014

THE LIBRARYS ROLE IN THE INSTITUTIONAL REPOSITORY

Abstract
Academic libraries initially developed institutional repositories (IRs) to preserve local
scholarship in special collections. IRs became a vehicle for disseminating digital scholarship to a
wider audience in an emerging new scholarly communications environment. The premise was
simple: support the institutional research community by providing permanent storage and global
access to materials. However, after some initial traction, many early adopters find themselves
unable to secure sustained faculty participation or administrative support. The perceived failure
or stagnation of many IRs may be due to a departure from its original service intentions, as IRs
must achieve a functionality exceeding mere archives. A return to the initially identified need can
help guide the second era of IRs. There is no indication the demand for a new model has waned:
digital scholarships trajectory is only beginning, and some scholarly communities are accepting
the impact factor achieved through open access (OA) publishing as a valid substitute for a
traditional journal brand. Recruitment issues can be resolved by realigning IR services with user
needs, and sustainability can be achieved by generating administrative buy-in. After revisiting
the original impetus for IRs, this paper inventories some of their unexpected complexities. By
taking a user needs perspective, it aims to help libraries clarify their vision for how to proceed
and share that vision with their communities to garner the support an IR requires.

Keywords: institutional repository, scholarly communications, digital scholarship, open

access, impact factor, library services, digital publishing

THE LIBRARYS ROLE IN THE INSTITUTIONAL REPOSITORY

The Librarys Role in the Institutional Repository

The Changing Realm of Scholarly Communications
After 15 years of implementation, libraries have discovered IRs can be more than just
special collections. In fact, researcher behaviors suggest a collection might be the wrong way to
think about an IR if libraries are to attract scholars to use it. While preservation is an ongoing
challenge, it occurs alongside retrieval, as research achieves value only when it can be accessed
and built upon. Concerns over intellectual property rights fuel the debate, but scholars seek
impact first. Thus, the IR becomes a key component in the publishing cycle, be it open access
(OA) or self-archiving alongside traditional journals. IRs must be structured and marketed to
reward faculty participation and connect institutional output to the global scholarly conversation.
Libraries have the expertise and position in the university to spearhead the IR, but they
often lack to resources required to bring the project to full fruition. They must now align services
with known stakeholders needs to advocate for support from top-level administrators. The IR
represents the interests of many factions of the university, which can serve to bolster its
importance but also fragment its execution. This research serves to outline the areas libraries
must develop in order to build robust and sustainable IRs. In the context of a serials crisis, it is
easy to make the argument that any contribution to a stable open access environment is in
everybodys best interest. What is less clear is how the IR can add to the global body of
scholarship while serving the scholars and the institutions that support them. The librarys role in
this shifting environment is to identify demand and tailor the IR agenda to develop to serve those
needs.

THE LIBRARYS ROLE IN THE INSTITUTIONAL REPOSITORY

How should libraries proceed in developing IRs?

This paper addresses some of the deficiencies identified by IR critics, namely the
unexpected challenge recruiting faculty submissions. The sum of findings is that IRs need to be
designed as services addressing specific user needs. Users of the IR have the interesting quality
of also being the contributors, which has implications for system design that are only starting to
be addressed in scholarly literature. IRs must be built seamlessly into researcher workflows and
provide an impact factor that rewards participation. An institutional mandate for submission
evidences a failure of service design. The first step in making an IR successful, then, is to do an
environmental scan of IR users to realign current services.
This paper operates on the assumption that IRs form a valuable service point for libraries
and the scholarly community, but require some modifications. By parsing the three main
stakeholders and their relation to the IR, it hopes to lay a groundwork for evaluating existing IRs
and designing them so they will serve their communities.
While this paper discusses functionality, it will not address in-depth systems questions
regarding platforms or technical operations. While it addresses advocacy, it does not insist on a
specific framework for selling the IR to administrators. This paper touches on open access and
the serials crisis, but it focuses on the IR as a user-centric vehicle. IRs and traditional publishing
are by no means exclusive, and while copyright is a concern, this paper does not delve into
specifics regarding levels of self-archiving.
Literature Review
The literature surrounding IRs displays an arc of early optimism (Young, 2002) followed
by revisionism (Furlough, 2009; Salo, 2008). The seemingly simple proposition to build digital

THE LIBRARYS ROLE IN THE INSTITUTIONAL REPOSITORY

warehouses for materials written by faculty of a particular institution (Gregory, 2011, 197)
quickly revealed its complexity. Amidst the nascent trajectory of digital scholarship, Lankes
agreed [IRs] in the abstract make great sense in the future of academic libraries (2011, 103).
Lankes tasked libraries to re-envision systems through the lens of user needs and community as
collection. Alongside the open access (OA) trend on the web, IRs present a way to ground
research in context and preserve some of the prestige inherent in titles (Jones et al., 2006).
IR advocates perceive failure in the lack of faculty participation. The growth rate of
materials in IRs slowed to such an extent that authors began doubting whether libraries should
provide IR services at all (Salo, 2008; Jones, Andrew, MacColl, 2006). The underlying
observation that the IR largely serves the librarys needs, rather than those of its scholars,
triggered a call for research regarding user experience and the incentives behind OA publishing
(Hswe & Tribone, 2014; Nariana & Fernandez, 2012). Part of this confusion resides in software,
which can become the golden calf of IR implementation. These locally designed and scaled
systems must be easy to use, have wide functionality, and be interoperable with IRs in other
institutions. An individual repository forms a part of an international system of distributed,
interoperable repositories available through the web. Thus, the repositories provide a foundation
for a new model of publishing centered on the author rather than the publisher (Gregory, 2011,
197-198). Chudnov noted that the necessary merger between IT and collection development
lacks a script, and often resulting in librarians fetishizing software products or projects (2008).
Salo (2008) and Hswe & Tribone (2014) echo the notion that software, rather than people, has
been the guiding mandate for IR development, and this might likely be due to underfunding and
a lack of clear vision. Compounding this is the problem of administrative support, as Salo wrote,

THE LIBRARYS ROLE IN THE INSTITUTIONAL REPOSITORY

most libraries consistently under-resources and understaff repositories, further worsening the
participation gap (2008, 98).
The Librarys Role in IRs
Definitions for services. An antidote to confusion is to revisit definitions to lay a
foundation for services. The Association of Research Libraries (ARL) defined criteria for
inclusion in its 2006 SPEC Kit as a permanent, institution-wide repository of diverse, locally
produced worksthat is available for public use and supports metadata harvesting (2006, 13).
While exact, this broad definition does not offer libraries much in the way of concrete services to
offer, and Furlough (2009) takes issue with the funereal implications of the term repository.
Gibbons (2004) takes the next step by focusing on scholarship as conversation and outlining five
characteristics which distinguish IR services. IRs contain digital content; they are communitydriven; they require long-term financial support; submissions are permanent; IRs universally
available without paywalls. These facets will frame an analysis of services following an
inventory of user needs.
Stakeholders Needs
Researchers and scholars. The major players in the IR are researchers, libraries, and
university administrations. While all must receive their due, recent history suggests it might be
best to begin with the researchers needs first. The IRs original selling points to scholars,
preservation and dissemination, have proved insufficient to sustain participation. While IRs also
offer scholars a publishing platform that allows them to retain copyright, most gladly forfeit
control for the prestige of a journal title. Studies (Nariana & Fernandez, 2012; Salo, 2008)
identify researchers number one priority is impact, often measured by publisher brand rather

THE LIBRARYS ROLE IN THE INSTITUTIONAL REPOSITORY

than impact factor (IF) or citation count. All else pales in comparison to this desire because
impact brings rewards: tenure, funding, and reputation. This fact ought to cue libraries in
designing their IR services towards dissemination, promotion, ease of access, and
communicating IF value to researchers. Another valuable consideration might be how to merge
collaborative authoring practices to resolve versioning issues. This consideration leads to a
systems analysis, and whether or not the library believes developing back-end software to
support creative processes is something their community desires.
Academic and research libraries. The IR offers the library is an opportunity to employ
its expertise in making an original, value-added contribution to its community. It is a way to
prove the librarys relevance and value. Building more complete collections representative of the
university environment is an exciting way to contribute to scholarly communications. Endorsing
OA and rerouting funds from overpriced serials and databases make budgetary and ideological
sense, and libraries must be prepared to allocate sufficient resources to the IR..
However, one underdeveloped step in IR administration involves generating sufficient
support for it. Libraries must demonstrate clearly to administrations that supporting digital
scholarship does not come at a cost to the institutional brand, but rather enhances efficiency and
competitiveness. This influences publication, academic rewards systems, and financially
supporting the IR. Libraries can ease administrative anxieties about preserving prestige in an OA
climate with some technical knowledge proving digital publishing does not reduce institutional
affiliation to mere bibliographic content. Metadata harvesters like Google Scholar and OAIster
bring IR content to the worldwide audience, but to retrieve the document itself, a searcher must
still enter the IR through the institutions site. Therefore, the reputation of the university is
inextricable from the IR materials, bolstering the argument that IRs are an excellent option for

THE LIBRARYS ROLE IN THE INSTITUTIONAL REPOSITORY

dissemination of output. Digital scholarship suffers some distrust in the academic administration,
as it is too newness makes promises not yet verified by the test of time. As Jones et al. noted in
2006, The greatest challenges of all for university librariesmay, therefore, be outreach and
liaison (21).
University administrations. The IR offers a way for universities to showcase their
scholarly output, which can influence potential students and faculty. Local preservation generates
greater control over institutional materials. Also, amidst the decline of university imprints, IR
software can also be extended publish institutional journals alongside the IR collection. From a
financial perspective, the funds rerouted from imprints and library subscriptions can alleviate the
perception that an IR is an added expense. The administration must understand fully the level of
commitment an IR requires and be willing to channel the necessary resources for the long term.
No doubt purchasing back access to institutional resources is a strong selling point, but it must be
bolstered by active appreciation of the IR as a replacement or supplement. Finally,
administrations must be willing to incentivize faculty to participate without enforcing an outright
mandate. This means restructuring the academic rewards system to include IR and OA
publishing as viable efforts for faculty working for tenure.
IR Facets and Challenges
This section attempts to deconstruct the spectrum of IR-related services and analyze
possible directions libraries can take. The sections have been harvested and modified from the
combined definitions of both Gibbons (2004) and Jones et al. (2006). Defining goals at the outset
will scale each activity to a particular institutions needs. The IR will require a collection
development policy that also accommodates retrieval and dissemination factors. The users of IR

THE LIBRARYS ROLE IN THE INSTITUTIONAL REPOSITORY

are also the publishers of content; therefore, the collection must be developed with utility in
mind. It must be easy to deposit as well as to access, link, share, and build off others work.
Material submission. An IR must enable but not require a researcher to upload their own
materials. Ironically, the best way to get faculty to participate is to minimize the actual amount of
participation required, therefore digitization services must be offered as well as easy batch
uploads. One of the IRs strengths is making available resources traditionally excluded from
journal publishing, including but not limited to theses, dissertations, pre-prints, technical reports,
white papers, gray literature, conference papers, working papers, A/V files, and datasets. Such a
dynamic collection will require creative stewardship that incorporates both librarian and author
perspectives. Post-submission, faculty must be able to manage their content, which can pose
challenges as libraries attempt to build documents into searchable collections. Gibbons (2004)
exposes some reluctance on the part of faculty to see their work as belonging to a pre-defined
collection, yet some balance between self- and mediated classification must be struck.
Metadata application. Good metadata built according to the Open Archives Initiative
Protocol for Metadata Harvesting (OAI-PMH) standards is essential for materials to be findable
outside the local system. This task requires library mediation to ensure interoperability. However,
librarians often lack the domain-specific knowledge required to produce accurate and deep
metadata content. Including faculty in the metadata creation process is essential, but must be
curated according standards set by information professionals.
Access control. While most institutions choose to make the majority if IR content freely
available, some level of access management is necessary. Back-end access control guarantees
only verified persons can post content. This concept of quality control calls into question who is

THE LIBRARYS ROLE IN THE INSTITUTIONAL REPOSITORY

10

allowed to post to the IR: faculty, librarians, researchers, graduate students, and in some cases,
even undergraduates. These are decisions that must be made on the local level amongst the
various stakeholders, and possibly designed with an override mechanism lest good material be
barred from inclusion. Access control and quality control are related, and it is at this point the
notion of peer review becomes problematic, as the IR skips that valuable step in the publishing
cycle. Gibbons (2004) suggests recruiting a committee of editors to screen for quality alongside
tasks like classification and metadata, but that does not seem feasible, responsible, or likely to be
well-received. How to filter for quality should be the subject of further inquiry, and it is possible
post-publication peer review will emerge according to citation count in the new OA environment.
Discovery support. The IR movement borrows some momentum from the driving force
of OA and web technologies, and the ease of discovery outside of proprietary boundaries forms
the crux of its success. To what level of sophistication will the content be indexed and
searchable? A simple search engine is the basic expectation, as well as a department or subject
heading list for browsing. Deeper levels of granularity could include faceted browsing or fulltext searching, depending on the limits of the institution and richness of content. Relationships
between files must also be established, as digital scholarship often crosses not only disciplinary
boundaries but also file types. Enabling a user to reference data sets, images, audio, video, and
other materials through linking as they consume research content is essential to harnessing the
full range of functionality from a web-based IR.
Distribution. While institutional software does not bear full responsibility for making
materials findable, it must make materials obtainable. Metadata harvesters do not extract the
contents of an IR, rather, they route the user to that institutions web site. Thus the material is
housed locally and accessed within the IRs domain, which aids in collecting accurate usage

THE LIBRARYS ROLE IN THE INSTITUTIONAL REPOSITORY

11

statistics. There might be technical issues with this process considering certain file types are
more common than others, and some require unique software to access. Therefore, some level of
integration of uncommon file formats into easily accessible version before publishing may be
necessary. This task is likely to fall to the library staff.
Preservation. Long-term preservation of digital resources is an entire field of
librarianship, however, for the purposes of this discussion suffice it to say an IR must be
designed to be permanent. If users are to entrust librarians as stewards of valuable research,
librarians must find a way to guarantee the longevity and integrity of information. Assigning
DOIs to documents, backing up on institution servers, and staying abreast of new software and
formats will fall to the library.
Rewarding. Scholars must be incentivized to contribute to the IR. The serials crisis
heralds the flattening of the scholarly publishing field, and young researchers especially are open
to new modes of building their CVs. Libraries can often find faculty support in advocating for
tenure reform to include OA and IR publishing. While forcing scholars to participate in an IR
implies its failure as a library service, some believe it to be a viable option, citing Harvard as an
example. If facultys primary concern about their publications is advancing up the academic
rewards chain, the IR must be woven into the fabric of the university system in such a way that it
supports prestige-building. To facilitate this, libraries should offer services that prove publishing
in the IR increases impact factor. The IR could have personalized faculty accounts wherein the
user can manage their materials, view citation counts, and see related publications to find other
scholars working in their field. An email or RSS alert for citations or newly published related
articles could minimize the amount of effort needed from the user.

THE LIBRARYS ROLE IN THE INSTITUTIONAL REPOSITORY

12

Misconceptions about IRs and OA form veritable barriers to participation. Libraries must
make proactive efforts to help researchers understand their options when it comes to intellectual
property rights. Likely it will be difficult to sustain much interest because faculty value impact
above all else, but the option must be there. This is an instance where offerings must be highly
tailored to needs, as understanding the value of self-archiving can entice IR support. If the goal
of IRs is to increase scholarship by increasing access to scholarship, the ability to locate funding
is a core component of that service. Grant writing services, indices of available funding, and
posting funders publishing requirements, are all part of the librarys role in offering the IR as a
comprehensive service.
Summary
The solution to some challenges in IR implementation may be addressed by applying the
question, How can this best serve IR users? If they are to be good investments for library
resources, IRs must not be standalone services but tools to facilitating knowledge creation
(Lankes, 2011) for users. Quoting Joan Bechtel from 1986, Lankes endorses a view of libraries
as centers for conversation and mediators in the conversations of the world (104).
If nothing else, this article has exposed areas ripe for further investigation. The most
intriguing challenge might be finding a substitute for the peer review process, considering much
of the prestige conferred by the journal article is its survival of rigorous refereeing. Arguably,
citation statistics will form a sort of post-publication, democratic peer review, but it is worth
exploring if some measure of quality control could be introduced early on. Lankes (2011)
captures the challenge in asking, Does your repository capture the richness of this [peer review
process] conversation or simply the artifacts? (103).

THE LIBRARYS ROLE IN THE INSTITUTIONAL REPOSITORY

13

While systems and software risk becoming the golden calf of IR study, they require
constant attention. One aspect might be to address the challenge of version control in
collaborative authoring. Could IR software be developed to link with data sharing systems like
SharePoint to help researchers coordinate their composition and writing efforts? Chudnov (2008)
notes it is best not to overprepare, but to [let] users drive what you do[and] build
incrementally. Lankes warns against confusing tools with mission (2011), but systems must
accommodate the range of interactive functionss IRs require.
Another worthy area of study would be assessment and measuring the return on
investment (ROI) of an IR. Citation analyses, impact factor studies, annual reports, and a citation
alert RSS might be necessary efforts. To garner support from both faculty and administration,
some system of validating the ROI must be developed. Assessment alongside predefined goals
for the IR will help librarians secure the means to sustain the service.

THE LIBRARYS ROLE IN THE INSTITUTIONAL REPOSITORY

14

References
Association of Research Libraries (ARL). (2006). SPEC Kit 292: Institutional Repositories.
Washington, DC: Association of Research Libraries.
Crawford, W. (2011). Open access: What you need to know now. Chicago, IL: American Library
Association.
Chudnov. D. (2008). The emperors new repository. Computers in Libraries, 28(9), 39-41.
Gibbons, S. (2004). Establishing an institutional repository. Library technology reports, 40(4), 618. doi: 10.5860.ltr.41n2
Jones, R., Andrew, T., & MacColl, J. (2006). The institutional respository. Oxford, UK: Chandos
Publishing.
Friend, L. (2011). Scholarly communication: Library as content providerDigital projects to
support reference and user services. In D. Zabel (Ed.), Reference reborn: Breathing new
life into public services librarianship (pp. 257-277). Santa Barbara, CA: Libraries
Unlimited.
Furlough, M. (2009). What we talk about when we talk about repositories. Reference & User
Services Quarterly, 49(1), 18-32.
Hswe, P. & Tribone, M. (2014). Hacking user experience in a repository service: ScholarSphere
as a case study. Retrieved from https://scholarsphere.psu.edu/files/6395wb17r
Lankes, D. (2011). The atlas of new librarianship. Cambridge, MA: The MIT Press.
Lewis, D. W. (2012). The inevitability of open access. College and Research Libraries, 73(5),
493-506.
Nariana, R., & Fernandez, L. (2012). Open access publishing: What authors want. College &
Research Libraries. Advance online publication retrieved at
http://crl.acrl.org/content/early/2011/06/10/crl-203.
Salo, D. (2002). Innkeeper at the RoachMotel. Library Trends, 57(2), 98-123. doi:
10.1353/lib.0.0031
Young, J. (2002). Superarchives could hold all scholarly output. The Chronicle of Higher
Education, 48(43), A29.