After a lengthy review process, the University of Toronto Libraries (UTL) have decided to discontinue several databases provided by ProQuest through a package known as ProQuest Central. This package, which aggregates content from many publishers, was originally subscribed to before direct-from-publisher subscriptions became the norm. Now that most of our ejournals come directly from publishers, aggegators like ProQuest Central add little unique content to our holdings. After considering the degree of overlap with other holdings, as well as usage and relevance to the university's teaching and research activities, the review group found that twenty-six of the thirty databases contain little unique content, show low usage by faculty and students, or have little relevant content and therefore do not seem to merit further expenditure by the Library. Instead, the Library intends to redeploy the funds to acquire new and unique content. Faculty and students are encouraged to identify any e-journal titles which will no longer be accessible through ProQuest Central so that we can investigate their reinstatement. Feedback from faculty and students is an important part of this process and may be submitted to firstname.lastname@example.org.
In its 2013-2018 Strategic Plan, the University of Toronto's Central Libraries express a commitment to expanding traditional practices to build collections that serve the needs of faculty and students at the university and support emerging directions in scholarship. The Libraries are committed to demonstrating effective and responsible stewardship of the library's acquisitions budget by using evidence to inform collection development decisions. With this in mind, the Libraries' Collection Development Management Committee recently conducted a review of one of our online database packages - ProQuest Central (PQC). PQC is a subset of our subscription from ProQuest. This package of 30 databases was chosen for review because it is an aggregator package and a big ticket item. The analysis undertaken and the findings of this review are outlined below.
We believe that scholarly content is the highest priority for researchers at the University of Toronto. In most cases, tools which "point" to content are useful but are now a lower priority for allocation of collection funds. Based on this premise, and because demand for access to new scholarly content continues, we plan to undertake analysis of individual products with the goal of redirecting money currently spent on indexing/abstracting and aggregating products towards new scholarly full-text content. We will continue to consult with colleagues, faculty and other users to gather input through this process.
What are aggregators?
To better understand the Library’s analysis of the PQC package under discussion, it is important to first understand how aggregators function and how we have used them in the past. (For more on aggregators please see this page.)
An aggregator is a company that aggregates eJournals from multiple publishers into one package. As the Web and e-journal publishing first developed, journal aggregators were advantageous because they gave the Library online access to electronic articles from a wide variety of publishers to which we did not at that time subscribe. Journal aggregators often include indexes that make it easier to search each bundle of journals. UTL now purchases most of our journal content directly from journal publishers, making most of the content within aggregator packages redundant. Further, multiple discovery tools, such as Summon, Google Scholar, Web of Knowledge, Scopus, and other online subject indexes reduce the need for a search engine devoted solely to specific e-journal packages.
Journal aggregators have several significant shortcomings:
- For large libraries like ours, there is often significant overlap between the journals offered through aggregator packages and journals we subscribe to directly from the publisher. This means that we are often paying for the same content twice.
- Aggregators often apply embargoes on recent content. This virtually eliminates any aggregator package from being the sole source of the most recently published content on a specific topic.
- They move titles in and out of their packages at their own discretion and without informing subscribing libraries in advance.
- They sometimes omit information which accompanies research articles, such as the data upon which conclusions are based or other supplementary materials.
- A significant number of the unique titles they provide are not peer-reviewed, or are otherwise of limited academic appeal to faculty and students.
- The proliferation of access points to the same title, but with different coverage dates and embargo conditions has meant that users searching library catalogues are often presented with more than one link to a particular journal. This is particularly frustrating for those searching for current content as the aggregator link will often lead the user to a record stating that the article they were seeking was unavailable due to an embargo.
- Finally, using the search interface in an aggregator does not result in a comprehensive search of journal literature on a topic since it is only searching across that aggregator’s package.
As a 2008 study of academic librarians’ attitudes to aggregator packages states, “While aggregator services offer extensive breadth and depth of coverage as well as valuable functionalities, they do not necessarily constitute a sustainable source of information for libraries.” (see http://www.pcgplus.com/wp-content/uploads/2013/03/aggregators_librarians_perspective.pdf )
Description of the package
The PQC package consists of the 30 databases below. Each of the databases comes with a bundle of aggregated content, including some indexed eJournals and some full-text eJournals. To see the title list associated with each database, consult the PQC full title list in Excel. Please note that PQC does not encompass all of UTL’s ProQuest subscription and there are many other ProQuest databases that were not part of this review. UTL originally subscribed to PQC more than a decade ago, when the package was offered to us with a bundle discount. Since then, UTL has renewed it annually and it is now several hundred thousand dollars to renew for 2016-2017.
As of June 2015, PQC contained 21,935 titles - 16,717 full-text and 5,218 abstracts only. This is a 'point in time' list because the title list of an aggregated package is subject to change. See explanation of aggregators above.
Of the 16,717 titles with full-text coverage, 2,413 are subject to embargo. The vast majority of these (77%) have embargoes on the last 12 months of content.
Databases in blue will be renewed.
ProQuest Nursing & Allied Health Source
ProQuest Health Management
Canadian Newsstand Complete
ProQuest Social Science Journals
ProQuest Science Journals
ProQuest Health and Medical Complete
Accounting & Tax Periodicals
ProQuest Military Collection
ProQuest Family Health
Banking Information Source
ProQuest Political Science
Career and Technical Education
ProQuest Health and Medical Complete
Hoover's Company Records
ProQuest Psychology Journals
ProQuest Criminal Justice
ProQuest Biology Journals
ProQuest Education Journals
ProQuest Asian Business and Reference
ProQuest European Business
The review process
A working group of 15 librarians representing various UTL units and subject areas was tasked with reviewing the package. The group worked for several months to develop and systematically carry out a review process based on the following criteria:
An analysis was carried out to determine the degree of overlap between PQC and the entirety of the libraries' electronic holdings. This analysis revealed that of the 21,935 titles available from PQC (full title list in Excel), only 1809 full-text journal or periodical titles (8%) were unique to this product and not available to us through other sources.
COUNTER standardized usage statistics (JR1) for unique titles were retrieved from Proquest and reviewed by selectors to determine which of the unique titles are heavily used by the University of Toronto community.
Relevance and value (qualitatively assessed by the subject librarian(s))
Selectors and subject specialists examined the unique content along with its usage statistics to determine whether the content was likely to be relevant to the research and teaching activities of the university. In many cases, although a title was unique, it was deemed non-essential because it was no longer being published,non-scholarly, obscure or on a subject matter that is not studied at the university. This careful scrutiny of the unique titles by members of the working group (who were often the subject selectors in the area they reviewed) was a check put in place to ensure that no core titles were missed and that unique or valuable titles were flagged for further consideration even if they had relatively low use.
Discovery and features
Members of the group also considered whether the database itself (and not just the content available within it) should be kept. In some cases, a database offered unique search capabilities and this was considered as part of the review process.
The recommendation of the PQC Review working group is to renew ABI/Inform, CBCA Complete, ProQuest Newsstand and Canadian Newsstand Complete and discontinue the remaining 25 databases.
The recommendation is based on the finding that approximately 92% of the content available from PQC is also available from other, more reliable sources. For those titles that were unique to PQC, some were embargoed or had limited coverage and for some subjects, the trade and non-scholarly publications that were unique were not deemed necessary.
While the focus of the review was on unique titles, for some databases other factors such as unique discovery or access features were also considered. For example, more than half of the content in the two Newsstand databases did not have associated ISSN so these products required a different approach. For ABI/Inform, the discontinued subscriptions of the other business subject databases meant that the unique content rose significantly. For CBCA, the trade publications covered are not necessarily available through other aggregators and due to individual subscriptions, may not be accessible to search through Summon, the Libraries' search interface, meaning that students could have difficulty both finding and searching them.
Please see the following table for the percentage of unique titles by database and brief review comments: PQC table.docx
Furthermore, as we found significant cost-saving opportunity in this case, the group believes that it is worthwhile to continue to review the content of our other aggregated products.
The digital world has brought new opportunities for research, study and publishing and so the libraries' priorities for collection development must evolve. While aggregated journal packages were a useful model of acquisitions in the early 21st century, their usefulness for a library with a large collection of subscription journals has diminished and as a result, we plan to review our aggregated databases with the goal of redirecting funds towards the new content and resource types. It is also important to note that while aggregated packages like PQC may be discontinued, valuable content they included can be reinstated through a direct subscription. Please alert us to any important lost content either through email@example.com or by contacting the subject selector.
If you have input regarding the review of the PQC package, please send comments to firstname.lastname@example.org.