URLs in this document have been updated. Links enclosed in {curly brackets} have been changed. If a replacement link was located, the new URL was added and the link is active; if a new site could not be identified, the broken link was removed.

Developing an Online Science Journal Collection: A Quick Tool for Assigning Priorities

Anne Christie
Life Sciences Coordinator
anne.christie@orst.edu

Laurel Kristick
Physical Sciences Coordinator
laurel.kristick@orst.edu

Collection Development Department
OSU Libraries
Oregon State University

Abstract

Oregon State University Libraries are committed to increasing electronic access to information. The catalog, myriad databases, and a small collection of e-books are accessible over the Internet. OSU Libraries are now focusing attention on access to primary science journals. Using an existing library liaison network, we conducted an informal e-mail poll of science faculty. We asked them for lists of the "top ten" journals that they would like to access electronically. These faculty lists have been compiled into an Excel spreadsheet for analysis. Other data added to the spreadsheet include the publisher, whether or not the Libraries have a current subscription (print or electronic), cost of the subscription, licensing conditions and the ISI impact factors and rankings. We are using this information to set priorities for the allocation of staff time and available funding to develop the online science journal collection.

Introduction

Oregon State University (OSU) Libraries are committed to the development of an electronic journal collection that provides 24-hour access for all of OSU's faculty and students. These people use library resources from their offices and residence halls and from OSU's agricultural research stations that are scattered across the state. The growing numbers of students enrolled in distance education programs also find an electronic journal collection convenient. Anecdotal evidence indicates that the faculty has become more receptive to electronic access for journals. They expect the library to provide this access although some faculty members are acquiring their own electronic subscriptions. Other reasons for an increased focus on electronic journals include lack of shelf space, reduction in damage to print journals, and reduced need for check-in, reshelving, and binding.

Background

Electronic Collection Development at OSU Libraries

Electronic collection development at OSU Libraries began in the mid-1980s with the acquisition of bibliographic databases on CD-ROM and the establishment of a very successful CD network (Butcher & Scott 1990). With the advent of the World Wide Web, the Collection Development Department's focus has been to convert the CD database collection to web versions. A database wish list was developed in conjunction with the subject librarians. The conversion to web-based databases has been achieved with a couple of notable exceptions (INSPEC and Web of Science).

OSU Libraries have also been developing a collection of electronic journals in a piecemeal fashion. The first paid subscription to an electronic journal was to the Journal of Biological Chemistry in 1996. Since then the Libraries have added electronic access to journals through the purchase of publisher packages (e.g., Project Muse and the American Chemical Society), titles that are "free with print" (e.g., American Institute of Physics and the American Mathematical Society), and aggregator databases from commercial vendors (e.g., EBSCO and the Gale Group). As the Collection Development Department became aware of "free with print" titles, links were placed in catalog records and in an alphabetical electronic journals list on the Libraries' web site. When some publishers began to charge for electronic access, OSU Libraries were forced to give up the web versions of many valuable titles (e.g., American Society for Microbiology publications).

Development of an electronic journal collection is expensive both in terms of the dollar cost of subscriptions and in the use of staff resources to negotiate licenses, verify URLs and provide access via cataloging and web page links. Libraries must deal directly with publishers making it a time consuming process. Subscription agents have yet to define and establish their role in this arena. Charles Hamaker described these issues a few years ago (Hamaker 1997) and little has changed since then. Given the costs of acquiring and processing electronic resources, the Collection Development Department decided to concentrate on those journals most in demand by faculty.

OSU Profile

Oregon State University offers a wide array of undergraduate majors but faculty members and graduate programs are heavily concentrated in the sciences (Table 1).

Table 1. Students and Faculty at Oregon State University
College	Undergraduates	% of Total	Graduates	% of Total	Faculty	% of Total
Business	1,796	13.04%	75	2.50%	41	3.25%
Graduate School*	0	0.00%	601	20.04%	N/A
Home Economics & Education	914	6.64%	413	13.77%	131	10.38%
Liberal Arts	2,608	18.94%	82	2.73%	210	16.64%
University Exploratory Studies*	998	7.25%	0	0.00%	N/A
Social Sciences & Humanities Subtotal	6,316	45.86%	1,171	39.05%	382	30.27%
Agriculture	1,072	7.78%	290	9.67%	283	22.42%
Engineering	2,890	20.98%	448	14.94%	127	10.06%
Forestry	378	2.74%	146	4.87%	100	7.92%
Health & Human Performance	872	6.33%	130	4.33%	45	3.57%
Oceanic & Atmospheric Sciences	0	0.00%	78	2.60%	65	5.15%
Pharmacy	231	1.68%	189	6.30%	32	2.54%
Science	2,013	14.62%	470	15.67%	193	15.29%
Veterinary	0	0.00%	77	2.57%	35	2.77%
Science Subtotal	7,456	54.14%	1,828	60.95%	880	69.73%
Total	13,772	100%	2,999	100%	1,262	100%
*Some of these students may be science majors; not counted in science subtotal

Seventy percent of the faculty is in the sciences and, according to statistics from OSU's Research Office, 92% of the research funds generated from outside sources in 1999/2000 was awarded to science faculty. Despite its relatively small size and the lack of a medical school, OSU has an impressive record in attracting outside research grants and is currently ranked 64th in federal funding (Chronicle of Higher Education March 16, 2001). OSU Libraries' current materials budget is approximately $4 million; serials account for $2.6 million and 80% of this is spent on science serials. There is little flexibility in the budget for decreasing the amount spent to support the social sciences and humanities collection. A recent citation study conducted by Hurd et al. (1999) at the University of Illinois, Chicago showed that for faculty in molecular biology 6% of the journals used accounted for 56% of articles cited. These factors prompted us to focus on science journals. We reasoned that we would make the best use of the serials budget by adding electronic access to journals already well used and highly cited by OSU faculty rather than by applying available money entirely to the purchase of new print subscriptions.

Method

We felt that the simplest approach would be best. We would ask faculty what journals they wanted. We had used a similar technique in a major review of our print journals collection in 1998, adapting the method used at Louisiana State University a few years previously (Kleiner & Hamaker 1997).

We conducted an informal e-mail poll of the science faculty in January 2001. The science librarians sent e-mail either to their departmental library representatives or directly to faculty e-mail lists. The message asked faculty members to submit a list of the "top ten" journals to which they would like to have electronic access (Figure 1).

Figure 1: Sample E-mail Message The library is investigating the possibility of providing increased electronic access to journals. In most cases we would have to pay extra for our paper subscription to a journal in order to have a site license to give you access to the electronic version as well. The advantage of electronic access is that it is available off-site (e.g., in faculty offices) and if the paper copy happens to be missing from the shelves electronic access is still available. We are polling faculty to determine which journals it would make "economic" sense to try and obtain electronically. Please send me your "Top Ten" list of journals that you would like to access electronically -- these should be journals which you use often and refer your students to regularly. In some cases, an electronic version may not yet be available -- don't let that stop you from suggesting a title. Please reply by Friday, February 2.

We did not ask faculty to check whether the print journal was in our collection or whether electronic access was already available before submitting their lists. We also did not ask them to verify titles on their lists. We wanted to make this exercise as easy as possible for them. We gave faculty a two-week deadline to reply. Some librarians sent a reminder message to departments where there was little response. The reminders generated an immediate reaction and a flurry of lists.

Gathering the Data

The data were compiled into an Excel spreadsheet containing the elements listed in Table 2.

Table 2: Elements of Spreadsheet Data
Column	Element
A	Title of journal
B	Publisher
C	Number of requests per title
D	Departments requesting
E	OSU print subscription
F	OSU electronic access linked from OASIS
G	OSU electronic access linked from e-journals list
H	Cost of electronic access - free with print, electronic not free,
I	Cost of print subscription
J	Cost of print + electronic
K	Comments (licensing conditions, etc.)
L	ISI category
M	ISI ranking within category
N	Number of journals within ISI category
O	ISI impact factor

Creating the spreadsheet and adding titles as the e-mail lists were received took approximately 15 hours spread over three weeks. Two additional hours were needed for bibliographic verification and consolidating duplicate entries.

OSU Libraries' serials vendor's database was used to determine publishers and availability of electronic access. Requested titles were checked against the online catalog and the alphabetical listing of electronic journals. This checking took 15 hours (spread among four people) and a further eight hours for the Collection Development Assistant to enter the information into the spreadsheet.

The 1999 ISI Science Citation Index Journal Citation Reports was used to gather the ISI ranking information. This took the two of us working together about six hours to collect. Eight hours were needed to add the data to the spreadsheet.

Compiling the data required a total of about 54 hours. The data can be sorted in the spreadsheet in a variety of ways (e.g., by number of requests per title, by publisher, by impact factor). For a relatively small investment of time, we have created a useful tool to help in making decisions about which electronic titles to pursue.

Review of Results

Number of faculty responses

The response by faculty to the e-mail poll was gratifying. The prospect of electronic access to journals engaged the faculty (204 lists received) more than any recent library issue including the last journals cancellation which triggered approximately 100 contacts with the Libraries. This response confirmed our sense that faculty members are much more interested in electronic access to journals now than they were when we conducted a substantial analysis of the serials collection three years ago.

The bulk of the responses were from faculty in the biological sciences. This was no surprise since the campus already has access to many titles in the physical sciences (e.g., American Physical Society and the Royal Society of Chemistry), making readily available respected journals that the faculty both publish in and cite. The journals listed by biological sciences faculty reflect areas in which OSU has strong research programs: forestry, fisheries and marine biology, ecology, and molecular biology.

Number of journals and requests

The number of journals requested (711) was much larger than we had expected despite the range of departments and faculty that we polled. OSU Libraries has current print subscriptions to 560 of the requested titles and electronic access to 103 of these. Of the 151 titles not owned, 23 were requested by more than one person. The Collection Development Department will consider purchasing subscriptions to some of these titles next year.

Of these 711 requested titles, 61 were requested by five or more faculty, and a further 152 were requested by three or more faculty. This is evidence that our strategy of focusing on electronic access to key scientific journals was on target. The most requested journals (ten or more requests) are listed in Table 3.

Table 3. Most Requested Titles
Title	# Requests	OSU Electronic	Availability	ISI Rank	ISI Impact
Nature	61	no	Additional cost for online	1	29.491
Science	47	yes	Additional cost for online	2	24.595
Ecology	38	no	Additional cost for online	8	3.573
Cell	24	no	Additional cost for online	2	36.242
Oecologia	21	no	Free with print	22	2.159
Ecological Applications	20	no	Additional cost for online	15	2.784
Canadian Journal of Forest Research	19	no	Additional cost for online	6	1.058
Proceedings of the National Academy of Sciences	19	yes	Free with print	3	10.260
Marine Ecology Progress Series	18	yes	Free with print	6	2.781
Journal of Geophysical Research	14	no	Not available online	6	2.781
Ecological Monographs	13	no	Additional cost for online	4	4.447
Forest Science	13	no	Not available online	7	1.034
Journal of Experimental Marine Biology & Ecology	13	yes	Free with print	21	1.348
Trends in Ecology and Evolution	12	no	Additional cost for online	2	7.621
American Naturalist	11	no	Free with print	6	3.928
Forest Ecology and Management	11	no	Additional cost for online	9	0.962
Marine Biology	11	no	Free with print	16	1.534
Oikos	11	no	Additional cost for online	17	2.566
Water Resources Research	11	no	Not available online	9	2.061
Canadian Journal of Fisheries and Aquatic Science	10	no	Additional cost for online	2	1.963
Journal of Biological Chemistry	10	yes	Additional cost for online	22	7.666

Of these journals, OSU Libraries provide electronic access to five titles. For two of these five titles, electronic access had been purchased before the e-mail poll was conducted.

Publishers

The requested journals are produced by 195 different publishers. The need to work with such a large number of publishers to establish electronic access to journals is a daunting task. Two publishers, however, account for 21% of the requested titles; 118 journals are Elsevier titles and 33 are Academic Press titles. This percentage does not include titles from imprints owned by Elsevier.

ISI Rankings

Twenty-two of the requested journals are ranked Number 1 in their ISI category (Table 4). Only one of these titles made the list of most requested titles (Table 3). These titles, however, are important to OSU's research programs. None of them is contained in large expensive electronic packages and thus can be acquired individually. Three of these titles are not available electronically. Of these 22 journals, OSU Libraries have 19 current subscriptions in print. Ten of these 19 were requested by more than one person. Of the three journals not owned, only one title is available electronically.

Table 4. Requested Titles Ranked #1 in Science Citation Index Journal Citation Reports
Title	OSU Subscription	Electronic Availability	# Requests	ISI Category	Impact Factor
Advanced Materials	yes	extra cost	1	Materials Science	5.415
Annual Review of Biochemistry	yes	extra cost	2	Biochemistry & Microbiology	37.111
Annual Review of Earth and Planetary Sciences	yes	extra cost	1	Geosciences	5.333
Annual Review of Entomology	yes	extra cost	2	Entomology	5.759
Annual Review of Pharmacology and Toxicology	yes	extra cost	1	Pharmacy	21.175
Endocrine Reviews	yes	extra cost	1	Endocrinology	20.250
Global Biogeochemical Cycles	yes	not available	6	Environmental Science	4.309
Health Affairs	yes	free with print	1	Health	5.079
Journal of Avian Biology	no	extra cost	1	Ornithology	1.532
Journal of Dairy Science	yes	free with print	3	Agriculture-Dairy & Animal	1.674
Journal of Quality Technology	yes	not available	2	Engineering-Industrial	1.133
Mass Spectrometry Reviews	no	free with print	1	Spectroscopy	6.885
Microbiology and Molecular Biology Reviews	yes	extra cost	1	Microbiology	17.773
Nature	yes	extra cost	61	Multidisciplinary Science	29.491
Nature Genetics	yes	not available	6	Genetics	30.693
Neural Computation (MIT Press)	no	free with print	1	Computer Science-Artificial Intelligence	2.829
New England Journal of Medicine	yes	free with print	5	Medicine-General	28.857
Operations Research	yes	free with print	1	Operations Research	1.268
Paleoceanography	yes	not available	1	Oceanography	3.967
Reviews in Fish Biology and Fisheries	yes	extra cost	1	Fish & Wildlife	4.444
Soil Science Society of America Journal	yes	free with print	7	Agriculture-Soil Science	1.604
Tree Physiology	yes	free with print	7	Forestry	2.042

Response to the Results

Clarifying Access to Existing Titles

Comments included with the responses to e-mail poll confirmed a long held suspicion. Faculty were often unaware that OSU Libraries provide access to electronic journals they are interested in, possibly because this information is in several places. For the 103 journals for which there is currently electronic access, 56 were listed in both the online catalog and the alphabetical electronic journals list, 16 titles were in the online catalog only, and 31 in the alphabetical list only. These latter 47 titles will be treated in both places.

Creating Criteria for Adding Electronic Access to Journals

The Collection Development Department has developed criteria for the addition of electronic journals to the collection. After a title has been selected based on content and the electronic version evaluated for coverage and currency, priority is given to titles by using the following criteria:

• Print plus free online version
• Online only and free
• Print or online, either one paid
• Print plus paid online version.

Increasing Electronic Access to Journals

Using the information obtained from the e-mail poll, the Collection Development Department has begun to work on expanding access to electronic journals. The first three steps are described below.

Adding "Free With Print" Titles

Of the 711 requested titles, 75 are available free with a print subscription but no links have been established in the online catalog or the alphabetical electronic journals list. The 75 titles have been sorted by publisher. Permission will be sought to add links to these titles as well as to any other titles from these publishers that are owned by OSU Libraries.

Subscribing to BioOne

OSU Libraries plan to subscribe to BioOne. The BioOne initiative, supported by several sponsoring organizations, will provide electronic access to a number of important scholarly journals published by professional societies. For OSU, BioOne will provide electronic access to 17 titles requested in the e-mail poll as well as to a further 19 titles owned in print now but for which electronic access was not requested by any faculty in this poll.

Adding "Most Requested" Paid Titles

We have sorted the list by the number of requests per title and by publisher and have begun gathering information on the cost of electronic access. Provided that the Collection Development Department is satisfied with the coverage and currency of the online versions, these titles will be added to the collection as funds permit. Although OSU Libraries will probably not move forward as boldly as Drexel University has done in developing an electronic journal collection (Montgomery 2000), the Collection Development Department hopes to substantially increase access to electronic journals in the future. It is likely that in order to contain costs we will need to investigate the feasibility of providing some journals in electronic format only.

What About the "Big Deal?"

As science librarians, we are concerned about the ability of the OSU faculty to remain competitive when they have less access to journal literature than faculty in those institutions whose libraries have subscribed to the "Big Deal." Kenneth Frazier recently exhorted libraries not to give in to the "Big Deal" which he describes as "an online aggregation of journals that publishers offer as a one price, one size fits all package" (Frazier 2001). Very limited cancellation rights are allowed in subsequent years and unimportant journals may be bundled in the electronic package. For these publishers, there is no way for a library to obtain access to individual titles contained in their packages.

OSU Libraries have explored the "Big Deal" as a means of increasing electronic access to science journals with little success. Oregon State University is a member of Orbis, a consortium of Oregon state-supported university libraries and private college libraries in Oregon and Washington. In 1998, Orbis pursued a consortial subscription to Academic IDEAL but was unable to negotiate a price that members could support. As the bulk of OSU Libraries' science journal subscriptions are held with one major publisher, Elsevier, the Libraries investigated a consortial subscription in cooperation with the largest academic libraries in Oregon. The goal was to reduce the extensive overlap between print collections and potentially increase access to titles that none of these libraries currently own. So far, this negotiation has not been fruitful.

The concept of the "Big Deal" has important implications for the further development of OSU Libraries' electronic journal collection. At least 21% of the requested titles are tied into these "Big Deal" packages. Should OSU Libraries hold out against the "Big Deal" and the resulting financial ramifications in anticipation of changes in the scholarly communication process? Oregon State University firmly supports alternative avenues for scholarly and electronic journal publishing and is a foundation member of the {Scholarly Publishing and Academic Resources Coalition} (SPARC). OSU Libraries has subscribed to SPARC sponsored journals and the next purchase will be BioOne. However, by not taking on the "Big Deal," are we hampering our faculty and their ability to succeed? Or, by holding out, are we staying in control of local collection management decisions and furthering the cause for new directions in scholarly publishing?

Conclusion

With significant annual budget increases unlikely, OSU Libraries will continue to develop a science electronic journal collection by focusing on the data collected in the e-mail poll. The survey has supplied useful and timely information for an effective plan to improve electronic access to journals needed by the faculty for research and teaching. We believe that other libraries can apply a similar technique for analyzing the needs of their faculty and making decisions about electronic journal purchases.

References

Butcher, K.S. & Scott, S.R. 1990. Effects of CD-ROM in a University Library. Journal of Educational Media & Library Sciences 27(3):257-269.

Frazier, K. 2001. The Librarians Dilemma: Contemplating the Costs of the "Big Deal." D-Lib Magazine 79(1) Available: http://www.dlib.org/dlib/march01/frazier/03frazier.html [March 19, 2001].

Hamaker, C. 1997. Chaos - Journals Electronic Style. Against the Grain 9(6):90-91.

Hurd, J.M., Blecic, D.D & Vishwanataham, R. 1999. Information Use by Molecular Biologists: Implications for Library Collections and Services. College and Research Libraries 60(1):31-43.

Kleiner, J.P. & Hamaker, C.A. 1997. Libraries 2000: Transforming Libraries Using Document Delivery, Needs Assessment, and Networked Resources. College and Research Libraries 58(4):355-374.

Montgomery, C.H. 2000. Measuring the Impact of an Electronic Journal Collection on Library Costs. D-Lib Magazine 6(10) Available: http://www.dlib.org/dlib/october00/montgomery/10montgomery.html [April 3, 2001].

Top Institutions in Federal Research-and-Development Expenditures, 1998 and 1999. 2001. Chronicle of Higher Education (March 16) 47(27):A26.

We thank Sue Goodson and Judith Langridge in Collection Development for their work on this project. Our colleagues, Bonnie Allen, Richard Brumley and Janet Webster, provided valuable editorial advice.

Previous

Contents

Next