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.

Aggregated Science: An Examination of Three Multi-Disciplinary Databases

J.B. Hill
Head of Reference
Southeastern Louisiana University
jbhill@selu.edu

Abstract

Multi-disciplinary aggregator databases index, abstract, and aggregate full-text electronic content from a wide variety of publishers on a wide variety of subjects. This study examines the scientific content of three leading aggregator databases: Bell & Howell's Research Library (Core and Sciences modules), Ebsco's Academic Search Elite, and Gale's Expanded Academic ASAP. While these databases are similar in scope, a detailed analysis of titles indicates that each database has relative strengths and weaknesses and that each database has a significant amount of unique full-text content.

Introduction

Aggregator databases vary greatly in size, origin, scope, and price. The only common feature among the different varieties of aggregator databases is that all aggregator databases do one thing: they "aggregate" or collect electronic publications into unique, identifiable and searchable databases.

Aggregator databases may be subject specific (e.g., Proquest Medical Library); they may be publisher specific (e.g., Academic Press' Ideal), or (in the case of association produced databases) they may be both subject and publisher specific (e.g., the ACM Digital Library). Aggregator databases also may be neither publisher nor subject specific.

The publisher-specific aggregator databases are much simpler from a collection management perspective. The title lists are known and relatively stable. For the titles included in the aggregation, complete issues with all of the published articles are the norm. The stability of both the titles and links allows libraries to confidently include these electronic holdings in the library catalog.

More problematic are the large, multi-disciplinary aggregator databases that aggregate electronic content from a wide variety of publishers and associations. They are what Calhoun and Kara (2000) appropriately call "tutti frutti surprise aggregations." Both librarians and users are never completely sure of what is included in the databases at any given moment. The producers of these databases provide title lists, but holdings information may be incomplete. Since the full-text content of these databases depends on the deals that the database producers can make with publishers (who are adopting and modifying their approach to electronic publishing and licensing content), there can be a significant amount of instability of the full-text content. New titles are routinely added; titles that were previously available are pulled from the database entirely or existing full text is left in the database but future issues are no longer made available in full text. Adding to the confusion is the fact that many of the database producers do not provide the option of title-level OPAC links to the specific publications included in their databases. Consequently, it takes a great deal of effort for libraries to integrate these holdings into their catalogs.

With the ever-changing content of many of the aggregators, it is important to periodically reexamine and reevaluate aggregator databases. There are many factors to consider in the evaluation of aggregator databases. Much of the recent literature on aggregator databases deals with providing access to the titles included in the databases. However, a few articles provide some insight on evaluation of the databases. In an article scheduled for an upcoming issue of The Serials Librarian, John Blosser, et. al. provide a model for aggregator evaluation that includes evaluation criteria (In press). Similarly, Scott L. Dennis (1999) provides a good discussion of features that should be expected in aggregator databases. More specifically, Lutishoor Salisbury, et. al. (2000) provide the results of a specific comparison of two aggregator products.

Evaluation of aggregator databases is a time-consuming but necessary undertaking. The information included in this report is a result of my participation in a lengthy evaluation of the aggregator databases conducted by the LALINC Database Selection Advisory Committee between August 2000 and April 2001. The LALINC Database Selection Advisory Committee is composed of librarians from Louisiana academic libraries who advise Louisiana's LOUIS Consortium on database selection for the 30 academic library members of LALINC (the Louisiana Academic Library Information Network Consortium). In the Committee's comparison of databases, factors considered were

Total number of publications indexed and abstracted

Number of full-text items

Number of peer-reviewed publications

Ability to easily customize searches (e.g. only full-text, peer-reviewed)

Years of backfiles

Ability to search multiple databases simultaneously

Frequency of updates

Title additions/deletions and method of notification

Server response time

Availability and usefulness of help screens

Ability to share full-text content between databases

Ability to link to additional electronic journals (Awagain 2001).

As part of the evaluation, committee members were assigned subject areas to study in the different databases offered by Bell & Howell (Proquest), Ebsco, and Gale (Infotrac). As a member of the committee, I undertook a comparison of database content for all areas of science and technology with the exception of medicine, nursing, and allied health.

Methodology

The methodology of this comparison was to examine the indexing, abstracting and full-text content of science and technology titles as they appeared in Bell & Howell's Research Library (Core and Sciences modules), Ebsco's Academic Search Elite and Gale's Expanded Academic ASAP in February of 2001. Title lists were downloaded from the database producers' Web sites and merged into a Microsoft Excel spreadsheet and then exported to a Microsoft Access database. Although each database's title list varies somewhat in content, the title lists generally include title, ISSN and year ranges for indexing and full-text.

Science titles were identified through the careful examination of the complete title lists. A subject code was assigned to each title based on the primary subject of the journal's content. Broad subject divisions identified were General Science, Aerospace, Agriculture, Astronomy, Biology, Chemistry, Computer Science, Engineering, Environmental Studies, Geology, Mathematics, and Physics. A second subject code was assigned as needed for some subjects, e.g. the Engineering divisions of Electrical, Civil, and Mechanical Engineering.

Journal Citation Reports was consulted and 1999 impact factors were entered. Impact factors are a measure of the relative importance of journals and are calculated by dividing the number of citations to a journal in a two-year period by the number of articles published in those years. The higher the impact factor is, the greater the relevance of a title within its discipline. Since the Institute for Scientific Information (ISI) only determines impact factors for those titles that are indexed in Science Citation Index, many of the titles in the aggregator databases did not have impact factors.

Ulrich's International Periodicals Directory (ulrichsweb.com) was consulted and a refereed on non-refereed status was entered for each title. For the most part, the focus of the analysis was on the scholarly titles found in the databases, i.e., those identified as peer-reviewed or having impact factors.

Title Comparison

AE = Academic Elite
EXP = Expanded Academic
RL = Research Library

All Science Titles
Indexing				Full Text				Image
AE	EXP	RL	All	AE	EXP	RL	All	AE	EXP	RL	All
618	551	340	884	277	261	199	526	201	148	173	390

Scholarly Science Titles*
Indexing				Full Text				Image
AE	EXP	RL	All	AE	EXP	RL	All	AE	EXP	RL	All
456	367	220	569	187	118	105	300	160	76	88	246
*Titles that are peer reviewed or have ISI impact factors.

Based on the three databases' title lists, 884 science and technology titles are indexed by one or more of the databases. Of these, 526 titles (59.5%) are available in full text in one or more of the databases.

Surprisingly, there is not a great deal of overlap in the full text included in the databases. Of the full-text titles, only 51 (9.7%) are available in full text in all three databases, while 105 (20%) have full text in two of the databases. The majority of titles, 370 (or 70.3%) have full text in only one database.

While there are a large number of popular and trade science magazines in the databases, the majority of science titles indexed in the databases are scholarly in nature. Of the 884 titles, 569 (or 64%) are peer reviewed or have an ISI assigned impact factor.

Of these scholarly titles, 300 (or 52.7%) have full text available. Once again, there is not a great deal of overlap in the scholarly full text included in the databases. Of these full-text titles, only 24 (8%) are available in full text in all three databases, while 68 (21%) have full text in two of the databases. The majority of titles, 216 (or 71%) have full text in only one database.

Of the three databases, Academic Elite has the largest number of indexed, full-text and full-image science titles. From the identified list of 884 science titles, Academic Elite indexes 618 titles, providing full text from 277 (44.8%) of the titles and page images from 201 (32.1%) of the titles. Of the 569 scholarly titles, Academic Elite indexes 456 titles and provides full text from 187 (41%) of the titles and page images from 160 (35.1%) of the titles.

From the identified list of 884 science titles, Expanded Academic indexes 551 titles and provides full text from 261 (47.4%) of the titles and page images from 148 (26.9%) of the titles. Of the 569 scholarly titles, Expanded Academic indexes 367 titles and provides full text from 118 (32.2%) of the titles and page images from 76 (20.7%) of the titles.

From the identified list of 884 science titles, Research Library indexes 340 titles, providing full text from 199 (58.5%) of the titles and page images from 173 (50.9%) of the titles. Of the 569 scholarly titles, Research Library indexes 220 and provides full text from 105 (47.7%) of the titles and page images from 88 (40.0%) of the titles.

Publisher Comparison

The scientific professional societies are major publishers of scientific literature and the three aggregators do provide indexing for many of the most significant titles. However, the three databases contain very little full-text content from the publications of these societies.

All of the databases have full text from a few of the American Statistical Association titles as well as one or more of the Institute of Industrial Engineers titles. Expanded Academic has full text from six Association for Computing Machinery (ACM) titles; Research Library has two of the ACM titles. Research Library has three full-text titles from the Mathematical Association of America. Academic Elite has full-text content from six American Institute of Physics titles (with a 12-month embargo period). Research Library has full text from seven publications from the American Meteorological Society and seven publications of the National Research Council of Canada.

Although the databases have indexing for the societies' publications, they do not have full text or page images from the publications of the American Chemical Society, American Institute of Aeronautics and Astronautics, American Mathematical Society, American Society of Microbiology, Geological Society of America, and the Institute of Electrical and Electronics Engineers.

Full text is available for only the most general publications of two of the engineering societies. Content from the American Society of Civil Engineers is limited to Civil Engineering (in Academic Elite and Research Library). Content from the American Society of Mechanical Engineers is limited to Mechanical Engineering (in Academic Elite and Research Library).

Expanded Academic has recently begun indexing and providing full-text articles from a number of the state academies of science journals, including those published by the Colorado-Wyoming, Michigan, Mississippi, New Mexico and Ohio academies of science. This gives libraries greater access to scientific literature that has been shown to be relatively less accessible and less widely indexed (Hill & Madarash-Hill 2000).

In addition to the limited full text available from professional association publications, there is also very little full-text content from some of the prominent science publishers. There is very little indexing and no full-text content from Academic Press and Kluwer titles. Only one Pergamon title and four Elsevier titles are available in full text (in Academic Elite). There are also relatively few titles available from Gordon & Breach, Springer Verlag, and Taylor & Francis. Many of the publishers' titles that are available in full text are restricted by three to twelve month embargo periods (i.e., delays after publication).

Embargo periods are a recent attempt by some publishers to provide content to aggregators while minimizing the risk of a loss of revenue through their own subscriptions and e-journal packages. Of the three databases studied, only Academic Elite includes science titles with embargo periods. The other database producers are not providing any titles with identified embargo periods. Publishers with content in Academic Elite that require embargo periods include:

American Institute of Physics (12 months)

Blackwell (12 months)

Carfax Publishing (3 months)

Gordon & Breach (6 months)

Oxford University Press (12 months)

Springer Verlag (12 months)

Taylor & Francis (12 months)

Full-text content from these publishers is not available in Expanded Academic or Research Library.

Subject Comparison

Within the scientific disciplines, each of the databases has strengths and weaknesses. Below are some observations about the scholarly subject content of the databases. The numbers listed in each subject and the titles mentioned are intended to be indicative rather than precise and exhaustive. Subjects were assigned for grouping purposes, but the assignments were sometimes arbitrary due to the interdisciplinary nature of the literature. Consequently, the numbers reported in each category may be smaller than what they could be. For example, a title that was assigned the subject "Environmental Studies" may also be relevant to geologists and civil engineers.

General Science

Indexing			Full Text
AE	EXP	RL	AE	EXP	RL
34	38	25	18	22	14

As would be expected, all three of the databases index, abstract and provide full text from a significant number of the general, multi-disciplinary science titles. There are a large number of popular science magazines, e.g. Discover, Popular Science, Popular Mechanics, and Poptronics. There are also the well known, relatively high impact general science titles, e.g. Nature, Science, Proceedings of the National Academy of Science, Scientific American, American Scientist, and New Scientist.

All three databases have some full text of the high impact Science. Expanded Academic has the greatest range (1984 to present); Research Library has full text only up to 1999. Academic Elite has only 1997 to present. Academic Elite and Research Library have full-text for Scientific American. Research Library holdings begin in 2000 whereas Academic Elite has the title from 1995 to present. Research Library has full text from American Scientist. Expanded Academic has full text from Scientist and New Scientist.

Aerospace

Indexing			Full Text
AE	EXP	RL	AE	EXP	RL
9	7	3	1	0	1

All three databases index Journal of Aircraft, Aerospace America, and Aviation Week & Space Technology. Although Academic Elite indexes more titles, Expanded Academic indexes more back years of the seven titles indexed by both; Aviation Week & Space Technology is indexed equally (1984 to present). For Aerospace America, Academic Elite indexes two more years (1980 to present). For the other five titles indexed by Academic Elite and Expanded Academic, Expanded Academic indexes four more years. Expanded Academic indexes further back than Research Library for two of the three titles. Research Library does not index AIAA Journal, Progress, Aerospace Science, or Journal of Aerospace Engineering, all relatively high impact journals.

Academic Elite provides full text from Space Communications, and Research Library provides full text from Aviation Week and Space Technology. Although Space Communications has a higher impact factor, Aviation Week and Space Technology may be more beneficial to most collections.

Agriculture

Indexing			Full Text
AE	EXP	RL	AE	EXP	RL
10	7	6	6	5	1

All three databases provide very limited indexing and full-text coverage of agricultural titles. Of the three databases, Academic Elite has a bit more coverage. All three databases have the full text of Agricultural Research. Both Academic Elite and Expanded Academic have American Journal of Agricultural Economics. Academic Elite has full-text of European Journal of Soil Science, Acta Agriculturae Scandinavica: A & B, Landscape Research, and Archives of Agronomy and Soil Science. Expanded Academic provides full text for Australian Journal of Soil Research, Crop Science, and Agricultural History.

Astronomy and Astrophysics

Indexing			Full Text
AE	EXP	RL	AE	EXP	RL
5	1	0	5	0	0

The three databases have very little astronomy/astrophysics coverage. Research Library indexes no titles. Expanded Academic indexes one title, the high impact Annual Review of Astronomy and Astrophysics, but only 1998 to present. Academic Elite indexes and provides full text for recent volumes of five titles: Astronomy and Astrophysics Review, Astronomy and Geophysics, Astronomical and Astrophysical Transactions, Astrophysical Letters and Communications, and Astrophysics and Space Physics Review.

Biology

Indexing			Full Text
AE	EXP	RL	AE	EXP	RL
98	66	62	43	23	25

Overall, Academic Elite provides more coverage of the biological literature, indexing 98 titles and providing full text from 43. Expanded Academic indexes 66 titles, providing full text from 23. Research Library indexes 62 titles and provides full text from 25 titles.

Within biology, each database has different strengths. Research Library is strong in animal sciences (e.g., zoology, marine science and veterinary medicine). Research Library indexes 20 titles and provides full text of 12 titles. Notable full-text titles are American Zoologist, Canadian Journal of Fisheries and Aquatic Science, Auk, Condor, Journal of Animal Science, Canadian Journal of Zoology, and Journal of Mammalogy.

Academic Elite indexes 23 titles and provides full text from eight titles. Full-text titles include Avian Pathology, South African Journal of Zoology, South African Journal of Wildlife Research, and Acta Agriculturae Scandinavia A. Expanded Academic indexes 12 titles and provides full text for American Zoologist.

Academic Elite covers a few more titles in the plant sciences as it indexes nine titles offering full text from five titles. Full-text titles include Plant Molecular Biology Reporter, International Journal of Plant Sciences, South African Journal of Botany, Acta Agriculturae Scandinavia B, and Landscape Research.

Expanded Academic indexes six titles while providing full text from four titles. Full-text titles include Botanical Review, International Review of Plant Science, Botanical Gazette (one issue only), and Crop Science. Research Library indexes six titles and provides full text from Canadian Journal of Botany.

Each of the databases provides limited coverage of microbiology, cell biology, and genetics. For microbiology, all three databases index four to six titles, including high impact titles such as Microbiology and Molecular Biology Reviews and Molecular Biology as well as a few other titles. As for full text, Academic Elite provides full text of Molecular Microbiology. Expanded Academic provides full text from Annual Review of Microbiology. Research Library provides full text from Canadian Journal of Microbiology.

Other high impact, full-text biology titles are included in the databases. Each has full text from Bioscience. Other notable full-text titles for Academic Elite include: Genes to Cells, Quarterly Review of Biology, Journal of Biological Rhythms, and American Naturalist. Expanded Academic has full text from Annual Review of Genetics, Gene Therapy Weekly, Quarterly Review of Biology, and Evolution. Research Library has full text from Biochemistry and Cell Biology, Systematic Biology, and Biophysical Journal.

Chemistry/Chemical Engineering

Indexing			Full Text
AE	EXP	RL	AE	EXP	RL
23	27	12	6	7	3

Of the three databases, Research Library indexes fewer titles in chemistry. However, Research Library has two unique full-text titles: Canadian Journal of Chemistry and Journal of Chemical Education.

Of the 23 titles indexed by Academic Elite and the 27 indexed by Expanded Academic, 15 (including 10 American Chemical Society titles) are indexed by both databases. For almost all titles, Expanded Academic indexes more years (from one to twelve years, with most being around four years more). For the uniquely indexed titles, Expanded Academic has higher impact journals including the high impact Annual Review of Biochemistry and Annual Review of Physical Chemistry. Expanded Academic is stronger in physical chemistry, indexing the Annual Review and Journal of Physical Chemistry A&B and in polymer chemistry, indexing Polymer Engineering and Science and Plastics Engineering.

Of the titles in full text in Expanded Academic and Academic Elite, only Chemical Week is common to both databases. Expanded Academic is stronger in chemical engineering as its six titles are focused on chemical engineering. Two of the titles are polymer science (Polymer Engineering & Science and Plastics Engineering). Academic Elite is focused more toward chemistry with titles such as International Reviews in Physical Chemistry, European Journal of Biochemistry, and South Africa Journal of Chemistry. While Academic Elite has higher impact titles in full-text, Expanded Academic has deeper backfiles of the titles, as Academic Elite has only very recent years with some titles having a twelve month embargo period.

Computer Science

Indexing			Full Text
AE	EXP	RL	AE	EXP	RL
68	46	27	30	20	11
IEEE Computer Society Titles
Indexing			Full Text
AE	EXP	RL	AE	EXP	RL
20	20	9	0	0	0
ACM Titles
Indexing			Full Text
AE	EXP	RL	AE	EXP	RL
6	6	3	0	6	1

Not surprisingly, the three databases provide substantial coverage of the computer literature. Many of these titles are popular or trade in nature, such as PC World, InfoWorld, or InternetWeek. There are 27 non-refereed, full-text titles in Academic Elite, 42 in Expanded Academic and 26 in Research Library.

The databases also have a significant amount of scholarly literature as well. All three databases index publications of the Association for Computing Machinery (ACM) and the IEEE Computer Society. Both Academic Elite and Expanded Academic index more than 60 IEEE titles, including 20 IEEE computer titles, as well as six ACM titles. Research Library indexes only nine IEEE computer science titles and three ACM titles.

Although year ranges vary among the titles, for most titles, Expanded Academic indexes the IEEE Computer Society journals from 1992 to present. Research Library indexes them from 1994 to present. Academic Elite indexes them from 1996 to the present. There is no full text for any of the IEEE titles.

Expanded Academic also indexes ACM titles for more years than Academic Elite and Research Library. For most titles, Expanded Academic indexes four to five more years than Academic Elite. In addition to indexing more years, Expanded Academic also provides full text of ACM articles, generally from the mid-90s to present. Titles include ACM Computing Surveys, ACM Transactions on Computer Systems, ACM Transactions on Database Systems, ACM Transactions on Information Systems, ACM Transactions on Mathematical Software, ACM Transactions on Programming Languages & Systems, and Communications of the ACM.

Academic Elite provides full text of a significant number of computer titles, including some relatively high impact titles, e.g. Neural Computation, Presence, IBM Journal of Research and Development, and IBM Systems Journal. However, many of the full-text titles in Academic Elite are less prominent than the ACM titles and do not have impact factors for comparison purposes.

Engineering

Indexing			Full Text
AE	EXP	RL	AE	EXP	RL
107	108	28	23	14	17

Academic Elite and Expanded Academic are significantly stronger in engineering indexing than Research Library as they provide more extensive coverage of the publications of the American Society of Civil Engineers (ASCE), the American Society of Mechanical Engineers (ASME), and the Institute of Electrical and Electronics Engineers (IEEE) journals. Expanded Academic generally indexes more years of the journals of these engineering societies than does Academic Elite. Although year ranges vary among titles, Expanded Academic generally indexes four years more than does Academic Elite.

While quite a few titles are indexed, the amount of full text from refereed engineering journals is very small. However, there are quite a few full-text engineering trade journals. Expanded Academic provides the most of this literature by offering 28 trade journals in full text. These trade titles are not included in the scholarly breakdowns below.

Civil Engineering

Indexing			Full Text
AE	EXP	RL	AE	EXP	RL
20	24	4	1	2	1

Academic Elite and Expanded Academic provide more coverage of the civil engineering literature. Both index ASCE journals (18 and 19 respectively) with Academic Elite and Research Library providing the full text of one: Civil Engineering. Expanded Academic indexes and additional four years for most of the ASCE titles (1992 to present compared to 1996 to present). Expanded Academic provides full text to Ground Water and Pollution Engineering.

Electrical Engineering

Indexing			Full Text
AE	EXP	RL	AE	EXP	RL
36	31	4	6	2	3

In the area of electrical engineering, Academic Elite and Expanded Academic are significantly stronger than Research Library as they index more than 60 IEEE titles (compared to 11 for Research Library). None of the databases has a significant amount of full text. Academic Elite has the most, providing full text to relatively low impact titles such as Electronic Design, Electronic Engineering, International Journal of Electronics, Bell Labs Technical Journal, Solid State Technology, and Power Engineering.

Mechanical Engineering

Indexing			Full Text
AE	EXP	RL	AE	EXP	RL
24	26	11	7	5	7

Academic Elite and Expanded Academic index more than twice the number of titles as Research Library. Academic Elite indexes 11 ASME titles and Expanded Academic indexes 16. Both Academic Elite and Research Library provide full text from the ASME's Mechanical Engineering. Expanded Academic indexes most titles from 1992 to present and Academic Elite indexes the same titles from 1996 to present.

All three provide full text from Journal of Vibration and Control, Advanced Materials and Processes, and Advanced Composites Materials. Expanded Academic also offers full text from IIE Transactions, Metallurgia, and Design News. Research Library also has full text of IIE Transactions, IIE Solutions, and Design News.

Environmental Studies

Indexing			Full Text
AE	EXP	RL	AE	EXP	RL
43	35	21	22	20	11

Environmental studies titles are relatively well represented in the databases. In addition to the scholarly literature (see title counts above), all three databases include the full text from a large number of popular, non-refereed titles, such as Mother Earth News, Sierra, Wilderness, and Whole Earth. Academic Elite includes 20 popular titles, Expanded Academic includes 26 titles and Research Library includes 16.

From the scholarly literature, Expanded Academic and Research Library provide full text from Ecological Monographs and Ecology, both high impact titles. It should be noted that Expanded Academic recently discontinued full text from some of its high impact titles (e.g. Ecological Monographs, Ecology, Ecological Applications, Ground Water Monitor, and Environmental Technology).

Geology

Indexing			Full Text
AE	EXP	RL	AE	EXP	RL
14	14	15	3	3	6

The three databases provide comparable coverage of geology, indexing 14 or 15 titles, including many of the same titles. Generally Expanded Academic and Research Library index a few more years than Academic Elite. Research Library provides a few more titles in full text than Academic Elite and Expanded Academic.

Academic Elite and Expanded Academic both provide full text from Journal of Geology and Rocks and Minerals. Academic Elite has full text from South Africa Journal of Geology and Expanded Academic has Mineralogical Record. Research Library titles include Rocks and Minerals, Mineralogical Record, and Canadian Journal of Earth Sciences.

Mathematics

Indexing			Full Text
AE	EXP	RL	AE	EXP	RL
30	16	13	18	6	7

Mathematics is generally not well covered by the databases, although Academic Elite provides better coverage than the Expanded Academic and Research Library. There is very little coverage of the society publications, as only one American Mathematical Society (AMS) title and one Society for Industrial and Applied Mathematics (SIAM) title are indexed. Statistics and econometrics are two of the stronger areas of mathematics coverage, as all of the databases have some full text from publications of the American Statistical Association.

Notable full-text titles in Academic Elite are The Journal of the American Statistical Association, American Statistician, Journal of Computational & Graphical Statistics, Mathematical Intelligencer, Quarterly Journal of Mathematics, IMA Journal of Numerical Analysis, and IMA Journal of Applied Mathematics. Research Library has the full text of three Mathematical Association of America titles, American Mathematical Monthly, Mathematics Magazine, and College Mathematics Journal.

Physics

Indexing			Full Text
AE	EXP	RL	AE	EXP	RL
17	7	5	14	0	2

Of the three databases, Academic Elite has greater coverage of physics literature. However, it should be noted that most of the unique titles in Academic Elite are relatively recent, some only going back as far as 1998. Also, the American Institute of Physics titles have a twelve month embargo period imposed by the AIP. High impact titles included in full text are Advances in Physics, Physics Today, Applied Physics Letters, Journal of Chemical Physics, Journal of Applied Physics, Contemporary Physics, Applied Physics A&B, Molecular Physics, Journal of Mathematical Physics, as well as AIP Conference Proceedings. Research Library has one unique title in full text: The Canadian Journal of Physics.

Additional Considerations

Academic Elite uses the Ebscohost interface that permits multiple database searching, allowing the databases to support and strengthen each other. While Ebscohost does not provide the clean issue/table of contents displays that Proquest does, it does allow linking from the catalog to the title level for the aggregated journals. Ebsco is also the only vendor of the three that allows full-text content to be automatically shared between databases. It is also is the only vendor to allow additional full-text content to be added to the databases via individual e-journal subscriptions (via Ebsco Online).

Expanded Academic uses the Infotrac interface and as such seems simpler and more intuitive than Ebsco or Proquest. Due to its simple interface and displays, Infotrac requires little explanation to even the most inexperienced users. Unfortunately, the Infotrac databases do not permit multiple database searching and do not enable linking from the library catalog to the aggregated journals included in the databases.

Research Library uses the Proquest interface that permits multiple database searching, allowing the databases to support and strengthen each other. Proquest also allows the aggregated journals to be displayed by title and table of contents. This enables libraries to add more effective direct links from the library catalog to the full-text content. Research Library is sold in modules, so that libraries have more control over which subjects are purchased. Since the Research Library Sciences module can be purchased as a separate package it permits libraries to add science content without purchasing the entire large database.

Conclusion

Academic Elite, Expanded Academic and Research all have significant science holdings (more than what might initially be expected). Each database has areas of strength and weakness. As an overall index of scientific literature, Expanded Academic is the strongest of the three databases due to its greater range of coverage. As an aggregator of full-text scholarly science content, Academic Elite is the strongest due to its more extensive title list. While the Research Library Sciences module does not contain as much science indexing and content as the other databases, it has the advantage of being available as a smaller, less-expensive, science-only package with a higher percentage of full text and page images. The database that is most appropriate is dependent on the needs of each institution.

References

Awagain, K. 2001. LALINC Database Comparison Subcommittee Report. Baton Rouge, LA: LALINC Database Selection Advisory Committee.

Blosser, J., et al. In Press. Aggregator services evaluation: not an easy comparison. The Serials Librarian 41(1).

Calhoun, K. and Kara, Bill. 2000. Aggregation or Aggravation? Optimizing access to full-text journals. ALCTS Online Newsletter 11(1). [Online]. Available: {http://www.ala.org/alcts/alcts_news/v11n1/gateway_pap15.html} [April 15, 2001].

Dennis, S.L. 1999. Aggravating or aggregating? Providing effective access to contents of aggregator databases: a reference/collection development librarian's perspective. Cataloging & Classification Quarterly 28(4): 15-30.

Hill, J.B. and Madarash-Hill, C. 2000. Publications of the state academies of science. Science and Technology Libraries 19(1): 21-37.

Salisbury, L., Davidson, B. and Bailey, A. 2000. Undergraduate full-text databases: Bell and Howell Medical Complete and InfoTrac Health Reference Center - Academic. The Journal of Southern Academic and Special Librarianship 2(1-2). [Online]. Available: http://SouthernLibrarianship.icaap.org/content/v02n01/salisbury_l01.html [April 15, 2001].

Acknowledgements

The author gratefully acknowledges the essential contributions of Katie Connell, Amy McCoy, Lorna Pillay and Keiko Koami in the laborious data collection and data entry phase of this study.

Previous

Contents

Next