A statistical study on global internet use shows that as of September 2002 there were approximately 605.60 million internet users worldwide ("NUA", 2002). The broad reach of technology increasingly effects people of all ages, cultures and demographic as more people are able to become connected to a global community and are able to access all types of information and easily exchange thoughts and ideas. Physical boundaries are of no concern in an online environment where someone can quickly use a search engine to find a product, have email conversations with people who own or know about the product, purchase and have it delivered without even leaving their home, or for that matter their chair.
A world of information is now just a click away and along with this growth of available information one can also see an increase in the need to study it. Scientific knowledge in the world and scientific journals that publish this knowledge through articles, have also shown steady increases as more subjects are being explored and more subfields being created as scientists of various disciplines work together to research all the new ideas and information being generated. Technological advancements in data storage and retrieval and virtual communication environments have forever changed how research is conducted and reviewed and how research results are communicated, disseminated and evaluated.This is truly an information age, but is access to so much information and technology always a good thing? There is a tendency to rely heavily on numbers and data available in databases in order to form opinions and judge quality. But databases aren't human minds and search capacities are limited and can't always capture all the information needed or the certain nuances embedded in the data that might make a big difference in establishing the truth.
This paper addresses the benefits and problems that have occurred with the union of technology and scholarly communication, with specific attention given to the Scientific Citation Index (SCI) created by Eugene Garfield, and how the relationship between the two has effected how scientific knowledge is ultimately created and dispersed.The major issues to be discussed are as follows:
1) Changes to scholarly communication due to the creation of the Science Citation Index (SCI)A review of papers from The Web of Knowledge: A Festschrift in Honor of Eugene Garfield, and various articles and literature related to scholarly communication and technology will be used to provide evidence and support to the overall focus which is how technology has influenced how scientific scholarly communication is evaluated and what are the benefits and problems that have resulted.
Changes to scholarly communication due to the creation of the Science citation Index (SCI)Scholarly Communication (SC) has been defined as a "system"..."the means in which the results of individual, private research projects are transformed into scientific knowledge" ("library.owu.edu"). One of the steps that start the process of acquiring new knowledge is to become aware of prior information that is available, to seek it out and make sense of it through an internal, cognitive process. The information seeking behavior of scientists involved in research will usually involve looking at prior research. As stated by Donald O. Case (2002) in Looking for Information, scientists "must keep abreast of what other researchers are doing in the field, what discoveries have been made, what equipment and techniques are available (p.11). One of the ways to acknowledge prior works that have impacted current research is to use citations. As stated by Blaise Cronin and Helen Barsky Atkins (2000), citations serve various purposes: "they can be used to bestow credit and repay intellectual debts; they guide the interested reader to related works and help locate ideas in larger social, conceptual and historical contexts" (p.4). By providing easily accessible bibliographical information (name, address, contact etc.), citations also help foster increased communication between scholars by giving information needed to contact others involved in similar research.
The idea of having a main source and index for citations in order to provide ease in access is not new. As discussed by Paul Wouters (2000), the concept can be initially seen as early as the 12th century (p. 66). But a citation index for science was a new concept and the uniqueness of its capabilities and uses has made significant impact on scholarly communication worldwide.A pioneer of science citation indexing, Eugene Garfield along with the Institute for Scientific Information (ISI), developed the first science literature index database called Science Citation Index (SCI) and the first index edition appeared in 1963 (Cronin & Atkins, 2000 p.3). One of the initial goals of SCI was to establish a tool in which scientists could access bibliographic information about others involved in related research (Wouters, 2000 p. 65). The potential for usage of this tool grew beyond access to names and contact data but into a formative information resource for scholars of various disciplines.
As stated by Arnold Thackray and David C. Brock (2000), SCI provided a "historiograph" or map of like-minded subjects that detailed a rich "outline" showing how a certain subject or field was ultimately established (p.18). Scientists could use a citation index to see how various disciplines might work together for current research or it might provide a springboard for ideas for future research. Providing connections in research being done in different disciplines was a primary concern for Garfield who advocates creating knowledge through cognitive awareness of related subjects. As stated by Robert Merton (1979), Garfield's development of SCI shows a "deep intuitive sense of the social, cultural and cognitive structures latent in the practice of science" (Merton, 1979 p. vii pg. 1).As described by Jack Meadows (2000), The need for establishing the first science database with access to bibliometric data can be seen as a response to the increasing "morass of information" available in the world (p. 87). Garfield's use of computer technology to combat the problems of the overload of information production can be seen as a unique vision of combining an older idea of citation indexing with science and the new technological advancement of computers (Wouters, 2000 p.67).
In addition to the initial uses of SCI, other uses also developed. Citations are used in the application review system for issuing patents. Both the applicant and the patent examiner will use citations in order to prove originality in invention (Oppenheim, 2000 p. 407) Citations are also largely used as an evaluative measure of influence and achievement which is an important part of scholarly communication. As stated by Cronin and Atkins (2000), citations and thusly citation indexes, have "become part of the evaluator's armamentarium, and are used to rank journals, research groups, laboratories, universities, and nation states" (p.4). Also mentioned was the use of citation counts in the decision process for awarding tenure and promotions. According to Cronin and Atkins, these uses were not a part of the original concept of SCI nor were these uses intended (p. 4).How the processes, products and trends of scholarly communication have been affected by technological advancements
There are various processes that make up the "system" of scholarly communication. The processes involve collaboration, communication, building of consensus for growing knowledge, through a complex communication and evaluation system of journal publishing and peer review" (Coleman, 2004). Scholars will work together at some stage of a research project, reach a level of consensus on their findings and communicate results by submitting their work to be reviewed. These processes can all lead to one of the main goals of scholarship which is publishing or some type of social acknowledgement of created work and a level of influence attributed to it.An important part of scholarship is the need to establish the influence of scholarly contributions and this is done by the evaluation and review of the products and outcome of scholarly work. These products can include: journal articles, papers and more currently e-print publishing's to name a few.
Each of these processes and products has been impacted in various ways by technological advancements and the creation of certain trends in scholarly communication has developed as a result. Collaboration in research is defined by Maria Bordons and Isabel Gomez (2000) as "two or more scientists working together on a joint research project, sharing intellectual, economic and/or physical resources" (p. 198). From this definition, many advantages in collaboration can be seen such as: sharing expenses of costly research instrumentation and research facilities; sharing labor and knowledge; alleviation of isolation; increased motivational benefits of working in a group environment and better research output (p. 199). Collaboration also becomes very important in reducing the labor and costs involved in research of "big science" which encompasses large and expensive fields of scientific disciplines such as oceanography (p. 199). Another benefit is the encouragement of internationalization (p. 204). Governments and policy makers can lay a foundation for good will and tolerance by fostering opportunities for international collaboration that can have mutual benefits for all countries involved that stretch beyond scientific research and into the realm of politics.A trend in scholarly communication that has occurred that has helped increase collaboration is the increase in the many subfields of scientific study which has caused many researchers to "over-specialize" in disciplines that seem endless in scope. This growth in sub-fields can be seen as a direct result of the information boom that has occurred since WWII and has caused an overall increase in the growth of journals (Meadows, 2000 p.87). A trend toward "interdisciplinarity" has developed due to the growth of overspecialization and can be described as the need for those working in various homogenous disciplines to work together in order to solve problems and conduct research (Coleman, 2002).
Communication and consensus has also been affected by technology. Communication technology advancements such as email have increased speed in interactions. As described by Christine L. Borgman, the "invisible college" which is a network of scientists who exchange thoughts and ideas often through informal means, has embraced the ease in communicating informally in a virtual environment (Borgman, 2000 p. 145).As information grows along with the many accesses to it, communication networks such as the invisible college become important in providing a forum for discussing all the research being done. As described by Stephen Cole (2000), individual scientists aren't able to read all the documentation being created (p. 111), so consensus of thought also occurs as individuals rely on the community network to discuss and select what research it deems important.
In addition to encouraging communication, internet technologies have also aided the information searches for those conducting research. As stated by Robert I. Berkman (2000) in Find it Fast, there's no question that the Internet has proven to be enormously beneficial for researchers. There are now more mountains more facts, data, and information easily and quickly available" (preface xi). The internet provides easy and usually inexpensive or free access to a multitude of subjects that may be of interest to someone conducting any type of research.
The products of scholarly communication have also been affected by advancements in technology. A trend toward "removing the middle man" or disintermediation has also developed as more products are offered less expensively online as opposed to the higher cost of traditional method ("Webopedia"). As discussed by Charles Burress (2004) in an article in the San Francisco Chronicle, there is a trend toward "open access" in which scholarly works can be available on the internet and thus allow users to avoid the high costs of journal subscriptions through publishing giants such as Elsevier. Libraries are also effected by the high cost of journal subscription and have had to "reduce or eliminate" their subscriptions (Burress, 2004). As an alternative to print journals, full text E-journals are increasingly available online. For a considerably reduced price, users can have an interactive experience with the material by quickly accessing hyperlinks to related information.
Online archives or "Digital Libraries" and digital repositories are increasing in popularity because they offer various types of digital data to a much larger audience and make information available to those who don't have access to a physical library and can't obtain the tangible product (Borgman, 2000 p.150).Another "open access" forum that is gaining popularity is Creative Commons ("CreativeCommons.org") which practices disintermediation by getting rid of the middle man by allowing for consented sharing of copyright material and encouraging collaboration of artists and scholars.
Disintermediation has also effected how libraries handle the new demands of their patrons. As stated by Elizabeth Thomsen in Rethinking Reference: The Reference Librarian's Practical Guide for Surviving Constant Change, libraries have had to add more computer stations to accommodate the demands of patrons who are using library internet access to participate in an online class or do research that would have traditionally been done at a business office such as a bank or a travel agency, for example, but now can be done online (Thomsen, 1999 pp.11-14).The ability to have open access to all types of information has an overall effect in helping to increase overall knowledge as more people have access to information.
How the process of evaluating scholarly communication has been affected by bibliometric databasesIn order to establish influence and reward excellence, evaluation of the products and processes of scholarly communication is needed. One of the main evaluation processes is journal peer review. Cole (2000) explains that the peer review process is important in the growth of scholarly knowledge because journals communicate the results of research on new and dynamic concepts or "frontier" research to the world. Cole argues that it is in this frontier that new knowledge is born and that these works have the potential of being accepted as "core" knowledge through a process of peer review and consensus (p. 110).
Not all journals are created equal and there are established evaluative procedures to determine if certain documents are right for certain publications. Journal peer review is a process by which editors and referees will review submitted articles and select which are suitable for publishing and thus available to the public. This is a subjective process and largely reliant on the biases of the peer review group. As described by Stephen Cole (2000), the ideal system would be if all articles could be "objectively evaluated" and only the best quality articles would be published quickly in journals that were highly visible (p. 112). But this is only an ideal and the reality of peer review is that there are inherent biases that make the system problematic. He describes the biases that can sway decisions of whether or not to publish as: against perceived negative results which are often shunned by editors of the author's research or particularistic characteristics of the author's cognitive style in the work and disagreement with the author's theoretical viewpoint (pp. 115-116).
In addition to being a subjective process in which biases can prevail, it is also time consuming and can keep breakthrough research from quickly reaching the public. This problem is minimized in open access forums like digital repositories in which the amount of peer review is reduced and works reach the public quickly.The perceived prestige of a journal also plays an important part in what articles are accepted for publishing and which articles authors choose to submit for review and publishing in certain journals. Cole (2000) shows that the "Halo Effect" in which a consensus in the scholarly community attributes merit to certain journals leads to higher visibility of that journal in the community (p. 118). Authors react to this established consensus by selecting their best articles to be published in high prestige journals through a process called self-selection. Cole argues that author's themselves are mostly responsible for the reason why high prestige journals end up producing the best quality and often the highest cited articles (p.125).
Even with the problems associated with these processes, Cole(2000) makes the point that due to the large quantity of publications now available in print and online, that most works have the opportunity to be published somewhere (pg. 113). Even though these processes can be seen to cause barriers to the goals of scholarly communication, Cole explains that these selection methods all work within the infrastructure of the system and make the overall process not too far removed from the ideal (p. 113)Due to the inherent biases of peer review, the use of database information is also used to determine scientific excellence and has been embraced by many in the scientific community. Referred to as "evaluative bibiliometrics", these mathematical formulas rely heavily on citation data in order to calculate the level of merit or excellence in a scholarly work. Impact factor of journals and citation counts are some of the data used in evaluative bibilometric methods.
One of the proponents of the use of an advanced bibliometric method to determine influence and scientific excellence is Anthony F.J. van Raan. van Raan (2000) describes a bibliometric method that can be used to best evaluate the "oeuvres" or body of works of groups that conduct research such as institutes and universities and how their works comparatively measure to the work of oeuvres being done worldwide (2000 p. 303). Largely based on the SCI indicators that measure "Impact factor" by tallying citation count history for journals and authors, van Raan's method relies on a "crown" indicator that measures the performance level of a research group as compared to the international "impact standard of the field" which is based on the western world average high (van Raan, 2000 p. 305). By establishing an international standard and indicator of performance, van Raan asserts that a more complete assessment of excellence can be made.There are also bibliometric methods used in evaluating collaboration. Co-authorship searches are available on the ISI database which tallies the number of authors per document. These searches can be used to show trends in collaboration. According to Bordons and Gomez (2000), this data confirms their assertion that collaboration is on the rise and single authored documents are on the decline (p. 201). Although the co-authorship index counts are considered reliable, limitations in co-author citation searches exist and will be discussed later in further detail.
A review of these evaluative methods shows that there is no one single way to measure excellence in science. Even with the sophistication of bibliometric methods, there will probably always be a need to incorporate subjective methods into the process. As stated by van Raan (2000), "Undoubtedly, the bibliometric approach is not an ideal instrument, working perfectly in all fields under all circumstances" but he feels that together with peer review a more complete assessment of excellence can be made (2000, p. 303). The limitations of international representation, scope and search capacity of the ISI databasesWith all the information available in databases, the question arises as to how complete is the data and how trustworthy is it? A review of the search capabilities on the CD-Rom version of SCI called Web of Science (WOS) and the Journal of Citation Reports (JCR), show that some difficulties arise when attempting different types of searches. One of the initial difficulties is that not all science journals are covered by SCI. Based on information on the SCI homepage, currently 5,000 journals are covered. When conducting a precursory search for various journals on WOS and JCR, there were several well established science journals that were not covered and there were many that weren't covered in both WOS and JCR, thus making citation activity research problematic. As described by Subbiah Arunachalam (2000), there has also been criticism that the journal coverage is slanted toward English language journals and that there is an under representation of developing countries (p. 216). One reason for this is that research is dominated by pursuits in "big science" which requires large funding dollars and the U.S. is one of the most sought after countries to collaborate with (Arunachalam, 2000 p. 222) which would be the likely reason English is the predominate language in SCI.
Other criticism of representation in SCI is the "over-representation" of journals published by Elsevier (Braun, Glanzel,Schubert, 2000 p. 273) who has been described earlier as operating like a monopoly in the journal publishing world (Burress, 2004). But Garfield himself expresses his own feelings on the matter of publishers dictating the cost of journals in the marketplace. In a 2000 interview in Information Today Garfield says, "In the marketplace, if someone has a monopoly, they will ordinarily behave like a monopoly. If nobody is willing or able to challenge them, they'll charge what the market will bear" (Information Today, 2000). Garfield, himself an entrepreneur, also comments that there is definitely room for new opportunities in publishing. He states, "It's very hard to compete with established journals, but publishing has always attracted new entrepreneurs" (Information Today, 2000).
Problems with various searches performed on the ISI databases have also been observed. Author name searches have posed several problems. As described by Howard White in a 2001 article in Journal of the American Society of Information Science and Technology entitled "Authors as Citers over Time", authors are listed by last name and only the first letter of their first name which causes problems in distinguishing between the list when there are multiple authors with the same last name and first initial. (JOASIST, 2001). Another problem with author name searches is with "homonyms and allonyms" which are same names for different people or different names for the same person (JOASIST, 2001). Garfield comments on these search problems in a 2002 interview in Password when he says that the time and expense in restructuring the SCI database to include first names would be "enormous" and he adds that a main point that is often forgotten is that the focus of the database is on the article and not the author and these problems don't exist when finding articles (JOASIST, 2001).
With the trend toward an increase in collaboration, there is also a growing problem in accessing citation information on co-authors. The ISI citivity searches are available for first authors only. ISI does provide one of the only databases for looking at bibliometric information on co-authors (Bordons & Gomez, 2000 p. 201), but the individual citivity counts for each author cannot be accessed separately so merit can not be assessed per individual author.
Some misleading citivity data is also incorporated in ISI searches. Negative counts also are included in citation count history, so if an author wrote a bad article and it was cited as being bad, that count would count in their overall citation history (Cole, 2000 p. 293). Self-citations are also included in citation counts. If an author cites themselves, that citation will be count for them and if a journal cites another article in that same journal that will count in the journal's impact factor.
These limitations in the search functions and scope of the ISI databases may cause some problems with performing certain information searches, but no computer or database has all the answers and there is always room for improvement. This begs the question: What do we ultimately want from computers? We increasingly rely on them because they are usually very good at storing information and unlike human brains, don't have difficulty making complex computations. As described by Andy Clark (2003) in Natural Born Cyborgs, the human brain is good at "recognizing patterns, at perception, and at controlling physical actions, but it is not so well designed for complex planning and long, intricate, derivations of consequences" (p.5). But problems can occur when we tip the balance and become overly reliant on computer technology and get frustrated that computers can't access information beyond their programs...in other words can't think. By relying too heavily on computers for data that is used to make certain decisions, we can fall prey to loosing our sensitivity to subtle nuances that could make significant impact on the level of truth in some decisions that we make.
Using a citation database as a guide to determining excellence and not as an absolute measure of excellence, will allow room for more variables that could establish a more complete picture of the truth.
Arunachalam, S. (2000). International Collaboration in Science: The Case of India and china. In B. Cronin & H.B. Atkins, (Eds.), The Web of Knowledge: A Festschrift in Honor of Eugene Garfield.(pp. 215-216,222). Medford, NJ: Information Today, Inc.
Berkman, R.I. (2000). Find it Fast: How to Uncover Expert Information on Any Subject Online or in Print.New York, NY: HarperCollins Publishers, Inc.,preface, xi.
Bordons,M. & Gomez, I.(2000). Collaboration Networks in Science. In B. Cronin & H.B. Atkins, (Eds.), The Web of Knowledge: A Festschrift in Honor of Eugene Garfield.(pp. 198-204). Medford, NJ: Information Today, Inc.
Borgman, C.L. (2000). Scholarly Communication and Bibliometrics Revisted. In B. Cronin & H.B. Atkins, (Eds.), The Web of Knowledge: A Festschrift in Honor of Eugene Garfield.(pp. 145,150). Medford, NJ: Information Today, Inc.
Braun, T., Glanzel, W. & Schubert. A (2000). How Balenced is the Science Citation Index's Journal Coverage? A Preliminary Overview of Macro-level Statistical Data. In B. Cronin & H.B. Atkins, (Eds.), The Web of Knowledge: A Festschrift in Honor of Eugene Garfield. (p. 273). Medford, NJ: Information Today, Inc.
Burress, C. (2004). The staggering price of world's best research, bay area universities leading charge against publishers, arguing the knowledge in academic journals must be kept within reach.San Francisco Chroniclehttp://www.sfgate.com/cgi-bin/article.cgi?file=/chronicle /archive/2004/03/28/BAGQE5SL3I1.DTL.
Case, D.O. (2002).Looking for Information: A Survey of Research on Information Seeking, Needs and Behavior.(p.11) Lexington: University of Kentucky Academic Press.
Clark, Andy (2003). Natural Born Cyborgs: Minds Technologies and the Future of Human Intelligence.(p.5).New York, NY: Oxford University Press.
Cole, J.R.(2000). A Short Hisory of the Use of Citations as a Measure of the Impact of Scientific and Scholarly Work. In B. Cronin & H.B. Atkins, (Eds.), The Web of Knowledge: A Festschrift in Honor of Eugene Garfield. (p. 293). Medford, NJ: Information Today, Inc.
Cole, S.(2000). The Role of Journals in the Growth of Scientific Knowledge. In B. Cronin & H.B. Atkins, (Eds.), The Web of Knowledge: A Festschrift in Honor of Eugene Garfield. (pp. 111-125). Medford, NJ: Information Today, Inc.
Coleman, A. (2004). Growth,Collaboration, Communication, Consensus and Conflict. University of Arizona.Spring IRLS 589.
Coleman, A. (2004). Interdisciplinarity.University of Arizona.Spring IRLS 589.
Creative Commons (2004). http://creativecommons.org/
Cronin, B. & Atkins, H.B.(2000). The Scholar's Spoor. In B. Cronin & H.B. Atkins, (Eds.), The Web of Knowledge: A Festschrift in Honor of Eugene Garfield.(pp. 3,4). Medford, NJ: Information Today, Inc.
Hane, P.J. (2000). Eugene Garfield celebrates a Birthday and a Career.Information Today 17(10)November http://www.garfield.library.upenn.edu/papers/informationtoday11172000.html
Libraries & Research (2004)Ohio Wesleyan University.http://library.owu.edu/.
Meadows, J. (2000). The Growth of Journal Literature: A Historical Perspective. In B. Cronin & H.B. Atkins, (Eds.), The Web of Knowledge: A Festschrift in Honor of Eugene Garfield.(p.87). Medford, NJ: Information Today, Inc.
Merton, R. (1979). Foreward in E. Garfield, Citation Indexing: It's Theory and Application in Science, Technology and Humanities (pp.vii-xi). In B. Cronin & H.B. Atkins, (Eds.), The Web of Knowledge: A Festschrift in Honor of Eugene Garfield.(p.1). Medford, NJ: Information Today, Inc.
NUA (2004). How Many Online?. Jupiter Media Corporationhttp://www.nua.com/surveys/how_many_online/
Oppenheim, C.(2000). Do Patent Citations Count? In B. Cronin & H.B. Atkins, (Eds.), The Web of Knowledge: A Festschrift in Honor of Eugene Garfield.(p. 407). Medford, NJ: Information Today, Inc.
Stock, W.G.(2002). Citation Consciousness: The Orgins of Citation Indexing in Science. Password.6/2002, p.22-25.http://garfield.library.upenn.edu/papers/passwordinterview062002.pdf
Thackray, A. & Brock, D.C.(2002). Eugene Garfield: History, Scientific Information and Chemical Endeavor. In B. Cronin & H.B. Atkins, (Eds.), The Web of Knowledge: A Festschrift in Honor of Eugene Garfield.(p.18). Medford, NJ: Information Today, Inc.
Thomsen, Elizabeth. (1999). Rethinking Reference: The Reference Librarian's Practical Guide for Surviving Constant Change.New York, NY: Neal-Schuman Publishers, Inc. (p. 11-14).
van Raan, A.F.J. (2000). The Pandora's Box of Citation Analysis: Measuring Scientific Excellence - The Last Evil?. In B. Cronin & H.B. Atkins, (Eds.), The Web of Knowledge: A Festschrift in Honor of Eugene Garfield.(pp.303-305). Medford, NJ: Information Today, Inc.
Webopedia (2004).Jupiter Media Corporation http://www.webopedia.com/TERM/d/disintermediation.html
White, Howard D. (2001). "Authors as Citers Over Time". Journal of The American Society For Information Science and Technology 52(2):87-108.
Wouters, P.(2000). Garfield as Alchemist. In B. Cronin & H.B. Atkins, (Eds.), The Web of Knowledge: A Festschrift in Honor of Eugene Garfield.(pp.65-67). Medford, NJ: Information Today, Inc. )