Investigating the Impact of 
Digital Communications on 
Scientific Research

(ldg, 4/99)

Background

Seminal work by William Garvey and Belver Griffith in the 1960s mapped the flow of information within science research including measurement of the timeline associated with the research and publication. They devised the following schematic model to characterize the key stages in paper-based publication:


(Click the above image for a detailed view)

 

The research timeline is particularly daunting. For example, they found that it required an average of four years between the initiation of research and the first journal publications reporting the findings. Once published, it typically required another year before it appeared in published indexes or abstract collections. When the scholarship is 7 to 10 years old, it appears in major review papers.  At 12 to 15 years of age, it becomes part of the discipline's canon.

Several authors (c.f. Hurd, 1996) have speculated about how today's Internet-based digital communications is changing the character of scientific communication, often pointing out demonstrations of on-line collaboration such as the human genome project.

Are the digital tools having a measureable impact? Are there other sociological forces that dilute the revolutionary changes claimed by technology enthousiasts?

Premises:

  1. Electronic communications accelerates the speed of information dissemination by facilitating the sharing of information.
Research scientists have long relied on an "underground" network to share preliminary information among colleagues through personal conversations, special seminars and colloquia, and the sharing of preliminary drafts of their research reports. Garvey & Griffith's research demonstrates that this informal network was highly efficient, even in the pre-Internet era of the 60's.  New technologies, such as electronic mail, listservs, and unannounced web sites should facilitate this exchange of information among active scholars.

This interaction should be observable by a compression of the time between publication of a research paper and when others cite that work in other publications. Since Science Citation Index has been published since the early 70s. It should be feasible to conduct longitudinal studies to determine if there exists a statistically significant compression in mean citation delay. Similarly, the growth in electronic communications should be signaled by a marked increase in citations of unpublished items (e.g. in press, preprints, and electronic communications).

In similar research, Stephen Harter (1996) examined the impact of electronic journals by measuring both the number of citations and the rapidity by which articles appearing in electronic journals are then cited in subsequent publications. His intent was to examine whether the rapid publication speed of electronic journals was having a measurable impact on the rate of information dissemination. He found that only a small impact from electronic journals as only a few of them are cited elsewhere. It is quite possible that reality has yet to catch up with the hype associated with electronic publication.

Subsequent research by Gregory Youngen (1998) demonstrates that the citation rate for electronic preprints is growing at a very rapid annual rate in areas of astronomy and physics where large on-line databases have been established. For example, a scan of citations shows a ten-fold increase in electronic preprint citations between 1994 and 1997.

  1. Electronic communications broadens public access to modern research information.
In the pre-digital world, the underground network of research colleagues creates an inherent bias in favor of those researchers in close physical and social proximity to each other. Indeed, this has been the major force leading to the rise of large research universities. Computer-based communication radically changes the economics of information flow. Using today's Internet, it is no more difficult or expensive for a researcher in Boston to converse with a colleague in Australia than with a colleague in the next city. Furthermore, it should be more feasible for researchers in small rural colleges to become more active participants in the information exchanges.

This effect should be observable through measures of the diversity of authors and institutions involved in research publications. For example:

    1. Has the diversity of institutional affiliations represented in today's co-authored papers increased since the pre-digital era?
    2. Do authors cite a wider variety of authors (and institutions) in recent publications compared to pre-digital publications?

      3.  
  1. The Internet improves the collaborative opportunities for researchers in small colleges.
Historically, the science faculty in small liberal arts colleges have been viewed as a research backwater. The heavy teaching responsibilities combined with inadequate laboratory resources and the lack of trained graduate students resulted in an environment that did not permit these faculty to keep abreast of their colleagues in the large research universities. The interconnected world of today's Internet offers several potential opportunities to reduce this schism between the teaching colleges and the research universities. For example, Garvey & Griffith's research showed that even in the 1960's, there existed a network of highly efficient informal communications among those scholars who are considered the leaders in their research discipline.  If so, it may well be that the Internet offers only minimal improvement to this core of scholars because they already possessed adequate access to the emerging research findings.  Conversely, the Internet may be particularly beneficial to those who are less central to the core research community, either because they work outside of the major universities or because they have not yet achieved broad recognition within their discipline.  This suggests that the Internet may be a much more important research tool within schools like Skidmore than for universities like Harvard and MIT.

On the other hand, the Internet will be effective for small colleges only if access to information is a major factor limiting scholarly productivity.  It is also possible that other factors, such as curricular responsibilities, the lack of graduate students, and limited facilities are far more important limits to faculty research.  Effective use of the Internet may also require a cultural shift among scholars, as faculty struggle to shift their focus from local facilities, students, and laboratories, to shared access to national and international resources.  Indeed, researchers at the major universities have a longer history of such collaboration, as shared time on nuclear reactors, linear accelerators, and large telescopes is much more common at universities than at small colleges.

This suggests a number of observable questions:

    1. Have the science faculty of small colleges increased their rate of research publication?
    2. Are these faculty more likely to co-author research with colleagues at research universities?
The need for a vetting process

The process of formal publication represents a major demarkation within traditional scientific communication.  The informal communications that precede publication rarely become part of the historical archives of the profession, and are frequently subject to substantial revision and expansion.  In the pre-Internet era, the demarkation between informal communications and the formal publications was obvious even to novices unfamiliar with the canons of the discipline.  Even undergraduates can easily recognized a typed informal draft from the typeset formal publications.

Perhaps the largest change introduced by the combination of desktop publishing and rapid Internet communication has been a blurring of the boundaries between informal and official communications.  It is much less obvious whether a well-done web page represents mainstream scientific thought or a collection of untested claims that have largely been rejected by the core scholars in that discipline.  Furthermore, the emergence of electronic journals and large central archives of electronic publications (e.g. the Los Alamos electronic  print archives for high energy physics) lead to situations where significant research never results in paper publications.

This blurring of the publication boundary presents a particular problem for librarians who are responsible for collection development.  Simply put, the potential to capture informal communications such as electronic mail, listserv discussions, and informal web pages, often tempt librarians into the belief that all of these items should be included into the collection archives for the discipline.  Such large scale capture of scientific scholarlship poses two major problems.

References:

Garvey, W.D. & Griffith, B. C.  Scientific communication:  its role in the conduct of research and the creation of knowlege. American Psychologist, 1971, 26, 349-62.

Garvey, W.D. & Griffith, B. C.  Communication and information processing within scientific disciplines:  Empirical findings from psychology.  In, W. D. Garvey, Communications: The Essence of Science. New York: Pergamon Press, 1979.

Harter, Stephen P. "The Impact of Electronic Journals on Scholarly Communication: A Citation Analysis." The Public-Access Computer Systems Review 7, no. 5 (1996): 5-34. (available at http://info.lib.uh.edu/pr/v7/n5/hart7n5.html)

Hurd, J. M.  Models of scientific communications systems.  In S. Y. Crawford, J. M. Hurd, & A. C. Weller, From print to electronic:  The transformation of scientific communication.  Medford, NJ, 1996  American Society for Information Science.

 

Younger, Gregory. Citation Patterns to Electronic Preprints in the Astronomy and Astrophysics Literature. In U. Grothkopf, H. Andernach, S. Stevens-Rayburn, and M. Gomez (Eds.) Library and Information Services in Astronomy III. San Francisco, Astronomical Society of the Pacific, 1998. (available at http://www.eso.org/gen-fac/libraries/lisa3/cover.html