Peer Review in the Internet Age: Five (5) Easy Pieces

 

Gerry McKiernan

 

"... [In] the digital world, the evaluation process stands ready to be reinvented in a clear, rational way by the relevant research communities themselves."1

Jean-Claude Guédon

 

Purpose of Peer Review

 

In general, peer review can be defined as "the assessment by an expert of material submitted for publication."2 Specifically, its purpose is to ensure that published research is important, internally-consistent, original, technically-reliable, timely, well-presented, and most importantly, benefited from guidance by experts.3 Overall, "the underlying strength of editorial peer review is the concerted effort by large numbers of researchers and scholars who work to assure that valid and valuable works are published, and conversely to assure that invalid or non-valuable works are not published."4

 

Problems with Classical Peer Review

 

While established peer review has its supporters, it has long been criticized as "... slow, expensive, profligate of academic time, highly subjective, prone to bias, easily abused, poor at detecting gross defects, and almost useless in detecting fraud."5 In a recent review article on the peer review process, Rowland analyzes and briefly characterizes many of the deficiencies of classic peer review as follows:

 

·        Subjectivity

Summary rejections by editor without sending the paper to referees; choice of referees by the editor (choosing for example, a known harsh referee for a paper the editor wishes to see rejected);

  • Bias

Discrimination against authors because of their nationality, native language, gender or host institution; situations where author and referee are competitors in some sense, or belong to competing schools of thought;

  • Abuse

Too many articles out of one piece of research, or duplicate publication; intellectual theft: omission or downgrading of junior staff by senior authors; plagiarism (stealing others yet unpublished work that has been sent for review), delaying publication of potentially competing research;

  • Detecting defects
Identification of factual errors within submission
  • Fraud misconduct

Fabrication of results; falsification of data false claim of authorship for results.6

 

Among the most common complaints about peer review is the long delay associated with the review process. As observed by Stevan Harnad, an active proponent of author-self-archiving and institutional repositories, "[t]here is much muttering about publication delay, a real enough problem, especially in paper publication, but peer review itself is often responsible for as much of the delay as the paper publication and distribution process itself."7

 

Scientific Publishing as Rhetoric

 

For some, the fundamental problems of peer review are inherent in the peer review process itself as it is currently implemented. As noted by Sosteric,

 

"the traditional mode of peer review obscures the problems of reference and the rhetorical dimension of science. The rhetorical process ... [that] is at the heart of science and peer review conveniently disappears with the final publication of the manuscript. In its place is an ideal typical representation (the scientific paper) of the realist assumptions about empirical reference. All the academic world sees is a polished manuscript where the personal involvement of the researcher and reviewers has been systematically eliminated.8

 

As an alternative to conventional peer review, Sosteric, Gross, and others, promote the framework of the 'ideal speech situation', a "theoretical construct that describes the ideal type of interpersonal interaction that should exist in a rhetorical situation” proposed and developed by Jürgen Habermas, the noted German philosopher and sociologist.9 Drawing upon Habermas, Gross describes the ideal speech situation in the following terms: 1) the ideal speech situation permits each interlocutor an equal opportunity to initiate speech. 2) there is mutual understanding between interlocutors; 3) there is space for clarification; 4) all interlocutors are equally free to use of any speech act; and, 5) there is equal power over the exchange.10

 

As applied in the context of peer review, Gross notes that ideally "scientific peer review would permit unimpeded authorial initiative, endless rounds of give and take, [and] unchecked openness among authors, editors, and referees.11

 

" ... The Times They Are A-Changin'"

 

In 2001, CERN, the European Organisation for Nuclear Research, in Geneva, Switzerland, served as the venue for the first Workshop on the Open Archives Initiative (OAI). Focused on OAI and 'Peer Review Journals in Europe,' the purpose of this workshop was to “mobilise a group of European scientists and librarians who want to play an active role in organising a self-managed system for electronic scholarly communication as a means to address the serials crisis. Such a system should be compliant to the technical standards proposed by the Open Archives Initiative (OAI) [www.openarchives.org].”12

 

Two years after the workshop, a policy briefing of the European Science Foundation was published. The publication not only profiled the variety of issues relating to Open Access and the OAI, but also summarized the themes of the first OAI workshop, and the second held at CERN in mid-October 2002, as well. In addition, the briefing included consensus recommendations from each workshop. While the "participants [of the first workshop] were unanimous in their belief that the certification of scholarly work remains a fundamental part of a system for scholarly communication," they also "believed that the electronic environment allows for novel approaches to accord a stamp of quality to scholarly works." The suggested 'new metrics' that could be extracted from a fully electronic communication system include the discussion level generated by a paper submitted to a publication system with open peer review and peer commentary features; automated citation indexing beyond the standard Institute for Scientific Information (ISI) print-focused service; and access statistics.13

 

LAMPSS: Lots of Alternative Models Provide Sensible Solutions

 

[1] Open Peer Review

 

During much of its recent history, conventional peer review has been wholly or partially anonymous. In the former arrangement, neither reviewer nor the author is known to each other; in the latter, the author is identified. Until five years ago, the British Medical Journal (BMJ), the high-impact, general medical journal of the British Medical Association, had "used a closed system of peer review, where the authors do not know who has reviewed their papers ... but the reviewers do, however, know the names of the authors."14 In announcing a change in its editorial policy, Richard Smith, the BMJ editor further observes that  “[m]ost medical journals use the same system, ... based on custom not evidence. Now we plan to let authors know the identity of reviewers. Soon we are likely to open up the whole system so that anybody interested can see the whole process on the World Wide Web. The change is based on evidence and an ethical argument.15,16 He further notes that "the primary argument against closed peer review is that it seems wrong for somebody making an important judgment on the work of others to do so in secret." In a supportive argument, Smith quotes Drummond Rennie, a deputy editor of JAMA: The Journal of the American Medical Association, stating that identifying the reviewer links "privilege and duty, by reminding the reviewer that with power comes responsibility: that the scientist invested with the mantle of the judge cannot be arbitrary in his or her judgment and must be a constructive critic."17

 

Journals that have implemented open peer review include not only the British Medical Journal (www.bmj.com), but also select journals published by BioMed Central (www.biomedcentral.com), as well as Internet Health: Journal of Research, Application, Communication & Ethics, (www.virtualmed.netfirms.com/internethealth/), among others.

 

[2] Commentary-Based

 

A journal that offers a novel form of 'open review' as well as conventional peer review is the Electronic Transactions on Artificial Intelligence (ETAI) (www.etaij.org). "The ETAI is an electronic journal that uses the Internet medium to the fullest extent for scientific communication: not merely for distributing the articles, but also for the review procedure, for bibliographic infrastructure, [and other applications]."18 With respect to reviewing, “it ... use[s] a novel, two-stage procedure where the first review phase is open and allows the peer community to ask questions to the author, and to create a discussion about the contribution. The second phase - called refereeing in the ETAI - is like conventional journal refereeing except that the major part of the required feedback is supposed to have occurred already in the first, review phase.”19 After refereeing, accepted articles are published in the ETAI journal in both a paper edition and an electronic edition where the latter is available free of charge, and contains a variety of other information, including past discussions.20 21

 

Launched in September 1996, the Journal of Interactive Media in Education (JIME) (www-jime.open.ac.uk), a borne-digital e-journal has incorporated one of the most varied forms of peer interaction and review. Its review environment provides an "opportunity to redesign the conventional journal review model to be more open, responsive and dynamic" In the JIME review model "authors have the right of reply", "reviewers are named and accountable for their comments, and their contribution acknowledged," and the wider research community has the chance to shape a submission before [forma] publication."22 Specifically,[a]rticles submitted to JIME are first reviewed by three reviewers who are named, and acknowledged for their contribution to a review. They post their reviews as threaded comments to a private [Web] site. Reviewers have the option of posting anonymously, but usually ... [they] are happy [to be] ... named... . [In turn], authors are encouraged to respond to [reviewer] comments. This [particular part of the process] takes place during an agreed period when authors and reviewers are able to respond in a timely manner.”23

 

On the basis of this interaction, if the assigned editor ascertains that the manuscript is of acceptable quality based on this discussion, the submission "will be published will then be published as a preprint for public open peer review, and announced to relevant communities to invite their participation." This second phase of the JIME open review is open for a month after formal posting. After public review and comment, the assigned JIME editor will "post to the discussion an editorial report summarising the most significant issues, and specifying change requirements to the authors." After formal acceptance and publication, JIME continues to support discussion about the revised, published article. In addition, the most interesting review comments/exchanges are published with the final version of the revised submission, "providing readers with insight into the issues that arose during review, and enabling them to build on those discussions."24

 

Within JIME, authors can “post links to publications to point to subsequent work. Readers can post comments and links to point to work which has not been referenced, or did not exist when the article was written. [In addition], [a]uthors, reviewers and anyone else who has subscribed to the article will receive email alerts to new postings to its discussion forum.”25

 

[3] Community Based

 

Paul Ginsparg, the creator and developer of arXiv.org, 26 the physics e-print service originally based at the Los Alamos National Laboratory, has proposed a tier-based hybrid review system that offers enhanced conventional review in conjunction with open submission and alternative assessment. In his model, any and all submissions are accepted after a cursory examination or other pro forma certification and assigned to a 'standard' tier. This phase of this review process would be "minimally labor-intensive, perhaps relying primarily on an automated check of author institutional affiliation, prior publication record, research grant status, or other related background; and involve human labor primarily to adjudicate incomplete or ambiguous results of an automated pass."27

 

"At some later point (which could vary from article to article, perhaps with no time limit), a much smaller set of articles would be selected for the full peer review process [and assigned to an upper tier]. The initial selection criteria for this smaller set could be any of a variety of impact measures, to be determined, and based explicitly on their prior widespread and systematic availability and citability: e.g., reader nomination or rating, citation impact, usage statistics, editorial selection, ... ."28

 

It is expected that this approach would lead to an overall efficiency in the review process by expending 'community intellectual resources' on submissions that are "most likely to be archivally useful, and hence merit the enhanced editorial treatment for upgrade into the upper tier, including, for example, text clarifications and other improvements."29 Upper tier enhancements could also include “anything from a thorough blind refereeing to open professional annotation and comment. The upper tier could also combine commentary on many related papers at once. The point is that it's possible to provide more signal of various sorts to users on a smaller subset of articles, without worry about fairness issues of limited dissemination for the rest, and this can be done at lower overall cost than the current system, both in time spent by editors and in elective time spent by referees.”30

 

As noted by Ginsparg, "the standard tier would provide a rapid distribution system only marginally less elite than much of the current publication system, and enormously useful to readers and authors," particularly in fields "in which the time to publication time is perceived to be too long." In addition, “the standard tier availability could also be used to collect confidential commentary from interested readers so that eventual referees would have access to a wealth of currently inaccessible information held by the community ... ."31

 

More than five years ago, David Stern, Director of Science Libraries and Information Services at Yale University, proposed a model with many of the components of the Ginsparg approach. Stern envisioned a system that would allow "the widest range of scientific manuscripts to be archived, searched, and distributed electronically at the lowest possible cost."32 As Stern notes, "this would be accomplished through very minimal filtering and subsequent placement of eprints on a non-commercial archival server by a subject-specific Moderator appointed by a society (or consortia of societies). A society-appointed Editorial Board (with double-blind peer review approved by the non-profit Peer Review Inc. organization) would then identify the most important materials from among these archived items, and the stamp of approval for these items would be included in a secondary Virtual Collection."33

 

In the Stern model, there are no direct submissions to the Editorial Board, instead manuscripts would be directed to the Editorial Board in one of three ways: “1. nominated by the eprint Moderator upon receipt for the archival server; 2. notification sent to the Editorial Board when a threshold number of hits are generated by any one manuscript on the archive server; or, 3. nominated by readers of material from the archive; this process requires a letter of support outlining the importance of the work to the Editorial Board.”34

 

[4] Usage-based

 

As noted several alternative peer review models recommend a metric that uses access statistics as an indicator of significance. While the European Science Foundation policy briefing observes that 'new metrics' could include the discussion level generated by a paper, as well as access statistics as well, the Ginsparg model, recommends that second tier review candidates might be selected on "a variety of impact measures, that would include reader nomination or rating, or usage statistics." In the Stern model, candidates for review for an editorial board would be identified "when a threshold number of hits are generated by any one manuscript on the archive server." More recently, Harnad, notes that a "new potential measure of on-line impact, not available in the on-paper era, is usage, in the form of 'hits.' This measure is noisy ([in that] it can be inflated by automated web-crawlers, short-changed by intermediate caches, abused by deliberate self-hits from authors, and undiscriminating between nonspecific site-browsing and item-specific reading) ..., [but] seems to have some signal-value too, partly correlated with and partly independent of citation impact ... ."35

 

[5] Citation-based

 

More than fifty years ago Eugene Garfield developed and applied a method of citation indexing to enhance access to the scholarly scientific journal literature.36 In October 1999, the Open Citation Project: Reference Linking and Citation Analysis for Open Archives (http://opcit.eprints.org/) was officially inaugurated. One of its most notable outcomes was the creation of Citebase (http://citebase.eprints.org/cgi-bin/search), a Web-based citation index for Web scholarship and "Google for the refereed research literature."37

 

Among its various functionalities, Citebase can provide several of the established scientometric measures of research impact, notably citation counts for a Web-based article, citation counts for the researcher, as well as co-citation analyses. In addition, it has enabled the creation, development, and implementation of new measures of 'impact', notably citation counts for the preprint phase of a publication, usage measures ('hits') for preprints and postprints, time-course analyses, and usage/citation correlators and predictors. Using Citebase, pre- and post-publication citations for individual papers can be measured against usage for the first time.38

 

More recently in February 2004, the Institute for Scientific Information (ISI) - the producer of the Web of Science and numerous specialized print and electronic citation indexes covering a variety of disciplines and subjects, announced a new initiative to create, test, and provide a 'Web Citation Index' for Web scholarship. In addition to tracking online publications, the new index will also document citations to print works by online sources.39

 

The new service will utilize the NEC CiteSeer technology (http://citeseer.ist.psu.edu/cis), which includes "extraction of bibliographic citations, autonomous citation indexing, calculating citation statistics and related documents, reference linking to cited articles, citation context display, automatic notification based on user profiles, correlation of related documents, full-text indexing, query-sensitive summaries of the context of search terms in an article, citation graph analysis, and targeted Web crawling."40

 

New Opportunities

 

"Let us be imaginative in exploring the remarkable possibilities of this brave new medium."41

                                                Sevan Harnad

 

As observed by Harnad, "the Net ...[not only] offers the possibility of implementing peer review more efficiently and equitably ...," but more significantly, provides a "real revolutionary dimension" with such features as "open peer commentary on published and ongoing work." In addition, the Net provides “room … for unrefereed discussion too, [notably] in high-level discussion forums ...."42 Such enhancements to conventional peer review need not, however, be limited to features that some may view as simple extensions of the traditional model. In addition to 'ideal' conversations, metrics such as access statistics, as well as citing and linking, can also offer impartial indicators of valid and significant scholarship in all its forms, at any and all stages.

 

Endnotes

 

1. Jean-Claude Guédon, In Oldenburg's Long Shadow: Librarians, Research Scientists, Publishers, and the Control of Scientific Publishing (Washington, D.C.: Association of Research Libraries, 2001), 54. Also available at: http://www.arl.org/arl/proceedings/138/guedon.html (accessed April 8, 2004).

 

2. Carin M. Olson, "Peer Review of the Biomedical Literature," American Journal of Emergency Medicine 8 no. 4 (July 1990): 356.

 

3. Ibid.

 

4. Anne C. Weller, Editorial Peer Review: Its Strengths and Weaknesses (Medford, NJ: Information Today, 2001), 307-308.

 

5. Richard Smith, "Opening Up BMJ Peer Review," BMJ 318 no. 7175 (January 2 1999): 4-5. Also available at: http://bmj.bmjjournals.com/cgi/content/full/318/7175/4 (accessed April 4, 2004).

 

6. Fytton Rowland, "The Peer-Review Process," Learned Publishing 15 no. 4 (October 2002): 247-258. Report version available at http://www.jisc.ac.uk/uploaded_documents/rowland.pdf (accessed April 4, 2004).

 

7. Stevan Harnad, "Implementing Peer Review on the Net: Scientific Quality Control in Scholarly Electronic Journals,” in Scholarly Publication: The Electronic Frontier, ed. by Robin P. Peek and Gregory B. Newby, 112 (Cambridge MA: MIT Press, 1996). Also available at: http://www.ecs.soton.ac.uk/~harnad/Papers/Harnad/harnad96.peer.review.html (accessed April 4, 2004).

 

8. Mike Sosteric, "Interactive Peer Review: A Research Note," Electronic Journal of Sociology 2 no. 1 (1996), http://socserv.socsci.mcmaster.ca/EJS/vol002.001/SostericNote.vol002.001.html (accessed April 4, 2004).

 

9. Antti Kauppinen, "Habermas Links," http://www.helsinki.fi/~amkauppi/hablinks.html (accessed April 4, 2004).

 

10. "Interactive Peer Review."

 

11.Alan G. Gross, The Rhetoric of Science (Cambridge, MA: Harvard University Press, 1990), 137.

 

12. Raf Dekeyser, Herbert Van de Sompel, and Corrado Pettenati, "Workshop on the Open Archives Initiative (OAI) and Peer Review Journals in Europe," http://preprints.cern.ch/OAi/ (accessed April 4, 2004).

 

13. Alison Buckholtz, Raf Dekeyser, Melissa Hagemann, Thomas Krichel, and Herbert Van de Sompel, "Open Access: Restoring Scientific Communication to Its Rightful Owners," European Science Foundation Policy Briefing 21 (April 2003): 1-8. Also available at: http://www.arl.org/sparc/SPB21_OAI.pdf (accessed April 4, 2004).

 

14. "Opening Up BMJ Peer Review.”

 

15. Ibid.

 

16. See also: bmj.com, "Our Peer Review Process," http://bmj.bmjjournals.com/advice/peer_review.shtml (accessed April 4, 2004).

 

17. "Opening Up BMJ Peer Review."

 

18. Electronic Transactions on Artificial Intelligence, "General Information about the ETAI," http://www.ida.liu.se/ext/etai/tempvis/welcome/sframe.html (accessed April 4, 2004).

 

19. Ibid.

 

20. Ibid.

 

21. See also Electronic Transactions on Artificial Intelligence, "How to Submit," http://www.ida.liu.se/ext/etai/orj-howsubmit.html (accessed April 4, 2004).

 

22. Journal of Interactive Media in Education, "About JIME," http://www-jime.open.ac.uk/about.html (accessed April 4, 2004).

 

23. Ibid.

 

24. Ibid.

 

25. Ibid.

 

26. Gerry McKiernan, "arXiv.org: The Los Alamos National Laboratory e-Print Server," International Journal on Grey Literature 1 no. 3 (2000): 127-138. Also available at: http://www.public.iastate.edu/~gerrymck/arXiv.org.pdf (accessed April 4, 2004).

 

27. Paul Ginsparg, "Can Peer Review be Better Focused?," Science & Technology Libraries 22 no. 3/4 (2004): 14. Preprint available at http://arxiv.org/blurb/pg02pr.html (accessed April 4, 2004).

 

28. Ibid.

 

29. Ibid.

 

30. Ibid.

 

31. Ibid., 14-15.

 

32. David Stern, "The eprint Moderator Model," in Guide to Information Sources in the Physical Sciences (Englewood: CO: Libraries Unlimited, Inc., 2000), 168. Also available at http://www.lib.unc.edu/prices/1999/PRIC214.HTML#214.5.

 

33. Ibid.

 

34. Ibid., 169.

 

35. Stevan Harnad, "Scientometric OAI Search Engines, "Posting to LIS-ELib [LIS-ELIB@JISCMAIL.AC.UK], Sun, 25 Aug 2002 15:28:55 +0100, http://www.jiscmail.ac.uk/cgi-bin/ wa.exe?A2=ind0208&L=lis-elib&T=0&F=&S&P =1362 (accessed April 4, 2004).

 

36. Eugene Garfield, Citation Indexing: Its Theory and Application in Science, Technology, and Humanities (New York: John Wiley & Sons, 1979).

 

37. Steve Hitchcock, Donna Bergmark, Tim Brody, Christopher Gutteridge, Les Carr, Wendy Hall, Carl Lagoze, Stevan Harnad, "Open Citation Linking: The Way Forward," D-Lib Magazine 8 no 10 (October 2002) http://www.dlib.org/dlib/october02/hitchcock/10hitchcock.html (accessed April 4, 2004).

 

38. Steve Hitchcock, Tim Brody, Christopher Gutteridge, Les Carr and Stevan Harnad, "The Impact of OAI-based Search on Access to Research Journal Papers," Serials 16 no. 3 (November 2003): 255-260. Also available at: http://opcit.eprints.org/serials-short/serials11.html (accessed April 4, 2004).

 

39. Vincent Kiernan, "Company to Track Citations of Online Scholarship," The Chronicle of Higher Education 50, no. 28, (March 19, 2004): A31.

 

40. Barbara Quint, "Thomson ISI to Track Web-based Scholarship with NEC's CiteSeer," Information Today, News Break & The Weekly News Digest, (March 1, 2004), http://www.infotoday.com/newsbreaks/nb040301-1.shtml (accessed April 4, 2004).

 

41. "Implementing Peer Review on the Net," 115.

 

42. Ibid., 116.

 

Note: This article is based on “Alternative Peer Review: Quality Management for 21st Century Scholarship,” an invited presentation delivered at the Workshop on Peer Review in the Age of Open Archives held May 23-24, 2003 at the International School for Advanced Studies (SISSA), Trieste, Italy A copy of the corrected and revised PowerPoint presentation is available at <http://www.public.iastate.edu/~gerrymck/APR.ppt> .

 

“Peer Review in the Internet Age: Five (5) Easy Pieces,” Against the Grain 16, no. 3

(June 2004): 50, 52-55.

| FINAL | 04-23-04 | 11:45| ~3,720 |