EPUBs are an experimental feature, and may not work in all readers.

This paper was refereed by the Journal of Electronic Publishing's peer reviewers.

Abstract

Electronic journals are offered in many forms, some available in both print and electronic form, and some in electronic form alone. Purely electronic journals have had the opportunity to develop new features and formats that take advantage of the online environment. This paper uses a genre-based analysis to look at online journals in two fields, physics and psychology, focusing on whether purely electronic journals do in fact feature elements and formats that would be impossible to present in print form, and the prevalence of such features across the fields. Journals were examined for the use of non-linearity, interactivity, multimedia, multiple use, and rapid publication. The results of the study indicate that additional features are only selectively used in electronic journals, and suggest that few incentives exist for authors and publishers to utilize additional features more, as conventionally structured articles fulfill the needs of the scholarly journal article genre.

Scholarly journals, along with monographic works and conference proceedings, are the primary medium for the formal display and dissemination of knowledge in the academic community, and have been for hundreds of years (Swan 2006). Over the past decade, as the utility of the electronic medium expanded, scholarly journals have increasingly become Internet-based. Initially, the perceived benefits of the electronic medium seemed to some observers to inevitably lead to the obsolescence of printed journals. The Internet was expected to facilitate greater “speed, scope, and interactiveness” in scholarly communication (Harnad 1991), allowing scholars to bat ideas back and forth, enabling rapid mutual feedback. New computer technology also allowed for the presentation of new forms of information, such as visual elements like charts and graphs, which could be displayed more easily and in greater detail in an electronic setting. Features such as audio or video could be inserted. Another incentive for the development of electronic journals was the perceived economic benefit of the new medium. Finally, it was thought that Internet journals might provide a means to skirt expensive conventional journal publishers, a prime incentive due to rapidly increasing journal prices (Harnad 1998).

Current journals come in a variety of forms, both electronic and print, but electronic publications have not superseded the print publishers, as some early observers had predicted. Statements in the vein of, “Electronic journals should not and will not be mere clones of paper journals” (Harnad 1991), have proven to be slightly off key. While it is true that Internet journals have very strongly taken root, and some electronic journal developers have established new formats and features specific to Internet technologies, not all have done so. There have been few studies of how electronic-only journals implement Internet-specific features. This study examines whether the majority of articles published in electronic-only journals do in fact utilize the advantages of the electronic medium, or whether articles still largely mimic the format of a printed journal article.

Background

Many different types of journals are available today. Kling and McKim (1999) separate electronic journals into four categories:

  • Pure E-journals, which were born digital and are only available in electronic form.
  • E-P journals, which are primarily distributed electronically, but may have limited distribution in paper form for archival or other reasons.
  • P-E journals, which are packages of peer-reviewed articles available through electronic channels, but whose primary distribution channels are paper based.
  • P+E journals, which were started as combined electronic and print journals.
The challenge with any publication model is attracting submissions and readers. Initially, electronic publications carried a stigma due to the fear that they wouldn’t be considered “real” publications. The relative success and influence a scholar enjoys in his or her field is often measured by the number of papers published, along with the number of citations received. Because of this, many universities use publication and citation data, specifically publication in prestigious journals, as a criterion for promotion and tenure decisions. Citation indexes such as the ISI Web of Science compile journal citation data to determine an impact factor for many individual journals, with the more widely cited publications having higher impact factors. Scholars therefore have had a great, and what Odlyzko calls a “perverse” (1997), career incentive to continue publishing in the established high-impact-factor journals, and little incentive to publish in newer low-impact electronic only journals. However, as early as 1995 studies showed that there was little actual negative response to electronic publications in the formal academic evaluation processes (Butler 1995), although there may have been “inconsistencies” in interpretation and practices within and across institutions (Cronin & Overfelt 1995). In a series of interview case studies to explore academic value systems and their influence on publishing behavior and attitudes of faculty at the University of California, Berkeley, Harley, et al. (2007) observed that these “inconsistencies” largely stem from differences in practices between disciplines and departments within institutions.

Despite some reservations, the electronic environment is now the fastest growing arena in scholarly publishing and there is great desire and incentive for journals to be accessible online. A recent survey by King, et al. indicated that scholars highly value electronic journal access; in fact, the vast majority of survey respondents, 85%, stated that they preferred electronic journals over print, primarily for the following reasons: electronic journals save time, make work easier, result in better quality research, and enable the scholar to find more materials (2003). Additionally, other research shows that this preference transfers into reading patterns (Tenopir, et al. 2003).

This does not mean that print publishers are going out of business in large numbers. Instead, many of them have followed the emerging trend and entered the electronic journal market. A 2001 study found that approximately two-thirds of print journal publishers offered an online version of their journals (Tenopir & King 2001). This number is certainly higher now. Some publishers offer online access only to more recent volumes, but many have digitally scanned previously printed volumes, in some cases providing access back to the origin of the journal. However, most P-E journals produce electronic versions that follow the same linear format as the printed journal. Because authors textually format articles for print prior to submission for publication, it is simple to reproduce the articles in the same form in both versions of the journal.

So how do researchers value electronic journals? Swan and Brown (2003) surveyed over 1,200 scholars around the world to obtain their views on electronic publishing and the usefulness and added value of additional electronic features. They found that linking from citations to cited articles was rated as the most valuable added feature (rated as “very important” or “important” by 88% of responders), followed by the inclusion of additional data (57%), additional or color images (45%), manipulatable content such as software, simulations, online experiments, etc. (23%), and video or sound (14%). The ability to publish articles as soon as they are finalized was valued as ”very important” or ”important” by 64% of responders. Peer review also ranked high in terms of value, receiving a ranking of “very important” or “important” by 94% of responders. Of the other types of electronic feedback included in the survey, the ability to submit comments about an article, availability of post-publication public commentary, publication of referee comments, referee identification, and availability of public commentary on preprints, all were valued as “not important” by a majority of responders (2003).

Harley, et al. (2007) found similar results. The interviewees in this study reported a number of the features of online publishing as useful, including the ability to reach a larger audience, ease of access by readers, more rapid publication even when a paper is peer reviewed, the ability to search within and across texts, and the opportunity to make use of hyperlinks. Peer review was again reported as an essential factor in how scholars appraise publications, and a critical factor in the acceptance of electronic journals. Many scholars interviewed in this study reported utilizing new modes of scholarly communication and publication, but most relied on traditional publishing formats for the formal dissemination of their scholarly products, as they felt that publications in traditional venues would carry more weight when the time came for career evaluations.

Both Swan and Brown (2003) and Harley, et al. (2007) reported that scholars valued electronic journals as a dissemination medium, but expressed concern about the ultimate longevity and accessibility of electronic publications. Print publications are highly valued by scholars because of the guarantee of long-term preservation. Many journal publishers have favored the print form as the means of permanent storage as well, as there is no widely used and reliable method for storing digital materials (Borgman & Furner 2002). This appears to be changing, however, as digital preservation techniques become a more significant issue. Moghaddam (2007) found that archival practices among electronic publishers vary, from CD-ROM archiving to DSpace repositories to digital archiving deals with the National Library of the Netherlands (Koninklijke Bibliotheek, or KB). Of the six major publishers surveyed by Moghaddam, all of the for-profit publishers favored archiving digital materials with the KB, while not-for-profit publishers utilized varied digital archiving practices.

Research

So why develop an electronic journal in the first place? Designing for a new medium “requires some rational understanding of who is using the materials, what they are doing with them, and how they fit into an overall way of life” (Agre 1995). What sets an electronic journal apart from a standard print journal? Approaching this question with the idea of genre can be useful in providing some answers. A genre is a “socially recognized, repeated strategy for achieving similar goals in situations socially perceived as being similar” (Miller 1984, from Bazerman 1988, pg. 62). Yates and Orlikowski define communicative genres more specifically as “typified rhetorical action in the context of socially defined recurrent situations” (1992, pg. 301). Journal articles perform a specific communicatory function within the academic community, and can be categorized as a genre of scholarly communication. Yates and Orlikowski also note that a medium of communication should not be confused with a genre of communication. Genres may be “physically created, transmitted, and stored in various media” (1992, pg. 310). Scholarly journal articles published both electronically and in print can be considered part of the same communicative genre.

Yoshioka, et al. (2001), in a discussion of genre taxonomy, detail a number of dimensions that can be used to analyze genres. Their core dimensions represent the why (purpose), what (content), who (participants), when (timing), where (location), and how (structure and medium) of communication. Applying these dimensions in an analysis of scholarly journal articles can help illustrate more clearly the difference between electronic and print mediums.

The “why” dimension, or purpose, of a scholarly journal article can be expressed generally as the desire of the author to transmit knowledge to a reader. There are other, more pragmatic reasons why authors want to publish articles, such as to help further their careers, or because a grant stipulation requires a certain number of publications from that research. But each of these cases also requires that the author express information in a way that will succeed in transferring knowledge to a reader. If an author is not successful in transferring knowledge to others, he or she will be much less likely to receive promotions or new positions; similarly, if a grant does not result in knowledge that can be displayed to others, it is very unlikely that the recipient will receive more grant money in the future.

It then follows that the “what” dimension, or content, of a scholarly journal article is the knowledge that is being transmitted. The “what” dimension is very wide in scope for scholarly articles, as the content of an article has a limitless range of possibilities. The content can take many forms. In the physical sciences it might be the results of an experiment, such as a newly discovered subatomic particle observed in an atomic collider. In mathematics the knowledge might be an original analysis of a proof, or a solution to a previously unsolved equation. An example in the social sciences might be a recommendation for new ways to implement low-cost housing when reconstructing run-down urban areas, or a study of the rhetoric of American foreign policy.

Peer-reviewed print and electronic journals do not differ in the “why” or “what” dimensions. They both have the same purpose for a specific content, and both engage the same participants. However, there are differences between the two mediums in the other four dimensions, the “who,” “when,” “where,” and “how.”

The “who” dimension, or participants, of a scholarly article consists of multiple groups: authors, editors and peer reviewers, publishers, and readers. Journal articles are written with the intention of communicating with other members of the academic community. Authors, editors, peer reviewers, and readers are derived from the same group of academics, who serve as the producers, quality controllers, and consumers of knowledge. Publishers sit between authors and readers, as intermediaries through which knowledge is transferred. Large publishing corporations maintain control of much of the scholarly publishing process, but the electronic environment has shifted some control over to the academy. One example of the changing face of scholarly publishing is the Scholarly Publishing and Academic Resources Coalition (SPARC). SPARC, an initiative of the Association of Research Libraries (ARL), endorses a number of journals for its Alternatives Program, which supports “publishing projects that represent a direct and strong competitive alternative to existing high-priced titles in important established [science, technology and medical] fields” (SPARC 2007). In a survey of 18 SPARC journals and partners, Lustria and Case found that the movement is largely scholar driven (2005). Most of the surveyed SPARC partners come from the academy, and joined SPARC after libraries demanded alternative publication models to counteract rising journal costs.

The “when,” or timing, of a printed journal article is restricted by the time constraints involved in producing a printed volume. The print process involves a series of steps. An appropriate number of articles must be collected and peer reviewed, followed by the mechanical production of typesetting, printing, and binding among other steps. Printed journals are then sent through the mail to subscribers. The printing and mailing lead time restricts how often a journal volume can be published. An electronic journal does not have many of those restrictions. The speed of electronic publication is essentially determined by the time required to put an article through the peer review process and conduct final editing and formatting. An article can be sent out to a peer reviewer electronically as soon as it is submitted, and as soon as it has been accepted by the peer reviewers and edited and formatted by the editor it can be posted on the journal Web page. In practice this isn’t always the case. E-journals often collect articles into groups and publish “volumes” like printed journals. But the capability for rapid article publication is a significant difference between the two mediums.

The “where” dimension, or location, is also different for online journals. Prior to the development of electronic journals, it was necessary either to go to the library or to request an article through interlibrary loan in order to acquire an article that wasn’t available through a scholar’s personal subscriptions. With the majority of journals now available electronically, the locus of scholarly research has shifted away from the library building. An article can be acquired much more conveniently online using an office or home computer, allowing scholars to access more articles than might otherwise be possible (Tenopir, et al. 2003; Brady, McCord, & Galbraith 2006).

The final dimension, the “how,” encompasses the structure or medium of a scholarly article. This is the primary dimension of interest, as print and electronic media have very different capabilities when it comes to the structure and display of an article. A main question has been whether the electronic medium will facilitate a move away from the linear article form. Kircz (1998) has put forth a model for electronic articles based on a modular structure instead of a linear one. In a modular structure, the individual facets of an article are separated, and each section of the article is placed in its own display page, or module. Kircz identified the following sections of an article to be the main elements of interest: bibliographic information (the standard information used to identify the article: author names and addresses, publication information, length, etc.), content, index terms, references, acknowledgements, and the abstract. He advocates sub-modules for the content section, each of which contains a sub-section of the article, including the research methods, measurements, results, and conclusions. Graphs or images may also be separated. Ideally, modules would allow a reader to move through the material in whatever fashion he or she desires.

This ability to jump from one area to the other with ease is described as one of the primary benefits of a modular form (Kircz 1998). Studies have shown that readers of scientific articles do not necessarily read through an article in a linear fashion (Bishop 1999). A reader might start with the abstract, jump to the bibliography, and only then get around to actually reading the article, if he or she reads it at all. Alternatively, readers might only read the introduction and conclusion, while skipping the methodology, analysis, or other main body sections. However, there have been no formal studies as to whether a modular publication model actually facilitates non-linear reading more than the standard linear article.

Another unique capability of the electronic medium is the ability to insert items that might not be presented in the main body of a conventional journal article, such as large data sets, video and audio segments, mathematical analyses, and derivations. The primary advantage of an electronic environment over print, according to Kircz (2002, pg. 28), is the ability to integrate “all possible expressions of scientific knowledge” into one carrier. Paper journals are constrained to textual articles, and have only limited ability to display charts, graphs, and other visually oriented material. Other methods for expressing information include texts, images, sounds, and simulations, all of which present knowledge in different forms. The same informational material could be displayed using a number of different mediums. For example, in physics a presentation on the Doppler effect could give a textual description of the phenomena—observable wavelength changes in emitted waves due to the relative motions of the emitting and observing bodies. The presenter could also create a graphical image displaying a waveform compressing or expanding, with arrows added to indicate motion between bodies. And finally, a sound or video recording of an approaching and subsequently receding train could also be used to illustrate the phenomenon. In an electronic environment, all of these items could be included, allowing a comprehensive illustration of the Doppler effect. A presentation of this phenomenon in a print setting would be limited to a textual description and derivation and, if the publication has the capability for visual images, perhaps diagrams or graphs.

A fascinating example of how non-textual material can be used to supplement traditional articles is SciVee (http://www.scivee.tv/), described in Fink and Bourne (2007). SciVee is a Web service that allows authors to upload already published papers and add a video or podcast to highlight the key points of the paper. The video can be synchronized with the paper to allow the relevant points to be displayed as they are being discussed in the video. Fink and Bourne also describe BioLit, a set of open source tools under development to enable the integration of open-access academic literature and biological data. The nascent e-science and cyberinfrastructure movement is predicated on the discovery and integration of related digital objects in just this fashion (Van de Sompel, et al. 2006). As e-science and cyberinfrastructure initiatives mature and extend to more and more disciplines, it is expected that scholarly communication will increasingly incorporate non-textual material (Lynch 2007).

The other notable feature of electronic publishing is “multiple use,” or linking to the full text of a cited article. In a traditional article, previous works are used as references through direct quotes or by citations at the relevant points in the paper, and a complete list of the references is given at the end in the bibliography. Multiple use differs from quotation in that “in multiple use the author can rely on the completeness and integrity of the quoted work” (Kircz 2002, pg. 28). In an electronic environment, links to references can be built into the article, enabling a reader full access to a cited work, whereas in the printed article the author either quotes or paraphrases small selections. Of course, multiple use requires that the articles listed in the bibliography are also available and accessible electronically, which is not always the case. More and more journals are available online as electronic journals become more prevalent and print journal publishers adapt themselves to the electronic environment, but a comprehensive article linking system does not yet exist. The SFX generic linking system of Van de Sompel and Hochstenbach (1999a and 1999b) is one example of a system developed to help libraries achieve full-text article linking. While not a perfect solution to the multiple use problem, SFX has proven to be a useful step forward in citation linking (Wakimoto, et al. 2006), and was the foundation of the subsequent OpenURL framework for open reference linking (Van de Sompel & Beit-Arie 2001). Other notable approaches to achieve reference linking include the DOI-based CrossRef Linking environment (CrossRef.org 2007) and metadata-based citation matching (Hitchcock, et al. 2002).

Another important aspect of the electronic medium is the increased ability of scholars to engage in interactive forms of communication (Harnad 1992). Scholars have always interacted with each other when developing new ideas for research and publication. Suggestions from colleagues are often instrumental in focusing the course of a research idea, and in facilitating necessary changes or improvements to a pre-published paper. Authors of academic papers are well aware of the benefits that this sort of interactive communication can offer. Over the past century this has solidified into the practice of including a short acknowledgement section in a published paper, giving credit and expressing indebtedness to colleagues who assisted in the formulation of the paper (Bazerman 1988, Cronin 2001). However, traditionally a scholar has been limited to only select groups of individuals for feedback on research ideas, such as local institution faculty members, formal peer reviewers, and other personal contacts. The electronic medium can facilitate additional modes of communication through e-mail and other interactive features like message boards, comment posting, and response forms.

Method

This brings us to the question of whether pure e-journals have in fact been taking advantage of the unique characteristics offered by the electronic medium in the ways suggested by early e-journal researchers, or whether they’ve been following the traditional printed journal format. This study looked at 21 electronic-only journals in two fields, 10 physics journals and 11 psychology journals. The study specifically looked for the inclusion of features that would not be possible in a printed journal.

The subject areas in the study, physics and psychology, were selected for their historical significance in the development of journal formats. Bazerman (1988) traced the course of scientific communication, and primarily the scientific journal article, from the inception of the first English language scientific journal, the Philosophic Transactions of the Royal Society of London in 1665, up to the current day. His study delineated the gradual development of journal articles over the centuries into the form prevalent today. Since the late nineteenth century, experimental physics journal articles have embedded “common theory” as a force in “structuring articles and binding articles together.” The centrality of theory directs physics scholars to “certain kinds of experiences, suggests the appropriate means of designing and interpreting empirical events, and allows results to be harmonized with the results and ideas of others” (pg. 157). Experiments are designed with the intention of proving or disproving theoretically predicted phenomena. Observations and data are gathered, analyzed, discussed, and only then is the article written. Articles “represent the consensus, the ‘facts,’ data with the noise removed” (Traweek 1988, pg. 122). Over time, the theory-based experimental structure resulted in an accepted format for physics journal articles.

The format of a natural science article, with its reliability and great capacity for detail, was desirable for its ability to build agreement among scientists. Other academic communities, influenced by this, developed similar communication forms. Among the human sciences, experimental psychology took the lead in developing its own format, and became a model for other groups, such as the social and political sciences. Today the American Psychological Association Publication Manual “symbolizes and instrumentally realizes the influence and power of the official style” (Bazerman 1988, pg. 259). Electronic journals therefore have centuries of precedent to consider when regarding the content, style, and formats of new publication models.

The scope of this study was limited to electronic-only journals in these two subject areas. Peer review was used as a selection criterion in determining the journals that were to be included in the study. Article repositories and non-peer reviewed journals were not included. Peer review is a well-established standard in the journal publication process, with a history that goes as far back as the Philosophic Transactions (Kronick 1990). It serves as an article-filtering process and helps a journal build scientific legitimacy.

The journals included in the study were identified using Ulrich’s Periodicals Directory (http://www.ulrichsweb.com/UlrichsWeb/), through a search for journals that were peer reviewed, online, and active in 2005. The initial search yielded 458 physics and 609 psychology journals of all varieties: P-E, E-P, and electronic-only or print-only. This list was then perused for electronic-only journals, which whittled down the list considerably to 20 physics and 16 psychology journals. Previous studies have shown that most journals are published in both print and electronic formats (Tenopir, et al. 2003), but the small number of electronic-only journals was surprising, especially given the large return of the initial search. Less than 5% (4.3% and 2.6% respectively) of online peer-reviewed physics and psychology journals listed in Ulrich’s were electronic-only when this search was conducted in early 2006.

A final list of 10 physics and 11 psychology journals was selected, based on a requirement of a minimum of five articles published in 2005. The final list of journals included in the study is provided in the Appendix of this article. Every article published by these journals in the year 2005 was examined, with the exception of the articles in two high-volume physics journals, where the number of published articles was very large. The two high-volume journals were systematically sampled for 23 articles, the average number of articles published per year per journal for the eight low-volume journals (min. 5 - max. 43). The psychology journals published an average of 18 articles per year per journal (min. 9 - max. 30).

Five journal features were analyzed: non-linearity, multimedia, multiple use, interactivity, and rapid publication. The definitions of these features follows:

Non-linearity – the presence of internal links within articles. Internal links within an article effectively break the text from one continuous body into parts that may be easily read out of order. Internal links were categorized as either navigation links, which link and move within the article, or citation links, which are links from in-text citations to the bibliography. In journals where non-linearity was observed, it was quantified by counting the number of internal links of each type per article.

Multimedia – the presence of non-textual forms of information, such as video or audio clips, simulations, and other downloadable materials. This does not include graphical or pictorial images, which could be included in a printed volume. In the journals where multimedia was observed, it was quantified by counting the number of instances per article.

Multiple use – the presence of external links in the text or bibliography of an article that provide access to the full texts or full text locations of cited works, or links to referenced Web sites. In the journals where this was observed, it was quantified by counting the number of external links per article. Links to abstracts were not counted, only links that led to full texts.

Interactivity – the ability for a reader to directly communicate with the article author through the journal itself without delay, through comments, e-mail, or forums. Letters to the editor were not considered for this category, as some time delay takes place between the submission and publication of an accepted letter.

Rapid publication – the publication of an article either immediately or within a very short time of acceptance. This was measured by looking at the acceptance and publication dates of articles to determine the average length of time between acceptance and publication within each journal.

Results

Cumulative

The cumulative findings indicate that pure e-journals on the whole have been publishing electronic-only features in their articles, but only selectively. Table 1 shows the per-article mean, standard deviation, and P-value of each feature, except interactivity, for the 21 journals.

TABLE 1
# of internal links # of multimedia # of external links

Pub gap *

(days)

Mean 63.65 0.24 14.05 8.24
StDev 123.83 0.46 27.90 16.61
P 0.0092 0.0082 0.0105 -
* Article acceptance and publication data available for only 10 out of 21 journals.

With α = 0.01, the P-values are statistically significant, with the external link category right on the edge of significance. However, Table 1 also shows the large amount of variation in the use of each feature. Internal links were found at an average rate of over 60 per article (Table 1 illustrates combined totals; the usage of the two types of internal links, navigational and citation, is shown in greater detail in Tables 2 & 3), external links at a rate of over 14 per article, while multimedia features were found in only one of four articles. Rapid publication appeared to be common among those journals that publish both acceptance and publication dates, but precise publication data, discussed in more detail below, was not available for all journals.

The interactivity feature was also utilized selectively. Author e-mail addresses were almost uniformly available; 97% of articles surveyed contained at least one author e-mail address, but aside from this feature, no other type of interaction was observed. Three of 21 journals had a comment feature that allows a reader to post public comments on an article, but no comments were left on any surveyed article.

Subject Breakdown

The cumulative view illustrates how these features are being implemented independent of subject area. Splitting the two subjects apart and examining the data more closely indicates definite differences across the subject areas in feature usage. Tables 2 and 3 show the breakdown in greater detail for the surveyed journals in the two subjects, detailing the average feature count per article for each journal, including the specific counts for the two different internal link categories. The types of multimedia found per journal are also illustrated.

TABLE 2
Physics Journal Averages / Article
Journal # of navigation links # of citation links total # of internal links # of external links # of multimedia Multimedia Type Pub. gap (days)
1 0.00 0.00 0.00 0.22 0.95 27 vid, 8 aud 32.00
2 1.00 22.48 23.48 2.57 0.00 - 14.00
3 0.00 0.00 0.00 1.29 0.00 - N/A
4 0.00 0.00 0.00 0.25 0.00 - 10.00
5 65.04 58.00 123.04 53.43 0.00 - 19.00
6 170.75 346.42 517.17 112.17 0.67 8 vid. 40.00
7 171.20 76.00 247.20 0.20 0.00 - N/A
8 51.86 67.35 119.21 42.09 0.19 8 vid. 45.00
9 0.30 0.00 0.30 7.43 0.35 8 vid. 13.00
10 62.92 68.12 131.04 45.52 0.84 21 vid. N/A
Mean 52.31 63.84 116.14 26.52 0.30 - 17.30
StDev 68.30 104.39 163.55 36.99 0.38 - 14.14
P 0.0077 0.0266 0.0124 0.0117 0.0066 - -
TABLE 3
Psychology Journal Averages / Article
Journal # of navigation links # of citation links total # of internal links # of external links # of multimedia Multimedia Type Pub. gap (days)
1 3.83 0.00 3.83 0.33 0.00 - N/A
2 0.00 0.00 0.00 0.00 0.00 - N/A
3 62.41 54.59 117.00 16.07 0.00 - 0.00
4 0.00 0.00 0.00 0.13 0.00 - 0.00
5 0.00 0.00 0.00 0.00 0.00 - N/A
6 0.00 0.00 0.00 0.04 0.00 - N/A
7 0.00 0.00 0.00 0.00 0.00 - N/A
8 1.76 46.90 48.67 9.90 1.76 36 vid, 1 ppt N/A
9 5.80 0.00 5.80 1.90 0.00 - N/A
10 0.00 0.00 0.00 0.00 0.27 3 vid 0.00
11 0.00 0.00 0.00 1.43 0.00 3 vid N/A
Mean 6.71 9.23 15.94 2.71 0.18 - 0.00
StDev 18.58 20.60 36.49 5.31 0.53 - 0.00
P 0.1155 0.0687 0.0737 0.0451 0.1233 - -

With α = 0.05, the physics journals exhibit statistical significance in all categories. The psychology journals, however, exhibit only marginal significance in the presence of external links, and no significance in any other category.

Discussion

The results shown in Tables 1-3 indicate that uniquely electronic features are only selectively used, and that physics e-journals exhibit electronic features more widely than psychology e-journals.

Non-linearity: In terms of linearity, only one journal out of 21 exhibited a modular form, the physics journal #6 in Table 2 above. Unsurprisingly, this journal also ranked at the top in internal links. Navigational links were used primarily in physics journals for linking to and from equations; other uses in both fields included links between sections and links to figures. Internal citation links were also heavily utilized in the physics journals, while only two psychology journals contained this feature. It is likely that implementing internal links for one function — citation linking or equation linking — leads to implementation for both. A total of 75% of journals (six of eight) that exhibit more than 10 internal links of either type per article exhibit significant use of both.

Multimedia: The instance rate of multimedia was relatively low, and, as indicated in Tables 2 and 3, it was almost entirely in the form of videos. The precise spread of content was as follows: 120 total instances, 111 videos (93 %), 8 audio files (6 %), 1 PowerPoint file (1%). The low incidence rate of multimedia does not necessarily reflect the ability of publishers to include multimedia material. Publishers may have the capability and desire for authors to submit different kinds of material, but authors may not be submitting multimedia for publication, regardless of the capabilities of the publisher. The prevalence of multimedia (in this case predominantly video) was similar across the two fields, but this is likely due to the fields studied. It would be expected that the type and volume of multimedia would differ in fields such as acoustic studies, where a prime source of data is in multimedia form, and authors are more active in submitting additional materials.

Multiple use: Most journals, 17 out of 21, exhibited external bibliography links. However, only five of those journals averaged over 10 external links per article. Given that most articles contain many references, and that citation chasing is a common technique for information seeking, this can be taken as an indication that many obstacles still stand in the way of comprehensive full-text external citation linking.

Interactivity: The presence of author e-mail addresses in such a high volume (97% of articles) is not surprising given the ubiquity of e-mail in every day communication, but given the number of other types of online interactivity available, the complete lack of any other type of interactivity raises questions. The answer may lie in the fact that most other types of interactivity – comments, message boards, etc. – are informal. Formal interaction usually takes the form of a response paper or a letter to the editor, in which the scholar has the opportunity to give a well-researched response. Informal scholarly communication usually takes place privately, either through direct discussions at conferences or meetings, or through phone or e-mail conversations. Comments and other types of informal publicly viewable interactivity may be slow to catch on within the academic community, where reputations and community perceptions are essential.

Rapid publication: Precise dates for both acceptance and publication were available for only seven physics journals and three psychology journals, making it difficult to make a definitive statement about the presence of rapid publication across these disciplines. At least one date, acceptance or publication, was published for all journals except one, but both dates are needed to make a quantitative comparison. Journals that did not provide both dates were contacted directly about the availability of the data, and while a number responded with relevant publication information, only one journal responded with precise dates (and in that case only acceptance dates). However, some conclusions can still be drawn from the publication practices, and from the very presence or absence of available data. Nine of the 10 physics journals and four of the 11 psychology journals publish articles individually as soon as the article is ready for publication. The remaining journals publish articles grouped into issues. All 10 journals that publish both acceptance and publication dates publish articles individually.

There is a clear difference between the two disciplines in the amount of publication data available. A total of 70% of the physics journals gave both dates, compared to only 27% of the psychology journals. Rapid publication is necessary to stake claim to a finding. The electronic medium has a definite advantage over print in this arena, but immediate publication may not be as crucial to psychology scholars, where the pace of research and discovery is slower than in physics. As shown in Table 2, the average publication gap for physics journals was 17.3 days. This gap reflects the period prior to publication when final article editing and formatting occurs. Seventeen days is considerably shorter than the publication gaps that previously existed in the print-only environment, where the gap is on the order of four to six months. As Table 3 shows, the three psychology journals that do publish both acceptance and publication dates show a publication gap of zero days. This should be taken not as an indicator of the publication practices of the psychology field, but as a clue to the marketing strategies of those three journals. Rapid publication may not be as critical in psychology as it is in physics, but it still likely has an appeal. Tables 2 and 3 indicate that journal publishers provide acceptance and publication dates as a selling point when the dates show a small publication gap.

It is interesting to compare these study findings with those reported by Swan and Brown (2003). They found that rapid publication, in which “[e]lectronic articles are published as soon as they are finalized,” was valued by 64% of the responding scholars. Multimedia items, referred to as “manipulatable content (software, simulations, online experiments, etc.),” and “video or sound,” were less highly valued, 23% and 14% respectively. Interactivity was also a low-value item, with all categories rated as “not valuable” by 44% of responders or less. Multiple use, on the other hand, which they called “[l]inking from citations to cited articles,” was valued as “very important” or “important” by 88% of responders. Though the current study focused on journals themselves as opposed to the scholars who use them, the findings show similar results to those found by Swan and Brown. Multiple use, while still fraught with complications, is being implemented on a fairly wide scale, a testament to the added value it brings to the user. Rapid publication seems to be a selling point for an electronic journal, and when it is present it is advertised to the user. Multimedia and interactive features, however, were not as common, consistent with Swan and Brown’s findings that these items have low value to scholars.

Conclusion

The results of this study indicate that the additional capabilities of the electronic medium are only selectively utilized in electronic-only journals in physics and psychology, and that there is a marked difference between the two fields in the extent to which the new capabilities are being used. Electronic-only physics journals tend more than electronic-only psychology journals to use the additional capabilities in four of the five fields studied: internal linking, multimedia, external linking, and rapid publication. Neither discipline used interactivity capabilities, with the exception of e-mail, which was included in 97% of all journal articles.

The ability to publish rapidly appears to be a primary advantage of the electronic medium, particularly for physics, where research and discovery moves at a rapid pace. Journal publishers have a great incentive to provide rapid publication in order to attract authors and readers, but the other uniquely electronic features examined in this study do not hold the same attraction. Non-linearity was used mainly as a tool for equation and citation linking. Modular design appears to be rarely used, as the large number of links necessary may make it more confusing than useful. Readers are accustomed to the form and shape of the print article; the majority of electronic journal users prefer to print out and read an article on a paper copy, as opposed to reading the article on line (Tenopir and King 2002), making it necessary for a publication that implements a modular form to also include a printable form of the article. Multimedia features can be an asset in more fully expounding on an idea, but these were seen in low incidence. Multiple use, or external reference linking, is also a great help for researchers, but requires the referred articles to also be openly available and accessible. This is not a simple requirement, as there are many impediments to a workable strategy for comprehensive inter-connectivity, including academic, economic, and social obstacles.

The relative importance of these features fits in with the analysis of the scholarly article as a genre of communication. The scholarly journal article has a long history, as Bazerman (1988) details. The linear experimental article has developed over centuries into its current form, one well suited for the delivery of information and knowledge. As Orlikowski and Yates state, “a community's genre repertoire indicates its established communicative practices” (1994). Genres of communication exist in the context of a given community. The scholarly article has gradually evolved to its current linear format, one that is highly effective in transferring knowledge in a rigorous and reliable fashion. Social practices shape how a new technology fits into a given community (Kling and McKim 2000); new publication models must fulfill the necessary social roles before they will be widely adopted. Fully embracing more features of the electronic medium would require that scholars “base their views of electronic writing on an oral metaphor because the computer is an interpersonal medium” (Ferris 2002), which may not be an easy perspective shift.

The additional electronic features can add to the appeal of an electronic journal, but are not essential to fulfill the purpose of this genre of communication. In addition, any additional features, such as multimedia presentations or large data sets, can increase the cost of publication (Kling & Callahan 2003). Journals with limited budgets are less likely to include these types of additional features, and have no practical academic incentives for finding a way around economic impediments in order to do so, as these items are not highly valued. The easiest features to implement are also those most likely to be found in electronic-only journals: internal linking, inclusion of author e-mail addresses, and rapid publication. These features add to the utility of the article, but do not change the form. It is likely that the traditional linear article will continue to be the prevalent format for scholarly journals, both print and electronic, for the foreseeable future, and while electronic features will garner more and more use as technology improves, they will continue to be used to supplement, and not supplant, the traditional article.

Acknowledgements

I would like to thank Prof. Christine Borgman and Prof. Jonathan Furner of UCLA, as well as the three anonymous reviewers, for reading and commenting on earlier drafts of this paper.

References

(All URLs last visited on August 1, 2007)

Agre, P.E. 1995. Designing genres for new media: social, economic and political contexts. The Network Observer (11). http://polaris.gseis.ucla.edu/pagre/genre.html

Bazerman, C. 1988. Shaping written knowledge: the genre and activity of the experimental article in science. Madison, WI: University of Wisconsin Press. http://wac.colostate.edu/books/bazerman_shaping/

Bishop, A.P. 1999. Document structure and digital libraries: how researchers mobilize information in journal articles. Information Processing and Management. (3): 255-279. [doi: 10.1016/S0306-4573(98)00061-2]

Borgman, C.L. and J. Furner. 2002. Scholarly communication and bibliometrics. Annual Review of Information Science and Technology: 3-72.

Brady, Eileen E., Sarah K. McCord, and Betty Galbraith. 2006. Print versus electronic journal use in three sci/tech disciplines: the cultural shift in process. College & Research Libraries, (4): 354-363.

Butler, H.J. 1995. Where does scholarly electronic publishing get you? Scholarly Publishing. (4): 234-246.

Cronin, B. 2001. Acknowledgement trends in the research literature of information science. Journal of Documentation, (3): 427-433. [doi: 10.1108/EUM0000000007089]

Cronin, B. and K. Overfelt. 1995. E-journals and tenure. Journal of the American Society for Information Science, (9): 700-703. [doi: 10.1002/(SICI)1097-4571(199510)46:9<700::AID-ASI6>3.0.CO;2-8]

CrossRef.org. 2007. DOIs for research content. http://www.crossref.org/

Ferris, S.P. 2002. The effects of computers on traditional writing. The Journal of Electronic Publishing, (1).

Fink, J.L and P.E. Bourne. 2007. Reinventing scholarly communication for the electronic age. CT Watch Quarterly, (3). http://www.ctwatch.org/quarterly/articles/2007/08/reinventing-scholarly-communication-for-the-electronic-age/

Harley, D., S. Earl-Novell, J. Arter, S. Lawrence, and C.J. King. 2007. The influence of academic values on scholarly publication and communication practices. Journal of Electronic Publishing, (2).

Harnad, S. 1991. Post-Gutenberg galaxy: the fourth revolution in the means of production of knowledge. Public-Access Computer Systems Review, (1): 39-53. http://www.ecs.soton.ac.uk/~harnad/Papers/Harnad/harnad91.postgutenberg.html

Harnad, S. 1992. Interactive publication: extending the american physical society's discipline-specific model for electronic publishing. Serials Review, Special Issue on Economics Models for Electronic Publishing, pp. 58-61. http://cogsci.soton.ac.uk/harnad/Papers/Harnad/harnad92.interactivpub.html

Harnad, S. 1998. On-line journals and financial fire walls. Nature (6698): 127-128. [doi: 10.1038/25878]

Hitchcock, S., D. Bergmark, T. Brody, C. Gutteridge, L. Carr, W. Hall, C. Lagoze, and S. Harnad. 2002. Open citation linking: the way forward. D-Lib Magazine, (10). http://www.dlib.org/dlib/october02/hitchcock/10hitchcock.html

King, D.W., P.B. Boyce, C.H. Montgomery, and C. Tenopir. 2003. Library economic metrics: examples of the comparison of electronic and print journal collections and collection services. Library Trends, (3): 376-400.

Kircz, J.G. 1998. Modularity: the next form of scientific information representation? Journal of Documentation, (2): 210-235. [doi: 10.1108/EUM0000000007185]

Kircz, J.G. 2002. New practices for electronic publishing 2: new forms of the scientific paper. Learned Publishing, (1): 27-32. [doi: 10.1087/095315102753303652]

Kling, R. and E. Callahan. 2003. Electronic journals, the internet, and scholarly communication. Annual Review of Information Science and Technology. :127-177. [doi: 10.1002/aris.1440370105]

Kling, R. and G. McKim. 1999. Scholarly communication and the continuum of electronic publishing. Journal of the American Society for Information Science, (10): 890-906. [doi: 10.1002/(SICI)1097-4571(1999)50:10<890::AID-ASI6>3.0.CO;2-8]

Kling, R. and G. McKim. 2000. Not just a matter of time: field differences and the shaping of electronic media in supporting scientific communication. Journal of the American Society for Information Science, (14): 1306-1320. [doi: 10.1002/1097-4571(2000)9999:9999<::AID-ASI1047>3.0.CO;2-T]

Kronick, D.A. 1990. Peer review in the 18th-century scientific journal. Journal of the American Medical Association, (10): 1321–1322. [doi: 10.1001/jama.263.10.1321]

Lustria, L.A. and D.O. Case. 2005. The SPARC initiative: a survey of participants and features analysis of their journals. The Journal of Academic Librarianship, (3): 236-246. [doi: 10.1016/j.acalib.2005.01.004]

Lynch, C.A. 2007. The shape of the scientific article in the developing cyberinfrastructure. CT Watch Quarterly, (3). http://www.ctwatch.org/quarterly/articles/2007/08/the-shape-of-the-scientific-article-in-the-developing-cyberinfrastructure/

Miller, C.R. 1984. Genre as social action. Quarterly Journal of Speech, (2): 151-167.

Moghaddam, G.G. 2007.Archiving challenges of scholarly electronic journals: how do publishers manage them? Serials Review (2): 81-90. [doi: 10.1016/j.serrev.2007.03.003]

Odlyzko, A. 1997. The economics of electronic journals. First Monday, (8). http://www.firstmonday.org/issues/issue2_8/odlyzko/index.html

Orlikowski, W., and J. Yates. 1994. Genre repertoire: the structuring of communicative practices in organizations. Administrative Science Quarterly (4): 542-574.

SPARC. 2007. Alternatives. http://www.arl.org/sparc/partner/alternative.html

Swan, A. 2006. Overview of scholarly communication. In Open Access: Key Strategic, Technical and Economic Aspects, edited by Neil Jacobs. Chandos Publishing: Oxford. http://eprints.ecs.soton.ac.uk/12427/

Swan, A. and S. Brown. 2003. Authors and electronic publishing: what authors want from the new technology. Learned Publishing, (1): 28-33. [doi: 10.1087/095315103320995069]

Tenopir, C. and D.W. King. 2001. Lessons for the future of journals. Nature, (6857): 672-674. [doi: 10.1038/35099602]

Tenopir, C. and D.W. King. 2002. Reading behaviour and electronic journals. Learned Publishing, (4): 259-265. [doi: 10.1087/095315102760319215]

Tenopir, C., D.W. King, P. Boyce, M. Grayson, Y. Zhang, and M. Ebuen. 2003. Patterns of journal use by scientists through three evolutionary phases. D-Lib Magazine, 9(5). [doi: 10.1045/may2003-king]

Traweek, S. 1988. Beamtimes and Lifetimes. Cambridge, Mass. Harvard University Press.

Van de Sompel, H. and P. Hochstenbach. 1999a. Reference linking in a hybrid library environment, part 1: frameworks for linking. D-Lib Magazine, (4). http://www.dlib.org/dlib/april99/van_de_sompel/04van_de_sompel-pt1.htmlVan de Sompel, H. and P. Hochstenbach. 1999b. Reference linking in a hybrid library environment, part 2: SFX, a generic linking solution. D-Lib Magazine, (4). http://www.dlib.org/dlib/april99/van_de_sompel/04van_de_sompel-pt2.html

Van de Sompel, H. and O. Beit-Arie. 2001. Open linking in the scholarly information environment using the OpenURL framework. D-Lib Magazine, (3). http://www.dlib.org/dlib/march01/vandesompel/03vandesompel.html

Van de Sompel, H., C. Lagoze, J. Bekaert, X. Liu, S. Payette, and S. Warner. 2006. An interoperable fabric for scholarly value chains. D-Lib Magazine, (10). http://www.dlib.org/dlib/october06/vandesompel/10vandesompel.html

Wakimoto J.C., D.S. Walker, and K.S. Dabbour. 2006. The myths and realities of SFX in academic libraries. Journal of Academic Librarianship, (2): 127-136. [doi: 10.1016/j.acalib.2005.12.008]

Yates, J. and W.J. Orlikowski. 1992. Genres of organizational communication: a structurational approach to studying communication and media. Academy of Management Review, (2): 299-326. [doi: 10.2307/258774]

Yoshioka T., G. Herman, J. Yates, and W. Orlikowski. 2001. Genre taxonomy: a knowledge repository of communicative actions ACM Transactions on Information Systems, 4): 431-456.

Appendix – Final list of journals included in this study

Psychology E-journals

1. Australian Journal of Educational and Developmental Psychology

Published by: University of Newcastle

http://www.newcastle.edu.au/group/ajedp/index.html

2. The Behavior Analyst Today

Published by: The Behavior Analyst Today

http://www.behavior-analyst-today.com/

3. Clinical Practice and Epidemiology in Mental Health

Published by: BioMed Central Ltd.

http://www.cpementalhealth.com/

4. Current Research in Social Psychology

Published by: University of Iowa

http://www.uiowa.edu/~grpproc/crisp/crisp.html

5. E - Journal of Applied Psychology

Published by: Swinburne University of Technology

http://ojs.lib.swin.edu.au/index.php/ejap

6. Journal of Indian Association for Child and Adolescent Mental Health

Published by: Indian Association for Child and Adolescent Mental Health

http://www.jiacam.org/

7. International Journal of Psychosocial Rehabilitation

Published by: Hampstead Psychological Associates

http://www.psychosocial.com/

8. Journal of Technology in Counseling

Published by: Columbus State University

http://jtc.colstate.edu/

9. Pragmatic Case Studies in Psychotherapy

Published by: Rutgers University Press

http://pcsp.libraries.rutgers.edu/

10. PsychNology

Published by: PsychNology

http://www.psychnology.org/

11. Revisita Electronica de Investigacion Psicoeducativa

Published by: Universidad de Almeria

http://www.investigacion-psicopedagogica.org/revista/new/index.php

Physics E-journals

1. Acoustics Research Letters Online

Published by: Acoustical Society of America

http://scitation.aip.org/arlo/

2. EARSeL eProceedings

Published by: European Association of Remote Sensing Laboratories

http://las.physik.uni-oldenburg.de/eProceedings/

3. Electronic Journal of Theoretical Physics

Published by: Electronic Journal of Theoretical Physics

http://www.ejtp.com/

4. Entropy

Published by: Molecular Diversity Preservation International

http://www.mdpi.org/entropy/

5. Journal of Cosmology and Astroparticle Physics

Published by: Institute of Physics Publishing

http://www.iop.org/EJ/journal/JCAP

6. Living Reviews in Relativity

Published by: Max Planck Institute for Gravitational Physics

http://relativity.livingreviews.org/

7. Mathematical Physics Electronic Journal

Published by: Angel Jorba & Jaume Timoneda, Eds. & Pubs.

http://www.ma.utexas.edu/mpej/MPEJ.html

8. Physical Biology

Published by: Institute of Physics Publishing

http://www.iop.org/EJ/physbio

9. Optics Express

Published by: Optical Society of America

http://www.opticsexpress.org/

10. The New Journal of Physics *

Published by: Institute of Physics Publishing

http://www.iop.org/EJ/njp

* Due to the extensive scope of the NJP, and the fact that articles were grouped by subject categories, I randomly selected one subject category to be the source of articles for this study. The randomly selected subject was “Structural, mechanical and thermal properties of materials.” The articles from this category can be found at the following URL:

http://www.iop.org/EJ/toc_topic/1367-2630?source=nj&code=17