SPARC Japan NewsLetter No.10 コンテンツ特集記事トピックス活動報告
line
menumenu menumenu
Methods of Achieving Open Access

Keiko Kurata
Faculty of Letters, Department of Humanities and Social Sciences, Keio University

● The Evolution of Open Access

Unsurprisingly, there is probably not a reader of this newsletter who has no idea what open access is all about. Open access has been one of the hottest topics in the fields of scholarly communication and academic libraries over the past ten years or so.

As for the definition of open access, however, there are actually rather notable differences from one to the next. To some people, the basic notion of open access is “improving access to the journal literature.”1 By that thinking, Elsevier in providing bibliographic information of millions of articles for free has contributed to “improving access,” compared to when there was only for-fee database searching. The focal points of discussion to date, however, have been about open access meaning the “free, unrestricted availability of academic journal articles to all.”

Open access with this narrower definition is raising fundamental questions about the current framework centered on subscription journals. Many of the articles and reports concerning open access up to now have discussed the meaning and potential of open access, or have talked about practices for promoting it. Moreover, investigation based on empirical data has tended to focus on determining the impact of open access, such as whether open access articles are actually being cited more frequently than non-OA articles.

Few studies have been tried more simply to determine empirically the overall trend (status) of open access, namely, to what extent it has progressed to date and by what means. This may have been due to recognition that open access had not yet advanced to the stage where proper results could be gained through these investigations. Nonetheless, open access has steadily progressed, such as the mandating of open access by research funding organizations including the NIH Public Access Policy, the emergence of open access mega-journals, and the growth of institutional repositories. As various proposals, strategies, and practices are tried in the future, the most basic need would be to have a grasp of the status of open access. This article is an attempt to look at the current state of open access, not in terms of progress of individual ventures such as open access journals or institutional repositories, but by introducing research investigating the percentage of journal articles available as open access.

● Means of Achieving Open Access

1. Green and Gold

As means of realizing open access, most of people mention “Green road” and “Gold road” as proposed in the Budapest Open Access Initiative (BOAI) issued in 2002.2 “Green OA” refers to so-called self-archiving, whereby articles originally published in traditional subscription journals are made available free on the website. “Gold OA” means publishing articles in open access journals, with payment of article-processing fee by authors.

When BOAI was issued, there were almost no actual institutional repositories and websites for self-archiving except for the current arXiv.org for Physics. This classification into Green and Gold was intended not to categorize the actual open access movements but to indicate directions for realizing open access. The purpose, in other words, was to point to the differences in direction, whether to enable articles available free while retaining the existing subscription journal framework, or to achieve open access in new framework around academic journals based on a business model different from traditional academic journals.

2. Actual means for realizing open access

The actual situation of open access was brought about by somewhat more complex and varied means. As will be explained more fully in the next section, in our research project we periodically surveyed the status of open access in the field of biomedicine from 2006 to 2012. We classified the means of realizing open access into the following seven types.

(a) Open access journals
(b) Publishing of open access articles in subscription journals
(c) PubMed Central (archives by subject)
(d) Institutional repositories/websites of organizations, etc.
(e) Personal websites
(f) Sites providing articles free of charge
(g) Other

This classification, it should be noted, was made for the purpose of grasping trends in the advancement of open access, and includes types of articles that some persons not consider as open access. For example, domestic academic societies using J-STAGE by Japan Science and Technology Agency (JST), distribute the printed journals for a fee whereas they provide e-journal from J-STAGE for free. Opinions are divided as to whether such journals can be called open access journals.

Note also that the articles in type (b) are published in traditional subscription journals but can in reality be obtained as open access articles. A typical example is embargo articles that are made available as open access a set time period after their initial publication. HighWire Press, founded in 1995 as a division of the Stanford University Libraries, provides e-publishing platform for relatively small academic societies and universities. In principle, articles in these journals are released as open access from six months to three years after initial publication. Currently more than two million of the 6.6 million articles that HighWire Press has been published can be accessed for free. Harnad does not include in open access such articles that can be obtained for free after an embargo period.

Also included in (b) are so-called open access choice, adopted by major commercial publishers and academic societies, whereby the author can choose to make his articles open access by paying himself. In addition there are subscription journals that make special issues available for anyone to read free of charge, or that offer certain kinds of articles for free (e.g., news, commentary, clinical case studies). These are examples of articles that raise questions as to whether or not they can truly be called open access.

In type (d), although institutional repositories are popular, this category also includes websites of organizations serving nurses, physical therapists, and other medical specialists, sites of NPOs or other organizations related to specific diseases, and sites of various organizations such as patient associations, which make journal articles available without charge.

The sites providing articles free (type f) are those like FindArticles that make available general magazine articles free. In the case of medical and health care fields, commentary or general discussion articles published not in major publishers or academic society journals but in magazines are posted on websites. Finally, Type (g) Other means mainly file-sharing sites and the like to which individuals upload their own articles or, in some cases, those written by other people.

● Share of Open Access Articles

The advancement of open access can be seen, for example, in the number of open access journals or the number of published articles.3 These numbers, however, tell us only about the status of open access journals. Only when we look at data showing the percentage of open access articles among all published articles, including those in traditional academic journals, can we be said to have an overall picture of the growth of open access. Here we will present some examples of research surveying the percentage of articles that are available as open access.

1. Open access rates in the initial period

Hajjem et al.4 investigated the percentage of open access articles out of all articles in 10 fields of Web of Science (i.e., biology, psychology, sociology, health care science, political science, economics, education, law, and management) from 1992 to 2003. The objective of their study, however, was to determine the advantage of open access from citations for those articles. For this reason, they merely indicated the open access trends on graphs, clearly specified only a small portion of the actual data.

The rates of open access articles rose gradually between 1992 and 2003, from 7 or 8 percent in 1992 to just over 10 percent in 2003. According to data by field in 2005, the smallest percentage was law at 5.1 percent, with sociology the highest at 16.0 percent (biology was 15.0 percent and health science was 6.0 percent).

2. Rates of open access more recently

Figure 1: Open access status in 2009

Figure 1: Open access status in 2009
Source: Björk, B.-C., et al.5 Figure 4. doi: 10.1371/journal.pone.0011273.g004

Björk et al. surveyed articles published in 2008 and indexed in Scopus which classifies them into 28 disciplines in 9 subject areas. They sampled a certain number of articles in each area and investigated whether those articles were open access as of 2009.5 The number of articles actually surveyed was 1,837. Adopting only the first method by which an article was found as open access, they divided the methods into Gold (open access journal sites, embargo, open access choice) and Green (subject-based archives, institutional repositories, other websites) in presenting the results (see Figure 1).

The overall open access percentage was highest for Earth Sciences at about 33 percent followed by Physics at 23 percent. For both of these subject areas, by far the majority of articles were made open access by Green methods. In medicine and biology (biochemistry, genetics & molecular biology), the percentage of open access articles was not particularly high at around 20 percent, but Gold OA was by far the favored method in these areas.

3. Open access to articles by UK academics

Table 1: Percentage of UK academic articles available as open access
  Gold Green
Mathematics 7% 60%
Professional fields 1% 29%
Engineering & Technology 1% 30%
Earth & Space 5% 47%
Arts & Humanities 1% 13%
Physics 4% 48%
Social Sciences 1% 34%
Chemistry 1% 15%
Biomedical Research 14% 43%
Health 8% 25%
Psychology 1% 34%
Biology 4% 38%
Clinical medicine 4% 34%

A Nature news article6 showed a graph, attributed to Thomson Reuters and Y. Gargouri, indicating the percentage of open access articles among some 85,000 articles published by UK academics in 2010. Since there is no reference and nothing mentioned about the specific research methods in this article, it is unclear how these results were obtained; but the overall percentages given are 5 percent for Gold and 35 percent for Green. Table 1 shows the percentages from the figure in table format. They are high for mathematics articles at 67 percent and for biological research at 57 percent. On the other hand, the figures are low for chemistry (16 percent) and arts & humanities (14 percent).

Figure 2: Advance of open access in biomedicine field Figure 2: Advance of open access in biomedicine field

4. Evolution (development) of open access in the biomedical field

Our research team investigated the status of open access successively in 2006,7 2008,8 2010,9 and 2012, limiting the study to the field of biomedicine. Using Google to search for articles sampled from articles published the previous year on PubMed (4,592, 1,908, and 1,942, respectively), we investigated whether they were available as open access and by what means.

From Figure 2 showed the percentages of open access articles at these three points in time, open access has been making steady progress. We can conclude that in the past several years it has become quite rare for a paper not to be made available in digital form from the noticeable reduction rates of “Not available online” articles.

Table 2: Open access realization methods in 2010 (multiple selection)
  OA
(%)
(a) Open access journals 26
(b) OA articles published in subscription journals 17
(c) PubMed Central 18
(d) Institutional repositories/websites of organizations, etc. 5
(e) Personal websites 1
(f) Sites providing articles free of charge 5
(g) Other 1

Table 2 shows the percentage of each method for realizing open access, out of all target articles published in 2010. When an article was made open access by multiple methods, it was counted multiple times. For example, an article which published in certain journal of BioMed Central was classified to be open access journals. The same article can be archived in PubMed Central and in some cases posted on the authors’ personal websites. For this reason the percentages of means realizing open access could not be shown divided simply into Gold and Green.

● In Conclusion

Three recent studies investigating the status of open access each indicated rather different results regarding the percentage of open access articles. The conceivable causes of this divergence are many, including differences in the methods of sampling target articles in the definitions of open access, and in the criteria used to decide whether an articles is open access. The extent to which open access has been achieved and by what means should be considered from diverse standpoints.

 


References

1. Willinsky, J. The Access Principle: The Case for Open Access to Research and Scholarship. MIT Press, 2005, pp. 307.
2. Budapest Open Access Initiative. http://www.soros.org/openaccess/
3. Laakso, M., et al. The Development of Open Access Journal Publishing from 1993 to 2009. PLoS ONE. 2011, vol. 6, no.6, e20961.
http://www.plosone.org/article/info:doi%2F10.1371%2Fjournal.pone.0020961
4. Hajjem, C., et al. Ten-year cross-disciplinary comparison of the growth of open access and how it increases research citation impact. IEEE Data Engineering Bulletin. 2005, vol. 28, no. 4, p. 39-47.
5. Björk, B.-C., et al. Open Access to the Scientific Journal Literature: Situation 2009. PLoS ONE. 2010, 5(6), e11273.
doi: 10.1371/journal.pone.0011273. http://www.plosone.org/article/info:doi/10.1371/journal.pone.0011273
6. Van Noorden, R. Britain aims for broad open access: But critics claim plan seeks to protect publishers’ interests. Nature. 2012, vol. 486, pp. 302–303. doi: 10.1038/ 486302a
7. Matsubayashi, M., et al. Status of open access in the biomedical field in 2005. Journal of Medical Library Association. 2009, vol. 97, no. 1, pp. 4–11. doi: 10.3163/1536-5050.97.1.002
8. Keiko, Kurata; Tomoko, Morioka; Keiko Inoguchi..Seibutsu igaku bunya ni okeru opun akusesu no shinten jokyo: 2005 nen to 2007 nen no deta no hikaku kara [Evolution of open access in the biomedical field: comparison of data from 2005 to 2007]. Proceedings of annual meeting of Mita Society for Library and Information Science, 2008, pp. 33–36. [in Japanese]
9. Kurata, K., et al. Proceedings of the American Society for Information Science and Technology. 2010 November/December, vol. 47, no. 1, pp. 1–2. doi: 10.1002/meet.14504701383