Image: iStockPhoto
"If I have seen further," claimed Isaac Newton, "it is by standing on the shoulders of giants." The key to comprehending such amazing vistas by all researchers is science's incrementalist tradition of building phenomenon-explaining grand theories finding by finding as facts accumulate across decades of previous research. In the past dozen years or so scientists have increasingly turned toward "open access" (OA) publication, where full-text articles, research results or complete journal issues are freely available online, rather than accessible only to subscribers who pay for a subscription, as a way to make these metaphorical giants' shoulders more widely accessible.
Historically, journals were mainly subscriber-supported, but high fees have made subscriptions prohibitive to many universities (let alone individuals) in the U.S. and abroad. In the past decade a new model has emerged. Some journals are now free to all, but the authors who publish their research in the volumes must pay a fee—often upward of $2,000. Currently, more than 5,600 OA journals are published around the world, and 8.5 percent of all scientific journals are estimated to be open. (The only major scientific publisher that is entirely open-access is the Public Library of Science, which produces the seven PLoS journals.) Much of the OA literature consists of individual articles that are made accessible either through a university database that can be accessed by anyone with sign-on privileges to that network or through standard subscription journals. If the trend continues, more of the major research journals may become open-access in the next decade, which should make scientific knowledge more accessible and advances happen more quickly. At least that is the hope and claim of OA advocates.
Dan Lee, director of the Office of Copyright Management and Scholarly Communications at the University of Arizona in Tucson, argued on November 2 that open-access publication of research articles and papers is key to the advancement of scientific creativity and innovation. "Without the barrier of subscription," Lee said, "the information is just out there on the Web and available to you. You get more research done if there aren't barriers to accessing previous research."
Such salutary claims of OA proponents are supported, at least in part. In the past several years Bo-Christer Björk of the Hanken School of Economics in Helsinki and Elena Giglia, a librarian at the University of Turin, have published research showing that OA articles are more widely read and cited than non-OA research. An article's number of citations is a rough measure of a finding's impact. Funding agencies such as the National Institutes of Health and major universities including Harvard, Cornell and the University of California, Berkeley, have all recently instituted policies that all research be made freely available.
And a recent study, led by Stevan Harnad of the University of Québec in Montréal and published on October 18 in PLoS ONE, found that open-access articles have higher citation rates whether the authors voluntarily submit their research to open-access university databases or the submission is mandated by universities and funding agencies. A statement released by PLoS said that "PLoS had no involvement in the research presented in this paper," and that the acceptance of the paper was at the sole discretion of the academic editor who was not affiliated with PLoS.
Harnad and colleagues had hypothesized that more accessible research would be more highly cited, simply because more scientists would have access to the findings. "It's as you would expect—the research is accessible to more potential users," Harnad says. "Scientists are doing research so that it can be used and built upon. They're not writing up their results just to make money," he adds. Because scientists are not motivated by profit, Harnad believes that open-access won't hurt the primary goal of advancing research. Open access, Harnad argues, opens the door of science to researchers in resource-poor settings and to citizen scientists and interested amateurs.
See what we're tweeting about
11 Comments
Add CommentDavis is mistaken in his premise that all decisions with social impact should wait for empirical evidence. The irony of such a position is that his reluctance to support open access reduces access to rigorous evidence where it is most needed; namely, in general scientific education and investigation, both professional and amateur.
Reply | Report Abuse | Link to thisThe decision to make science open is an ethical issue, because there is no way for Davis or anyone else to asses the potential benefits of open access; because society is an open system. So what if it's not cited as much. That is one dimension of a much larger system whose potential is absolutely predicated on the dissemination of rigorous peer reviewed science. The quality of average citizen's understanding of science is just one of many more possible benefits.
Again, Davis's reluctance to support open access is illogical as it restricts the very information that he is claiming to base his position. I'm not going to pay $30 to rent an article for 48 hours to see if his argument is valid. In such a case, progress is thwarted.
Furthermore, if any research facility receives a dime of government money it needs to be illegal to restrict access. This doesn't mean that a publisher can't recover their printing costs by charging for printed copies. PDFs, however, absolutely need to be free.
This mishmash of misunderstandings is not worthy of Scientific American. Arnold leaves five false impressions: (1) that all OA journals charge author-side publication fees, when 70% do not; (2) that PLoS is the "only major" OA journal publisher, when there are many others, including BioMed Central; (3) that articles on deposit in OA repositories are only accessible to users "with sign-on privileges", when they are freely accessible to everyone on Earth with an internet connection; (4) that OA mandates at funding agencies and universities require gold OA (published in OA journals) when they only require green OA (deposited in repositories); and (5) that disagreements about whether OA increases citation tallies are disagreements about whether OA advances scientific research.
Reply | Report Abuse | Link to thisPeter Suber has said it all. The only thing I can add is some comments on Phil Davis's studies:
Reply | Report Abuse | Link to thisPhil's dissertation results are welcome and interesting, and include some good theoretical insights, but insofar as the OA Citation Advantage is concerned, his empirical findings turn out to be just a failure to replicate the OA Citation Advantage in one small sample and time-span. Phil's original 2008 sample of 247 OA and 1372 non-OA articles in 11 journals one year after publication has been expanded in his dissertation to 712 OA and 2533 non-OA articles in 36 journals two years after publication. The finding is a significant download advantage for OA articles but no significant citation advantage.
The only way to describe such an outcome in the face of the many much larger studies that have reported the positive effect is as one non-replication; it is certainly not a demonstration that the OA Advantage is an artifact of self-selection, since there is no control group, across the same sample and time-span, demonstrating the presence of the citation advantage with self-selected OA and the absence of the citation advantage with randomized OA : There is simply a failure to detect any citation advantage at all.
The replication failure is almost certainly due to the small sample size and the short time-span. It take a few years of accumulating citations and a lot of articles for the OA Advantage to grow big enough to reach statistical significance. To get a sense of this, compare the sample size of Davis's negative results with the sample-sizes and time-spans of some of the many studies that have found positive results. (Google: "Correlation, Causation, and the Weight of Evidence".)
This article, while intending to summarize the topic, confused several related issues and misattributed the work of scholars.
Reply | Report Abuse | Link to thisI am not aware of any published research by either Bjork or Giglia on the effect of access on readership or downloads. Nor am I aware that Cornell University has implemented a policy mandating open access. And while I appreciate being cited so frequently in this piece, the author misunderstood my position on science policy.
Linking to the cited articles may have reduced the misattribution and allowed readers to learn more about the complex issues.
The article says, "The only major scientific publisher that is entirely open-access is the Public Library of Science." My understanding is that BioMed Central is not entirely open access. Does it not also have some subscription journals?
Reply | Report Abuse | Link to thisAnyway, scientific results being more accessible would be beneficial for its influences. Though the subscription fee is called a "porous sieve",it is still there. Removing the porous barrier would be still helpful for the amateurs who want to keep track and contribute on the scientific process.
Reply | Report Abuse | Link to thisScience should be open to all rather than only the professionals. Anyone can contribute to science, maybe there is another "Einstein" somewhere, who knows?
I think for anyone that WAS advocating against OA the question would really be "isn't the burden of proof on you as well?" While there may not be much solid evidence that OA is beneficial to science or even to society at large I'd venture to say that the claim it is better than 'closed access' is a less radical claim and thus aught to be seen as the position requiring less supporting evidence in order to be taken as the default policy.
Reply | Report Abuse | Link to thisI agree with Li Changjian. Although the majority of experts in their fields have access to these journals through their company or university, many "rising stars" attempting to educate themselves in their area of interest may not have the same access.
Reply | Report Abuse | Link to thisThe article relies too much on academic citations. I think one of the real gains with open access, and a terrible loss of closed access, is the potential for an impact outside of academic institutions. This is particularly the case in many developing countries where even academic institutions have only limited access to closed-access journals. If you want to have an impact in places that are not solely academic institutions within the OECD, then open access is the clear choice.
Reply | Report Abuse | Link to thisI have found open access a big plus in geology.
Reply | Report Abuse | Link to thisAlso, equally valuable has been 'recent' papers from the last couple decades made available. Much less of my time is spent 'chasing down' information.
Digital 'everything' and open access is just a matter of time. Sure wish I had this when I was a student 35 years ago.
A couple of points not mentioned in the article nor the comments:
Reply | Report Abuse | Link to this1. Most scientific research, at least in the US, is in some way funded by taxpayer monies. For that final research to not be available to the people who paid for it is absurd. I get that it's not always 100% funded, but it's my understanding that very few research projects do not have ANY Federally-backed funding of some sort.
2. While professional citations might not be high, personals might be. Case in point: there's a misguided claim that Climate Change is caused by the Sun and is "proved" by temperatures rising on other planets. I went out and found 28 open-access papers that I then forward on to the people making these claims with the note "Here are 28 papers that say your claim is wrong. They have evidence and data, you have an opinion. Once you've disproved every single one of them, then we can talk. Until then, you're regurgitating Faux News."
(AGW dimwits, just to be clear: not interested in your rebuttals for the purposes of these comments.)
Too often we hear "the science says this" or "the science shows that"...part of the distrust of the general public is due to the inaccessibility of the research data. We tell people "look at the science!"...and then don't give them access to it. Irrational and paranoid people are hard enough to convince WITH actual data, it's impossible without it.
3. This one was brought up, but bears repeating: having access to the science makes it possible for someone else to build on it. It's practically mantra in the scientific community that you never know where the next big thing is coming from...it's entirely possible some kid could stumble on it as part of a science fair entry to recreate some paper they read during their research. Without access to it, though, t'ain't gonna happen!