By Peter Binfield, co-founder of PeerJ and previously the Publisher of PLOS ONE.
This post follows a talk at the University of British Columbia, which can be recorded and can be viewed here. The slides for the talk, synched with the audio of the talk, is available on Slideshare. The original dataset for this article, alongside the graphs and the slides, can be found at Figshare.
‘MegaJournals’ are a reasonably recent phenomena in the history of scholarly publishing, however their popularity (as evidenced by the number of articles they publish) as well as their continuing growth (in title launches, and total output) are creating a new class of journal which has the potential to dramatically change the publishing landscape. This post, which is based on themes that I developed at the UBC Open meeting earlier this week, explores this phenomena and its potential for change.
A ‘megajournal’ is widely understood to be an online-only open access journal, that covers a very broad subject area and selects content based only on scientific and methodological soundness (or some variation on that statement), with a business model which allows each article to cover its own costs. With these attributes, megajournals are not limited in their potential output and as such are able to grow commensurate with any growth in submissions.
It is worth spending a little bit of time on the editorial criteria that these journals are applying. To be explicitly clear, these journals perform extremely rigorous peer-review (and several go so far as to publish the peer review process that each article went through). They do not, however, use notions of ‘impact’ to inform their decision to publish. They critically, and formally, peer-review articles to determine whether or not they deserve to join the scientific literature; they then allow the readership to make their own decisions about the interest level (to them) of any given article once it is published.
To put it another way, they make their publication decisions only on the basis of whether or not a submission deserves to join the scholarly literature. They believe that this approach puts science into the world more rapidly, more efficiently, and more transparently than the ‘traditional’ process of rejecting otherwise publishable work in an attempt to filter articles into specific journal titles. The result is a net benefit to academia.
PLOS ONE is the most visible success story in this category of journal. PLOS ONE is currently expected to publish more than 30,000 articles in 2013, approaching 3% of all STM articles published that year (PubMed indexes approximately 1 million new articles each year). Recognizing the success of this model, many other Publishing Companies and Academic Societies (such as Nature, Springer, SAGE, BioONE, PeerJ, BMJ, F1000, the American Institute of Physics, the American Society of Microbiology, the Genetics Society of America and so on) have launched similar journals and each of them are seeing their megajournals grow in volume, month on month (see the end of this post for a list of all known ‘megajournals’). Because of the size and the growth of titles like PLOS ONE (a growth which is shown in the graph below), much of the attention around the megajournal story has focused on these large broad scope titles.
Normally ‘mega’ is taken to mean ‘million’ and so ‘megajournal’ is clearly somewhat of a misnomer, even for PLOS ONE! Nonetheless, the term has stuck because of the sheer size of the most successful megajournals (titles which aim to publish across an extremely wide subject area).
However, if we put aside the issue of size, then there is another attribute of this type of journal which is important to consider – that of the editorial criteria that is applied. Megajournals such as PLOS ONE, Scientific Reports, BMJ Open, PeerJ and so on peer-review for scientific validity, but they do not pre-judge articles based on subjective notions of ‘impact’, ‘reach’, or ‘degree of advance.’ This editorial criteroa is also being applied by journals in much smaller, or more ‘niche’, subject areas. Clearly, a niche journal (for example, “Frontiers in Neurorobotics”) that uses this editorial criteria would not expect to become extremely large (i.e. ‘mega’) due to the fact that it is in a somewhat small field.
Therefore, if we define this ‘class’ of journal as being based on the editorial criteria outlined above, then it is clear that both megajournals and many smaller ‘niche’ journals share the same class. The essential difference between them is simply whether or not they pre-judge the readership of their articles. A truly large megajournal does not need to do that (e.g. anything in the whole of biology, or the whole of medicine, or the whole of science is in scope), but one of the ‘smaller’ journals could reject an otherwise publishable article because it doesn’t fit their more limited subject-area scope (for example, the Frontiers in Neurorobotics journal would not publish a neuroeconomics article).
All of this is a long-winded way of saying that perhaps the term ‘megajournal’ (which refers to the potential size of the journal and only applies to the broad scope titles) is not the best one to help us understand the ‘megajournal’ phenomena. Another way to think about it is to look at the editorial criteria which is applied, regardless of the size (or scope) of the journal. If ‘megajournal’ is a misnomer in many respects, then what would be a better (yet still succinct and understandable) term? Some have suggested ‘non-selective’, or ‘impact neutral’, or ‘rigorous but inclusive review’ but none of these really capture the phenomena. Perhaps that is the topic for an OA Week Competition, or the comments area of this blog post…
When the definition of this journal class is expanded in this way, then it is clear that there are very large programs of journals which apply an editorial criteria similar to that of PLOS ONE – for example, the whole of the “Frontiers In…” series of journals, all of the BMC Series journals (which make up approximately half of the output of BioMed Central), as well as and titles making up approximately 1/3 of the output of Hindawi. To be fair, this point has been made in the past (e.g. by Matt Cockerill and Paul Peters at the COASP meeting in 2011), but it is my belief that most people’s attention has been focused on PLOS ONE, and similar ‘broad scope’ launches such as Scientific Reports, BMJ Open or PeerJ, with the result that the simultaneous rise of many other ‘impact neutral’ journals has perhaps been overlooked.
Regardless of name, or type, another defining characteristic of these journals is the fact that almost without fail, all journals using this editorial model grow in output, month on month. The overall effect is that the output of the entire class of journals is growing extremely rapidly, and in fact much more significantly than might have been assumed if you simply look at the PLOS ONE output. An illustrative graph is shown below, using data which was compiled with the input of most of the publishers of each of the titles listed in the appendix to this post.
So this class of journal is clearly very successful and rapidly growing. In 2012 alone, as can be seen from the graph, approximately 47,000 articles were published using this editorial model, and by extrapolation 2013 could see as many as 75,000 articles published in a model which intentionally makes no ‘pre-publication’ judgments of their significance / impact / degree of advance etc. (note: 75,000 articles is approximately 8% of all STM journal output). And this growth is being driven by a genuine author desire – each of these growing publications have clearly been able to fill a previously unmet need for their authors to the extent that they are flocking to the model, and reporting extremely positive experiences.
But Have They ‘Changed Everything’?
The fact that a journal does not use ‘significance’, ‘impact’, or ‘degree of advance’ to determine whether an article should be published does not mean that those aspects are not important (it is simply that they aren’t important to the decision to publish). If subjective filtering (on whatever criteria) has not happened ‘pre-publication’ for as much as 8% of the academic corpus, then clearly the community needs to apply new tools ‘post publication’ to try to provide these types of signals based on the reception of the article in the real world. To take this a step further, it is also becoming apparent that these journals are changing the way that people think about articles themselves – increasingly, people are coming to understand that the article itself is more important than the journal in which it happens to be published.
This is one of the key reasons, I believe, that we are currently seeing such an explosion in interest in ‘altmetrics’ and why this field is becoming more mainstream. For example, at a recent PLOS-sponsored ALM meeting, representatives from publishers such as Elsevier and Springer, Universities such as Harvard and funders such as Wellcome Trust and Sloan were present. In recent years, we have also seen the creation of article-level metrics programs at many publishers (for example PLOS, and Frontiers), as well as the formation of several start-up companies in the ‘alt-metric’ space such as Impact Story, Altmetric, and Plum Analytics.
Another way in which these journals are causing changes in the publishing world is their ability to publish negative results, or replication studies. Historically, it has been very hard to get studies of this nature published; however, if they had been published, then the community would not need to waste time repeating the mistakes that others had made before (but not made public). These journals are ideal venues for this kind of material. As more and more articles of this type are published, the net benefit to the academic community will be great and much time and energy will be saved.
But perhaps most importantly, as implied at the start of this post, the existence of these journals contributes to a considerable net increase in the speed and efficiency of the overall publishing ecosystem. Traditionally an article might be sent to a ‘first choice’ journal and even if it were publishable it could be rejected based on reasons such as ‘lack of interest’ or ‘insufficient advance in the field’ or ‘lack of novelty‘. That article will have spent weeks or months being peer reviewed, and once rejected it will simply be sent by the authors to their ‘second choice’ journal, where it will again spend weeks or months in the process, and be evaluated again by new peer reviewers (for examples, see this collection of ‘serial rejections’).
Eventually the article will be published (unless the authors give up in despair, which many do), but in the intervening time it will have been delayed by months or years, and it will have wasted the time of multiple reviewers and editors. And all of this time and energy will have been spent in the name of ‘filtering’ the article into a specific journal ‘bucket’. It has been estimated by Rubriq that as many as 15 million hours a year are wasted on ‘redundant review’.
By contrast, if an article is peer-reviewed once in a megajournal model, and then (dependent on suitable revisions) it is published, then that article will have entered the public sphere much more rapidly and without wasting the time of additional ‘redundant’ reviewers. If all articles were published in this model then the net benefit to academia (due to increased speed to publication, and reduced duplication of effort by reviewers to name just two effects) would be dramatic.
As Open Access itself grows in importance, we can expect to see more journals of this type launched; however, it is apparent that the growth of this model cannot continue indefinitely. If PLOS ONE already publishes 3% (and growing) of the STM literature, then it does not take many journals (or programs of journals) publishing at that scale before a significant proportion of the scholarly literature is being published in this way with all the advantages that this entails. Put simply, as more and more content is published using this kind of editorial model, then the net result will be that:
- New business models, new innovations and new thinking are able to flourish in a new publication ecosystem;
- ‘Mistakes’ or ‘non-results’(so called ‘negative results’) are actually reported, saving future researchers time, energy, and resources;
- Previously ‘uninteresting’ results can now be reported, providing the potential to incrementally build on these ‘micro findings’;
- Reporting standards can be more easily standardized and the levels more easily raised;
- Less time is wasted by multiple reviewers on the same content;
- The process of publication is made more transparent and fair for the author;
- Better methods of filtering, evaluating and sorting publications will evolve; and finally,
- Science will be published more rapidly, saving author time and improving the overall speed of discovery.
And when that happens, then I think it is fair to say that the rise of this editorial model will have changed everything…
MegaJournals Launched To Date:
AIP Advances – 973 articles – launched 2011
Biology Open (the Company of Biologists) – 252 articles – launched 2012
BMJ Open – 1,540 articles – launched 2011
CMAJ Open (Canadian Medical Association) – 15 articles – launched 2013
Cureus – 57 articles – launched 2012
Ecosphere (the Ecological Society of America) – 399 articles – launched 2010
EPJ-Plus (part of the European Physics Journal) – unknown articles – launched 2011
F1000 Research – 225 indexed articles – launched 2012
FEBS Open Bio (Federation of European Biochemical Societies) – 129 articles – launched 2011
G3 (the Genetics Society of America) – 383 articles – launched 2011
mBio (the American Society of Microbiology) – 601 articles – launched 2010
Optics Express (the Optical Society of America) – unknown articles – launched 1997 in some form
PeerJ – 171 articles – launched 2013
PLOS ONE – 75,382 articles – launched 2006
QScience Connect – 53 articles – launched 2011
SAGE Open – 371 articles – launched 2011
SAGE Open Medicine – 12 articles – launched 2013
Scientific Reports (Nature) – 2,731 articles – launched 2011
Springer Plus – 548 articles – launched 2012
The Scientific World Journal (Hindawi) – 1,860 articles – (re)launched 2012
Megajournals ‘Coming Soon’:
BMJ Open Respiratory Research – 2013
BMJ Open Diabetes Research & Care – 2013
Open Heart (BMJ) – 2013
Elementa (BioONE) – 2013
IEEE Access – 2013
OpenLibHums – 2014
The Cogent Series (T&F) – 2014
The Winnower – 2014
Existing large programs of ‘niche’ journals which apply the same editorial criteria (with output through end-2012):
The BMC Series of journals (approx. half of the annual output of BMC) – 56,000 articles
The“Frontiers in…” Series (part of Nature Publishing Group) – 9,921
Hindawi: the ISRN series; the “Case Reports in Medicine” series; the “Conference Papers in Science” series; the “Dataset Papers in Science” journal, and the “Scientifica” journal – 6,713 articles
Acknowledgements – I would like to acknowledge the input and help of most of the publishers of the journals listed above, as well as Michael Habib of Scopus, for providing publication data and other ‘food for thought’.