TL;DR: this post explains the basics of academic publishing; highlights several severe problems; discusses the recent activities of the American Psychology Association (APA) targeting psychological researchers; suggests some ways forward; and ends with an unexpected plot twist: an APA journal invited me to join their editorial board while I was writing this APA-critical blog post. Also, Sci-Hub.
Friends who do not work in academia regularly expect me to throw parties when I have a good month, with multiple published papers. Why? The money I make with several publications, especially when they are published in prestigious journals, must be considerable. Right?
This is not the case, because we work in the upside down of publishing: academia.
How academic publishing works in a nutshell
Here is a short summary of how academic publishing commonly works.
- As a tax-payer funded researcher, Bob works for a year or two on a research project, together with his colleagues, and writes it up as a scientific paper.
- Bob submits the paper to a highly prestigious scientific journal. This journal is owned by a publisher, such as Elsevier or Springer.
- The handling editor at the journal is Susan, a psychology professor. Meet Susan — she’s great! Susan works as editor for the journal without payment. She browses the paper, decides it deserves a chance, and sends it out to 2 peer-reviewers, Andy and Claire.
- Researches Andy and Claire (a post-doc and a professor) review the paper, without payment. So far, the journal has done nothing, except for running an online submission system where authors can submit manuscripts.
- The majority of papers are rejected the first time around (meaning the free work Susan, Andy and Claire have put in did not lead to a publication, and happened invisible to the public), but Bob is lucky: his paper is accepted after several revisions. Revisions means there are several rounds of work for our four dramatis personae Bob, Susan, Andy, and Claire. Bob 1 now pays US$1,500 to the journal, because color figures cost extra, and are often quite relevant to convey important information (imagine this figure in black and white). In this process, Bob also signs away many of his rights: It’s required to publish academic papers in most journals.
- So far, Bob has earned -US$1,500 with his paper. If you meet Bob, please invite him to a beer when he tells you his paper got accepted. Bob needs you right now. The journal has done nearly nothing so far. The publisher now moves the paper into a PDF file with the journal layout, puts the paper online, and includes it in a printed version of the journal that is released e.g. once a month.
- After the paper is published, Bob wants to read his own paper, but he cannot: No University can afford subscriptions to all journals, and the journal Bob submitted his paper to, a highly respectable journal in his field, is too expensive for his University library to purchase. Note that he works at the top-rated University in the world, Harvard University, but even Harvard can no longer afford the absurd fees publishers charge (about $3.5m per year in 2012). To read his own paper, Bob now is forced to obtain it illegally, by asking his colleagues to share it with him, or by using the service sci-hub. Bob is also not allowed to put the final version of his own paper on his website.
This is how things have worked for many of my papers. The procedure varies a bit, but in clinical psychology, psychiatry, and methodology, the above is standard practice and representative of publishing in these fields.
The system is broken
Obviously, no sane person (especially no scientist) would ever design such a system2. I see four main problems.
There is a group of mega-corporations that make a ton of profit from researchers working for them for free, as writers, reviewers, and editors3. Put differently, tax-payers pay the salary of most researchers, but researchers spend a considerable time of their careers working for free for academic publishers who make a ton of profit, meaning they have less time for their research. This is somewhat of a simplification of course, but we take a lot of taxpayer money that was set aside for research, and give that money to scientific publishers. That’s odd.
While most research is funded by tax-payers, most research papers are not available to the tax-payers; research is usually hidden behind paywalls of scientific publishers. Universities ‘rent’ subscriptions to journals, and these subscription fees amount to about US$10b a year globally4.
Subscription charges are overprized. Björn Brembs has summarized this recently: About US$10b are spent annually on subscriptions world-wide, for about 2 million papers, which means publishers charge about US$5,000 per article. About 10 publishers5 have estimated that the price for cost-neutral publishing is a few hundred dollars per paper, and independent analyses agree with this figure. This means that governments and societies could save between 72% and 90% of the US$10b per year if we published cost-neutrally.
Finally, many scientific publishers are highly controversial. I will not go into detail here. Instead, I’ll post a screenshot of Elsevier’s Wikipedia page that I annotated with subtle visual pointers to some of the problems.
American Psychological Association
Given this context, I want to specifically discuss actions taken by the American Psychological Association (APA) in 2017. The APA is the largest scientific and professional organization of psychologists in the US, with ~117,500 members and an annual budget of around US$115m6.
The APA is known for three things: (1) weird websites that often have a distinct 90s vibe7; (2) numerous prestigious scientific journals such as Emotion, Journal of Abnormal Psychology, Journal of Personality and Social Psychology, Psychological Bulletin, Psychological Methods, or Psych Review; and (3) alleged involvement in torture. The story is long and complicated, but the short version is that APA had refused to advise its members not to participate in torture. This contrasts with the American Psychiatric Association’s ban of all direct participation in interrogations by psychiatrists in May 2006, and the American Medical Association’s ban in June 20068. The APA itself ordered an independent review of potential malpractice in 2014, which concluded a year later that the APA had secretly collaborated with the Bush administration to bolster a legal and ethical justification for the torture of prisoners, and that several high-ranking members had colluded with government officials to loosen APA ethical guidelines9.
Early 2017, the APA made the news once again: For sending DMCA Notices of Copyright Infringement to numerous researchers.
Now, we all sign agreements with the APA when they decide to publish our papers that we cannot put our own papers in final APA layout online, so what the APA did was not legally wrong10. But I do wonder what the idea behind this is: people make free advertisement for your journals, help you increase your visibility, improve your impact factors — and we can share the content of the papers anyway because APA journals usually allow to put papers online in a layout that differs from APA format.
For Psych Review, a prestigious APA journal, you can upload both preprint and postprint, as long as the PDF is not in the final layout:
Now, I understand that publishers sometimes need to threaten and pretend they would actually follow-up on their policies. But the reason I’m blogging this is that I discovered this a few days ago on the personal website of a friend and colleague of mine:
You can find these notices on multiple sites. This means that the APA actually spent resources and time (i.e. parts of your membership fees) to use the Digital Millennium Copyright Act (DMCA) to have WordPress as blog hoster access the blogs of scientists without their permission (!), and forcefully remove content from these blogs. The content, in this case, was the researchers’ very own papers.
Again, that is something you can do legally, but why in the world would you ever spend resources on enforcing this? What do you gain? You used a copyright law to forcefully remove papers from the personal websites of the very people who wrote them, just because they use a font type and header slightly different from the preprint. I’m not sure this aligns with APA’s main mission:
Now, it’s obvious what there is to gain: Money11. But please remember that the very people you’re going after are the those who work for you for free, as reviewers and editors. So maybe cut us some slack here.
Considerable criticism erupted after APA’s warnings were sent out in June 2017, and the APA released a press statement on June 15, explaining that they would “refocus” their program to target commercial piracy sites instead of individual authors. Note that the forced WordPress edits I mention above happened after this press release, but I guess there can be delays in communication12
The public, and also researchers, including those in first-world countries working at top tier Universities, often have no legal access to crucial scientific publications. And if I have problems, you can imagine how severe this restriction is for scholars from poorer countries.
This situation was the main reason for the development of Sci-Hub, a repository of over 65 million academic papers available for direct download. Sci-Hub bypasses publisher paywalls by enabling direct access via educational institution proxies. It was founded by a student who could not afford paying the fees for papers, Alexandra Elbakyan. Nature listed her in their “Top 10 people who mattered most” in 2016. Sci-Hub often switches domain, and while I’m writing this, Sci-Hub.tw is available; once this does not work anymore, check the Sci-Hub Wikipedia page that usually contains a valid domain that is currently up and running. I will leave it to you whether the site is worth a small donation, but I do think that sci-hub can not be the final solution to this problem, and can at best be a temporary fix.
So what can we as researchers do to improve the situation? Here is a brief list.
- Be dissatisfied with the current state of things. It’s not ok. Don’t accept it as a given.
- Don’t review or work in other ways for journals or organizations you are critical of, unless you do so to try to change things. The Alliance of Science Organisations in Germany published a press release in 2016, with juicy quotes such as: “[Elsevier do] not comply with the principles of open access and fair pricing. […] The publisher rejects more transparent business models that are based on the publication service and would make publications more openly accessible. […] Elsevier is trying to exploit its dominant market position and has threatened all scientific institutions whose contracts expire at the end of 2016 that it will disable their access to its services”. And several prominent scientists resigned from positions in Elsevier journals only a few months ago in protest.
- Work for journals and organizations that are worth supporting. I personally accepted a position as Associate Editor for Collabra Clinical Psychology recently13, because they have (a) a transparent open access model, (b) focus on open data, open code, and open review, (c) reimburse editors and reviewers for their services, and (d) have lower open access fees than many other open access publishers (details 1, details 2).
- Make sure to share your research papers — before and after publication — on your website and other repositories such as the arXiv (mostly for Physics, Mathematics, Computer Science, Biology etc.) or PsyArXiv (psychology and social sciences). If you want to know whether a journal allows you to deposit preprints and postprints, google “Romeo Sherpa” and the journal name.
- Preprints can look somewhat ugly, and it can be a bit of work to get them to look pretty using programs such as LateX. It would be great if someone put together an R-package that transforms a standard Word preprint into a fancy preprint format. Maybe folks at PsyArxiv are working on this already?
- Talk to your librarian and tell them that subscription cancellations may be reasonable if the saved funds go into supporting a more modern infrastructure (Björn Brembs describes this point in more detail in this blog post).
- Finally, there might be economic disadvantages we need to consider when discussing gold (article processing charge) open access publishing; it may backfire.
While writing this blog post, I was asked to join the Editorial Board of an APA journal. I thought about it, and sent a critical email to the incoming editor in chief (who’s work I like a lot; see his vision for the journal here). He sent back a strong email, with a vision how to move forward, and the notion that the first special issue of the journal would be about increasing replicability, transparency, and openness in clinical psychological research. The call went out last week:
I thought about it, talked to mentors and colleagues, and decided to give it a try for a year to see if I can help improve things from the inside. I’ll report back in 2019.
- His University = tax-payers
- The Guardian featured a long article in 2017 highlighting some of the core issues
- Elsevier had a profit margin of 37% in 2016
- Yes, that’s a b, which stands for bazillion.
- SciELO, Ubiquity, Hindawi, PeerJ, Scholastica, RIO Journal, Science Open, F1000Research, and arpha.
- I wouldn’t be surprised if there are still hardcoded html frames in there…
- https://www.aclu.org/blog/national-security/american-psychological-association-sees-no-evil; http://ethicalpsychology.org/materials/Coalition-Response-to-Div42-Board.pdf
- https://en.wikipedia.org/wiki/American_Psychological_Association#Warfare_and_the_use_of_torture; https://www.nytimes.com/2015/07/11/us/psychologists-shielded-us-torture-program-report-finds.html?emc=edit_na_20150710&nlid=68650512&ref=cta&_r=0
- When I was a young graduate student about 8 years ago, at least in my academic environment is was expected to submit papers to prestigious APA journals, so I personally never signed away my rights because I thought that was a good idea, but because I didn’t see alternatives.
- An anonymous source sent me an email saying that in 2013, US$86m of US$126m annual APA budget came from academic publishing, but there was no reference. Does somebody have insight into this, or can confirm/disconfirm?
- Unrelated to this, but consistent with APA policies over the last years, read this Kafkaesque horror story of how the APA understands the term ‘open data’.
- Thanks to Jennifer Tackett!