New international investigation tackles ‘fake science’ and its poisonous effects - ICIJ

Hundreds of thousands of scientists worldwide have published studies in self-described scientific journals that don’t provide traditional checks for accuracy and quality, according to a new journalistic investigation.

Dozens of reporters from media outlets in Europe, Asia and the United States have analysed 175,000 scientific articles published by five of the world’s largest pseudo-scientific platforms including India-based Omics Publishing Group and the Turkey-based World Academy of Science, Engineering and Technology, or Waset.  In addition to failing to perform peer or editorial committee reviews of articles, the companies charge to publish articles, accept papers by employees of pharmaceutical and other companies as well as by climate-change skeptics promoting questionable theories.

Some of those publishers send targeted emails to scientists who are under pressure to publish as many articles as possible in order to obtain promotions and improve their curriculum, according to the findings by Norddeutscher Rundfunk (NDR), WDR and Süddeutsche Zeitung.

In addition to the German outlets, a group of more than a dozen media organizations including the New Yorker, Le Monde, the Indian Express and the Korean outlet Newstapa took part in the investigation. The International Consortium of Investigative Journalists facilitated the collaboration.

Although the existence of these internet-based pseudo-scientific journals is not new and has been warned against by universities and research institutions, its recent rapid growth — with the number of publications put out by the top publishers tripling since 2013 and involving some 400,000 scientists – set off alarms among former Nobel Prize winners.

The credibility of science is at stake, said U.S. physician Ferid Murad, the 1998 winner of the prize in physiology or medicine. Randy Schekman, a U.S. cell biologist who was among the 2013 winners of the Nobel prize, said that he was horrified that scientists were publishing in such journals. “This kind of thing has to be stopped,” said Robert Huber of Munich, who was awarded the prize in 1988. “If there is a system behind it, and there are people who aren’t just duped by it but who take advantage of it, then it has to be shut down,” said Stefan Hell, a Nobel laureate in chemistry.

Those journals contribute to the production and dissemination of “fake science” by failing to uphold basic standards of quality control, the report said. In Germany alone, more than 5,000 scientists — including those supported by public funding — have published their articles in such predatory journals, which have been increasing for the past five years.

While those journals’ publishers claimed that a panel of scientists is in charge of verifying the accuracy of the papers, the investigation showed that articles are published within a few days of submission without any vetting process.

In one case, an article in the Journal of Integrative Oncology stated that a clinical study had shown the extract of propolis, a secretion that bees use to glue hives together, was more effective than chemotherapy in treating colorectal cancer. The study was fake and the authors were affiliated with a research center that doesn’t exist, Le Monde reported.

After the journalists questioned the journal about those findings, the article was deleted but an archived version is still available online.

Omics, which published the journal in question, claims to have published over 1 million articles and is currently being investigated by the U.S. Federal Trade Commission for alleged fraudulent claims, according to the Indian Express. A spokesman has denied any wrongdoing and defended the integrity of its publications.

Reporters from the media outlets involved in the investigation successfully published numerous non-scientific papers with the publishers whose practices they were examining and also participated in several of their conferences.

Mock Conferences - Copy, Shake, and Paste

After being forced to remove pages from my blog dealing with what I called a "fake conference" and naming a name, the lawyer who tried (unsuccessfully) to discredit me at my university wrote a "thank you" note to the university stating what a fine person I am and then tried to get me to give her some specific information. I said "No", and remembered that I wanted to revisit the topic of fake conferences. But since the name "fake" seems to be hotly contested because the conferences do tend to take place, I am now using the term "mock conference", in addition to "junk journals" and "pretend publishers" for things I want to be writing about in the near future.

What is a mock conference? Here's the discussion from one of the pages I had to remove, minus the reference to a particular conference and enhanced by points from the discussion that ensued.

I feel that a mock conference is one that has some (or all) of the following properties:

1. Has an extremely wide call for papers.
2. Is co-located with many other conferences that are all in the same manner, but with another field, or is located in the same place a similar conference happened a few days before (see my table about the suspicious Chinese conferences from 2009).
3. Is located in a place people would want to visit as a tourist (Las Vegas, Orlando, Hong Kong, etc.) or even at a tourist hotel.
4. The same person organizes multiple international conferences in one year (one national conference is enough to tire anyone).
5. The sponsors are dodgy - for example, IEEE seems to sponsor anything that pays for the use of the logo. IEEE has, however, begun to crack down on mock conferences and has decided not to publish the proceedings from quite a number of conferences in 2010 and 2011.
6. Or the "sponsors" are just the department that specific professors are associated with, but the advertising is done with the university logo. Sometimes logos are just used without the institution involved knowing about its so-called sponsorship.
7. Even though they may brag about the number of citations they have (and in my book, if you have to announce that people have cited papers from the conference, then it is not an important conference), one needs to factor out the self-citations. These are when the author of a paper at the conference is citing own work submitted to a previous version of the conference.
8. Makes sure you pay your fee before the paper is published. Although it seems that there have been to many authors not showing up at conferences after getting a paper accepted, which rather defeats the purpose of a conference. Having paid the conference fee is supposed to increase the chance of actually presenting the paper.
9. Offers a special deal if you "take" two papers.
10. Accepts papers just days before the conference as long as you pay the fee.
11. Accepts papers only on the basis of an abstract.
12. Often chooses a publisher that sounds very similar to a renowned publisher, or publishes at a print-on-demand house. Some even just publish online (but with ISBN number) to save trees.
13. Accepts papers without sending out reviews. Many of these conferences insist that they "do" peer review, but there are often no substantial comments made about the individual papers. Or the reviews only come back when explicitly requested.
14. Has many, many parallel sessions that are only sparsely attended, usually because they are on such vastly different topics.
15. The program committee of the conference is unreasonably large, e.g., more than 100 members.
16. The number of accepted papers is in the 100s.
17. Anything else?

Panos Ipeirotis had also noted: "The way that you separate the legitimate from the fraudulent event is through the community. Unfortunately, if there are academics that form a mutual admiration clique and decide to meet once a year, exchanging citations, it is very difficult to separate an event like that from other legitimate fields that are rather insular and do not communicate much with other fields."

I hope we can continue discussing the properties of mock conferences, without resorting to names.

Updates: Split 1. into 1. and 2. Maybe I need to start sorting the properties into categories?

TURKISH MOCK CONFERENCES