Sunday, March 16, 2014

COPE European Seminar 2014

I gave a talk at the COPE (Committee on Publication Ethics) European Seminar 2014 in Brussels yesterday on manual versus automatic plagiarism detection.

The first talk was by Simon Godecharle, from the Center for Biomedical Ethics and Law, Faculty of Medicine, University of Leuven, Belgium. He is a PhD student who is collecting policy documents on guidance about research and publication ethics in the EU. He was wondering why there were so many scientific scandals in Europe and decided to go have a look at the policy documents available. Out of 31 nations looked at, only 19 had any kind of policy document at all, although these were the major research nations. He found 49 national level guidance documents in English, French, German, Dutch or Italian, i.e. languages he could read.

The only countries with a national policy that has been implemented in law are Norway and Denmark. Germany, England Sweden, Finland, Austria, and the Netherlands at least have a national policy, even if it is not encoded in law. The wide variation of the topics and the severity of the different transgressions surprised him. He noted that too many rules make it harder for a scientist to decide what to do when faced with an ethical dilemma. He will be publishing his collection, I look forward to seeing it.

Ana Marusic, a professor for anatomy from the University of Split School of Medicine, Croatia, Ed. in Chief of Journal of Global Health spoke about Differences in publication ethics in Central and Eastern Europe. This was important in showing the difficulties that researchers from countries that do not speak English have in getting their voice heard and their research published. 

Christopher Baethge, the Chief Scientific Editor of the German monthly journal for doctors,  Deutsches Ärzteblatt explained their novel bi-lingual concept. They now publish all research papers in a translated version online at They pay medical translators to translate the papers in an effort both to make German research more visible and also to help German-speaking doctors get more comfortable with English-language research.

One of the major problems they encountered was their conflict of interest form. They print the coi now, and have been receiving complaints that they are publishing too much research financed by the pharmaceutical industry. Apparently, that hasn't really changed but is being made transparent now. They have tried to get other publishers such as Springer to use their coi form, they refused.

In the discussion after my talk a few very good points were made that plagiarism detection system should take to heart in order for them to be useful for journal editors:
  • Kill the numbers ("originality index"). They are usually wrong and don't mean more than an potential alarm. A color-coded system might be more effective.
  • Let the user grow a list of common terminology that is not flagged as plagiarism. Especially for journals there are a number of phrases peculiar to each area that almost every paper needs to have.
  • Learn how to ignore the references, they are supposed to be exact copies of references given elsewhere! It's not enough to look for the reference section and then ignore it. Some texts have inline information, others footnotes or endnotes. Yes, I realize that Citavi alone has 6000 methods of noting references. This needs work.
We then had a workshop discussing 4 cases from the COPE database. They have over 400 cases there, all based on real cases. They are fascinating to read and to discuss with others.

Finally, Irene Hames, who is pretty much reachable all over the Internet, gave some statistics on the kinds of misconduct that are increasing, staying stable, and decreasing, according to their perception. She also presented their forthcoming taxonomy.

1 comment:

  1. Are you aware that COPE has essentially abolished its complaint procedure? COPE policy now officially states that COPE member journals will not face consequences for violating COPE guidelines. COPE membership entails no responsibility whatsoever. If you are not aware of the policy change at COPE, I can give you details.

    A complaint against two member journals filed with COPE in compliance with COPE's published complaint procedure was met with the response that COPE doesn't process complaints any more. COPE chair Ginny Barbour wrote that COPE's role is now restricted in "fostering a dialogue" with journal editors. This doesn't sound so bad except that it takes two for a dialogue.

    Three months (!) after the complaint was filed, a COPE assistant contacted the editors, who responded in some way (of which the complainant was not informed) about three months later. The assistant then wrote back to say that the complaint was closed. There was no "dialogue" and no consideration of the ethics violations. COPE always was a tool of the publishing industry and now it has become a blatant whitewash. As long as journals don't wish to clean up their act - and most don't - nothing will change.