Text recycling by Dutch researchers

On September 24, 2017 the Dutch newspaper de Volkskrant reported on an investigation into self-plagiarism (zelfplagiaat) that was conducted by a Nijmengen research group. The sociologist of science Willem Halffman and his PhD student Serge Horbach analysed 922 publications by Dutch researchers from recent years. In economics, 14 % of the papers contained text from previous publications of the author(s), in psychology the figure was 5 %.

Without naming the scientists involved, Halffman recounts that they even found a duplicate article republished with just one small change, and two highly similar articles by the same author in the same issue of a journal. They also found that authors who publish more papers are more likely to reuse text.

The use of the term selfplagiaat in Holland appears to have originated in 2013, the report states, in connection with a scandal at the Free University (VU) in Amsterdam. The Royal Netherlands Academy of Sciences (KNAW) now includes the reuse of own text without reference to the source as a questionable research practice, except for minor bits such as definitions.

Lex Bouter, a professor for scientific methodology and research integrity at the VU notes in the report that if you are re-publishing your data, you run the risk of deforming reality in a meta-analysis by having one study counted as two.

The paper is in press at Research Policy, a corrected proof is available at Science Direct. 

Self-Plagiarism

There has been quite a discussion on self-plagiarism in Germany the past few weeks, based mostly on a discussion about a paper that the current Minister of Education published 32 years ago parallel with her thesis that contains much identical material and neither refers to the other. I've been on the radio on the topic and there have been quite a number of articles published on the question.

There is an interesting legal article written by an Austrian law professor about the question of self-plagiarism: Gamper, Anna: Das so genannte „Selbstplagiat“ im Lichte des § 103 UG 2002 sowie der „guten wissenschaftlichen Praxis“. In: zfhr 8 (2009) 1, S. 2-10. DOI: 10.1007/s00741-008-0204-5

Conclusion (my translation)

A self-plagiarism is apparently only frowned upon in those cases, in which it is in essence identical to an original work, without referring to that work. [...] In such a case of sameness it cannot be the case that these are "new scientific findings [...]". 

(Wenn ein „Selbstplagiat“ verpönt ist, dann offenbar nur in Fällen, in denen eine damit im Wesentlichen idente Originalarbeit bereits veröffentlicht wurde, ohne dass auf diese gleichzeitig hingewiesen wird. [...] In einem solchen Fall der Identität liegen jedenfalls keine „neuen wissenschaftlichen Erkenntnisse“ iSd § 103 Abs 3 Z 2 UG 2002 vor.)

Research Cloning

The German business daily newspaper Handelsblatt reports on the case of Bruno Frey, a well-known economics professor in Zürich. There had been a bit of fuss in July when colleagues discovered that he had published four papers about the sinking of the Titanic with co-authors Benno Torgler and David Savage without mentioning that pretty much the same papers had already been published elsewhere. The texts were not identical, but the research was, and each paper suggests that it is the only one presenting the results.

The "Journal of Economic Perspectives" (JEP) has formally censured him, the "Journal of Economic Behaviour and Organization" (JEBO) has blacklisted the authors and will not accept any further papers from them. Frey and Torgler have said that Savage is not at fault and have tendered excuses at 3 of the 4 journals [German language detail: The article says that they "excused themselves", I always thought you had to ask the other party to excuse you --dww]. Apparently, Frey had not gotten around to writing to the "Proceedings of the National Academy of Sciences" (PNAS) before the Handelsblatt  started its investigations.

The whole issue seems to have started with the blog Economic Logic and an entry entitled "On the ethics of research cloning". The author of the blog had a good look at the CVs of the senior authors and finds evidence both of slicing results very thin in order to get much publication mileage out of them, as well as republishing the same results multiple times. In the comments a number of other clones showed up, and a FreyPlagWiki (the currently popular German way to collect evidence on scientific misconduct and plagiarism) was set up.

Interesting things have popped up, such as Frey exempting his doctoral students from coursework now required by the University in Zürich, or his being dropped from an editorial board without explaination, according to a followup blog entry.

The Handelsblatt author Olaf Storbeck has set up a Google Table documenting the cloning in 5 papers - it would be great if this and FreyPlagWiki were unified. It is a wiki, after all.

Update: "Ich habe mich zu wenig selbst zitiert"

Researchers admit to self-plagiarism, withdraw paper

[finally clearing out a large backlog, sorry about that - dww]

Spiegel online reports that stem cell researchers Karim Nayernia from Newcastle University and Wolfgang Engel from the University of Göttingen (yes, the German university of fabricated publications fame) have withdrawn a paper they published in July 2009 in "Stem Cells and Development" about producing male sperm from stem cells.

The research has caused quite a stir, so the withdrawal has also been disturbing. It appears that a large part of the introduction is a self-plagiarism without source from a previous paper published by the research group. Additionally, one of the authors listed on the paper states that he did not read the paper before it was published.

Nayernia and Engel insist that the data is valid, and regret the unintended self-plagiarism. Further reports say that the researcher responsible for the plagiarism has been fired.

CV Bloat with Recycled Papers

I happened across a paper today and saw that there was a similar paper by the same German authors, just in a different order of authorship:

• Lemke C., Loose H.
  A Low-Budget Autonomous Mobile Robot for Outdoor Tasks. 7th International Workshop on Research and Education in Mechatronics REM 2006, Stockholm, 15.6.-16.6.2006.
• Loose H., Lemke C.
  Autonomous Mobile Robots for Outdoor Tasks. 2. Wernigeröder Automatisierungs- und Informatiktage, Wernigerode, 12.-13.10.2006, S. 161-168.

I wondered what the difference was. So I downloaded them, and discovered that the major difference is in the formatting. Oh, there is a figure dropped and a sentence added and a paragraph or so removed, but even a misspelled word such as "Acknoledgement" is in both papers. Oh, wait - the list of students contributing to the work is missing in the Wernigeröde paper, and the word "proposed" has changed to "developed". It is, after all, three months later.

At Harald Loose's web site there is a third reference to be found:

• Loose H.
  Autonomous Mobile Robots for Outdoor Tasks. MSM 2007 - 3rd International Conference on Mechatronics and Materials, Kaunas, 27.-29.9.2007.

Hmm. Prof. Loose has lost his co-author - but except for formatting, this paper is identical to the Wernigeröde one. Same text, same figures, same data. The only difference seems to be that he's used a spelling checker and fixed the spelling of "Acknowledgement".

But for all practical purposes, we have three papers that are identical material. Two have the same name, but different authors (one being dropped). They are identical, except for the formatting. Two have the same authors, but in different order with different paper titles, but they are for all practical purposes the same.

Prof. Loose has bloated his CV with three publications by recycling not just the content, but also the wording. Ms. Lemke does not seem to be at the FH Brandenburg any more, although her CV would have two papers for the work of one. Neither of the later papers refer to the previously published ones.

This kind of self-plagiarism is not acceptable according to the Deutsche Forschungsgemeinschaft (DFG) rules for good scientific practice:

"... bereits früher veröffentlichte Ergebnisse nur in klar ausgewiesener Form und nur insoweit wiederholen, wie es für das Verständnis des Zusammenhangs notwendig ist. " (previously published work is only to be reused when clearly labeled an only when necessary in order to understand the context)

Perhaps the papers were never "really" published and are only available online. But I do feel that the newer ones should refer to the older ones, and the identical papers should have identical authors. I have asked the author if he cares to comment.

Update: I wrote to Prof. Loose and asked if he could give any clarification. He wrote a very nice letter back, I've decided to write a new blog entry on a point that he raised there.

He explained that the first two papers were not to be considered scientific papers. They were just the written form for talks given at small workshops. And the order of the authors was switched so that the first author was the one giving the talk. He himself is mystified by the third paper as to why the second author is no longer on the page and will check it up when he gets back home. I wish to thank him for a speedy answer!

Update 2012-07-15: Prof. Loose's home page is linked from his school, but completely unaccessable.