01 June 2009

Cheaters in white coats

Falsification of scientific results has become the normAsya Parfenova, Infox.ru
Cases of falsification of scientific results are not the exception, but the rule.

And researchers from medicine and pharmacology lie the most, says Scottish sociologist Daniel Fanelli.

The image of scientists as unselfish, unbiased seekers of truth has recently been increasingly violated by scientists themselves. The review system does not always work against talented forgers. Even in the most prestigious journals, such as Science, Nature or Physical Review Letters. A wave of controversy and research on this topic was caused by the last two particularly high-profile scandals related to the names of Wu Suk Hwan and Jan Hendrik Shen.

The first (Hwang Woo-Suk) is a South Korean researcher, professor of genetic engineering and biotechnology at Seoul National University. He became famous for his work in the field of stem cells. In particular, two of his articles published in 2004 and 2005 in the journal Science reported on the creation of human stem cells by cloning. In 2006, he was convicted of fraud and deprived of his professorship, which does not prevent him from continuing his scientific activities.

The second (Jan Hendrik Schön) is a German physicist working in the USA at the famous Bell Laboratory. He became famous for his work in the field of microelectronics, in particular organic semiconductors and superconducting fullerenes. In 2002, he was convicted of multiple falsifications. His eight papers in Science, seven in Nature and six in the journals of the Physical Review series were found to be fake.

Subject for disputesBoth scientists and many media prefer to consider these cases unpleasant exceptions.

They base their belief on the fact that the profession of a scientist presupposes impartiality and a critical approach, which is incompatible with data forgery. However, many direct surveys of scientists suggest that known cases of fraud are only the tip of the iceberg. And most of the scams in science remain unsolved.

Assumptions and estimates of the number of such cases are usually based on indirect data. For example, on official abstracts of scientific articles or on reproductions of the results of articles from a random sample. Such methods give rather contradictory results. Therefore, many researchers directly interview scientists from different fields. But since researchers use different survey methods and formulate their questions differently, their results are quite difficult to compare with each other, so the overall situation remains unclear.
Scientific generalization

Daniele Fanelli from the University of Edinburgh decided to fill the gap in this area. He collected, systematized and processed with the help of meta-analysis more than 20 papers based on direct surveys of scientists on the topic of fraud in science, and presented the results of his work in the public journal PLoS ONE.

To begin with, Fanelli divided examples of fraudulent behavior into those that affect the final scientific knowledge, and those that are considered fraud only from the point of view of ethics (for example, plagiarism). From the studies published in various scientific journals, the sociologist selected 21 papers where it was possible to distinguish examples of scientific deception in the content according to the selected criteria and where the results met the requirements of statistical reliability.

Disappointing conclusionsGeneralization of the results of the selected works gave disappointing results.

About 2% of scientists report that at least once they invented or falsified their data in order to demonstrate not the real, but the desired result. Another 34% of scientists reported that they had committed "minor" frauds once or more often: they ignored values or data that fell out of the general concept of the study, relying on the fact that it was the result of a carelessly performed experiment. Or they preferred not to show data that questioned or refuted the results of their previous work.

The results of a survey of scientists about known cases of fraud by colleagues also turned out to be not encouraging. 14% of respondents report that they are aware of at least one case of a major intentional falsification committed by a familiar researcher. And 72% of respondents speak about minor frauds of colleagues.

Cheaters in white coatsThe selected works showed fairly similar results.

Fanelli notes that the results of the surveys do not depend on the country where the scientist worked. The situation in different fields of scientific activity is also approximately the same, with the exception of medicine and pharmacology. Scientists, on the results of whose work our health, and sometimes our lives, depend, most often confessed both to their misdeeds and to what they know about the falsifications of colleagues.

One can, of course, assume that doctors and pharmacists are the most open and honest in their answers. But it is also possible that fraud and intentional distortion or concealment of results are most frequent in this area. Then the fears of many reputable scientists that the material support of manufacturers contributes to the distortion of scientific results for the promotion of new medicinal products and medical services of commercial companies have valid grounds.

By the way, the latest scandal in the scientific press is connected with medicine and pharmacology. Elsevier, one of the largest publishers of scientific literature, recently published the results of an internal investigation, admitting the fact of placing hidden advertising of medicines under the guise of scientific publications. The Australian-Asian division of the publishing house was engaged in disguised advertising without the knowledge of the management, which produced magazines published with the money of pharmaceutical companies without specifying sponsorship, and published advertising "research" in them.

Everyone liesDaniel Fanelli believes that the figures obtained, although shocking, are probably underestimated.

The questions asked by scientists are quite delicate. Moreover, probably not all respondents answered honestly, especially to questions about their own practice. It is also eloquent that in some surveys, only 20-25% of respondents eventually sent completed questionnaires. However, the sociologist suggests treating the statistics on major lies of colleagues (14%) with great confidence.

Things are hardly better in Russia. The Commission on Combating Pseudoscience and Falsification of Scientific Research of the Russian Academy of Sciences deals mainly with obvious fraudsters. The verification of the authenticity of the content of articles published by employees of reputable scientific organizations falls, as in the whole world, on the shoulders of reviewers. And the latter sometimes approach their work in bad faith, and this fact was vividly demonstrated by the sensational story with the "grubber". However, this is a topic for a separate study. In the near future Infox.ru he will try to find out how acute this problem is in Russia.

Portal "Eternal youth" http://vechnayamolodost.ru/01.06.2009

Found a typo? Select it and press ctrl + enter Print version