The pressure to publish pushes down quality

The pressure to publish pushes down quality

I am pleased to announce that as of mid-April, my Elsevier publications had received 30,752 page views and 2,025 citations. I found these numbers in a promotional e-mail from Elsevier, and although I’m not sure what they mean, I think it would be better to have an even larger number.

In fact, the wide availability of bibliographic data from sources such as Elsevier, Google Scholar and Thomson Reuters ISI makes it possible for scientists (with their employers to focus on their shoulders) to obsess about their productivity and effectiveness and compare their numbers to those. Makes it easy to do with other scientists.

And if more is better, the trends are favorable for science. The number of publications is growing rapidly; By 2012 it was already approaching two million per year. More importantly, contrary to common mythology, most papers are cited. In fact, more papers are being cited more frequently, over time, from more journals. One possible reason for the increasing citations is the incredible search capabilities that the web now offers. This would seem like good news.

But what if more is worse? In 1963, physicist and historian of science Derek de Sola Price observed development trends in the research enterprise and saw the threat of “scientific doomsday”. The number of scientists and publications had been growing rapidly for 250 years, and Price felt this trend was unsustainable. Within a few generations, he said, this would lead to a world in which “we have two scientists for every man, woman, child and dog in the population”.

Price was also an elite class who believed that quality could not be maintained in the midst of such growth. He showed that scientific prestige was concentrated in a very small percentage of researchers, and therefore the number of prominent scientists would grow much more slowly than the number of only good ones, and would lead to “an even greater importance of manpower capable of writing”. scientific paper, but not able to write specific”.

The quality problem reared its head in such a way that the price could not have been estimated. Mainstream scientific leaders increasingly acknowledge that large bodies of published research are unreliable. But what seems to have escaped general notice is a disastrous backlash between the production of poor-quality science, the responsibility to cite past work, and the compulsion to publish.

The quality problem has been widely recognized in oncology, in which many cell lines used for research become contaminated. For example, the breast-cancer cell line used in more than 1,000 published studies actually turned out to be a melanoma cell line.

The average biomedical research paper is cited 10 to 20 times over 5 years, and one-third of all cell lines used in research are known to be contaminated, so the arithmetic is fairly easy: by one estimate, 10,000 The papers published in the year cite work based on contaminated cancer cell lines. Metastasis has spread in the cancer literature.

The enterprise of science is developing towards something different and has so far only been seen slowly.

Similar negative feedback occurs in other areas of research. Biomarkers for neurological diseases, cancer and other diseases, and experimental psychology have revealed widespread quality problems for rodent studies amid the publication of thousands of papers.

So yes, the web makes it more efficient to identify relevant published studies, but it also makes it easier to troll for supporting papers, whether they are good ones or not. No wonder quote rates are on the rise.

The problem is likely to be worse in policy-relevant fields such as nutrition, education, epidemiology and economics, in which the science is often uncertain and the social stakes may be high.

The never-ending debate about the health effects of dietary salt, or how to structure foreign aid, or how to measure ecosystem services, are typical of areas in which there is abundant peer-reviewed support for taking either position. could – a condition that then justifies the call for more research.

More than 50 years ago, Price predicted that the scientific enterprise would soon undergo a transition from exponential growth to “something fundamentally different”, unknown and potentially dangerous. Today, the interconnected problems of scientific quantity and quality are a frightening expression of their predicament.

It seems extraordinarily unlikely that these problems will be solved through improved statistics and home remedies of laboratory practice, as important as it may be. Rather, they seem – and that’s what Price believes – to declare that the enterprise of science is developing into something different and so far only slow.

Current trajectories threaten to drown science in the noise of its increasing productivity, a future that Price describes as “aging.”

Leave a Reply

Your email address will not be published. Required fields are marked *