Researcher says most scientific studies "are false"



[ad_1]

The dissemination of scientific research through articles, conferences and conferences, among other mechanisms, contributes to the collective construction of knowledge and the promotion of continuing education. It allows the entire population to access important information related to health and well-being

However, for some, this would also have a negative side. The pressure on researchers, the competition between scientific journals and the insatiable appetite of media for studies that announce revolutions, great discoveries or the latest news derive from information being published-sometimes-little verified

Why? Because many studies use samples too small to arrive at conclusions that can be used to generalize to a total population, since it is few selected participants are likely be representative.

This was the goal of a 2013 study in which two researchers selected the 50 most used ingredients in a cookbook and badyzed how many of them had been badociated with a risk. or a benefit against cancer. in several studies published in scientific journals.

The answer they got was striking: 40 out of 50 [19459] 004] a list that included salt, flour, parsley and sugar. "All we eat is cancer-related?" They wondered and that was the title of their work

" Most of the articles published, even in serious journals, are lazy ", considered in a dialogue with the press agency AFP one of the authors, John Ioannidis, professor of medicine at Stanford, specializing in the study of studies. [19659007] This detractor of the bad scientific research demonstrated in a 2005 article "why most published studies are false ." From that moment, he states that only some progress has been made. [19659002Forexamplesomejournalsrequireauthors to provide their raw data and publish their protocol previously, and this transparency prevents researchers from twisting their methods and data to find a result. ;at to verify or "replicate" the study.

It often happens that when they are redone, the experiments rarely lead to the same results. Only one third of the 100 published studies in the three most prestigious psychology journals could be reproduced by researchers, in an badysis published in 2015. For Ioannidis, there is the same problem in others branches of science, such as medicine, epidemiology, drug clinical trials and nutritional studies. "In the biomedical sciences and elsewhere, scientists or are not sufficiently trained in statistics and methodology ," he said.

How to choose between studies

So what to choose between significant number of studies published each day? Ioannidis recommends to ask the following questions : Is this an isolated study or does it strengthen existing work? Is the sample small or big? Is it a random experience? Are researchers transparent?

These precautions are fundamental in medicine, where bad studies contribute to the adoption of treatments at best ineffective and, at worst, harmful .

The solution goes through a collective adjustment of the joint criteria of all those involved in the process of publishing not only journals, but also universities, public funding bodies and laboratories, according to experts in the field.

The media also have their share of responsibility, they must better explain to their readers the uncertainties inherent in scientific research, and avoid sensationalism, to understand by Ivan Oransky, journalist co-founder of the site Retraction Watch ] which covers withdrawals of scientific articles. " The problem is the endless succession of studies on coffee, chocolate and red wine, we have to stop ," he told the same news agency.

[ad_2]
Source link