The term scientific fact has been coined throughout history to stamp a sure-fire statement as a hard-nosed fact. This fact describes a truth that has evolved out of meticulously collected evidence, to support a well-grounded theory. Today, scientific evidence is deemed so reliable that it largely dictates how we live our lives. It somewhat uproots the system a little, then, to say that science may not be reliably robust at all.
Today’s hard scientific facts, against a vast increase in scientific knowledge, seem to be slipping into a valley of subjective inaccuracies that only hinder our understanding of the modern world. Unfortunately, there is not just one simple reason why, but a medley of sources of misinformation, often used to support policy changing theories, via poor scientific practice.
“reproducible published science is a much rarer occurrence in modern literature than you would think”
In science, a phenomenon must be demonstrated multiple times before a trend amongst your collected evidence is deemed significant enough to support a theory. To do this, a researcher must perform an experiment repeatedly, otherwise one result could have arisen by chance, and not even be representative of a normal outcome under the circumstances. This is a principle taught to budding scientists at the age of about 11, and so should be relatively straight forward to abide by, surely?
Amazingly and unfortunately, reproducible published science is a much rarer occurrence in modern literature than you would think. Worryingly, in pre-clinical trials for oncological studies this is particularly evident. Failure rate of clinical trials in oncology is much higher than in other areas of science, and many failures have been attributed to poorly supported scientific evidence at the pre-clinical stage. A study looking at 53 landmark pre-clinical cancer papers found only 6 that could be reproduced, whilst another found that 75% of 67 published studies could not be reproduced without major inconsistencies.
“scientists often only publish partial datasets, abolishing data which contradicts their original theory”
With two out of two reproduced studies suggesting a lack of reproducibility in science, it may seem that we have quite quickly found a simple source of the problem. Unfortunately, the reasons behind poor science in today’s literature stems far deeper than that.
So how is evidence published that cannot be reproduced and not reliable enough to draw any conclusion from? A good chunk of scientific evidence (truly reliable evidence this time, I assure you) has found flaws in research originating from ignorance towards data not agreeing with a hypothesis.
Studies on the representation of results have demonstrated that scientists often only publish partial datasets, abolishing data which contradicts their original theory. If only 30% of your data supports a hypothesis, and only that data is used in a final publication, you can quite easily produce evidence to support most hypotheses you could wish for, it is just a case of what replicates you choose to notice.
“it is worth remembering that the scientific fact is not a fact at all”
As well as intentional flaws in scientific practice, unintentional mistakes are often not noticed before publishing such blemished science to support a theory. Before a piece of scientific research is published, it is rigorously checked, and checked again by peer reviewers to be deemed worthy of the public eye. This should be the gold standard of scientific research, but consistently throughout my university career, I am shown example after example of poor yet published science. When flawed science is found frequently in leading journals such as Nature, one starts to wonder how on earth it got there.
Unfortunately, bad research is published partially as an ill-thought through publicity stunt. When one finds the common denominator of a striking and eye catching title to catch the media’s eye, that could potentially be a source of further funding and publicity, this conundrum is answered very quickly and very simply. Accuracy is unfortunately often traded for a big and exciting story. Clearly The Daily Mail isn’t the only publication in existence to allow questionable facts in their pages.
So it is worth remembering that the scientific fact is not a fact at all, and although it should be known amongst the scientists and the public as a mere evidence-based opinion, it appears that this has been ignored. In forgetting what a true scientific fact is, people have been inclined to follow down two worryingly slippery slopes: either towards forgetting the need to critique an evidence based opinion, or towards forming an opinion and adopting tunnel vision in the hope of supporting it.
“scientists are promoting an increase in knowledge that isn’t knowledge at all”
Of course, the general public can be forgiven for failing to critique every scientific fact by checking the methods of every piece of research that is ever trusted. It is partially accepted, even, that most people must put trust in science for getting us to where we are today, and hope that good science prevails and continues to progress our lives.
What is troubling however, is that scientists themselves are also forgetting to not fall down either slippery slope. Essentially, scientists are promoting an increase in knowledge that isn’t knowledge at all. Instead it is the promotion of an opinion based and dangerous belief system in science, thanks to a combination of poor practice, ignorance, and desperate efforts to find the next big story – be it a scientific fact or not.
Georgina Bray
Image courtesy of poisson lucas via Flickr. License here.
For more information on science and technology please follow Impact’s Science and Technology Facebook page
