Scientific experiments must be reproducible to prove its claims and for future use. However, there has been a "reproducibility crisis" that has been bothering scientists. The irreproducibility of science studies has become a major concern among drug developers who considers pre-clinical research very important.
According to a research, more than two-thirds of researchers can't reproduce another researcher's science studies, BBC reported. Immunologists Dr. Tim Errington carries out The Reproducibility Project, which tries to recreate the findings on five important cancer studies. He does the experiments at the University of Virginia's Center for Open Science.
The main goal of the project is to try to do the same exact things from various experiments and determine whether they can get the same results, said Errington. Though it may not be easy, science studies and experiments should be replicable. The original researchers of these experiments should have conducted the replication before they publish their works.
The published works should include the step by step methods of the experiments. And other scientists should be able to simply follow the instructions. However, Errington found out that most of these experiments are sadly irreproducible.
It is important to prove that scientific claims are true by reproducing it, The Conversation reported. However, a survey conducted by Nature shows that 52 percent of researchers believe there is a problem in reproducing these experiments and it has become a crisis. Meanwhile, 38 percent of the scientists believe there is a slight crisis in reproducing these experiments.
Dr Errington believes this is a cause for alarm since replication is a hallmark of scientific integrity. Reproducibility crisis has been growing for some time now. One scientist at the Bristol University even almost quit his career since he always fail to reproduce an experiment on anxiety. Marcus Munafo, who is a professor of biology at the Bristol University said he had problems with confidence thinking he didn't conduct his study well.
