Data are ‘cleaned’, methods and protocols are shortened and neatened, and conclusions are aggrandized to fit with exacting specifications while less than scrupulous journals spring up, willing to accept papers that don't walk the tight rope stretched between interest, impact and presentability, and credible scientific results.

This positive feedback loop, using publications as currency for the prestige of an institution and the success of an academic career, leads to a spiral of increasing numbers of ‘impactful’ papers that are less likely to be reproducible. Instead, academics are incentivized to chase publication targets, meanwhile, many journals will prioritize the acceptance and publication of a paper that will make a big impact in a field over a reproducible but less dramatic paper, leading to citations and the attainment of an ever-increasing impact factor – the leading metric by which a journal's worth is valued by industry and academic institutions alike. The very basis of the scientific method is not set up to facilitate or encourage reproducibility. It should be clear by now that issues of reproducibility in science run much deeper than simply unreliable experimental reagents, poor documentation and shaky protocols. The results from Nature's survey showing that researchers from many walks of research had issues reproducing others and their own work. So, with the abundance of information available and global awareness of the importance of reproducibility increasing, why did a 2019 survey of 243 researchers from 21 countries – 79% of whom worked in the USA – find that only 47% of the participants were very aware of US National Institutes of Health (NIH MD, USA) research rigor initiatives, with a shocking 26% completely unaware of the initiatives set up by the institute in order to improve reproducibility in the scientific community ? In combination with an earlier 2016 study that indicated 70% of 1576 researchers had been unable to reproduce another's experiments and over half had even been unable to reproduce their own ( Figure 1), this indicates that attitudes to reproducibility and meaningful changes are currently slow in coming. Papers have also rocked the research world that highlight the catastrophic waste of time and money that could be incurred if incorrect agents are used or detailed documentation of protocols and reagents are not provided, leading to fundamental research that is not reproducible. Improvements in the standards and quality control of study reagents, such as antibodies and cell lines, have been marked, with nonprofit organizations such as ATCC (VA, USA) offering authenticated cell lines and numerous industry and academic papers providing information on the validation of antibodies. In recent years, however, gains in reproducibility remain marginal. Reproducibility is a topic that has been much discussed, debated and repeated over the last decade as the scale of the ‘reproducibility crisis’ has become apparent.

Scite zoomed in