A major step in scientific research is the confirmation of experimental results. Studies are replicated to ensure that their results are not due to errors or fraud. However, this process is time-consuming, expensive, and potentially biased. Having an accepted and inexpensive way to independently confirm findings would accelerate scientific research, leading to faster understanding of disease and more certainty in reporting.
A solution to reproducibility
Biologist Elizabeth Iorns recognized this issue and started the company Science Exchange, which in the Fall of 2012 spearheaded the Reproducibility Initiative. Via this pay-for-replication model researchers can have their results independently confirmed for, as estimated by Slate, approximately one-tenth the cost of the original research.
A number of journals have given a nod of approval to this initiative, including the Public Library of Science (PLoS) journal PLoS ONE. For some journals, including Nature, scientific papers that are confirmed by the service will link to the reproduced data. Yet other may receive a badge alerting readers and fellow scientists that the results were corroborated, which is considered a sign of trustworthiness and validation.
How the Reproducibility Initiative works
When a researcher wants to have their study validated, they send their study to Science Exchange. The company then matches the researcher with one of a thousand participating biotech companies and labs. The researcher pays that lab to redo the study in a blind fashion (i.e. they do not know the original findings). The researcher is then provided the outcome to know whether the results were reproduced. If the study is confirmed, the researcher receives a Certificate of Validation and writes a second paper heralding the confirmation of the information they originally published. In fact the PLoS is starting a new journal specifically for these results – the PLoS Reproducibility Collection.
A step forward in research
According to Business Insider, the company Amgen was unable to reproduce the findings of 88 percent of landmark cancer studies. This translates into wasted funds and time because erroneous findings lead to dead ends and unusable, sometimes simply inaccurate, information. Now researchers will know how reliable data and protocols are when they start, saving them the task of validation before building on that work. The broad scope of this project will take it beyond medical research, though the initial beneficiaries look to be the best funded areas – cancer, stem cells, and genetics.
At a time when information is just a mouse click away, and yet the number of retractions is increasing among scientific journals, the Reproducibility Initiative of Science Exchange, PLoS and other journals is attempting to increase the reliability of reported scientific findings and alert the community to experimental issues. Science Exchange is requesting a limited number of studies to start the project to ensure the approach works before opening it up to all researchers in their collaborators' fields.