The Replication Crisis of Science Right Now

The reproducibility of experiments is an important part of the scientific process. When researchers prepare the methods and materials sections in their manuscripts, it’s supposed to be drafted in a way that helps other scientists reproduce their results. But recently, the scientific community realized that the findings of many experiments were in fact not reproducible, by neither the original researchers themselves or independent researchers. They rightly became concerned that there may be a significant amount of false data published – something that became so widespread it was named “the replication crisis.”

1500 Scientists Polled

In a 2016 poll conducted by the journal Nature, it was revealed that 70% of the 1500 scientists who were polled failed to replicate a colleague’s results, and half could not reproduce their own data. More than 50% of those polled noted that the results were a significant crisis in all areas of science. The better news is that less than about a third of those scientists believed that failure to reproduce published results doesn’t necessarily mean the result is probably wrong – the majority stated that they still trust the published literature.

What Scientists are Doing to Fix the Problem

Scientists themselves say that it’s critical for those who are new to the field to learn to follow more rigorous standards that include learning more about statistics. They hope to mentor new scientists, who, by becoming more familiar with this data and other methods, as well as collaborating with statisticians, will help to make a dramatic improvement in the way data is analyzed and presented, naturally leading to more reliable data getting out to the public. They also recommend conducting validations within the lab, having another expert validate the experiment – if it can’t be reproduced the initial results are problematic.

Independent labs are important too – if another lab can validate a scientist’s findings, they’re likely to be accurate. Before any data is published, it should be replicated in a different environment, the scientists say.

Data is Important for Just About Everyone

You don’t have to be a scientist to use data that can help you make better decisions. Utilizing data can be essential financially and otherwise. For example, using a home affordability calculator to answer the question “How Much House Can I Afford?” Buying a house is a major financial decision that can affect your bottom line for three decades or even longer – using data like this will help you understand how much house you can afford so you don’t get in over your head.

This article does not necessarily reflect the opinions of the editors or management of EconoTimes.