Replication study seeks repeats
A new study has attempted to replicate 21 social science research projects.
Replication is a key part of the scientific method, solidifying findings and ensuring results are not just one-offs or flukes.
But many areas of research are in the midst of a 'replication crisis', realising that many studies cannot be repeated with the same results.
In a new study, researchers attempted to reproduce 21 studies published in the two top journals: Nature and Science.
The primary replication method used in the study found that one out of the four Nature papers and seven of the seventeen Science papers evaluated did not replicate.
Researchers in the US conducted high-powered replications of the 21 experimental social science studies — using sample sizes around five times larger than the original sample sizes.
They found that 62 per cent of the replications show an effect in the same direction as the original studies.
This effect varies between 57 per cent and 67 per cent, depending on how replicability is determined.
The authors suggest that the original studies are likely to have contained both false positives and inflated effect sizes.
Based on a survey of 400 social scientists, the authors also report a strong correlation between the research community’s expectations of each study’s replicability and the reality.
Authors of the non-replicating studies have been invited to discuss why their results did not replicate, in the hope of finding potential sources of irreproducibility.