What is it about?
In this paper, we develop a method to standardize the reproduction of publications in the most simplest way, by "pressing go" on the statistical coding file using the original data. We find that the majority of authors will not share their data and code with us, which means that the majority of papers in our sample are not reproducible, even at this very basic level. The authors who refused to share their data include papers published in journals with data sharing requirements, and papers that say data will be made available upon request.
Featured Image
Why is it important?
We show in this paper that scientific publications continue to be mostly not reproducible.
Perspectives
We were very happy to publish this paper. This is kind of the capstone of my work with 3ie. The results are somewhat sad but true.
BENJAMIN WOOD
Integra LLC
Read the Original
This page is a summary of: Push button replication: Is impact evaluation evidence for international development verifiable?, PLoS ONE, December 2018, PLOS,
DOI: 10.1371/journal.pone.0209416.
You can read the full text:
Contributors
The following have contributed to this page
