Computationally reproducing results from meta-analyses in Ecology and Evolutionary Biology using shared code and data

This is a Preprint and has not been peer reviewed. The published version of this Preprint is available: This is version 2 of this Preprint.

Add a Comment

You must log in to post a comment.


There are no comments or no comments have been made public for this article.


Download Preprint


Steven Kambouris , David Peter Wilkinson , Eden T. Smith, Fiona Fidler


Many journals in ecology and evolutionary biology encourage or require authors to make their data and code available alongside articles. In this study we investigated how often this data and code could be used together, when both were available, to computationally reproduce results published in articles. We surveyed the data and code sharing practices of 177 meta-analyses published in ecology and evolutionary biology journals published between 2015–17: 60% of articles shared data only, 1% shared code only, and 15% shared both data and code. In each of the articles which had shared both (n = 26), we selected a target result and attempted to reproduce it. Using the shared data and code files, we successfully reproduced the targeted results in 27–73% of the 26 articles, depending on the stringency of the criteria applied for a successful reproduction. The results from this sample of meta-analyses in the 2015–17 literature can provide a benchmark for future meta-research studies gauging the computational reproducibility of published research in ecology and evolutionary biology.



Ecology and Evolutionary Biology


computational reproducibility, meta-analysis, data sharing, code sharing


Published: 2023-07-04 18:05

Last Updated: 2024-03-14 08:39

Older Versions

CC BY Attribution 4.0 International

Additional Metadata


Conflict of interest statement:

Data and Code Availability Statement:
The data and code files to reproduce all results reported in this article are available on Zenodo at