What you will find on this blog
I blog about reproducibility, replication, pre-registration, research transparency and open peer review. The blog started as a personal project with a focus on political science, and it is now used by many scholars to exchange ideas about replication in the wider social and natural sciences, for example in a guest post series.
Why I started this blog
I started the blog in January 2013 because I was frustrated about the state of reproducibility in political science:
- many journals did not publish replication policies online
- many journals did not require authors to upload replication data when they submit the article, or at least upon acceptance
- many authors did not keep replication data files on their computers to send them out on request
- many authors did not keep clearly commented R code or STATA .do files for their models
- some authors did not answer to emails about replication, or they refused to send the data
I should stress that I have also exchanged many fruitful emails with researchers who were helpful and willing to provide their files and help me and my students to replicate their work. But this does not seem to be the standard yet.
About publishing my emails with original authors
I sometimes publish personal correspondence – helpful and frustrating – about my own replication endeavors. I do this rarely, and I make sure that these emails are anonymized. This is not to offend my colleagues. I do understand that everyone is busy and might not be able to dig up old files and clean them up. But I do hope that pointing to difficulties of obtaining data on this blog contributes to a change in attitude about replication.
My wish list
Ideally, I would like to see the following changes in the social sciences:
- All journals should require that authors upload the data sets and software code for all models used upon acceptance of their article.
- Authors should keep detailed records about data sets, models, R code and .do files for all results presented in the full text, footnotes and appendix.
- Authors should not just ‘promise’ to send data ‘on request’, but do so. Ideally, they would upload their data to journal archives or data repositories.
- There should be an open source Journal of Social Science Replication to boost replications not only for learning purposes but also as a valuable contribution to the field.
Replication workshop at Cambridge
To integrate replication practice into teaching at Cambridge, I teach the Cambridge Replication Workshop. In 8 weeks, graduate students replicate a paper. The course runs a third time in 2015; several students have gone on and published their replication study in a peer-reviewed journal.
Research on reproducibility
My article “Bringing the Gold Standard Into the Class Room: Replication in University Teaching” is now published in International Studies Perspectives.
Please contact me if you have questions about the blog, if you would like to write a guest post, or if you would like to share ideas about reproducibility.
Research Methods Associate at the Social Sciences Research Methods Centre
University of Cambridge
Email: firstname.lastname@example.org or via my uni email (faster).