I consider reproducible analysis to be really important. In particular, I'd like to see more researchers in psychology:
- sharing their data
- producing their journal articles using literate programming technologies like Sweave and knitr that combine data analysis code and text into an integrated documenst.
I consider this particularly important in light of concerns over research fraud (e.g., data fabrication) and various forms of data fishing (e.g., testing many hypotheses and only reporting the significant one).
Certainly in statistical fields, there is a lot of discussion about the importance of reproducible analyses (e.g., see here, and here). However, I sometimes feel that in psychology this is only starting to filter through.
Thus, my question is:
- Are there any journal articles in psychology that have advocated reproducible analysis?
There are many angles that such articles might take: reward structures that discourage open science and reproducible analysis; technical challenges which make it difficult to do such analyses; how-to articles that outline how to perform reproducible analyses in psychology. I'd be interested in any articles that address these topics.