This study by EK Samota and RP Davey investigated the knowledge of researchers about issues related to data reproducibility as well as factors that can improve reproducibility.
The authors conducted a survey on 251 researchers who have published in the eLIFE journal, and researchers from the Norwich Biosciences Institutes (NBI).
In general, according to this survey, scientific publications are still lacking sufficient information, which allow to reproduce research outcomes. The majority of respondents stated that data underlying a published study are either not available in publications and/or are not provided upon request. Notably, 91% of respondents specified that detailed description of methods, including data, analyses and code, would enable a better reproducibility of published research. Moreover, the majority of the participants rated that interactive figures in published articles a well as the automatic reproduction of computational experiments are beneficial for both readers and authors: “The figure would give access to the raw data, code and detailed data analysis steps, allow for in situ reproducing computational experiments by re-running code including statistical analyses “live” within the paper”. Moreover, the authors of this article suggest that “this may be a good starting point for improving research reproducibility by reproducing experiments in situ”.