Assessing Reproducibility

Reproducibility can be defined as the ability for independent scientists to obtain the same quantitative results using the same datasets and methods as the original study. According to the US National Science Foundation, reproducibility is a minimum necessary condition for a finding to be believable and informative. For a study to be reproducible, the reader must be able to:

• Understand what was done by the researchers
• Understand how the data was analyzed
• Understand how to obtain full access to data generated by the experiment

 Questions to ask when assessing reproducibility include:

• Have the authors detailed all of the materials used in the study?
• Have the authors reported the identity of all instruments used and provided all instrument settings?
• Have the number of repetitions of the experiment been clearly stated, and has an estimate of experimental error been clearly provided?