This note discusses terminology, but it isn’t just semantic. It is important to make a meaningful distinction between the two terms.

Scientists usually discuss the need to “replicate” an experiment, not to “reproduce” it. But somehow, when it comes to discussing results, the terms “reproducible finding” and “replicable finding” seem to be interchanged. And when it comes to the property of good research the term “reproducibility” overtakes “replicability”. For example, ‘The Reproducibility of Psychological Science’ initiative reports about the ‘replicated‘ studies and not about the reproducible ones.

“Reproducible result” has a very clear meaning in some parts of science. Here is an explanation endorsed by the founding editors of ‘Biostatistics’ journal:

“The replication of scientific findings using independent investigators, methods, data, equipment, and protocols has long been, and will continue to be, the standard by which scientific claims are evaluated. However, in many fields of study there are examples of scientific investigations that cannot be fully replicated because of a lack of time or resources. In such a situation, there is a need for a minimum standard that can fill the void between full replication and nothing. One candidate for this minimum standard is “reproducible research”, which requires that data sets and computer code be made available to others for verifying published results and conducting alternative analyses. Peng, R. (2009). Reproducible research and Biostatistics Biostatistics, 10 (3): 405-408 .

Thus “reproducibility” means that if we start from the data gathered by the scientist we can reproduce the same results, p-values, confidence intervals, tables and figures as those reported by the scientist. Anyone involved in research knows this is not a trivial requirement. It is a requirement for drug registering but a very expensive one. Efforts to make this easier are on-going: see for example the Reproducible Research Planet and the Reproducible Research Net efforts.