fixedpoint.jp - Repeatability, replicability, reproducibility, oh my!




Repeatability, replicability, reproducibility, oh my!

We often see a word "reproducibility" in a scientific context, as well as in a bug report of a software, and we know it plays an important role at analysing an uncertainty. But wait, there are a couple of similar things aside; "repeatability" and "replicability". How they are different from each other?

Here are corresponding articles in Wikipedia:

Do you distinguish between them clearly now? Not really, at least to me. It seems harder than expected as the following papers matter:

The last one indicates that it is helpful for reproducible research to make repeating/replicating an experiment easier with make(1).


© 2006-2014 fixedpoint.jp