Producing Reproducibility

A new initiative will tackle the reproducibility problem.

Full-text access for registered users only. Existing users login here.
New to GenomeWeb? Register here quickly for free access.

Sincere appreciation for the

Sincere appreciation for the efforts of eLife. There's only one way to document and subsequently insure reproducibility: redo the work by a third party.

Sir Francis at the NIH will not solve the problem with postdoc training, checklists and trying to rewrite publishers review policies. His club has dollar signs attached - until the NIH openly funds third-party reproduction of results (a trivial expense) no progress will be made.

There are two types of

There are two types of research irreproducibility: one is deliberate fraud, the other results from insufficient statistic. In the latter case increasing the numbers could prove the conclusions correct. It is known that Mendel's original data ware fraudulent and would not support his laws. When Mendel's laws were rediscovered the results were reproduced at a large scale and the laws proven correct. Michael Lerman, Ph.D., M.D.