3 ms·

If an experiment is not repeated, it's not important.

• jzachary 17y ago
  That comment suggests an interesting study: how many experiments published in papers are actually repeated by other researchers. I think in computer science it is very low. I would bet that in the physical sciences it could be even lower given the cost of physical laboratory equipment and time constraints to publish "original" research. Does anybody have data or experience to suggest repeat rates?