3 ms·

And remember 50% of published scientific studies can't be successfully reproduced! 100% confidence.

• dekhn 9y ago
  50% of studies can't be reproduced, and 90% are outright false (which makes you wonder about the percent that can be reproduced, but are false) :-)
  • kazinator 9y ago
    "can be reproduced" means true. If you can reproduce the experiment, so that approximately the same data comes out, but the paper turns out false, that can only be because the paper makes a flawed interpretation of the data. But that has nothing to do with reproducibility; you can tell that a flawed conclusion is made from some data without reproducing the process which produces the data. If the analysis is correct and the empirical results can be reproduced, then the research is true. Or something like that.
    • sheepmullet 9y ago
      > "can be reproduced" means true. Not really. I know what you mean but that's not the same as what you are saying. You mean consistently and reliably over a large number of studies.