Reproducibility and the Philosophy of Data with Clare Gollnick - TWiML Talk #121


:star: :star:

Notes

  • Guest: Clare Gollnick has neuro background
  • P hacking: reanalyzing data to try and find an explanation. Icky
  • All data is from the past. Assumption that it’s predictive is faulty
  • Better model is “I looked at it differently” instead of “I tried a ton of stuff and one happened to work”