4 ms·
>In what situation would you trust a study based on one person? Anytime someone else makes a blanket statement with no study (example: amphetamine doesn't make
by throwaway_tech 7y ago
>In what situation would you trust a study based on one person?
Anytime someone else makes a blanket statement with no study (example: amphetamine doesn't make you crazy)...all you need is n=1 to disprove that.
- drewsmahgoos 7y agoBut that goes both ways...
- throwaway_tech 7y agoProving and disproving a hypothesis are entirely different standards. For example I have a coin with heads/tails, I hypothesis if I flip it it always lands on heads and we flip it 1 time (n=1), and sure enough it lands on heads that doesn't prove my hypothesis (its very weak support at best), whereas is we flip the coin 1 time (n=1) and its tails that disproves my hypothesis (or at least is strong support my hypothesis is wrong).
- vidarh 7y agoBut that requires demonstrating an actual casual relationship, not just a correlation.