Rezultati pretraživanja
  1. Monty has questions about the article saying you shouldn't sleep next to your dog, like "is a 3 pt difference in sleep efficiency (out of 100) even statistically significant? He suggests you foster him to test it out together! 🐶💗

    Prikaži ovu nit
  2. 28. stu 2019.

    Best evidence I know against current selection pressures in academia: 1) Selecting on number of publications rewards unreliable science: 2) Selecting on journal rewards unreliable science:

  3. 20. ruj 2017.

    🤨According to Elsevier’s Editorial System, a gets the lowest rating for impact on the field by default.

  4. 31. srp 2018.

    "So you banned p-values, how’s that working out for you?" has had a look at whether you can really improve papers by leaving out the p-values. (Spoiler: no).

  5. Odgovor korisniku/ci

    Even research-based social sciences are going through a right now. "Theory" means nothing in these fields.

  6. 20. velj 2018.

    What started out as a blog post is now an invited review article: The data suggest that hiring, funding and promoting scientists who publish much and in high ranking journals contributes to the

  7. If u can’t measure it, u can’t study it It u can’t analyze it appropriately, u can’t understand it If u can’t reproduce it, u shouldn’t publish it

  8. 9. kol 2019.

    🎙 Two psychologists, one having published several studies on meditation, discuss the state of research. Spoiler: thousands of studies and decades of work have found pretty much nothing, partly but not only for reasons.

    Prikaži ovu nit
  9. 13. velj 2018.

    Looks like Brian Wansink's "bottomless bowl" study is about to go down in flames. Who wants to try to replicate? Hypothesis seems plausible.

  10. 1. pro 2018.

    Our new paper is out in JEP General. We argue that averaging over effect sizes in meta-analyses can be problematic and can lead to accuracy loss /thread

    Prikaži ovu nit
  11. 2. pro 2017.
  12. 25. svi 2018.

    If you can, make time to listen to Alexander Bird's talk on the so-called . There's a link from my blog on :

  13. 8. srp 2014.
  14. 4. lip 2019.

    Given the string of recent, costly failures of Alzheimer's trials (see ) it's worth re-emphasizing that biased results reporting in preclinical work is the root cause of the (see this review from 2012 as example )

    Prikaži ovu nit
  15. 1. svi 2016.

    Fresh interview w/ : How good Management leads to better Science

  16. “Creating a culture of replication takes prizes, grants and magnanimity — as well as publications”. weighs in on the

  17. Experiments show that Inuit man was just shitting with the gullible ethnographer: knives manufactured from frozen human feces do not work, apparently. h/t ⁦

  18. The familiar pattern: pro-left-wing finding from social psychology shows very questionable pattern in funnel plot analysis with ~0 effect size for largest studies.

  19. 2. lip 2017.
  20. 29. ruj 2016.

    Science's replication crisis described in a single table. Source: Sterling (1959).

Čini se da učitavanje traje već neko vrijeme.

Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.