Twitter | Search | |
Javier Benítez / Pr(Dz | Data) ≠ Pr(Data | Dz)
Medical doctor, tweets not medical advice or endorsements. Interested in & technology. Always learning. I'm no expert. No financial conflict of interest.
40,045
Tweets
1,460
Following
3,351
Followers
Tweets
Javier Benítez / Pr(Dz | Data) ≠ Pr(Data | Dz) 7h
Replying to @PaulLantos
Really?! I could be wrong, but they don’t sound romantic to me.
Reply Retweet Like
Javier Benítez / Pr(Dz | Data) ≠ Pr(Data | Dz) 9h
Gigerenzer on NHST and causality
Reply Retweet Like
Javier Benítez / Pr(Dz | Data) ≠ Pr(Data | Dz) 9h
Reply Retweet Like
Javier Benítez / Pr(Dz | Data) ≠ Pr(Data | Dz) 11h
Two things I like most about statistics: 1. The debates are about substantive matter 2. They do not romanticize about stats practice It’s almost impossible to discuss matters of scientific practice when people have a romantic-fairytale view & approach about how the world works
Reply Retweet Like
Javier Benítez / Pr(Dz | Data) ≠ Pr(Data | Dz) 11h
What is forgetting for?
Reply Retweet Like
Javier Benítez / Pr(Dz | Data) ≠ Pr(Data | Dz) 13h
Free is advertisement for paid services you’re not being made aware of.
Reply Retweet Like
Javier Benítez / Pr(Dz | Data) ≠ Pr(Data | Dz) 19h
Yes, people will value nonsense & others will cash in on that by trying to make nonsense appear scientific. Some are ignorant while others are willfully ignorant of these processes. This is why it’s so important to teach critical thinking & how science works, to be less fooled.
Reply Retweet Like
Javier Benítez / Pr(Dz | Data) ≠ Pr(Data | Dz) retweeted
Javier Benítez / Pr(Dz | Data) ≠ Pr(Data | Dz) Oct 17
Some people’s job security depends on reinventing the wheel of scientific reasoning without calling it scientific reasoning.
Reply Retweet Like
Javier Benítez / Pr(Dz | Data) ≠ Pr(Data | Dz) retweeted
John Antonakis Oct 18
Two big issues. First ethical. Parading pseudoscience as science is wrong and muddies the signalling system. Second economical. Paying top dollar for something that can be achieved in a more cost effective way is folly. One can show diversity in teams using other valid measures.
Reply Retweet Like
Javier Benítez / Pr(Dz | Data) ≠ Pr(Data | Dz) 21h
“Remember that all models are wrong; the practical question is how wrong do they have to be to not be useful."—George E. P. Box
Reply Retweet Like
Javier Benítez / Pr(Dz | Data) ≠ Pr(Data | Dz) 22h
Reply Retweet Like
Javier Benítez / Pr(Dz | Data) ≠ Pr(Data | Dz) Oct 17
I don’t think people are born knowing this, so to me it is a sign of bad training when people cannot distinguish between the mechanism of disease (a model) with its epidemiological evidence, and these from clinical decision-making under uncertainty.
Reply Retweet Like
Javier Benítez / Pr(Dz | Data) ≠ Pr(Data | Dz) retweeted
Bryan Carmody Oct 17
Replying to @jbcarmody
I bring this up as yet another example of the unchecked power we’ve unwittingly granted the NBME and FSMB in running the USMLE monopoly. When you’re the only game in town, you can do what you want.
Reply Retweet Like
Javier Benítez / Pr(Dz | Data) ≠ Pr(Data | Dz) Oct 17
How can you talk to people about the uncertainty of scientific knowledge and decision-making when they have been conditioned to have faith in just-so stories?
Reply Retweet Like
Javier Benítez / Pr(Dz | Data) ≠ Pr(Data | Dz) Oct 17
Some people’s job security depends on reinventing the wheel of scientific reasoning without calling it scientific reasoning.
Reply Retweet Like
Javier Benítez / Pr(Dz | Data) ≠ Pr(Data | Dz) Oct 17
If you were not anointed by the powers-that-be you don’t get to criticize research practice, so they say. On the term “self-appointed” . . . « Statistical Modeling, Causal Inference, and Social Science
Reply Retweet Like
Javier Benítez / Pr(Dz | Data) ≠ Pr(Data | Dz) Oct 17
Why do they keep it a secret?
Reply Retweet Like
Javier Benítez / Pr(Dz | Data) ≠ Pr(Data | Dz) retweeted
Javier Benítez / Pr(Dz | Data) ≠ Pr(Data | Dz) Oct 12
“A culture of publication-worship unwittingly incentivizes questionable scientific practices and gluts the economy of scientific papers, argues Ava Kiai. To protect trust in science, we must focus on methodological rigour, rather than publishability.”
Reply Retweet Like
Javier Benítez / Pr(Dz | Data) ≠ Pr(Data | Dz) retweeted
Javier Benítez / Pr(Dz | Data) ≠ Pr(Data | Dz) 12 Mar 15
Reminds me about the famous Yerkes and Dodson curve used in stress experiments
Reply Retweet Like
Javier Benítez / Pr(Dz | Data) ≠ Pr(Data | Dz) retweeted
Rob Briner 28 Mar 17
Turns out the Yerkes-Dodson Law is more folklore. But still in the slide deck of every stress/well-being workshop.
Reply Retweet Like