works cited: @sTeamTraen @maltoesermalte @patmarkey @uri_sohn @KateLaskowski @MicrobiomDigest @jamesheathers
-
-
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
-
-
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
-
-
Great post! But: *Eysenck
- Kraj razgovora
Novi razgovor -
-
-
One reason outright faking data may be so rare: it's unnecessary and psychologically unpleasant. Why fake data, forcing yourself to acknowledge 'I am the baddie', when p-hacking some random noise and underpowered samples is so easy and so truthy?
-
Consider Rosenthal's Pygmalion experiment: the raw data shows gains like +100 IQ points - and only in some classrooms/grades. The data has to be real: no one would ever fake such absurdities! But p-hacking garbage data lets Rosenthal sleep at night after fighting the good fight.
- Još 2 druga odgovora
Novi razgovor -
-
-
Agree with most of the blog, even if the pessimism is a little exaggerated. A big step forward would be to require all data and code to be published. Sure it's an arms race, but that's no reason not to make fraud as difficult and dangerous as possible.https://referee3.org/2019/12/25/data-sharing-should-be-mandatory-public-and-immediate/ …
- Još 1 odgovor
Novi razgovor -
-
-
This made me Lol a lot. Especially the editor & university. Painfully accurate
pic.twitter.com/1QlbUGHn26
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
-
-
I never had any doubt that quality data faking is hard to detect.
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
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.