The more I learn about the mathematics behind interactions, the more they scare me. They're definitely not something we should be doing willy-nilly.
-
-
-
And yet, a recent Nature Perspectives article argues that scientists should check routinely for interactions with gender. And that this can be done without requiring larger sample sizes :-/
- Još 3 druga odgovora
Novi razgovor -
-
-
That Bayesian statistics is 'better'. Rarely said in those words, but often implied and never really explained. *runs for cover*
-
*grins*
Kraj razgovora
Novi razgovor -
-
-
That the mean is the best statistic for characterizing central tendency, and quadratic loss is normatively desirable
- Kraj razgovora
Novi razgovor -
-
-
Many of the times we assume i.i.d.
-
So many terrors tucked away in such an innocuous acronym...
Kraj razgovora
Novi razgovor -
-
-
Ive always question the accuracy of the xaxis of Kaplan Meier curves... and the profound influence they have in healthcare
-
I had not thought much about that one but omg yeah... I guess simple Kaplan Meier curves don't allow noise on the x-axis, which does seem... bad
- Još 2 druga odgovora
Novi razgovor -
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.
)