This ties in with current ideas and visions about machine-readable and actionable data and code, as well as current debates on statistical inference (p-values, alpha levels etc.). (2/4)
Prikaži ovu nit
I've just read this preprint on machine-readable hypotheses by @lakens and @lisadebruine: https://psyarxiv.com/5xcda/ . What they propose is very precise specification not only of hypotheses as such but also of how to decide when the hypotheses are corroborated. (1/4)
-
-
-
One can envision a future space of research objects, callable by anyone, to connect data, analysis code, and inference procedures. (3/4)
Prikaži ovu nit -
Importantly, in my view, this highlights a partitioning of value to the different components. You don't need to invent a new theory to do science. (4/4)
Prikaži ovu nit
Kraj razgovora
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.