I don't have an informed opinion about whether strong theory would increase or decrease the replication rate, but I'm fairly confident it would substantially decrease the need for, and value of, replication studieshttps://twitter.com/hardsci/status/1050157233659498496 …
-
Show this thread
-
Replying to @talyarkoni
What’s a good theory, in this context? Most widely accepted? Fits more evidence? Generates best predictions? Is more parsimonious with other theories? All of these are measures of strength of theories, but they need not go hand in hand.
1 reply 0 retweets 2 likes -
Replying to @felipedebrigard
fair point. I used the term "strong" to avoid some of the ambiguity that surrounds "good". I'd say a strong theory is one that makes clear, precise predictions. if I predict and confirm an effect of 0.4481 +/- 0.02, fewer people are going to yell "okay but does it replicate?"
2 replies 0 retweets 3 likes -
Replying to @talyarkoni @felipedebrigard
But (strong) theories are not just about predictions; you need good evidence for them. That is why strong theory and replicability go hand in hand, because they are mostly the same thing but with a different name.
3 replies 0 retweets 2 likes -
Not sure I entirely agree here. Strong theories will predict otherwise nonobvious things that will turn out to be replicable. But I think it's entirely possible to have highly replicable findings with basically no underlying theory. They're just good observations.
2 replies 0 retweets 6 likes -
Replying to @wgervais @Research_Tim and
Highly replicable findings = theory.
1 reply 0 retweets 0 likes
This is really confusing. What does equals mean here? Do you think phlogiston theory is "highly replicable"? Or do you think it's not a theory?
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.