Just as certain Buddhist teachings aren't available to people until after transmission, I think "science is broken" is a message that should be carefully deployed so you don't empower quack doctors, vaccine/climate/etc denialism, and other harmful memes.
-
-
Replying to @mrgunn @jamesheathers
Yes, that's a hard call. I think I'm in the "it's better for us to yell about it first and try to fix it, because fixing it quietly is infeasible, and the bad guys are already on the case" camp, but I can see arguments on both sides.
2 replies 0 retweets 10 likes -
This Tweet is unavailable.
-
This Tweet is unavailable.
-
Yes… and the problem is (probably) mostly not “sketchiness,” it’s “cluelessness” and “doesn’t really care.” We don’t know quite what the right incentive gradient is, which is why
@michael_nielsen’s proposal to create many and see which work is so attractive.2 replies 0 retweets 5 likes -
Replying to @Meaningness @dailectic and
I weakly think the existing institutions are unfixable. They can be outcompeted, and then they will either collapse or fix themselves. But either reform or creating alternatives will probably involve considerable hue and cry. It’s a big job that needs noise to mobilize resources.
2 replies 0 retweets 4 likes -
Replying to @Meaningness @dailectic and
I’m no longer close enough to the institutions to have a well-informed opinion about this, however.
2 replies 0 retweets 0 likes -
Replying to @Meaningness @dailectic and
There's a large body of research (see
@METRICStanford and links therefrom) on what incentives exist, how they're responded to, what correctives exist & how well they work.1 reply 0 retweets 0 likes -
Replying to @mrgunn @Meaningness and
One of the most promising interventions is
@RegReports. Essentially, you publish your experimental design, hypothesis, analysis plan & then publish again when you have the data. This limits "P-hacking".1 reply 0 retweets 0 likes -
Replying to @mrgunn @Meaningness and
In general, it helps to limit HARKing (sorry for all the jargon) & limits the effect of certain kinds of bias. If you know a reported result was unexpected, it changes how you interpret the finding. Another intervention is open peer review. Early days, but this shows promise.
2 replies 0 retweets 0 likes
Yes, thanks, I’m thoroughly familiar (at an interested-outsider level) with these approaches, and I think they’re fantastic and loudly applaud them. I’m not sure they are enough, for the reasons I went into here… https://meaningness.com/metablog/upgrade-your-cargo-cult#replication …pic.twitter.com/gocos09k8D
-
-
Replying to @Meaningness @dailectic and
Will read with interest, but for now, I agree it's probably not enough. Funders & institutions are now willing to adopt them & fund research into interventions, so the noisemaking worked. We'll see how much impact they have.
1 reply 0 retweets 0 likes -
Replying to @mrgunn @Meaningness and
Don't get me wrong; my instincts are very much "burn it down", but I am in a position to see many aspects of how things work. Lots of things are well designed & conducted by professionals. It's hard to think those parts need burning down, rather incremental improvement.
1 reply 0 retweets 0 likes - 3 more replies
New conversation -
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.