I’m not saying gut instinct is a substitute for careful analysis, but you can rule out a lot of papers as basically-null results spun as significant ones, just by using your eyes.
-
-
Show this thread
-
(This thread was brought to you by a 15% median survival increase in C. elegans. For a mouse that would be huge; for a worm, it’s easily noise. But the headline is “increases lifespan!” This is mild puffery as things go, but it probably gives some people the wrong idea.)
Show this thread
End of conversation
New conversation -
-
-
As social scientists, we're taught to look for significant effects (confidence that effect is non-zero), but in the grand scheme of things and understanding the world a very small effect of something is not very different from no effect. That's important to understand.
-
It’s a long time I’m out of universities, but i believe the line between a small effect and unknown biases is a fine one to walk, right? Or there are more “sophisticated” statistical tools I didn’t study?
End of conversation
New conversation -
-
-
Do you have an example?
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Sameeeee, although I haven't read nearly enough scientific papers to be able to judge, but I've experienced in statistics classes that once you have n=10…20, it's *extremely* easy to generate data that rejects H_0 at p=0.05 but still makes the viewer go "there's nothing here".
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
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.