randomizeR FTW
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Sometimes I get annoyed that randomized trials dedicate so much of their methods section to how they generated & distributed the random numbers, and then I see something like this

-
much yikes
End of conversation
New conversation -
-
-
Ok, I'll bite. 1) If you are recruiting on the spot, it's sometimes convenient to also randomize on the spot. Yes, you can certainly weigh randomization by # of people already enrolled in each arm, but you can run into even bigger problems that way, particularly with small n's.
-
2) Different # of people in each arm is a problem for statistical efficiency (i.e. uncertainty), but not really for the sorts of biases you're looking to avoid in an RCT, so maybe not such a big deal. 3) The problem here isn't so much the randomization procedure as it is small n.
- Show replies
New conversation -
-
-
At the end of my maths degree, I thought I would have much deeper and more arcane issues with randomisation than... this sort of thing.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Were the descriptive characteristics similar between arms?
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.
