Effect sizes of preschool interventionspic.twitter.com/GXeVP4hpvQ
@JamesPsychol Try plotting the sample sizes vs. effect size. Very clear publication bias.
-
-
@KirkegaardEmil Yes, and even bigger publicity biasThanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
@KirkegaardEmil@JamesPsychol That is essentially what is done with the bubble sizes -
@Billare@JamesPsychol Yes, but not very well. Just plot the data the right way. Probably someone should publish that. I have the data. :) -
@KirkegaardEmil@Billare Publish, and I will link to it. -
@JamesPsychol@Billare I did post the funnel plot before https://twitter.com/KirkegaardEmil/status/591582923905708032 … but I will write it up and make a better one.
End of conversation
New conversation -
-
-
@KirkegaardEmil@JamesPsychol Yes, a PET-PEESE meta-regression would be very telling. -
@JoeHilgard@JamesPsychol I have the data, but no one have written PET-PEESE for R yet. Can you help? -
@KirkegaardEmil@JamesPsychol I have a (very rough!) PET-PEESE function written up at https://github.com/Joe-Hilgard/PETPEESE …. You're welcome to it. -
@KirkegaardEmil@JamesPsychol It assumes column names generated by Comprehensive Meta-Analysis so you may need to edit colnames or code. -
@KirkegaardEmil@JamesPsychol I'd be willing to help you if you need, as best I can (in the middle of a move right now) -
@JoeHilgard@JamesPsychol I never got around to trying this. Are you done moving now? :) -
@KirkegaardEmil@JamesPsychol I am, but still up to my neck in alligators. Try me Jan 17th.
End of conversation
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.