« Probably Truncated | Main | 9-8-7 »

September 21, 2012


Feed You can follow this conversation by subscribing to the comment feed for this post.


You're just so stinkin' smart that the most difficult thing for you must be how you remain so humble as you respond to simple folk like me. I am an admirer.

Fancypants Journal reviewers need a little lesson in basic statistical analysis and you are just the person for the job!


The scientific-type talk is interesting if over my head. But chocolate-applesauce muffins? Now you're speaking my language!

I hope you like them!

Small samples are certainly more sensitive to type I errors. I don't know about the details of your experiment or analysis, but the story I remember from grad school involved the unscrupulous researcher saying "Stop taking data! The effect might go away!"

On the other hand, I think that absolutely everybody gets tripped up by statistics every now and again.

There's an intentional false-positive result awarded this year's Ig Nobel Prize for Neuroscience (statistically significant brain activity observed using fMRI on a dead fish). The poster that received the citation is worth a read. I liked that they gave the fish a task.

The comments to this entry are closed.