Power, p-values, publication bias and statistical evidence

I created this video to show as part of a presentation I’ll be giving next week. Your comments welcome, either here or at Youtube.

This entry was posted on Tuesday, February 8th, 2011 at 8:39 pm and is filed under epistomology, feature, stats. You can follow any comments to this entry through the RSS 2.0 feed. Both comments and pings are currently closed.

4 comments

  1. Basil
    February 18, 2011 at 2:26 pm

    Looks like an informative cartoon. I showed it to my students and it held their interest for about 10 seconds. A problem that the characters did not address was over sampling. Large samples can produce low p-values.

  2. Matt Asher
    February 18, 2011 at 6:38 pm

    Hi Basil,

    Yikes! I suppose 10 seconds is better than nothing….

    FWIW the vid got a mild response during my presentation, a few chuckles, but otherwise not much. Maybe not as amusing as I thought it was. Oh well…

    Cheers!

    • Basil
      February 22, 2011 at 10:40 am

      If it makes you feel better, I watched the entire video. I thought it slightly entertaining and possibly a good way to have students ask questions about what a p-value really means.

  3. DCR
    March 27, 2011 at 2:27 am

    Really great video.