Skip to main content

View Diary: Not So Fast Summers (270 comments)

Comment Preferences

  •  Junk science strikes again. (none)
    She correlated unrelated studies to "prove" her point there was no difference. Considering that each study would have a different sample number, different methology and aim, it's hard to see what the value of such an exercise could be.

    Someone from the Blank Slate crew wanted a soundbite for the media, and got one.

    •  often used (none)
      meta-anaylsis, as it is refered to, is often used in studies where it might be too difficult or too expensive to repeat the work. statstical techniques, which are probably too difficult for me to do, are used to account for the differences in study conditions.
      •  Doesn't help much. (none)

        You still have to select studies and assign a weight to the results of each one.  This results in a conclusion that's almost completely arbitrary, based on the whims of the one conducting it. There are several reasonable criterion for weighing each one, and you can pick and chose between which one you want until you produce the final result you desire.

        So long as you picked what appears to be a "reasonable" criterion, you can claim that you weren't aiming for the specific result.

        More "consensus" science.  Beautiful.

Subscribe or Donate to support Daily Kos.

Click here for the mobile view of the site