Oil Spill Pie Charts Suck 'Outliers'
Mark Blumenthal | June 14, 2010
Topics: Outliers Feature
John Harwood sees little impact from the oil spill on Obama approval (with an assist from Charles Franklin).
Joel Benenson frames climate change legislation as a political winner (via Smith)
Pete Brodnitz had a good night last week.
Mike Huckabee asks pollsters to include him on 2012 trial heat questions.
Marco Rubio is not worried about polls.
PPP invites you to vote on where they poll this weekend.
Lymari Morales counts the many ways Gallup posts updates of Obama job approval.
Bob Groves reports that the Census non-response follow-up is "about 93% complete.. somewhat ahead of schedule and certainly under-budget."
Junk charts says the BP oil spill brings out the worst in pie charts.
Comments
Lymari Morales says that what Gallup posts on its home page is the three-day rolling average. That's what I thought but how do you explain a three-day rolling average shifting as much as 5 or 6 points in a single day (which I have seen more than once in the past several weeks).
Doesn't that mean that there would have to be one-day swings in the 10 to 18 point range in order to net out to a 6 point change in the three day average? 18 point difference between one day and the next? Something is out of whack.
Posted on June 15, 2010 3:58 PM
Post a comment