Debate Bumps, 1988-2008: The Chart

by John Sides on October 2, 2012 · 1 comment

in Campaigns and elections

Courtesy of Thomas Holbrook.  More discussion is in his post.  These numbers show why, based on recent history, it would be unusual for the debates alone to close the gap between Obama and Romney.

{ 1 comment }

Andrew Gelman October 2, 2012 at 2:22 pm

Just one thing. Please please please round all percentages to the nearest integer. Hundredths of a percentage point (or even tenths of a percentage point) are meaningless in surveys. Not just because of sampling error (although that would be enough of a reason to round to the nearest percentage point) but also because of whatever variation there is between polls, between days, between hours of the day, etc.

Comments on this entry are closed.

Previous post:

Next post: