Some notes on future improvements to the calculation. David Kline points out that to calculate the single-state probabilities, for small (<30) numbers of polls, a t-distribution is more appropriate than a normal distribution. This distribution has longer tails and will give less certainty in the estimates. Going in the converse direction, in retrospect some states were more certain than the one-week snapshot indicated, and outlier polls tended to introduce occasional inaccuracies. A more sophisticated averaging procedure is needed, one that uses more polls but gives more weight to recent ones.
Princeton Election Consortium
A first draft of electoral history
-
Poll Updates- US: Health Care, Fort Hood (Fox 11/17-18) November 20, 2009
- CA: 2010 Sen (Rasmussen 11/17) November 20, 2009
- US: 2012 Pres (PPP 11/13-15) November 20, 2009
- NY: 2010 Gov (Marist 11/12-17) November 20, 2009
- US: National Survey (Kos 11/16-19) November 20, 2009
History and snapshot
State-by-State Maps
Senate and House
Meta-Analysis Thread
- Spacetime rift delays update until noon
October 7, 2008 | 11:42 amAndrew says there was “a routing problem this morning between the University and quite a few outside websites — a split in the internet, if you will.” For the next EV estimator update, check back at noon…
- VP debate: big audience but race unchanged
October 5, 2008 | 3:00 pmBiden v. Palin drew 70 million viewers, one-third more than Obama v. McCain. VP debates usually don’t matter, but I wondered… Read more »
- For locals: a talk on the Meta-Analysis
October 2, 2008 | 9:00 amTomorrow (Friday) at noon, I’ll give a talk in Kosuke Imai and Dustin Tingley’s political methodology seminar. If you’re nearby, you’re welcome to attend. Of course, the math will all be included. The blurb is here (PDF).
Update: Here’s another local event, happening next Tuesday: a panel on the reliability of state polls featuring both pollsters and academics, including Andrew Gelman, author of Red State, Blue State, Rich State, Poor State. I hear the panel will be webcast. If so I’ll post a link.
-
Archive for Meta-Analysis Thread »
- Spacetime rift delays update until noon
No Comments so far ↓
Like gas stations in rural Texas after 10 pm, comments are closed.