The nonpuzzle of the close opinion polls
Categories:
6 Comments
Leave a comment
Subscribe to Entry
Subscribe!
Andrew Gelman's Books
For more info on our research:
Blogroll
Sister Blogs:
Statistics:
- Chance News
- The Endeavour
- Christian Robert
- Revolution Computing
- Yu-Sung Su's Blog
- The Numbers Guy
- Messy Matters
Visualization:
Cognitive and Behavioral Science:
- Decision Science News
- British Psychological Society Research Digest
- Seth Roberts [experimental psychology]
- Criteria's Employee Testing Blog
Social and Political Science:
- Monthly Labor Review Precis
- Marginal Revolution [economics]
- Language Log
- Social Science Statistics
- The Baby Name Wizard
- Vox EU
Machine Learning:
Cultural:
Pages
Research supported by the National Science Foundation
National Institutes of Health
Yahoo Research
Search
Recent Comments
- Hal Varian: Here's a related piece about which party is better for read more
- Andrew: Steve, The Abramowitz et al. article you link to is read more
- Brian: I noticed that the 2 exceptions (where the incumbent party read more
- Steve: Here's an interesting related link. http://www.centerforpolitics.org/crystalball/article.php?id=AIA2008072401 I need to read read more
- Andrew: Derek, Yes, I rank ordered them in the display because read more
- derek: Is that income growth figure a *rank*? read more
Recent Entries
- More healthy, happy conservatives
- What statistics courses should a methods-interested poli sci Ph.D. student take
- Updated MICE package for multiple imputations
- Race, region, and vote choice in the 2008 election: implications for the future of the Voting Rights Act
- More on the problem with email
- A Quinn Martin production
- Heads I win, tails I don't lose
- OmniGraphSketcher functionality for Windows?
- Hollywood-style lectures? Maybe not.
- Bayesian jobs at SAS
- The divergence of the intellectual upper class
- Inbox zero
- Big juicy datasets
- A horror story involving the correction of a published scientific article
- A question about poker
- Seniors Skeptical on Health Care Spending
- Linf: An L-infi nity Classifi er
- Election 2008: What really happened
- Baseball stats: innovation, randomness, and other issues
- A cool bit of experimental research on group threat and voter mobilization
Categories
- Administrative (15)
- Art (36)
- Bayesian Statistics (240)
- Causal Inference (93)
- Decision Theory (151)
- Economics (200)
- Literature (135)
- Miscellaneous Science (120)
- Miscellaneous Statistics (424)
- Multilevel Modeling (171)
- Political Science (640)
- Public Health (134)
- Sociology (248)
- Sports (27)
- Statistical computing (123)
- Statistical graphics (153)
- Teaching (151)
Is that income growth figure a *rank*?
Derek,
Yes, I rank ordered them in the display because I thought that would be the most readable way to present the data in the book. Follow the links to see some alternative versions of the graph.
Here's an interesting related link.
http://www.centerforpolitics.org/crystalball/article.php?id=AIA2008072401
I need to read through both yours and this one in more detail before I comment, but I thought it was relevant to this thread. What's your take on Professor Sabato's Center's article Professor Gelman?
I noticed that the 2 exceptions (where the incumbent party lost despite good growth) were 1952 and 1968--war years. 2008 is also arguably a war year, and I think that explains part of the 'puzzle' of the close polls.
Steve,
The Abramowitz et al. article you link to is fine, but it's a bit too deterministic for my taste. It's probably a bad sign that the website it's on is labeled "Larry J. Sabato's Crystal Ball." Larry Sabato knows a lot, but nobody has a crystal ball.
Obama is forecasted to win, but there's always uncertainty. They cite lots of recent polls but that's not always a perfect guide to what will happen in November.
Here's a related piece about which party is better for the stock market. Take a guess before you click...
Which Party in the White House Means Good Times for Investors? [NYTimes]