- 'An End to the Southern Strategy, But No Post-Racial America' says David Love
- "A Question of Place": An essay on the power of community
- Just Equally Speaking….
- Eagles owe Philadelphia the 8 million it needs to keep libraries open
- who would like to see Verizon offer cable TV in Phila?
- Council Committee Passed the Freeze
- Carol Campbell Passes Away
- My first trip to the public library
- Fight digital exclusion
- What if half of Philadelphia didn't have roads?
Looking deeper at the latest Daily News/ Keystone Poll
Submitted by Dan U-A on Thu, 08/14/2008 - 1:29pm.
link:
See the recent Daily News article about the Presidential race in Pa tightening? Check out Adam B.'s analysis of it at Daily Kos. His basic take:
It's utter bullshit. Barack Obama -- based on this poll's own raw data -- actually leads John McCain among likely voters by a ten point margin. Let me explain.
Barring a miracle, including voters going to sleep, and volunteers deciding not to help, Obama is going to kick McCain's ass in PA.
»
- Login or register to post comments
- Email this Quick Hits
- Printer-friendly version
Undercounting young voters
I have always suspected that voters under thirty, a big Obama demographic is undercounted in polls. This is due to the fact that many do not have landlines and rely solely on cell phones. People who use their cell phones as their primary phones are never polled.
Your suspicion is correct.
Your suspicion is correct. Young, new and minority voters are under-represented in polls.
New voters are the most
New voters are the most likely to be excluded if the poll is on likely, rather than registered voters, because one of the questions is usually have you voted before. They would not be undercounted particularly in a registered voter poll.
In terms of race, most pollsters seem to roughly adjust their proportions of respondents based on the historical turnout of each race. That said, this year may be tougher than any to poll, with increases expected in turnout of young people and African-Americans, above and beyond their normal levels of turnout.
Pew looked at this...
Pew studied this, and at least in their poll, didn't find much of an effect:
Three points is pretty big.
And just so I'm clear: my adjustments to the Keystone Poll don't require any speculation; it's all about correcting for actual voter turnout. Terry Madonna keeps sampling the wrong voters.