- 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?
- You know, let's not even worry about the City Commissioners office messing up voter registration processing
- Bold ideas to fix the budget
- Mayor Nutter's Town Hall Meeting Schedule
- City Releases Library Information to City Council
- Size of Philadelphia government?
More on the poll
Submitted by Dan U-A on Wed, 01/31/2007 - 2:13pm.
Not to get too dorky, but...
The Keystone poll is interesting, but, after looking at it, I think you have to take it with a grain of salt.
Chris at MyDD caught this, but, here is the sampling of those polled, by race:
54% White
39% Black or African American
2% Asian
1% American Indian, Alaska Native
4% Other
It appears that Latinos (6%) were potentially lumped in with whites, which would mitigate some of the difference. But either way the poll seems to have a pretty big skew in its ratio of White to African-American voters. (Most people predict a roughly even split of votes between white and African-American voters.) Obviously, not everything is racial, but, it would seem to alter the results. (And, I am unclear if the those numbers above are for the people identified as Dems or just everybody.)
Also, the stuff about which African-American candidate gets the most white votes, and so on? Really don't pay attention to those numbers, except in a very general sense. The sample size there (registered, white Democrats) is about 130 voters. That would put the margin of error into totally unacceptable ranges.
Also noted in the poll- public financing of elections- at least when asked about as stand alone measure- is not popular. I believe that when contextualized, it certainly will be. But, those who believe in it have some work to do.
The poll is good for some trends- Knox is on his way up and violence is still overwhelmingly the number one concern of voters. But as we get into smaller subsets, it really does not hold up statistically.
Bookmark/Search this post with:
Dan: Let me clarify the poll a bit
In our January poll we're more interested in a birds eye view of things than making a careful measure of the horse race. This far out, there is a high Undecided factor and the contest is more of a name-recognition game than anything else. Thus this poll (on the non-horse race stuff) included Republicans and those not yet registered to vote. That accounts for the racial breakdown being closer to the city as a whole rather than just Democrats. We polled 436 Philadelphians, of whom 252 were registered Dems, which accounts for the higher MOE on that subset.
As we move along in time we will poll only registered Democratic voters to get a closer look at the primary race.
Although there are more African-American Democrats, the turnout usually splits close to even or slightly more whites.
In 1999 our first 2 polls (in which registered Dems were a smaller subset just like this one) showed the only candidate with TV ads, Marty Weinberg, rising from 3% to 23%, second to John Street. Subsequent polls with double the sample size confirmed this move. As we all know, that's exactly how it finished, although there were some exciting twists and turns along the way.
Thanks for the info.
Thanks for the info. Basically, good for trends (Knox up, violence swamps everything).
As an aside, if you want to make your poll cutting edge: instead of asking people whether they are liberal, ask them whether they are progressive.
I know the benefit of liberal is that you have long trend lines- but- get on the cutting edge (I think Ray, for example, doesn't like to self identify that way), and either ask progressive, or liberal/progressive. That would make it pretty interesting. Nationally, people are trying to get pollsters to do this- you could provide some initial evidence of whether it actually makes any difference.
good start, more details
I was glad to see a poll that actually asked about the casinos proposed for Philadelphia (it is the first poll that I know about since the sites were imposed on us 13 months ago). I was also very encouraged by the results. In the future it would be good to see detailed questions that provide for a more thoughtful policy debate about casinos, given the complexity of the issue(s), for example:
-should casinos be located in residential neighborhoods(within 1,500 feet of a home, school, park or religious institution)?
-would you support casinos in Philadelphia if your tax money was being spent on increased police presence and other public services that may result from locating casinos in Philadelphia?
-would you support casinos in Philadelphia if it led to a loss in jobs for city residents?
-should Pennsylvanians have been given the opportunity to vote on slots and other gaming before their introduction in our state?
-would you be more likely to support a Mayoral candidate if they were opposed to casinos in residential neighborhoods in Philadelphia?
I am not a pollster but wanted to provide some thoughts for future polling on this issue. I am sure there are many other questions that could be developed to help inform an even more vibrant civic discussion about the issues that arise from proposing casinos in our city. There could be other questions that more directly relate to revenue and taxes, public health, property values and issues of crime, traffic, environment, planning and zoning, riverfront issues and so much more. Thanks, Jethro Heiko, Casino Free Philadelphia