PA Dem Primary-- Apr 16, Apr 9, Apr 2 trend: Shares by Region, Weightings, Undecideds, Defs (PPP)

[tiptoe]

                                                 April 16 / April 9 / April 2

Trend				+ Obama      + Obama	  + Obama     + Clinton	    + Clinton
         
             Pennsylvania       Northeast    Southeast   SouthCentral  SouthWest    WestCentral
Weight (Avg)     100               9 %          44%         10%          28%           9 %    
  
Clinton        42/46/43         57/64/62     33/35/37     48/60/53     41/43/35     52/50/45
Obama          45/43/45         28/30/26     53/53/53     39/31/30     46/42/50     37/37/39
Undecided      13/11/13         15/ 6/11     14/12/10     13/ 9/18     12/15/15     11/13/16
  
  
  
Northeast PA     defined as area codes 570 
    Scranton, Williamsport
Southeast PA     defined as area codes 215, 267, 484, 610 
    Philadelphia 
South Central PA defined as area codes 717 
    Harrisburg, Lancaster, York
Southwest PA     defined as area codes 412, 724 
    Pittsburgh, Bethel Park, Monroeville, New Castle
West Central PA  defined as area codes 814 
    State College, Erie, Altoona

(major cities)

Data source: Public Policy Polling
April 16, April 9 / April 2

April 16:

PPP surveyed 1095 likely Democratic primary voters on April 14th and 15th. The survey’s margin of error is +/- 3.0%. Other factors, such as refusal to be interviewed and weighting, may introduce additional error that is more difficult to quantify.

April 9:

PPP surveyed 1124 likely Democratic primary voters on April 7th and 8th. The survey’s margin of error is +/- 2.9%. Other factors, such as refusal to be interviewed and weighting, may introduce additional error that is more difficult to quantify.

April 2:

PPP surveyed 1224 likely Democratic primary voters on March 31st and April 1st. The survey’s margin of errors is +/- 2.8%. Other factors, such as refusal to be interviewed and weighting, may introduce additional error that is more difficult to quantify.

Public Policy Polling had the most accurate numbers of any company in the country for the Democratic primaries in South Carolina and Wisconsin, as well as the closest numbers for any organization that polled the contests in both Texas and Ohio.

[Yael]

1. I am in amazement as to the numbers of undecideds growing in the east

From 6 to 15 in the Northeast?

Wow.

[tiptoe]

4. the movement "away from Clinton" towards Undecideds/Obama was also mentioned here:

http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=132x5531529

The MOST RECENT PA polls (with all polling conducted on the 12th, 13th, 14th, and 15th) reveal an amazing shift. Each consecutive day, more PA voters have left Clinton, more have become undecided, and more have gone to Obama. This snapshot demonstrates what a shifting electorate should look like: the collective change of voters from pro-Hillary to not-Hillary/not-Obama to pro-Obama.

Disclaimer: conclusions from a meta-look like this across pollsters/methodologies is far from a sure thing.

With that caveat, here's how it looks (and these are the ONLY three PA polls we currently have for polling wholly within the 12-15th that I can locate):

12th-14: SUSA has it Clinton 54, Obama 40, Undecided 3
14th: Rasmussen has it Clinton 50 (down 4), Obama 41 (up 1), Undecided 9 (up 6)
14-15th: PPD has it Clinton 42 (down 8 more), Obama 45 (up 4 more), Undecided 13 (up another 4)

This four-day snapshot shows Clinton support moving to Undecided and to Obama. If Obama can capture only 38% of this growing batch of Undecideds as reported by PPP, he'll have a PA win wrapped.

http://www.realclearpolitics.com/epolls/2008/president/pa/pennsylvania_democratic_primary-240.html

[Yael]

6. Looks that way!

:hi:

[tekisui]

2. Interesting, big differences across the state.

[Bucky]

3. While I'd like to believe these numbers...

This is a fucking robo-call survey! These always skew to younger voters, who are less apt to hang up on impersonal recorded messages. Even with 1095 survey respondents, there's no way that a polling supervisor can catch any demographic slants in the data and correct the sampling accordingly. I want to see an Obama upset in PA, but too many other polls by reputable firms are saying this survey is an outlier.

[tiptoe]

5. PPP claims (and I don't know if true or not) "the most accurate numbers...for Dem primaries in..."

"Public Policy Polling had the most accurate numbers of any company in the country for
the Democratic primaries in South Carolina and Wisconsin, as well as the closest
numbers for any organization that polled the contests in both Texas and Ohio."

http://www.publicpolicypolling.com/pdf/PPP_Penn_Release_040208_1.pdf

[tiptoe]

7. There's no guarantee a "reputable" sponsored poll isn't being conducted other than as client wants

or in a pollster's continuing business self-interest.

SUSA PA-polling is "sponsored" by 4 television stations (which have means to influence Public Opinion per whomever might be controlling their productions).

[tiptoe]

8. Kick for substance, and where to look for vote suppression nt

[tiptoe]

9. Latest Zogby PA Poll also has it "too close to call"

Newsmax/Zogby Poll: Deadlocked in Pennsylvania!
http://zogby.com/news/ReadNews.dbm?ID=1481

UTICA, New York—With just five days left before Democratic primary voters go to polls to decide who they want to be their presidential nominee, Hillary Clinton of New York and Barack Obama of Illinois are locked in a battle that is too close to call, the latest Newsmax/Zogby telephone poll shows.

The survey, which was conducted April 15-16, 2008 and came out of the field midway through Wednesday's contentious debate between the two candidates in Philadelphia, shows Clinton at 45% and Obama at 44%, with 12% either wanting someone else or left undecided.
...