Guess what? ABC poll oversampled Repukes!

[lancdem]

I don't have a link, but sonicx mentioned this in another thread. They sampled 6 percent more Repubs than Dems, and in the last election likely voters were Dem 39, Repub 35. That's a 10-point swing. They also polled too many Dems after the DNC (+8), so I'm certain the numbers will settle down. Even the number of Dems voting for Bush (12 percent) isn't as bad as some of the other post-RNC polls and will improve.

[tedoll78]

1. My suspicions are confirmed.

If we were to re-weight the polling, it would seem that ABC's poll results would be more aligned with the results from Rasmussen and Zogby. I'm soooo surprised (/sarcasm) :)

[wurzel]

2. Didn't stop ABC using them tonight. As for Stephanoupolis! Total whore!

[Nicholas_J]

3. It is amusing to note that non-media created polls show different results

Presidential Tracking Poll: Bush-Kerry

Updated Daily by Noon Eastern Election 2004

Presidential Ballot

Bush 47.5%
Kerry 46.8%
Other 2.0%
Not Sure 3.6%
RasmussenReports.com



Thursday September 09, 2004--The Rasmussen Reports Presidential Tracking Poll shows President George W. Bush with 48% of the vote and Senator John Kerry with 47%. The Tracking Poll is updated daily by noon Eastern.

Senator Kerry has not been ahead in the Tracking Poll since August 23. President Bush leads 86% to 11% among Republicans. Senator Kerry leads 80% to 17% among Democrats. As for those who are unaffiliated, 45% prefer Kerry and 43% Bush.

Our data is collected via nightly telephone interviews and reported on a three-day rolling average basis. Today's report includes interviews conducted on Monday, Tuesday and Wednesday. Tomorrow will be the first report based entirely upon interviews conducted after both the Republican Convention and the holiday weekend.



http://www.rasmussenreports.com/Presidential_Tracking_Poll.htm

It seems odd that polls can have such wide differnces in their results. So far, all of the polls done by the news media have been reportedly oversampling Republicans. For whatever inscrutable reasons.

[qanda]

4. Why do they keep pulling this crap?

When a Gallup poll sampled more Dems, it seemed that everyone had a collective fit. When are we going to start screaming about this?

[althecat]

5. On the washpost site there is a scheduled online discussion

I submitted a question on this very matter after looking through the various details they had posted and finding nothing.

That makes three polls where this has been done if this report proves correct.

And that is completely outrageous. It smacks of a conspiracy to distort public opinion.

[Nicholas_J]

8. Not a conspiracy

but creating melodrama. The media may well know that the race is far closer, and Bush jumping ahead after his convention is news, but reading the same old "the race is still too close to tell" is not news. The public needs red meat, some change in the political environment in order to get them to buy newpapers and watch the news.

Then later on when things seem to be going very different and it seems that Kerry is moving ahead, that is news, but a headline that still reads "the race is still to close to tell" is not.

The media should be kept out of this stuff. They create far too much problems.

[sonicx]

9. not conspiracy or drama, i think its...

stupidity. they need to learn a lesson from zogby. he was the most accurate in 2000.

[althecat]

10. I don't think its that simple.... sabotaging US democracy

Firstly the media generally like being right rather than newsy.. especially if the newsyness is at the expense of accuracy.

Secondly these polls have a large number of dangerous consequences.

1. They potentially put off Dem voters from voting
2. They cause panic in the Dem ranks concerning their candidates prospects
3. This in turn leads to revisions in strategy that may not be warranted.
4. They also play to the "back the winner" mentality where people decide to go with the herd and start to question their own doubts about *.

Finally.

5. In the event that this all gets found out it undermines public belief in polling and prepares the ground for the stealing of the election via electronic means. Q: "But the result is so far away from the polls" a: "Oh but polls are extremely unreliable remember three months back..." etc.

The fact that this has now happened three times with three of the most widely read and publicised polls in the US indicates that there is some concerted action going on here.

This is nothing short of YET another attempt to sabotage US democracy.

[Gloria]

11. Yes, these are polls created by the media, used to create news that

they then can use selectively...by ignoring legitimate, independent polls and creating "winners and losers" before the election.

Polling by news organizations should be eliminated, esp. if it is done solely to push their agenda...

The sampling methods are unprofessional and are unexplained to the public. It really sucks. I think there is much more polling being done this year, and I think it's deliberate....never a day without a new poll....

[althecat]

6. Here's the link and reference from sonicx...

Thanks sonicx..

http://abcnews.go.com/sections/politics/Vote2004/bush_c...

snip

The fact that underlying views have changed means they can change again. After the Democratic convention there were eight points more Democrats than Republicans among likely voters; today, there are six points more Republicans than Democrats. Party allegiance has been +3 or +4 Democratic in exit polls since 1988; Kerry would gain ground by moving the alignment back to its Election Day norm. But shifting party allegiance is not his only problem: He's losing 12 percent of Democrats to Bush, twice Bush's loss of Republicans to Kerry.

snip

codeword: more repubs sampled than dems. again.

[Chomskyite]

7. A continued dead heat would not sell papers or get ratings

But sudden movement on either side would be newsworthy (if true) and would attract readers and viewers.