There are so many examples of how the polling industry needs reform. They have a glut of polls from non-consequential times, no polling at crucial moments, and then misreport results at other times.

The California Field poll is an example:
http://www.baynews9.com/content/36/2008/ 1/30/321121.html

The poll was based on a telephone survey of 511 likely voters in the Democratic primary was conducted between Jan. 25 -Feb. 1.

The results show Clinton 36 obama 34.
Here's the lead in the newspaper:

""A startling surge of support for Barack Obama has catapulted the Illinois senator into a virtual tie with Hillary Rodham Clinton in California's Democratic presidential primary, a Field Poll released Saturday shows."""

Here are my issues:

1. This poll began surveys 5 days before Edwards left the race
2. This poll began 6 days before a historic debate
3 This poll spans ONE FULL WEEK

4. This poll reports the full Obama South Carolina/Kennedy bounce at a time when it has faded
   
5. People are using this poll to corroborate Zogby's polls which themselves are unbelievable
   
6. It wasn't startling after the 2 weeks Obama had in Jan.

The end result is that it inspires Senator Obama's people to work harder. It gives the media their most desired "neck and neck" storyline.

The problem is: IT'S NOT NECK AND NECK from all available evidence.
I haven't seem any post debate polls both on the state or national level that haven't shown a Hillary surge.
We all know before the debate IT WAS NECK AND NECK and I SAID SO. Since then, to my delight the evidence from Rasmussen , Gallup, and a slew of state polls conducted after the debate show good news.
Of course Zogby could be right and the others wrong about the last 2 days. But on some blogs people are using the Field Poll as evidence that Zogby's not alone. Zogby tested 1/31 - 2/2 while the field tested 1/25 - 2/1.
The Field Poll is just advertising for Senator Obama. It is political spin.
That is one reason why I don't support campaign finance reform because it doesn't include the media. Plus I believe in freedom of speech. But if everyone else is going to be regulated, the media and the polling outfits should also.

Here is a "nonpartisan" poll which has a great reputation that is injecting itself into the debate by releasing data from a favorable Obama time into a period that is favorable for Clinton.
Look at the manipulation that can occur:
Since no one knows when they are "in the field" they can extend or shorten their work to "shape" the results. In this case, pro Obama would be older dates, and pro Clinton would be newer dates.
Had we known beforehand what they were going to do then they would be have to stay with that plan. Also, they don't have to report the results if they don't like them, or they can minimize their importance.

Either way, the papers that report this without the big caveats that came with Hillary's Florida win (remember no delegates) are contributing to the Obama campaign and should have to follow the same rules as everyone else.

I don't think this other polls (I think are bad) can change an election. It will probably help Senator Clinton in the end if it motivates her people even more.
If this is the worst the MSM does until Tuesday, I'll sign for that right now.

I have the credibility to write this diary as on the day of the debate, my diary was:
"dogfight for dem nom"
http://www.mydd.com/story/2008/1/31/1393 7/5316#commenttop

I can honestly say, I feel the race is over, and we should rally early around the ticket of
Clinton-Obama 2008.

(update) msnbc and mason-dixon have done a similar thing: They released a batch of state polls from 1/30 - feb1. The surveys start before the debate. http://www.newsobserver.com/politics/story/920176.html They all show clinton leading except Georgia. But it makes no sense to release this information, unless you let people know that the poll began before the biggest event of the campaign