As a Publisher of an Interactive web site which does polls, it came as a shock to me that the John Kennedy for US Senate would be promoting an online interactive poll that shows the Treasurer is leading US Senator Mary Landrieu 47% to 41%.

All prior polls showed Landrieu ahead although the differences were within the margin of error.   I am also aware that the only poll that matters is “election day” and that polls have been wrong on numerous occasions.  Obviously, they have also been correct on numerous occasions.

Political polls are a snapshot in time.  Based upon different polls in a matter of weeks, Bobby Jindal’s “favorables” see sawed substantially as he went through the legislative pay raise crises and the veto session issue.  There were extreme highs and then some substantial lows (for Jindal) with a return to the highs as the public supported his veto decision. 

Polls are very important from a fundraising standpoint and for internal reasons to help candidates know the political waterfront.

Obviously, based upon the polls, the Landrieu campaign is in a fierce fight.  The immediate question is, however, is she suddenly six points below John Kennedy and if so, what is the significance?

The Landrieu campaign has forwarded a statement saying  “American Association For Public Opinion Research Condemns "Self-Selected" Surveys: The American Association For Public Opinion Research (AAPOR) believes that any publicizing or promotion of self-selected "polls" which report only the people who call in or write-in, "not only damages legitimate market and survey research, but can be very misleading when used to influence public policy or simply to disseminate information about the general public." The organization lists the practice under "Survey Practices the AAOPR Condemns" on its website. (AAOPR, "Survey Practices that AAOPR Condemns," Accessed 7/31/08).

I contacted the Kennedy campaign for more information about the poll and I am waiting for their response. 

I also contacted the Zogby polling service about the accuracy of the poll and received this reply:

I am the director of communications at Zogby International, and you can quote me as spokesman or whatever title works best for you. Please feel free to use this comment below on our interactive polling practices and on the race in LA:

We at Zogby are very confident of the accuracy of our interactive online polling methodology, particularly as it has to do with U.S. Senate races - here is why: in the 2006 cycle, we polled 18 Senate races around the country, and ended up nailing 17 of them, even though several of them were very, very close contests. The 18th race - the one in Missouri between incumbent Republican Jim Talent and Democratic challenger Claire McCaskill - we had Talent winning by the slimmest of margins, while McCaskill was actually able to pull off a narrow upset. Even so, our polling in that race was within the margin of error. Since then, our interactive polling processes have gotten even better.

Our interactive polling technology has been in development for a decade now, and we are an acknowledged leader in the industry in online polling. We pull a random sampling of respondents from a pool of several hundred thousand people nationwide to take part in our surveys. Respondents never know when they will be polled, and they never know what the subject of the survey might be. They are also NOT compensated for participating in Zogby polls, a key feature that means we do not have "professional" poll takers - we have average U.S. citizens. It is important to remember that now, in 2008, about 95 percent of likely voters nationwide have regular contact with the Internet, so this polling process is particularly effective in measuring political races. Respondents are directed to our Zogby secure servers, and can take the survey only once, so it's not like these polls you see on some websites. Zogby maintains datapoints on each respondent to guarantee accuracy over time, and we also follow up our interactive online polling with a random verification process using our own telephone call center to make sure our poll takers are who they say they are. No one else in the industry does this, but it is another reason we are so accurate.

All this said, it is understandable that the Landrieu campaign would discount our most recent poll showing her trailing her challenger. That is part of the campaign game, and we have certainly seen it before and are not surprised by it. But the poll reveals that Democratic candidates for Congress in general in Louisiana are not as popular as Republicans right now.. The Landrieu campaign has a simple choice - they can either deny the stark reality that the Zogby poll reveals, or they can campaign harder to try to convince Louisiana voters to change their minds. It doesn't matter to us either way - we don't have a dog in this fight - but we stand by our survey results and our past record of dramatic success.

Best,

Fritz Wenzel

Director of Communications

Zogby International

Utica, New York

However, it should be noted that the poll in question was taken during the period Jun 11 - Jun 30.  This means that the sample is dated by over one month.

Since that time, Landrieu has been running television ads statewide and has a war chest advantage.

So, the interactive poll raises some very interesting questions:  Is the American Association For Public Opinion Research   correct in condemning interactive polls or is Zogby right, in general?  Even if the poll is accurate for that time period, is it accurate for today, August 1.

It is absolutely wonderful that the Internet is becoming more accurate and firms such as Zogby are willing to invest their reputations on the results.  It is also important to recognizing the dating of the poll and how event can cause tremendous ups and downs such as what occurred with Governor Jindal.  As the public focuses more on the election, it is possible that we might see major swings over short periods of times for either or both candidates.

The Internet is and has without doubt changing the political landscape.  I know that Bayoubuzz and other online publications play a major role in the equation.  If the Kennedy campaign and the public wants to rely upon the Zogby technology versus the American Association For Public Opinion Research, that is their choice and their right.   Zogby stands by their results although one can question whether the numbers contemplate more recent events.  However, the main goal is accuracy in polls and in media.  And, obviously, the ultimate decision is what occurs on “election day” where in Louisiana, we have had quite a number of recent surprises, in part, to the growing use of the Internet.

Advertise on Bayoubuzz.  Be seen by a great audience
            Louisiana Calendar  Post your own events.  Over 1000 visitors/day
           Join BayoubuzzDance.com & input your own content

Sign up for our Email Newsletter

 
For Email Marketing you can trust

          Do you want to write for Bayoubuzz?  Email us.
          See the "hot" topics on our discussion boards and chime in