At his Big Question blog at the Star Tribune site, reporter Eric Black interviews Rob Daves of the Star Tribune Minnesota Poll. I interviewed Daves in 2000 and 2002 regarding the Minnesota Poll. Following the Coleman-Mondale Senate election in 2002, I wrote a column incorporating my interviews with Daves and commenting on the discrepancy between the final published Star Tribune Minnesota Poll and election results: "The trouble with the Star Trib poll." Daves explains the mysteriously appearing Republican voters that traditionally make it to the polls in defiance of the Star Tribune's final published pre-election polls as an empirically verifiable closing "Republican surge." For some reason, it is a "surge" that generally is not demonstrated by polls other than the Minnesota Poll. In the comments section of the Daves interview, reader Mark Liveringhouse outlines all of the
other polls over time for the 2004 race showing Bush with 47-48 percent of the vote versus the 41 percent at which the Minnesota Poll had him pegged. He then pulled the exit polling from the 2004 election that shows, contrary to Daves's asertion, that late deciding voters broke for Kerry:

Just to show you the how ridiculous it is for Rob Daves to claim that the reason for the problems in the Minnesota Poll is this mysterious REPUBLICAN surge that they have “empirical evidence” for here are some polling sequences for the state during the 2004 Presidential Election.

Remember, the actual results in the state were [Bush 47.61% Kerry 51.09%].

Here are the series of Minnesota Polls done by the Star Tribune:

3/28 Bush 38 Kerry 50
9/7 Bush 41 Kerry 50
10/9 Bush 43 Kerry 48
10/26 Bush 41 Kerry 49

Now here is the Mason Dixon/Pioneer Press polls:

5/25 Bush 41 Kerry 44
7/12 Bush 44 Kerry 45
9/11 Bush 46 Kerry 44
10/15 Bush 47 Kerry 45
10/27 Bush 48 Kerry 47

The SurveyUSA polls:

7/10 Bush 47 Kerry 47
7/31 Bush 45 Kerry 47
8/28 Bush 46 Kerry 47
9/26 Bush 45 Kerry 47
10/12 Bush 45 Kerry 47
10/29 Bush 48 Kerry 48

And last but most accurate, Scott Rassmussen’s polls:

3/23 Bush 44 Kerry 47
5/1 Bush 43 Kerry 48
6/1 Bush 41 Kerry 50
7/1 Bush 42 Kerry 49
8/1 Bush 44 Kerry 48
9/18 Bush 46 Kerry 46
10/8 Bush 45 Kerry 48
10/25 Bush 47 Kerry 48

All of the other polling affiliates seemed to have already captured the REPUBLICAN SURGE that Rob Daves reports. How did they do this before it happened?

Again, anyone who concludes or talks about the so called last minute surge is being ridiculous. The other polls show just the opposite, that the late undecided broke for John Kerry.

Further, anyone who bothered to look at the exit polling will show this is confirmed.

When did you decide to vote for?

Today Bush 49% Kerry 48%
Last 3 days Bush 37% Kerry 57%
Last week Bush 46% Kerry 49%
Last Month Bush 44$ Kerry 54%
Before that Bush 49% Kerry 51%

***

Of the people making their decision close to election day (and AFTER the Strib Poll) the majority of them broke for Kerry.

If I were the editor or publisher of the Star Tribune, I would be seriously concerned about, if not mightily embarrassed by, the quality of my product. If the Star Tribune's poll product were edible instead of legible, it would long ago have been recalled as dangerous to human health, or it would have killed off its customers. We can only hope that some day the Star Tribune cares as much about the quality of its news product as McDonald's does about the quality of its hamburgers.

Posted by Scott at 05:44 AM  |    |