Rasmussen

[Alcon]

I was just going through the polling for Gubernatorial and Senatorial races to update my predictions, and looking at Rasmussen really worries me.  In a lot of states, they are the odd man out in predicting Democratic victories.  In fact, in most instances, his one-day polls are finding results that are distressingly conflicted with other polls, mostly in the Dem's favour.

While Rasmussen does have a history of slight Republican lean, I've never seen Scotty move around this much.  This one-day polling is really bothering me and, despite good 2004 results, I'm having trouble trusting Rasmussen - a shame, since they are so prolific.

Any comments on this?

[○∙◄☻¥tπ[╪AV┼cVê└]

Rasmussen totally blew the 2000 election.

[Alcon]

Rasmussen totally blew the 2000 election.

I did trust that they "fixed" things before 2000 and 2004.  But my question is, what broke?

[Sam Spade]

I was just going through the polling for Gubernatorial and Senatorial races to update my predictions, and looking at Rasmussen really worries me.  In a lot of states, they are the odd man out in predicting Democratic victories.  In fact, in most instances, his one-day polls are finding results that are distressingly conflicted with other polls, mostly in the Dem's favour.

While Rasmussen does have a history of slight Republican lean, I've never seen Scotty move around this much.  This one-day polling is really bothering me and, despite good 2004 results, I'm having trouble trusting Rasmussen - a shame, since they are so prolific.

Any comments on this?

I reposted some old Vorlon commentary about his discussion with Rasmussen per 2004 state and national polls.  Hope you noticed it.

I'm basically going to repost some of my own research in combination with his knowledge to decipher a conclusion.

The big difference between 2000 and 2004 is that he stabilized or weighted his national turnout model towards expected turnout numbers (sort of like how Zogby does it, but without as many useless weights as Zogby) and nailed it. 

In 2000, he admitted that his turnout model underestimated Dem turnout.  He wasn't the only one, as Dem turnout was underestimated in that race till the end.  Same thing happened with the Battleground poll, which had been the most accurate in the 1992/1996 election period, got the turnout model wrong in 2000, but refigured and got it right in 2004.

But anyway, this has nothing to do with his state polls, which operate much differently.

Remember, Rasmussen conducts his polls very similar to SUSA, using "cluster samples" to generate his polling sample.  I've tended to find that this method can jump around like crazy early in the campaign, but is usually fairly accurate in the end.  Similarly, I don't trust his one-day polls right now, unless they show the same thing consistently.  Same can be said for SUSA.  I also think that the Rasmussen method works better when conducted as a tracking poll, rather than one one-day poll per month.

Mason-Dixon is much more accurate over the whole of the campaign because their method is simply much more solid and reliable.

Actually in 2000, Rasmussen's state polls performed very well as I recall, getting 15 out of 17 states correct and most within MOE.  This was, of course, overshadowed by the national problems.  He did well in 2004 also, but we all know that.

So, in the end, that's what I look at.  My predictions are giving some credence to polls right now, but I'm also keeping in mind the weakness of polls this early in the campaign and concerns with regards to wild jumps.  My predictions haven't changed as wildly either because of this.

[Alcon]

Hmm, very interesting stuff.  What is it that Mason-Dixon does that gives such perpetually stable results?

[ATFFL]

Rass is also doing what amounts to one day snap polls using a method designed for a three day tracking poll.  In summer.  Not the best time for any pollster and he compounds it by not playing to his polls strength.

[CARLHAYDEN]

I was just going through the polling for Gubernatorial and Senatorial races to update my predictions, and looking at Rasmussen really worries me.  In a lot of states, they are the odd man out in predicting Democratic victories.  In fact, in most instances, his one-day polls are finding results that are distressingly conflicted with other polls, mostly in the Dem's favour.

While Rasmussen does have a history of slight Republican lean, I've never seen Scotty move around this much.  This one-day polling is really bothering me and, despite good 2004 results, I'm having trouble trusting Rasmussen - a shame, since they are so prolific.

Any comments on this?

There is a reason (and a story) behind this.

In 2000 the Democrats did an incredibly good job in the voter turnout effort while there was NO nationally coordinated Republican turnout effort.  The result was that in the popular vote Gore did significantly better than he would have done in a normal turnout election.

This messed up Scott's preduction (at that time Rasmussen was know as 'Portrait of America').

Well, Rasmussen changed his weighting to include more Democrats (particularly Black Democrats) and fewer Republicans than in his prior methodolgy.

Well, this din't work too badly in 2004 because Soros et al spent millions on turnout for the Democrats (but were countered by a Republican turnout effort, especially in key states like Iowa and Ohio).

This year, the Democrats do NOT have a party directed/funded turnout effort (they had a good one in 1998), Soros and Co. are not paying for such an effort this year, so a normal distribution seems likely (which means a higher Republican and lower Democrat turnout ratio than in 2004).

So, this explains why Scott's polls seem skewed in favor of the Democrats.

[Verily]

I was just going through the polling for Gubernatorial and Senatorial races to update my predictions, and looking at Rasmussen really worries me.  In a lot of states, they are the odd man out in predicting Democratic victories.  In fact, in most instances, his one-day polls are finding results that are distressingly conflicted with other polls, mostly in the Dem's favour.

While Rasmussen does have a history of slight Republican lean, I've never seen Scotty move around this much.  This one-day polling is really bothering me and, despite good 2004 results, I'm having trouble trusting Rasmussen - a shame, since they are so prolific.

Any comments on this?

There is a reason (and a story) behind this.

In 2000 the Democrats did an incredibly good job in the voter turnout effort while there was NO nationally coordinated Republican turnout effort.  The result was that in the popular vote Gore did significantly better than he would have done in a normal turnout election.

This messed up Scott's preduction (at that time Rasmussen was know as 'Portrait of America').

Well, Rasmussen changed his weighting to include more Democrats (particularly Black Democrats) and fewer Republicans than in his prior methodolgy.

Well, this din't work too badly in 2004 because Soros et al spent millions on turnout for the Democrats (but were countered by a Republican turnout effort, especially in key states like Iowa and Ohio).

This year, the Democrats do NOT have a party directed/funded turnout effort (they had a good one in 1998), Soros and Co. are not paying for such an effort this year, so a normal distribution seems likely (which means a higher Republican and lower Democrat turnout ratio than in 2004).

So, this explains why Scott's polls seem skewed in favor of the Democrats.

This is all bullocks, and Rasmussen ended up spot-on everywhere.

[CARLHAYDEN]

I was just going through the polling for Gubernatorial and Senatorial races to update my predictions, and looking at Rasmussen really worries me.  In a lot of states, they are the odd man out in predicting Democratic victories.  In fact, in most instances, his one-day polls are finding results that are distressingly conflicted with other polls, mostly in the Dem's favour.

While Rasmussen does have a history of slight Republican lean, I've never seen Scotty move around this much.  This one-day polling is really bothering me and, despite good 2004 results, I'm having trouble trusting Rasmussen - a shame, since they are so prolific.

Any comments on this?

There is a reason (and a story) behind this.

In 2000 the Democrats did an incredibly good job in the voter turnout effort while there was NO nationally coordinated Republican turnout effort.  The result was that in the popular vote Gore did significantly better than he would have done in a normal turnout election.

This messed up Scott's preduction (at that time Rasmussen was know as 'Portrait of America').

Well, Rasmussen changed his weighting to include more Democrats (particularly Black Democrats) and fewer Republicans than in his prior methodolgy.

Well, this din't work too badly in 2004 because Soros et al spent millions on turnout for the Democrats (but were countered by a Republican turnout effort, especially in key states like Iowa and Ohio).

This year, the Democrats do NOT have a party directed/funded turnout effort (they had a good one in 1998), Soros and Co. are not paying for such an effort this year, so a normal distribution seems likely (which means a higher Republican and lower Democrat turnout ratio than in 2004).

So, this explains why Scott's polls seem skewed in favor of the Democrats.

This is all bullocks, and Rasmussen ended up spot-on everywhere.

Lets stop and take a look at the facts.

First, Rasmussen did a good job, but had a slight aggregate advantage to the Democrat candidate relative to the Republican candidate:

Party                    Actual %                    Rasmussen Projection     Difference

Republican            50.7                              50.2                              -0.5
Democrat              48.2                              48.2                             +0.3

Second, Rasmussen did NOT project the winner in all states.  They Scotty labeled eleven states as "tossups." 

In Florida, Bush carried the state by 5.01% and in Hawaii, Kerry carried the state by 8.74%.  Not what most people would call a "tossup."

A study published by Slate magazine found that Mason-Dixon was the most accurate of all the major polling organizations.

[Gabu]

Hmm, very interesting stuff.  What is it that Mason-Dixon does that gives such perpetually stable results?

This post is from a while ago, but this seems as good a place as any to ask it.  It seems to be pretty well unanimous that Mason-Dixon is the best and most trustworthy pollster around, especially in Florida.  Whenever a poll from Florida comes out and it's not by Mason-Dixon, everyone always says to trash it.

My question is in two parts: if Mason-Dixon is so good, 1) what do they do, and 2) why doesn't everyone just copy it?  On the face of it, it seems kind of silly for people to all be doing their own thing if one polling agency consistently does better than the others.

[Alcon]

I think Rasmussen may have done better than M-D in 2006, amusingly enough.

[CARLHAYDEN]

Hmm, very interesting stuff.  What is it that Mason-Dixon does that gives such perpetually stable results?

This post is from a while ago, but this seems as good a place as any to ask it.  It seems to be pretty well unanimous that Mason-Dixon is the best and most trustworthy pollster around, especially in Florida.  Whenever a poll from Florida comes out and it's not by Mason-Dixon, everyone always says to trash it.

My question is in two parts: if Mason-Dixon is so good, 1) what do they do, and 2) why doesn't everyone just copy it?  On the face of it, it seems kind of silly for people to all be doing their own thing if one polling agency consistently does better than the others.

First, there are unique factors in just about every state.

In my state, random surveys in July are next to worthless.

In several states, if you survey dates include the first day of deer season, you will also get very skewed results.

Second, it costs money and takes time to compensate for these factors.  Most survey research firms don't even try.  Mason-Dixon does, to its credit.

 

[Gustaf]

Hmm, very interesting stuff.  What is it that Mason-Dixon does that gives such perpetually stable results?

This post is from a while ago, but this seems as good a place as any to ask it.  It seems to be pretty well unanimous that Mason-Dixon is the best and most trustworthy pollster around, especially in Florida.  Whenever a poll from Florida comes out and it's not by Mason-Dixon, everyone always says to trash it.

My question is in two parts: if Mason-Dixon is so good, 1) what do they do, and 2) why doesn't everyone just copy it?  On the face of it, it seems kind of silly for people to all be doing their own thing if one polling agency consistently does better than the others.

You don't seem to aware of a little word called "business secrets". Well, two words, but you know...basically, people don't necessarily know what Mason-Dixon does. Secondly, as Carl pointed out, a good poll costs money and not everyone wants to pay. As to bad pollsters surviving, it's a tough business and it's hard to be good at it, obviously.

But every media outlet wants to have it's own poll, so that they can get exclusive news. Whether the poll is actually good is less important, since most consumers have no idea anyway. So if media outlet X has taken good pollster A, media outlet Y has to settle for semi-good pollster B and so on, and evnetually someone has to take what's left over.