Polls

[muon2]

I just read this morning that pollster Robert Teeter died. He's been one of Vorlon's top rated pollsters, and works with Hart for the WSJ poll. Will his death have an impact on that poll's performance?

[??????????]

I watch it every day and it appears that the undecideds are becoming decided. This is the second day in a row its been Bush 47 Kerry 44. Bush is about to break out IMHO and take off with it. Barring another scandal.

StatesRights, I'm sure you already noticed this, but today Kerry pulled back into the lead on Rasmussen.  So much for your theory.

My theory is not dead. Some one will break out in 2-3 more months.

[The Vorlon]

I just read this morning that pollster Robert Teeter died. He's been one of Vorlon's top rated pollsters, and works with Hart for the WSJ poll. Will his death have an impact on that poll's performance?

Yes, I noticed that also.  I worked with Mr. Teeter for 1 election cycle in 1994 and he was truly a wonderful man as well as a great pollster.

He had an absolute dedication to "getting it right" and making sure all the tiny details of how you did things were all done properly - it was annoying at the time, but in retrospect he was just about always correct in what he said.

Bob has taught so many people over his life that I don't think the accuracy of his firms polls will go down, he has many many proteges who will carry the torch.  

More than perhaps any other, he will be remembered as "teacher" as much as a pollster.  You can look at many of the top firms today on the GOP side (Ayres McHenry, POS etc) and they all have direct links back to Bob.

http://www.freep.com/news/latestnews/pm20244_20040614.htm

[CARLHAYDEN]

Don't place too much credance in polls between now and labor day.

Summer polls typically give Democrats a few extra points for a number of technical reasons.

Polls which use registered voters typically also have a Democrat bias.

Polls which base their 'likely voter screen' based on following the election (prior to labor day) also have a Democrat bias.

National surveys with fewer than 800 respondents should also be discounted.

I could go on, but, Vorlon has made many of the same points, and could probably give you even more.

I would in conclusion suggest that you need to take polls between now and labor day with more than a pinch of salt.

[phillies]

The process you are describing will skew in odd ways some polling data.  The correct process would be always to merge three four hundred person polls into a 1200 person poll, on an overlapping basis if it is a time series, even when the sets differed,  recalling that part of the time the data sets are supposed to be outside the margin of error of each other.  Rasmussen does this.  If the data sets are always within the MOE of each other, and enough sets have been taken, the data is very likely to be wrong.

If nobody minds me going back to somethign eariler, a poll of 500k peopel would pass the law of diminishing returns on MoE, but would really boost the confidence level in the poll.  A problem would be that it would be very hard to do a sister poll or two  to confirm the confidence.  It would be much more cost effective to sponsor 3 polls of 10,000 or even 5,000 voters and get a similar MoE and have checks to confirm your confidence in the poll.

A lot of polsl you see released with 1200 respondents are 3 polls of 400 merged.  The the 400 polls are taken and if they are all in MoE to each other, they are released as one poll.  If 2 are similar and one not, you can eithe rget anothe rsample of 400 or release it as 800 and lower your MoE and confidence a bit.

[ATFFL]

The process you are describing will skew in odd ways some polling data.  The correct process would be always to merge three four hundred person polls into a 1200 person poll, on an overlapping basis if it is a time series, even when the sets differed,  recalling that part of the time the data sets are supposed to be outside the margin of error of each other.  Rasmussen does this.  If the data sets are always within the MOE of each other, and enough sets have been taken, the data is very likely to be wrong.

If nobody minds me going back to somethign eariler, a poll of 500k peopel would pass the law of diminishing returns on MoE, but would really boost the confidence level in the poll.  A problem would be that it would be very hard to do a sister poll or two  to confirm the confidence.  It would be much more cost effective to sponsor 3 polls of 10,000 or even 5,000 voters and get a similar MoE and have checks to confirm your confidence in the poll.

A lot of polsl you see released with 1200 respondents are 3 polls of 400 merged.  The the 400 polls are taken and if they are all in MoE to each other, they are released as one poll.  If 2 are similar and one not, you can eithe rget anothe rsample of 400 or release it as 800 and lower your MoE and confidence a bit.

You are describing a tracking poll, I am describing an internal check method.

[khirkhib]

The Poll map is better but it still needs to be improved.  It states that the toss up, slight or lean status of each state is determined by the average of the last three polls.  Slight is between 2-5% and lean over 5%.  Oregon, Iowa and New Hampshire though are all labeled as toss-up states now though Kerry has an averaged lead of 2.6 in Oregon, 2.3 in Iowa, and 3.3 in NH (if you counted the Zogby poll).  

I feel that there has been more than 2 polls out of Nevada.  Looks like you didn't include any of Zogby's last three there because if you had Kerry would have the average lead of 1.6 and the state would be a toss up.  Even if you had just averaged Zogby's last three and put it against the Mason Dixon and  even if you just used Zogby's most recent which had Bush at a 2.5 lead the average would be 4.8 and make the state just slight Bush Lead. Which seems to make sence since the poll that is pushing the three, the Mason Dixon, gave bush an 11 point lead but was done back in March, doesn't seem to be a timely measure.

The other toss ups checked out.  It doesn't seem like all the polls that are being put out are being integrated into the equation either.  What are one of those good poll compilation sites?  It'd be nice to see.    ISome polls seem to get entered in really quicky and others not at all.  Just trying to keep it honest.

[ATFFL]

Zogby's Internet polls are a bad joke and including them would invalidate the entire point of the polls page.  I'm pretty sure the only Zogby polls included are his traditional polls.

SLight requires a 2-5% lead AND to have won all 3 of the last polls.  The sates you cite fail the last requirement.

[khirkhib]

Well in doesn't seem right that both Survey USA polls be used in Iowa when the only one that gives Bush a lead is the hypotetical, if Vilsack was president this is not the case and unlikely and with out that survey Kerry would have all three wins and a 4.3 averaged advantage.  You could count this survey if Kerry did pick Vilsack.

And in Florida Bush did not win all of the last three but he is still painted as lean there.  It still seems somewhat arbitrary which survey's are counted and which are not and the system should be standardized so it will be honest.  

In Dale's Electoral College Breakdown the blogger decides off the cuff if he will consider the survey or not.  He generally doesn't change the Surveys in Kerry's favor.  http://home.comcast.net/~gerrydal/

Where as www.electionprojection.com showed Kerry with a huge lead and winning Missouri and Arizona etc and than 2 weeks Bush took back a big lead and not jut becauseof the Regan funeral but because the blogger had changed the calculation that he had been using and when more in favor of state polls.  Just trying to keep things honest here.

Here is one site I remembered for pollig info.  www.realclearpolitics.com.  
____________________________________________________

Your right it is www.electionprojection.com

The complaint still stands though, but what are the other best poll and projection sites that people use.

[Akno21]

Well in doesn't seem right that both Survey USA polls be used in Iowa when the only one that gives Bush a lead is the hypotetical, if Vilsack was president this is not the case and unlikely and with out that survey Kerry would have all three wins and a 4.3 averaged advantage.  You could count this survey if Kerry did pick Vilsack.

And in Florida Bush did not win all of the last three but he is still painted as lean there.  It still seems somewhat arbitrary which survey's are counted and which are not and the system should be standardized so it will be honest.  

In Dale's Electoral College Breakdown the blogger decides off the cuff if he will consider the survey or not.  He generally doesn't change the Surveys in Kerry's favor.  http://home.comcast.net/~gerrydal/

Where as www.electionprojection.org showed Kerry with a huge lead and winning Missouri and Arizona etc and than 2 weeks Bush took back a big lead and not jut becauseof the Regan funeral but because the blogger had changed the calculation that he had been using and when more in favor of state polls.  Just trying to keep things honest here.

Here is one site I remembered for pollig info.  www.realclearpolitics.com.  

Those sites are pretty good.

[stry_cat]

www.electionprojection.org

Try http://www.electionprojection.com/  The .org link doesn't work for me

[Nation]

New Quinnipiac (the only worthwhile University poll) poll of LVs in Florida shows the race there still essentially tied -- good news for Kerry, I suppose. Better to be tied than to be behind, especially in a state where he's not expected to win.

[ATFFL]

One more note on the polls page:  If you win 2 of 3 polls but still have an average of 5% or greater, it will show the candidate with a slight lead.

[khirkhib]

Do you know the rules of the polling section for sure or you explaining your observations of them.

[ATFFL]

Do you know the rules of the polling secetion for sure or you explaining your observations of them.

As best I can figure out.  Only Dave knows exactly how they work, Vorlon has said he is pretty sure it uses only the three most recent, and I number crunched the states when questions came up.

[opebo]

I just clicked on the poll map and it has Kansas as a tossup.  Little mistake there?

[ijohn57s]

Has anyone else signed up for the Zogby online polling?   I did today, and already took completed my first poll.

It takes so little to make me happy some days...  :-)

Yes, I participate in those Zogby online polls as well.

[khirkhib]

I just wanted to express my joy that Oregon is no longer a toss-up in the polling section.  Yeah.

[stry_cat]

By the way, is anyone wondering about why the Libertarian vote is so high on the Mock Election Page?

'Cause we're everwhere

I would actually guess that Libertarians are more politically active and would be more interested in finding a place like this than the average joe.  Plus there are a large number of Libertarians in geekdom so an online poll might be just a tad skewed.  

Now isn't there a thread else where for the discussion of the mock election?  Oh yes here it is: http://www.uselectionatlas.org/FORUM/index.php?board=7;action=display;threadid=3972

[classical liberal]

The only poll in Nevada taken in the past 3 months shows the race at a 1 point margin.  Why is the state listed as a lean?

[Floridude]

Why isnt Florida listed as a toss up on the board on our page?  A very recent poll from strategic vision shows Bush up by 2.  It has not been put up.

[muon2]

I've noticed on the polls page the national Rasmussen is slipping further down the list. It's a tracking poll, and updated every day. It would be useful if Rasmussen was updated on a somewhat regular basis on the polls page as well, perhaps monthly.

This will probably become a greater concern as more pollsters turn to daily tracking after Labor Day.

[ATFFL]

The Rasmussen tracking poll should be updated weekly, with his weekly numbers.

[Floridude]

Has anyone heard of the strategic vision company?  They have been doing some polls of swing states.  Does anyone know if they have party affiliation, how reliable they are, et cetera?  Their polls seem a little bush positive but not outlandish.

[kevinatcausa]

Strategic Vision seems to be a firm whose primary purpose is to provide "Political Consulting", "Campaign Message", "Opposition Research", "Polling", "Media Training", etc. for their clients.

Their website is at http://www.strategicvision.biz/political/index.html .  I looked at their list of clients and every name I checked was a Republican.