Posted on 06/10/2004 6:33:13 AM PDT by Dales
Edited on 06/10/2004 8:28:27 AM PDT by Admin Moderator. [history]
The battlegrounds:
| State | 2000 Result | Media Battleground | Current Status |
|---|---|---|---|
| New Jersey | Gore +16 | No | Kerry, Lean |
| Iowa | Gore +0.3 | Yes | Kerry, Lean |
| Michigan | Gore +5 | Yes | Kerry, Slight |
| Oregon | Gore +0.4 | Yes | Kerry, Slight |
| New Hampshire | Bush +1 | Yes | Kerry, Slight |
| Minnesota | Gore +2 | Yes | Kerry, Slight |
| Pennsylvania | Gore +4 | Yes | Kerry, Slight |
| Wisconsin | Gore +0.2 | Yes | Kerry, Tossup |
| Florida | Bush +0.01 | Yes | Tossup |
| New Mexico | Gore +0.06 | Yes | Tossup |
| West Virginia | Bush +6 | Yes | Tossup |
| Ohio | Bush +4 | Yes | Tossup |
| Arkansas | Bush +5 | Yes | Bush, Slight |
| Virginia | Bush +8 | No | Bush, Slight |
| Nevada | Bush +4 | Yes | Bush, Lean |
| Arizona | Bush +6 | Yes | Bush, Lean |
| Missouri | Bush +3 | Yes | Bush, Lean |
The scoreboard:
| ECB | Bush 203, Kerry 190 | Bush 222, Kerry 221 with tossups |
| ECB Classic | Kerry 208, Bush 179 | Kerry 316, Bush 212 |
| ECB Classic Sans Zogby Interactive | Kerry 200, Bush 178 | Bush 280, Kerry 231 |
| Calculated National Result | Bush 45.5%, Kerry 44.5% |
One can really see the impact of Zogby from the tossup numbers there.
It is getting to be too much to include all of the updated states in totality in the weekly update. As such, I will just provide the links to the state pages. States updated this week were:
Missouri, West Virginia, Ohio, Wisconsin, Kentucky, Michigan, Florida, Massachusetts, Iowa, Oklahoma, Arkansas, South Carolina, Maine, Alabama, Minnesota, Washington, New Jersey, Virginia, Pennsylvania, Louisiana, Georgia (again), Connecticut, North Carolina, Oregon, Illinois, New York, Texas, California.
| Effective National Popular Results: Bush 45.5%, Kerry 44.5% |
|---|
|
| Kerry E | F Bush | |||||||
|---|---|---|---|---|---|---|---|---|
| Safe | Strong | Lean | Slight | Tossup | Slight | Lean | Strong | Safe |
| DC (3) | DE (3) | MD (10) K48-B43 3/24/04 |
NH (4) K49-B45 4/26/04 |
NM (5) B46-K45 4/1/04 |
VA (13) B47-K45 5/31/04 |
NV (5) B49-K38 3/17/04 |
MS (6) B49-UD29 12/22/03 |
AK (3) |
| HI (4) | VT (3) K51-B36 5/1/04 |
CA (55) K49-B41 5/31/04 |
MN (10) K48-B43 5/31/04 |
OR (7) B46-K45 5/31/04 |
AR (6) B48-K43 5/31/04 |
TN (11) B52-K41 3/22/04 |
KS (6) B57-K39 3/4/04 |
ND (3) |
| RI (4) K53-B31 2/7/04 |
IL (21) K54-B38 5/31/04 |
NJ (15) K51-B39 5/31/04 |
MI (17) K47-B45 5/31/04 |
PA (21) B45-K44 5/31/04 |
- | CO (9) B49-K44 4/14/04 |
OK (7) B53-K34 5/20/04 |
NE (5) |
| NY (31) K57-B34 5/31/04 |
CT (7) K46-B36 6/1/04 |
WA (11) K49-B44 5/3/04 |
- | FL (27) B46-K46 5/31/04 |
- | AZ (10) B43-K38 5/23/04 |
SD (3) B50-K35 5/21/04 |
WY (3) |
| MA (12) K58-B33 5/31/04 |
ME (4) K54-B35 5/31/04 |
IA (7) K49-B41 5/31/04 |
- | WV (5) K47-B41 5/28/04 |
- | NC (15) B48-K44 5/31/04 |
LA (9) B48-K42 5/31/04 |
ID (4) B55-K23 3/17/04 |
| - | - | - | - | WI (10) B44-K42 B44-K42 6/8/04 |
- | MO (11) B48-K37 6/8/04 |
GA (15) B49-K32 6/2/04 |
UT (5) B67-K22 5/10/04 |
| - | - | - | - | OH (20) K45-B42 6/8/04 |
- | - | SC (8) B49-K39 5/31/04 |
IN (11) B54-K33 5/19/04 |
| - | - | - | - | - | - | - | KY (8) B52-K39 6/8/04 |
MT (3) B53-K33 5/26/04 |
| - | - | - | - | - | - | - | - | AL (9) B57-K36 5/31/04 |
| - | - | - | - | - | - | - | - | TX (34) B55-K38 5/31/04 |
| Totals | ||||||||
| Kerry States | Battleground States | Bush States | ||||||
|
|
||||||||
| 54 | 38 | 98 | 31 | 95 | 19 | 61 | 62 | 80 |
|
|
||||||||
| 190 | 145 | 203 | ||||||
Since 1948, one state has gone every time to the candidate who won the national popular vote. Which?Delaware!
No quiz this week due to me being on the road.
Ping! I decided to not wait for the red-eye out of here to do the writing.
Good morning, my friend...I have one question I've meant to ask for a few weeks...even those who feel the race won't be that close, feel that it WILL be a late-breaking move. However, i the last 4years we've seen some states, among them Washington and Oregon, where vote-by mail has becomevery popular, something like 40% of voters doing so 3 or more weeks before the election....could this be an impact in a few states?
Thank you,,you do a great job and I look forward to your posts.
You say that there are lots of polls this week with motion towards Kerry. You do realize you cannot make a trend declaration if the pollsters choose to modify their methodology in the midst of the supposed trend?
With Reagan's death, it seems credible to me that the pollsters would use the occasion as an excuse to tweak their sampling mix in some well intentioned attempt to extract the Reagan event's effect.
Have a safe trip back toots.
222-221
Election years are boring in Texas.
you are to be congradulated for this work. Thanks.
You bring up an interesting point though- I will have to look for evidence of such tweaks. If I were a pollster I would avoid making any like that since I would want to measure the change in attitudes because of the event and a methodology change would work to distort that somewhat.
Sure, it could.
Thanks for the breakdown.
Word has it the GOP did a dry run on their turnout efforts. HHopefully, this is a good sign - if they have a GOTV effort about as good as the Rats, then the Rats are toast.
Pollsters are using the Reagan death to tweak their results - my god, we sound as looney as those on the Left sometimes -
Movement is going towards Kerry (movement has been going towards the Dem's since last year if you look at any grahpical chart) - and this is happening BECAUSE the GWB team has allowed false premise after false premise to be set - be it on the economy (not being strong) or on the war with Iraq (no WMD & no need to have removed saddam and no post-war plan).
The fact is, if the GWB reelection team does not get on the offensive with regard to the facts about our robust U.S. economy and our successes in the WOT & Iraq - he will lose.
We must STOP with all the excuses - the GWB reelection team is doing an awful job - playing defense and completely inept at getting any successful coherent message out. (not to mention not having any RNC 527's up and running).
Pollsters are using the Reagan death to tweak their results - my god, we sound as loony as those on the Left sometimes -
Movement is going toward Kerry (movement has been going toward the Dem's since last year if you look at any grahpical chart) - and this is happening BECAUSE the GWB team has allowed false premise after false premise to be set - be it on the economy (not being strong) or on the war with Iraq (no WMD & no need to have removed Saddam and no post-war plan).
The fact is, if the GWB reelection team does not get on the offensive with regard to the facts about our robust U.S. economy and our successes in the WOT & Iraq - he will lose.
We must STOP with all the excuses - the GWB reelection team is doing an awful job - playing defense and completely inept at getting any successful coherent message out. (not to mention not having any RNC 527's up and running).
why poll republicans and democrats? you pretty much know how they are going to vote.In a close election, the loss of precision between knowing pretty much and knowing can be critical.
A new ARG poll is out for New Hampshire:
Bush 46, Kerry 46
http://www.americanresearchgroup.com/
How right you are.
Anyway, I just wanted to point out that the ISMAP interactive map references freerepublic.com (ergo, "site not found" upon clicking any state) and not your home directory (http://home.comcast.net/~gerrydal/). You may want to manually insert your home directory before all those state sub-directories when including the ISMAP html into your FR thread.
You may even want to consider forcing a new window for each state with target="_blank", given that most people probably pick and choose their home state, then one or two battleground states for an update. Just a thought. Thanks for your hard work.
I do that normally. I did that this time as well, I just messed something up. Fortunately, I can fix it.
Question. I believe pollsters adjust the results based on conceived demographics. Does this apply to all the questions or just the ones involving a head to head match-up?
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.
I Wouldn't Touch It With a 10 Foot Poll
Apples to Oranges
A few weeks back, I wrote about what can be gleaned by comparing Apples to Apples. This week, I will be comparing apples to oranges-- how different polling companies involved in multiple states match up. I began writing this article on Sunday, so the data I am using is from before then. For each company examined, I will show their most recent numbers in any state that another multi-state polling company has released numbers. For this, I have limited the information to those polls conducted since Kerry became the Democrat's nominee. Since that time, while there has been some apparent movement in some states, overall the race has been remarkably stable, which should minimize distortions in this view caused by different polling dates. The companies I will be looking at are Rasmussen, Zogby, SurveyUSA, Research2000, Mason-Dixon, Quinnipiac, and American Research Group.
Towards the left, we have Zogby, whose polls have generally been the most favorable towards Kerry when there were multiple polling companies hitting the same state, with only one or two exceptions. On the other end of the spectrum has been Mason-Dixon, which has been the most favorable to Bush in most states they have polled. Rasmussen and SurveyUSA, the two automated polling companies, have been both to the left and to the right of each other, with Rasmussen being to the left just slightly more often. With the exception of Ohio, American Research Group has been more favorable to the President than most of the others.
Polling companies all poll the same populations (except for Zogby Interactive which polls a subset that had registered on his website). The different results, even when conducted during the same timeframes, can be random chance but the odds of that drop significantly when the same company keeps coming out to the left or the right time after time.
Each of these companies tries to poll "Likely Voters", but each's definition of who is a likely voter can be significantly different. Zogby tends to predict turnout in different areas and within different demographics and ensures that his samples reflect this, either directly in the number of those polled or by performing some scaling of the results to fit the model. His method can be deadly accurate when he has his turnout predictions nailed, and it can make him look really bad when he does not. Mason-Dixon uses a much simpler method and one that is less subjective: their polls are of registered voters who say they are likely to vote. This is a relatively inexpensive method, because it only requires two questions in the script. However, studies performed by Gallup and others have shown that while a person saying they plan on voting is an indication that they are more likely than the typical registered voter to turn out, it should be used with other indicators as well for best accuracy. The other polling companies listed have their own ways of determining who is a likely voter, and they guard their methodology since it is a prime differentiator between them and their competition.
Which method is right? In the past few years, SurveyUSA has had one of the more impressive records. However, each election is different, and until election day we will not know which method is going to be most accurate for this year's dynamics. It is best to consider them each as unique snapshots of the race taken from different angles. But sometimes using them in conjunction can tell us things that we may otherwise miss. Take Mason-Dixon, for example. This year their polls have been favorable to President Bush. In 2000, this was not the case; more times than not their polls were towards the middle of the results from other companies. But keep in mind their methodology. If they are coming consistently to the right of the other polling firms, that indicates that Bush has an advantage among those who say they are planning to vote but are not being captured by the more restrictive methods for determining likely voters or by the methods based primarily over past election turnouts. This hints that this year the Republicans may find success by stressing, even more than usual, get out the vote efforts designed to mine for votes.
There are other indications I have seen that indicate that the turnout this election may favor the Republicans more than the Democrats. Rasmussen's state-by-state polling consistently shows that Bush has a higher percentage of his support saying they are certain to vote for him than Kerry gets from his support. A Pew poll earlier this year found that by a two to one margin, Kerry supporters say they are primarily voting against Bush than for Kerry; this is a trait that has happened only twice in recent Presidential elections with both times resulting in a loss for that candidate where they had difficulty getting their base to turnout (Dole and Dukakis).
Last night provided another possible indication in this regard. New Jersey's primary was last night. On the Republican side, President Bush received 211,218 votes. On the Democrat side, Kerry took 191,816 votes. It was not that the Democrat vote was divided, either-- Bush took more than all of the Democrats (Kerry, Dennis Kucinich, George Ballard, and Lyndon LaRouche) combined. In 2000, Bush took about the same number of votes, But Gore took nearly double what Kerry took. Part of the difference between Kerry and Gore was undoubtably due to the Senate primary in 2000, but it should be remembered that the Republicans also had a contested Senate primary that year (although one that spent considerably less money). Be that as it may, it still remains that with neither side spending money to drive up turnout, in a state where Democrats hold a 6% registration advantage, Bush outpolled Kerry on primary day.