In the immediate aftermath of the 2006 election results, there was sigh of relief as the returns alleviated fears of another stolen election. In any normal year, the tsunami that developed, as voters expressed their disgust with the Republican party, would have swept a large number of Republican candidates — and, indeed, it had.
But lost in the celebration was an ugly fact, one that had emerged in aftermath of 2004 — the exit polls had told a different story than the ‘official’ results. It was a tsunami, indeed, but one of even greater proportions than we had been led to believe.
Analysis of the exit data shows a wave that should have swamped even more Republican candidates, and would have if official returns had reflected the will of the people. And, again, much like in 2004, canards are being proferred as part of a disinformation technique to distract the public from the pertinent methodological questions — such as; exit polls are unreliable, the Rove turnout machine neutralized voter fallout, and so on.
Although “safe districts” insulated Republican candidates in some cases, even those safe districts came into play, as even the most heavily Republican districts, ones that no one would have dreamed could flip, were swamped by a combination of forces; Democrats had turned out in greater numbers, a large contingent of independents had voted two to one for Democratic candidates, the youth (heavily Democratic) had turned out in record numbers, and so on. The damage had been limited, somewhat, by twenty five years of gerrymandering, but the tidal wave of discontent had been too high. Much like Katrina, even well-constructed electoral levees stood no chance in the onslaught.
In spite of this perfect storm, some of the Republican candidates remained standing, as if protected by a some kind of miracle. In places such as in Florida’s 13th Congressional District, where touchscreens rule and paper trails are nonexistent, 20% of the early voters, who voted on touchscreens, didn’t register a vote for the Congressional race. Meanwhile, the absentee voters, who differ from the early voters in that they voted on a paper ballot, failed to register a vote at a more typical rate of 2%. Furthermore, the electoral miracle confined itself to a single county; Sarasota, as the neighboring counties in the five county district experienced non-vote rates close to the historical average.
Perhaps these events explain why Karl Rove was so confident in the run-up to the election — is there something he knew that we didn’t?
Again, lost in the celebrations was the ugly fact that the tsunami had been much bigger than indicated by official results — which, of course, are officially registered on electronic machines that were failing at astronomical rates in polling done before the election. In this age of ‘faith-based’ elections, we have only faith to reassure us, since the evidence — which is far closer to my religion — shows something quite odd was happening.
The following article examines a problem that first reared its ugly head in 2004, that is; a huge difference between the exit polls and the official results.
http://electiondefensealliance.org/landslide_denied_exit_polls_vs_vote_count_2006
Thank you,
In the aftermath of the democrat’s win the election protection advocates have been fairly successfully silenced. It’s as if this win vindicates all the anti-fraud screamers. Or so they seem to believe. Unfortunately, the campaign to discredit exit polls as a viable guage of the integrity of our elections has been largely successful. Short of a reversal of that meme, there needs to be another statistical way to show when the elections are tainted or corrupted. We still don’t have access to the machine’s code and this past election will make it harder for activists to gain access to it. Further disasters loom on the horizon, sad to say.
interesting that FL 13th congressional is Katherine Harris’ district Wouldn’t you know a whopping 18,000 under votes. It appears another court battle looms.
RE: exit polls.
Listening to the absurd canards about the so-called problem with exit polls makes me cringe, as do other obviously made-up tales which have become a staple in today’s political milieu.
Exit polls are more reliable than the more common pre-election poll, simply because the voters are answering about an act that they’ve just completed, rather than merely announcing their preferences about an act which may never occur.
The two types of voter classifications; likely and registered, are used because they can predict fairly well who might actually vote, as well as their preferences in that vote. However, alhough pre-election statistical samples can accurately predict how voters will vote, they can falter at predicting turnout, i.e., the intensity of feelings, which drives prospective voters to follow through on their intentions.
A particular theory that was put forth in 2004 by the Edison/Mitofsy consortium; the ‘reluctant Republican respondent,’ could have some merit if it was born out by other data — and it was not. As such, it is a form of special pleading, since there is no indication of a change in methodology to produce this effect. Pre-election polling would also suffer from this same bias, too, but any competant pollster would have taken steps to ensure that the sample reflected the distribution of the population as a whole. Although it’s tempting to accept this ad hoc hypothesis, since it basically says that Republicans lie more than Democrats, this methodological problem is routine, expected and easily counteracted through substitution.
Stratified samples, which attempt to correct for inbalances in the numbers of Republicans and Democrats polled, can have other problems. Eliminating the bias that results when reluctant voters opt out, thus underrepresenting their numbers, can create a new bias, of underestimating high turnout groups. A true random sample will reveal when one group in outperforming others, but, as stated, this makes the sample susceptible to yet another bias.
http://en.wikipedia.org/wiki/Stratified_sampling
The real problem with the exit polls revolves around the use of “forcing,” in which Edison/Mitofsky replaces the actual exit poll with the poll results themselves. Presuming that the official tallies are accurate assumes that electronic machines are also accurate, and that nobody would rig or hack the election. It also underestimates overperforming groups, as can be seen from the following quote from the Election Defense Alliance’s paper:
——————————————————-
[Note: Edison/Mitofsy asked voters a background question in 2006; “who they voted for in 2004,” and the final proportions of adjusted totals — which are “forced” to conform to official poll tallies — show this unusual result.]
” – the final, adjusted exit poll showed a margin of 43% Kerry to 49% Bush. This 6% margin in favor of Bush was a dramatic distortion of the 2.8% margin actually recorded in E2004. – “
——————————————————-
The charge that too many Democrats were included in the exit data v. the official returns, is based of the divergence of the two. In other words, exit polls ‘must’ be oversampling Dems because they’re at odds with the official results. What the background question shows us is that 2006 exit poll (+/- 1%, N= 10,207), in which Bush voters outnumbered Kerry voters by 6%, was highly skewed. Apparently, Edison/Mitofsky had used the official results from 2004, which in turn led them to de-weight parameters in the 2006 poll, and that in turn led to their dramatic underrepresentation in 2006 and the now snowballing error. Each adjustment in the data creates an even larger problem when that data is adjusted to bring it back in line. The problem is not that the national exit poll is wrong (it is), even though it’s so far out of its confidence interval that its probability is getting infinitesimally small, the problem is coming from relying on parameters that were based on official results that were themselves wrong.
The interesting thing is that the Democrats would not even have to “protest.” They could merely point out that it is now time for the US to adopt voting procedures meeting the standards of reliability and varifiability of your average third world country.
Or perhaps even the standard of the old mechanical machines, which–although they could be gimmicked–under ordinary disinterested/or biparty supervision worked fairly well.
Even this is too much to say. I think this may be a hint of the full extent to which American politics–over its entire specrum–is owned.
That the Dems are okay with this suggests that they are owned, too.
Well. Who knew?
The advantage of replacing politics with theater is that it makes it more predictable–for the owners who pay for the theater production.
So. It is the oligarchs who have taken the lesson from Sun Tzu this time. (“It is better to take over a unit than destroy it; it is better to take over an army than destroy it; it is better to take over a country than destroy it.”) They have taken a deadly opponent–free popular elections–and by taking it over by buying it and then subverting it (and buying it again) have turned it into a reliable ally.
Well, this would all make sense would it not? Was this election a test of what they could get away with?
Rove knew the rpubs were going to get their asses kicked big time in ’06. So he sets out to test what people are willing to tolerate when they win. How much could they get away with and see if people will stop screaming voter fraud/voter reform. Nobaody is saying anything so that sets the table for 2008 if they need to fix it, which they will. Just a few rambling thoughts on the subject this morning.
Was the fix in? We really won’t know until at least January when Democrats take power in the Congress and can issue subpoenas to investigate.
If it was, the netroots insistence on turning out numbers to overwhelm the fix was the correct strategy for the Democrats.
The fix was definitely in for some locations like parts of Sarasota in FL-13. Probably some parts of Maryland as well. In Virginia, the efforts at turnout suppression were blatant. And in Maryland, the Steele campaign tried to confuse voters in Democratic neighborhoods in Baltimore. It was not just about the machines. The attack was on every mechanism involved in the voting process from registration (voter ID and felons lists), to turnout (phony information, phony phone calls from phony police, phony flyers), to the set up of ballots (both paper and electronic). I think that we will find that any hacking of voting machines was a very small percentage of incidents. It’s so much easier to “butterfly ballot” the presentation of choices on the voting machine.
I can simply and briefly state an alternative proposal for voting in the U.S. that is detailed, comprehensive, and workable:
Use the Canadian system.
Paper ballots.
Detailed, debugged rules.
Distributed hand-counting, with observers.
Transparency at all stages.
Hard to corrupt.
Fast results.
And, it’s easy to state the idea as a question that demands an answer:
“Why not use the Canadian system — wouldn’t it be better?”
A description here.
Detailed rules here.
In America…the fix is ALWAYS in.
“Vote early and often.”
A truth from the ward heeler years of America.
So-called “literacy” tests for black people in the South.
JFK’s mafia connected primary win in West Virginia.
Richard Daley’s Chicago.
The 2000 + 2004 vote counts.
The “Arrrgh”-ing of Howard Dean and the ensuing coronation of Small K kerry at 2004 Dem Convention.
And so on.
Bet on it.
The only difference? Ever?
How DEEPLY the fix is in.
This year…? They didn’t screw it in all the way to the base.
Their bad.
2008?
If we are not careful?
“Brace yourself, Bridget!!!”
Bet on THAT, too.
AG