Powerful Government Accountability Office report confirms key 2004 stolen election fi

there has been alot of talk about the statistics. The more I study statistics, the more I've seen that you can really make them say whatever you want... work with the right groups, word your questions appropriately, look at the right data or with the right angle, you get dramatically different results... It bothers me when I see certain polls online. For instance, you see an online poll asking something like "Do you approve of Bush's policy on so and so?". They request a simple yes or no. No varience. So, you answer No. You don't approve. Do you think he is too far right? Too far left? There is no difference. There are alot of polls on CNN that do this (I tend to frequent that web site). I'd rather a poll have more options.

Consider another poll... you go to DC and ask what they think of some kind of policy (lets say gun control). Depending on the policy or question, you get dramatically differing results than if you asked in Texas or Kansas.

So, point is... lets see where these statistics are coming from. Is it good science just because a few statisticians have problems? are there more reasonable explanations for some of the anamolies? Would you call it good statistics if you get your results from a narrow group? *scratches head* you -might-, and of course, the results will not be presented in such a fashion... so, in order to make decisions regarding the validity, I'd like to see sample groups.

For instance, this was mentioned before, and laughed off, but it actually has a degree of merit. Alot of people that are on welfare are likely not working, and would possibly vote earlier. Alot of retired people would vote in the morning, and the fear tactics of Social Security might possibly push them democratic. Is it that impossible that Republicans vote late? There are alot of polls going on, informal of course, but exit polls that give an idea of how the election is going. they are not the FINAL vote, but a predictor. If I saw my candidate was winning by 20 points, would I be quite as likely to go vote? If I saw it was a close vote, would I be as likely to vote? What -would- be compelling is if this never happened before. I've not read all the web pages presented (not had time, trying to graduate soon!), so if I missed this, please forgive and point out.

Anyways, I've always had a beef with certain statistics... this discussion is not necessarily relevant, just a beef I have with the presented "facts" I often see...

MrH
 
You bring up a lot of good points and I think that you have illustrated why it is good to actually look at the studies that were done. The original data is there, the models they used, and the results. The bottom line is that I've run regressions on the original exit poll data and the results. I can replicate the results. Am I an expert with statistics? No. Does it take an expert to see this? No. The results are always "significantly" different and they shouldn't be.
 
upnorthkyosa said:
Who? And where are there counter studies? What numbers did they use?
"
"What they did with their model is wrong, and their results are flawed," McCullough said. "They claim those results have some meaning, but I don't know how they can do that."
McCullough said they focused on one statistical model to conduct their analysis while ignoring other statistical models that would have produced opposite results."

http://www.wired.com/news/evote/0,2645,65896,00.html

upnorthkyosa said:
How do you know that no crime has been committed? If we look at the evidence in the form of verified allegations and discrepincies in the vote and then we compare it to areas that had e-voting and we take into account the security problems the GAO highlighted, it sure as hell looks as if something fishy happened. Finding who put vote switching software in place and who hacked the system will be the only way to prove that a crime was committed in this situation. THAT will require an investigation.
Because there is no evidence of it. That's how I know no crime has been committed. Your evidence obviously isn't attracting any special prosecutors....prosecutors, i might add, who often pursue people on some pretty shaky grounds. If a crime was committed, provided REAL evidence of it, not a sham, wack-job 'statistical model' laughed at by REAL scientists.

In short, you've got no crime and no evidence of a crime.

upnorthkyosa said:
Who authorized these special prosecutors? Who runs the Justice Dept?
The same people who allowed Scooter Libby to be indicted for perjury.

The reason there is no criminal investigation being conducted is very simple...THERE IS NO EVIDENCE A CRIME HAS BEEN COMMITTED.


upnorthkyosa said:
Exactly how is the information bogus?
It's bogus, because the numbers are pure jibberish.


upnorthkyosa said:
Who? And what exactly did they say? Where are the counter studies?
They don't need a counter-study to examine the 'study' you provided, and conclude it has absolutely no meaning. But if you want a counter-study, here you go.

http://election04.ssrc.org/research/critique-of-hmcb.pdf

I think they rather clearly disassemble that biased piece of, dare I call it, 'research'.
 
upnorthkyosa said:
You bring up a lot of good points and I think that you have illustrated why it is good to actually look at the studies that were done. The original data is there, the models they used, and the results. The bottom line is that I've run regressions on the original exit poll data and the results. I can replicate the results. Am I an expert with statistics? No. Does it take an expert to see this? No. The results are always "significantly" different and they shouldn't be.
So what you're saying is, you took the numbers 'figured' by Berkley, figured them the same way, added them up, and came to the same conclusion? lol.

"An easy way to show that there is something seriously wrong with a statistical study is to use the same data and the same approach to reach the opposite conclusion. HMCB apparently never bothered to check this aspect of their model, esle they'd have easily found it: we show that HMCB's modelling approach also supports the contention that electronic voting favored Kerry."

"We are generally concerned about the formal method- or lack thereof- whereby HMCB settled on their models. It appears that they just ran regressions until they got the answer they wanted." http://election04.ssrc.org/research/critique-of-hmcb.pdf
 
sgtmac_46 said:
What they did with their model is wrong, and their results are flawed," McCullough said. "They claim those results have some meaning, but I don't know how they can do that."

Not everyone agrees with McCullough (only 2 of the 7 profs that reviewed the paper) and McCullough himself did not address every aspect of what the previous paper addressed. Besides, even if they did make a mistake, its only one paper...not all of them.

"McCullough said they focused on one statistical model to conduct their analysis while ignoring other statistical models that would have produced opposite results."

Yet, he gives to rationale as to why the other model is better then the one chosen.

Because there is no evidence of it. That's how I know no crime has been committed. Your evidence obviously isn't attracting any special prosecutors....prosecutors, i might add, who often pursue people on some pretty shaky grounds. If a crime was committed, provided REAL evidence of it, not a sham, wack-job 'statistical model' laughed at by REAL scientists.

All you, Tom Delay and the rest of the GOP have been able to do is ridicule the oppostion. There has not been one single peice of evidence that has refuted the allegations thus far. The people who testified under oath are telling the truth for all anyone knows.

In short, you've got no crime and no evidence of a crime.

That depends on what you think is evidence. These statistical studies constitute a dead body.

It's bogus, because the numbers are pure jibberish.

Again, absolutely no rationate behind this statement. Show me how it is jibberish please.
 
sgtmac_46 said:
So what you're saying is, you took the numbers 'figured' by Berkley, figured them the same way, added them up, and came to the same conclusion? lol.

Nope, I took the numbers from the official record of votes tallied. I took the numbers released by the networks. I took numbers from the US census. This is the exact place that others found them...how do I know? I read the studies.
 
You dont have to "refute allegations". Allegations are not "proof" or "evidence" to be refuted. I allege that you are wearing a tin hat..refute me. ;)
 
Tgace said:
You dont have to "refute allegations". Allegations are not "proof" or "evidence" to be refuted. I allege that you are wearing a tin hat..refute me. ;)

If one is going to accuse someone of purjury, then one needs evidence that a crime has been committed, do they not?

btw - you just might be right...;)
 
No..only that what was said was untrue. It does not "prove" any underlying crime.
 
Tgace said:
No..only that what was said was untrue. It does not "prove" any underlying crime.

I agree, however, like I said above, if one takes the numbers into account, the allegations, the GAOs report, and the testimony of individuals who wrote software for the machines, then one has to at least conclude that we need an official idependent investigation into this matter. Hopefully this will either uncover or finally put to rest allegations of criminal activity.
 
Than why hasnt there been an invesitgation...beyond bloggers and internet hackers..yet?
 
Tgace said:
Than why hasnt there been an invesitgation...beyond bloggers and internet hackers..yet?

One of the sources I posted above explains how the system works and goes through all of the steps in which the GOP has stood in the way.
 
upnorthkyosa said:
http://www.appliedresearch.us/sf/Documents/Hypotheses-2004%20US%20Pres%20Election%20Exit%20Poll%20Discrepancy.pdf

This study takes on all of the rediculous explanations that have been proposed to explain the discrepency in the numbers.

"I think they're wasting everybody's time, frankly," Mitofsky said. "I know they're very serious about believing that there was fraud, but I don't happen to share their view. I find myself in the awkward position of having to argue that the exit polls were wrong." Warren Mitofsky, the president of Mitofsky International and a pioneer of the art and science of exit polling.

"This is not the first election with errors -- and the simplest explanation is probably the right one. I think fraud on a massive scale that their conclusion essentially requires is totally implausible. To make it plausible it would have a lot of people working together, and you know from being in the news business how hard it is to keep something secret. I just think their whole explanation is implausible."

Polling, Mitosfky argues, is not Mitteldorf's area of expertise. He and others have taken the USCV statisticians to task for shoddy work.
"The trouble is they make their case very passionately and not very scholarly," Mitofsky says. "I don't get the impression that any of these people have conducted surveys on a large scale."
Others have questioned the methodology and conclusions of the USCV report as well -- and not just Republicans. Democratic pollster Mark Blumenthal, who writes the mysterypollster.com blog, accuses USCV of blithely ignoring crucial parts of the Edison-Mitofsky report that attempted to explain the reasons for statistical error -- particularly the failure of polltakers to follow crucial directions.

http://www.washingtonpost.com/wp-dyn/content/article/2005/04/24/AR2005042401545.html





It seems that with ever conspiracy theory that gets debunked, the leftists get ever more desperate.
 
So, do we actually have a criticism that deals with the numbers used? The math is simple enough that I can do it and you've claimed the numbers that I and other are using are gibberish. I'm still waiting.
 
William C. Velasquez Institute

Dec. 3 - The William C. Velasquez Institute (WCVI) announced that NBC said that it studied the discrepancy between exit poll figures for Latino votes from the WCVI and NEP (Edison/Mitofsky), and now believes WCVI's figures were more accurate and that NEP under-represented urban areas (leaning Democratic) and over-represented non-urban areas (leaning Republican). WCVI claimed that NBC has therefore adjusted its figures and believes that in fact Kerry had 58% not 53% of the popular Latino vote and Bush has 40% and not 44% as previously stated. NBC's estimate for Hispanic support in Bush's home state of Texas was also revised, turning a reported 18-point lead for Bush into a 2-point win for Kerry among Hispanics, a "remarkable" 20-point turnaround from figures reported on election night.
 
upnorthkyosa said:
I agree, however, like I said above, if one takes the numbers into account, the allegations, the GAOs report, and the testimony of individuals who wrote software for the machines, then one has to at least conclude that we need an official idependent investigation into this matter. Hopefully this will either uncover or finally put to rest allegations of criminal activity.


It will never put to rest the allegations, as you already have no proof, and STILL believe it's true. It's entrenched itself like any good conspiracy theory. That is because, your beliefs are not based on any statistics, any rationale, but have entered the realm of dogma.

Further, by what do you mean 'independent investigation'? I've heard this one before. I assume it will be headed up by an 'impartial panel' headed by Bill Clinton and Janet Reno, presided over by 'UN observers'? What a farce. Then, no matter what's discovered, unless it's 'BUSH CHEATED', you won't believe it.

Again, upnorth, it's become my conclusion that apparently, some people believe that if they weave an intricate enough web of BS, that it suddenly turns in to silk by volume. The reason no one is taking your 'evidence' seriously, is the majority of Americans, much to your dismay, have a pretty decent built in BS detector, and mines going off the scale right now.

Like the old saying goes "if you can't dazzle them with brilliance, baffle them with BS."
icon12.gif
 
Back
Top