4. Keep up the good work TIA. This is the only way to wake people up.
I can't see any other way except statistics to make people aware of the fraud. Every other avenue is closed off or so hard to travel thru that it's essentially impossible to make any inroads. Look at how hard it was to get the fake recount in OH. A lot of money, speeches, pressure, and in the end the recount went as the RW people in power wanted it to go: only 3% of the precincts hand recounted and those 3% cherry picked by the Blackwell folks in one way or another. None of the really crucial precincts hand recounted. Same with FL and NM. There are other law suits pending and going on and that's important too, but the real breadth and depth of the fraud can only be revealed by stats and you're providing an invaluable service.
Keep yourself well. Get plenty of exercise, eat the right foods, and think positive! I want you to stay around a long time.
May you live to be 150, TIA, and may you have many offspring.
11. But the 2.24% can come about in numerous ways.
You can't do stats, which looks at aggregate distribution and characteristics, and then say the s.d. is useless.
S.d. characterizes the distribution, as does skew. Give me skew, s.d., average, and kurtosis, and I have a good idea what the curve looks like. Give me an average, and I know very little. Give me lots of individual tokens, and I'm likely to get lost in the details while wondering why anybody provided an average, or I'll look at the s.d. and skew.
27. When did I say the S.D. was useless. I'm only saying
Edited on Sun Jan-16-05 07:57 PM by TruthIsAll
that when there's a 2.24% average deviation for all 50 states, it is significant. This is confirmed by calculating the MOE for each state poll. If there was an outlier or two, we would see it. But when TWENTY-ONE states fall outside or are equal to the MOE, we have a problem, Houston. The average deviation is not sure not due to an outlier.
7. These numbers are dramatically different from 2004.
Are these based on final adjusted numbers, or like the 2004 data? If this is a stupid question, sorry. As I mentioned in another thread, I'm just starting to wrap my statistically disabled brain around data like this. I guess what I'm trying to say is, can this set of data be compared to the data in your other thread on an equal basis (apples to apples), or are they different? Do you ever sleep, TIA? Nice work, again! :toast:
8. Just looking at the data from Michigan, my state....
there is no possible way. Kerry was much more popular and supported than Gore was, at least in the SE part of the state. Kerry rallies drew huge masses of people, with several hour waits, and many turned away. That did not happen when Gore visited the state. I'm comparing the turnout at the same facility in the state. The differences in the turnout were huge.
The exit polls showed Gore with a 2.3% lead. After the requisite number of Gore ballots were spoiled Gore had only a 0.5% lead. QED
Note from the data in TIA's excellent find that Arkansas may have been stolen from Gore in 2000, along with Florida. And the data in the Washington Monthly/Kevin Drum article shows that Repubs may have fraudulently rigged the presidential elections in 1988, 1992, and 1996.
It is very rare that I would call someone crazy here.
I'm doing it now...at least, about your idea. It's crazy. Totally insane. True blue conspiracy tin foil hat stuff. To imagine that the Republicans rigged the election against Clinton twice (yet lost), rigged the election against Gore (lost that too, for the popular vote) and rigged the Dukakis election (which they would have won even without the rigging) as the only logical explination for the Washington Monthly data, is lunacy. They would have had to employ a massive, nation-wide conspiracy to do it. In those earlier elections, there was no software to rig...they didn't even have electronic voting. Nor was there a handful of voting machines like now..back then, there were hundreds. So, it would have had to be a conspiracy on a bible scale of production...with NOBODY talking about it all these years.
The obvious FIRST logical explanation is the one offerred by the creator of exit polling himself, who is also a lifelong Democrat: that the polls NEED TO BE WEIGHTED TO BE ACCURATE, AND WITHOUT WEIGHTING THEY ARTIFICIALLY DEVIATE TO DEMOCRATS DUE TO POLLING IN URBAN LOCATIONS MORE THAN RURAL LOCATIONS.
Seriously...that you would even think to conclude that massive conspiracy is the more logical explanation than unweighted polls are flawed is scary to me.
No one cares. If the math is right, and you know how paste data into Excel, your underying assumptions can be true or false without affecting the validity of the argument. The calculations are CORRECT. End of story.
"Do you believe the 2004 exit poll numbers? I do."
Which ones...the weighted or the unweighted ones? You seem to believe the unweighted ones...and yet are now citing weighted exit polling data to claim that exit polls are accurate. Make up your mind TIA...what's your position...is it weighted polls are right, or unweighted polls are right?
"Do I believe these are comparable? Who really knows?"
You know, for sure, that they are the weighted numbers. You have, in your hands, the unweighted numbers, and they do NOT match the election. So, who are you kidding?
Seems no one can be sure right now, but if I had to guess they're probably weighted (aren't they usually weighted before being made public?) but not corrected to agree with the vote counts, or at least not consistently corrected. Otherwise how does one explain the states with the larger deviations?
Powered by DCForum+ Version 1.1 Copyright 1997-2002 DCScripts.com
Software has been extensively modified by the DU administrators
Important Notices: By participating on this discussion
board, visitors agree to abide by the rules outlined on our Rules
page. Messages posted on the Democratic Underground Discussion Forums are the
opinions of the individuals who post them, and do not necessarily represent
the opinions of Democratic Underground, LLC.