Sunday, September 05, 2004

A summary of problems with recall vote A very brief summary of what we know: -Exit polls: Very low probability of sample being taken from those final results according to Sanso and Prado, using three different exit polls. -Hot Audit: Incomplete, opposition present in only 27, in those Si won handily. Most never completed. -Cold Audit: Random Sample being questioned, may not have been that random. CD not allowed to choose boxes. -Coincidences: Very low probability of happening, 1 to 3 in 1000 according to Prof. Taylor. -Quick Poll: Only thing that there is agreement it was OK by all parties. What do you think? Paranoia? Just coincidences? De que vuelan, vuelan? (In Spanish there is a saying: I don't believe in witches, but that they fly, they fly) 11:28:27 PM    comment []

Letter to Prof. Johnathan Taylor I just sent this letter to Prof. Johnathan Taylor of Stanford University (Use links in text, line links don't work) Dear Prof. Taylor: On Aug. 16th. Venezuelan opposition analysts began charging that there were some surprising numerical coincidences in the number of SI votes within voting centers in Venezuela. The Carter Center and the Electoral Board said on the 18th. that there were only 402 and this was reasonable. But this was not what the opposition was charging, the 402 repetitions were across the "mesas" (tables) and not across the machines. Thus, apparently the Carter Center had consulted on the  wrong problem. In my blog, I explained the difference on Aug. 18th.: http://blogs.salon.com/0001330/2004/08/18.html#a1702 pointing out that a center may have a number of mesas, but that each mesa may have a variable number of "cuadernos". In centers with electronic voting, each cuaderno is equals to one machine. Thus, the lowest equivalent unit is the cuaderno (machine) not the mesa. On the 22nd. I reiterated the issue: http://blogs.salon.com/0001330/2004/08/22.html pointing out that the 402  SI coincidences at the mesa level, became 805 coincidences at the center level with 1879 machines involved in these coincidences. Elio Valladares took a first stab at the mesa problem: http://blogs.salon.com/0001330/2004/09/01.html#a1732 reaching conclusions that you now agree with. However, it was not until yesterday that Elio solved the problem in what I believe is the correct way, he did his simulations at the cuaderno (machine) level and from there calculated the coincidences for the SI vote at the mesa level. His results (attached) indicate that the probability of the number of coincidences reported is about 1 in 10,000 when the problem is considered at the machine (cuaderno) level. Certainly a worrisome result given what is at stake. While no mathematical study will prove that there was fraud, it certainly seems suspicious that these coincidences apparently have such low probability, that a statistical study of exit polls by Prof. Sanso and Prof. Prado of the University of California at Santa Cruz: http://blogs.salon.com/0001330/2004/08/26.html#a1720 also finds very little probability of those exit polls matching the vote at each voting center and that the live audit of comparing 199 ballot boxes the night of the vote was never completed. The Carter Center, based on your earlier conclusion disregarded the problem of coincidences. I believe it is only fair that you calculate the same problem Elio Valladares calculated at the machine level and using your conclusions help decide whether the Carter Center and the OAS should or should not reopen the case for fraud in the Venezuelan recall vote. Respectfully. 8:46:34 PM    comment []

Carter Center trying to look good, defends results from Rubin now I guess I was not the only one paying attention, here JorgeRodriguez says they talked to Prof. Taylor and he saw the same text I posted, but now Taylor is being pressured. Meanwhile Jennifer MCoy continues skirting the issue and now talks about other experts, without realizing that they have not even solved the correct problem. They even issue a new report and now side with Rubin's data. This is now getting interesting as I added in the note below, Elio Valldares has solved a more complete problem and he gets a probability of one in 10,000 for the coincidences observed. Tulio Alvarez, the lawyer preparing the case, now says the Carter Center played the role of useful idiots. The attempts today by the Carter Center to me are more irresponsible than idiotic. I think both Taylor and Rubin, should work on the real problem at the machine level to calculate what Elio Valladares did and shut up the Carter Center for once and for all. Enough! 8:36:44 PM    comment []

On Mathematical models of the recall vote and fraud, part V: Prof. Taylor posts correction, changing Carter Center conclusions A reader, Mercedes Rosas,  has pointed out in the comments below to a correction in the webpage by Prof. Jonathan Taylor of Stanford University and an e-mail from Prof. Taylor to her, on the results of the recall vote in which he says he made an error earlier and corrects hs results. Prof. Taylor’s work was used by the Carter Center to “show” that the number of “Si” coincidences in the mesa (table) votes was reasonable. This result was widely used and quoted by the international press as part for the “evidence” that there was no evidence of fraud in the recall vote. As I have reported elsewhere, Elio Valladares got quite a different result, suggesting that the probability was not that “reasonable”, in fact Elio obtained that it was small, if not miniscule. Now Prof. Taylor has corrected his work on his web page and I would like to quote him so that there is no misinterpretation of what he says or not:   “ It seems that an expected number of ties between 345 and 350 is reasonable, as it came out from many different models. Using the Poisson assumption to estimate the standard error, it seems then that the probability of observing 402 or more ties for SI is between 1 and 3 in 1000. While this probability is small, I do not feel that it should be interpreted as overwhelming evidence of fraud.”   Yes, is not overwhelming, but we have gone from reasonable to small, but it was the reasonable that led the Carter Center to its conclusion. What would they say now?. By the way, the CNE also used this result by Prof. Taylor to say that the Si vote coincidences were irrelevant.   Prof. Taylor has acted with the integrity characteristic of scientists, I wonder if the Carter Center will post a clarification to their conclusions, but doubt it. I sure hope Prof. Taylor will now calculate the coincidences at the machine level. I believe in that case he will find that the probability is even lower, if not impossible!!! That should have been the case that the Carter Center should have had him study to begin with!   Note added: I have now received a new study by Elio Valladares in which, if I understand correctly, he simulates the vote at the level of the cuadernos (each cuaderno is a machine in centers with electronic voting) rather than at the mesa level. He then calculates then the probabilities at the mesa level using the results from the simulation of the cuadernos. His conclusion from this is that the number of observed SI vote coincidences should be 1 in 10,000, which I think even Prof. Taylor would consider we are getting into the realm of the "overwhelming evidence of fraud"   Second note added: Prof. Taylor has now removed the word overwhelming from his conclusions. I wish that rather than worry about words, they would work on the real problem at the machine level. 12:01:02 AM    comment []