HomeLatest ThreadsGreatest ThreadsForums & GroupsMy SubscriptionsMy Posts
DU Home » Latest Threads » Forums & Groups » Places » International » Latin America (Group) » Venezuelan Audit Can’t Fi...

Fri Apr 26, 2013, 08:21 PM

Venezuelan Audit Can’t Find Any Different Result in Presidential Election, Statistical Analysis Show

Venezuelan Audit Can’t Find Any Different Result in Presidential Election, Statistical Analysis Shows
Send to friend Printer-friendly version

By Center for Economic and Policy Research, April 26th 2013

A statistical analysis by the Center for Economic and Policy Research(CEPR) has shown that if Venezuelan opposition claims that Nicolás Maduro's victory was obtained by fraud were true, it is practically impossible to have obtained the result that was found in an audit of 53% of electronic voting machines that took place on the evening of Venezuela’s April 14 elections. The odds of this occurring would be far less than one in 25 thousand trillion.



“The U.S. government must know this, too,” said CEPR Co-Director Mark Weisbrot, economist and co-author of a forthcoming paper with economist and computer scientist David Rosnick. “So it is difficult to explain why they are refusing to recognize the elected president – in opposition to all of the countries in Latin America and most of the world.”

The results of Venezuela’s April 14 presidential election returned 7,575,506 votes for Nicolás Maduro, and 7,302,641 votes for challenger Henrique Capriles Radonski. This is a difference of 272,865 votes, or 1.8 percent of the two-way total between the candidates.

In this election, voters express their preference by pressing a computer touch-screen, which then prints out a paper receipt of their vote. The voter then checks to make sure that the receipt was the same as her choice, and deposits the paper receipt in a sealed box.

When the polls closed, a random sample of 53 percent <1> of all the machines (20,825 out of 39,303) was chosen, and a manual tally was made of the paper receipts. This “hot audit” was done on site, in the presence of the observers from both campaigns, as well as witnesses from the community. There were no reports from witnesses or election officials on site of discrepancies between the machine totals and the hand count.

Immediately after the election results were announced on the night of April 14th, the Venezuelan opposition demanded a full “recount” of all of the voting machines’ paper receipts and subsequently called for an audit – or manual count – of the 46% of the sealed boxes containing the paper receipts that had not yet been audited. After the Venezuelan Electoral Council’s (CNE’s) decision to grant their request, on April 18th, the main opposition party came up with a series of new demands suggesting that they did not believe that a full audit would provide evidence of any significant fraud. On April 26 they announced that they would “boycott” the audit that they had requested the previous week.

What if it were true that there were enough mismatches in the 39,303 machines to have given Maduro a 50.8 percent majority, when Capriles had been the true winner? CEPR calculated that the probability of getting the results of the first audit would then have been less than one in 25 thousand trillion.

“The results are pretty much intuitive,” said Weisbrot. “With a sample that huge verified during the April 14 ‘hot audit,’ if there were any discrepancies between the machine count and the paper ballots, it would have shown up somewhere. But it didn’t.”

It is therefore practically impossible that an audit of the remaining 46 percent of ballot boxes could find enough discrepancies to reverse the result of the election.

The forthcoming paper also calculates the probability that the remaining 46 percent of ballot boxes, if audited, could change the outcome. It also looks at other possible scenarios, including allegations from Capriles that there were irregularities in some 12,000 of the remaining machines, and other ways that the unaudited machines could have enough errors to change the result. The above calculation can be seen here. The full paper will be available next week.

<1> Another 1 percent was audited the next day.
Source: CEPR

This work is licensed under a Attribution Non-commercial No Derivatives Creative Commons license

http://venezuelanalysis.com/analysis/8867

7 replies, 1059 views

Reply to this thread

Back to top Alert abuse

Always highlight: 10 newest replies | Replies posted after I mark a forum
Replies to this discussion thread
Arrow 7 replies Author Time Post
Reply Venezuelan Audit Can’t Find Any Different Result in Presidential Election, Statistical Analysis Show (Original post)
Catherina Apr 2013 OP
bananas Apr 2013 #1
BethanyQuartz Apr 2013 #2
bemildred Apr 2013 #3
joshcryer Apr 2013 #5
bemildred Apr 2013 #6
naaman fletcher Apr 2013 #7
Peace Patriot Apr 2013 #4

Response to Catherina (Original post)

Fri Apr 26, 2013, 10:01 PM

1. “The U.S. government must know this, too,”

Of course they do.

Reply to this post

Back to top Alert abuse Link here Permalink


Response to Catherina (Original post)

Fri Apr 26, 2013, 10:30 PM

2. Hopefully

 

Hopefully this audit (which we sure could have used in 2000 and 2004 in the USA!) will put this to rest now and the opposition can go back to sucking on their diamond studded pacifiers. They still mostly get their way in Venezuela as it is.

Reply to this post

Back to top Alert abuse Link here Permalink


Response to Catherina (Original post)

Fri Apr 26, 2013, 10:55 PM

3. 53% is usually considered to be a pretty good sample size.

Reply to this post

Back to top Alert abuse Link here Permalink


Response to bemildred (Reply #3)

Sat Apr 27, 2013, 10:30 AM

5. 10% would be enough if observers were at every station.

But polling stations representing 700k observers didn't have observers from both sides.

Mark keeps harping on this but is neglecting that without observers the hot audit can trivially be gamed.

Reply to this post

Back to top Alert abuse Link here Permalink


Response to joshcryer (Reply #5)

Sat Apr 27, 2013, 12:49 PM

6. Better to randomize it (independently), you don't want to be able to predict the audited stations.

Or anything else about the audit.

But given that, what you need is a good sample size, say 1000 or so, any percentage that does that is sufficient (statistically).

And all the software should be open source with closed access, and every voter should be able to make a copy of the voting software they just voted on if they want it (transparency).

And then there are certain uses of encryption to assure identity and security, etc. Every machine, voter, and vote must have a unique identity, assigned as part of the preparation for the election.

And the machines should be "read only" and lacking in all network capability. Running off identical SIM cards would be good, for example, then you could just pull the SIM and send it to the counting venue, they check it's not been tampered with, count the votes on it, verify any audit checks, etc.

I don't know anybody that is doing it right now though, it's almost like politicians don't WANT to give up their ability to fudge elections.

Reply to this post

Back to top Alert abuse Link here Permalink


Response to bemildred (Reply #6)

Sat Apr 27, 2013, 12:52 PM

7. exactly

 

it's almost like politicians don't WANT to give up their ability to fudge elections.

yup

Reply to this post

Back to top Alert abuse Link here Permalink


Response to Catherina (Original post)

Sat Apr 27, 2013, 02:54 AM

4. Know what kind of audit we have here? NO AUDIT AT ALL in half the states and...

...a miserable 1% audit in the other half.

Know how much usually gets audited in a recount, here (if you can afford the money and lawyers to get a recount)? 3%.

Know who owns the programming code in these UNAUDITED and miserably inadequately audited voting machines, all over the U.S., in every state? 75% owned and controlled by ONE, PRIVATE, FAR RIGHTWING CONNECTED CORPORATION--ES&S which bought out Diebold.

Know what this code is called? It is called 'TRADE SECRET' code! The public is forbidden to review it!

Know what Venezuela's code is called? OPEN SOURCE code! It is owned by the PUBLIC and anyone may review it.

Know how much of an audit experts say is needed to detect fraud in electronic systems? At least 10%!

Venezuela is doing an audit that is more than FIVE TIMES the amount needed to detect fraud!

The U.S. government demanding a 100% audit in Venezuela is "Alice in Wonderland-ish" hypocrisy.

“'The U.S. government must know this...' said CEPR Co-Director Mark Weisbrot...".

He's damn right they know it. EVERY Democratic Party/Obama administration POLICY is designed with one thing in mind: the noose that ES&S/Diebold has around their necks.

One, private corporation, with far rightwing connections that would make your hair stand on end, designed the U.S. vote counting system (um, vote rigging system) for profit and for power, and operates it as a monopoly!

And they dare to criticize Venezuela!?

Gawd.





Reply to this post

Back to top Alert abuse Link here Permalink

Reply to this thread