Chatham House Study: The 2009 Iran Presidential Election

[Winehole23]

This paper is published by Chatham House
and the Institute of Iranian Studies,
University of St Andrews

Preliminary Analysis of the Voting
Figures in Iran’s 2009 Presidential Election
(http://www.chathamhouse.org.uk/files/14234_iranelection0609.pdf)
Editor:
Professor Ali Ansari, Director, Institute of Iranian Studies,
University of St Andrews; Associate Fellow, Middle East and
North Africa Programme, Chatham House; author, ‘Iran,
Islam and Democracy’

19 pages. Via NYT.





1. Irregularities in Voter Turnout

Two provinces show a turnout of over 100% and four more show a turnout of over 90%. Regional variations in participation have disappeared. There is no correlation between the increase in participation, and the swing to
Ahmadinejad. Firstly, there is a massive across the board increase in turnout, with several provinces increasing their participation rate by nearly 75%. This increase results in substantially less variation in turnout between provinces, with the standard deviation amongst provincial turnouts falling by just over 23% since 2005. The 2005 results show a substantial turnout gap, with seven provinces recording turnout below 60%, and ten above 70%. In 2009, only two were below 70% and 24 were above 80%. In fact, 21 out of 30 provinces had turnouts within 5% of 83%. The data seems to suggest that regional variations in participation have suddenly disappeared.

This makes the evident lack of any sort of correlation between those provinces that saw an increase in turnout and those that saw a swing to Ahmadinejad(see Fig.1) all the more unusual. There is no significant correlation between the increase in participation for a given province, and the swing to Ahmadinejad (see Fig.1). This lack of correlation makes the argument that increased participation by a previously silent conservative majority won the election for Ahmadinejad somewhat problematic.

Furthermore, there are concerns about the numbers themselves. Two
provinces, Mazandaran and Yazd, have results which indicate that more votes were cast on 12 June than there were eligible voters, and that four more provinces had turnouts in the mid-nineties.

In a country where allegations of ‘tombstone voting’ – the practice of using the identity documents of the deceased to cast additional ballots – are both longstanding and widespread, this result is troubling but perhaps not unexpected. This problem did not start with Ahmadinejad; according to official statistics gathered by the International institute for Democracy and Electoral Assistance in Stockholm, there were 12.9% more registered voters at the time of Mohammed Khatami’s 2001 victory than there were citizens of voting age.

[Winehole23]

Nyt:


the study also showed that in a third of all provinces, the official results, if true, would have required that mr. Ahmadinejad win not only all conservative voters and all former centrist voters and all new voters, but up to 44 percent of formerly reformist voters.

[Winehole23]

Raimundo (http://original.antiwar.com/justin/2009/06/21/iran%E2%80%99s-green-revolution-made-in-america/) on the election.

[Winehole23]

Election numerology (http://www.washingtonpost.com/wp-dyn/content/article/2009/06/20/AR2009062000004.html):


We'll concentrate on vote counts -- the number of votes received by different candidates in different provinces -- and in particular the last and second-to-last digits of these numbers. For example, if a candidate received 14,579 votes in a province (Mr. Karroubi's actual vote count in Isfahan), we'll focus on digits 7 and 9.

This may seem strange, because these digits usually don't change who wins. In fact, last digits in a fair election don't tell us anything about the candidates, the make-up of the electorate or the context of the election. They are random noise in the sense that a fair vote count is as likely to end in 1 as it is to end in 2, 3, 4, or any other numeral. But that's exactly why they can serve as a litmus test for election fraud. For example, an election in which a majority of provincial vote counts ended in 5 would surely raise red flags.

Why would fraudulent numbers look any different? The reason is that humans are bad at making up numbers. Cognitive psychologists have found that study participants in lab experiments asked to write sequences of random digits will tend to select some digits more frequently than others.

So what can we make of Iran's election results? We used the results released by the Ministry of the Interior and published on the web site of Press TV, a news channel funded by Iran's government. The ministry provided data for 29 provinces, and we examined the number of votes each of the four main candidates -- Ahmadinejad, Mousavi, Karroubi and Mohsen Rezai -- is reported to have received in each of the provinces -- a total of 116 numbers.

The numbers look suspicious. We find too many 7s and not enough 5s in the last digit. We expect each digit (0, 1, 2, and so on) to appear at the end of 10 percent of the vote counts. But in Iran's provincial results, the digit 7 appears 17 percent of the time, and only 4 percent of the results end in the number 5. Two such departures from the average -- a spike of 17 percent or more in one digit and a drop to 4 percent or less in another -- are extremely unlikely. Fewer than four in a hundred non-fraudulent elections would produce such numbers.

As a point of comparison, we can analyze the state-by-state vote counts for John McCain and Barack Obama in last year's U.S. presidential election. The frequencies of last digits in these election returns never rise above 14 percent or fall below 6 percent, a pattern we would expect to see in seventy out of a hundred fair elections. But that's not all. Psychologists have also found that humans have trouble generating non-adjacent digits (such as 64 or 17, as opposed to 23) as frequently as one would expect in a sequence of random numbers. To check for deviations of this type, we examined the pairs of last and second-to-last digits in Iran's vote counts. On average, if the results had not been manipulated, 70 percent of these pairs should consist of distinct, non-adjacent digits.

Not so in the data from Iran: Only 62 percent of the pairs contain non-adjacent digits. This may not sound so different from 70 percent, but the probability that a fair election would produce a difference this large is less than 4.2 percent. And while our first test -- variation in last-digit frequencies -- suggests that Rezai's vote counts are the most irregular, the lack of non-adjacent digits is most striking in the results reported for Ahmadinejad.

[Winehole23]

Guardian Council, via Iranian state press :



Iran's Guardian Council has suggested that the number of votes collected in 50 cities surpass the number of people eligible to cast ballot in those areas.

The council's Spokesman Abbas-Ali Kadkhodaei, who was speaking on the Islamic Republic of Iran Broadcasting (IRIB) Channel 2 on Sunday, made the remarks in response to complaints filed by Mohsen Rezaei -- a defeated candidate in the June 12 Presidential election.

"Statistics provided by the candidates, who claim more than 100% of those eligible have cast their ballot in 80-170 cities are not accurate -- the incident has happened in only 50 cities," Kadkhodaei said.

The spokesman, however, said that although the vote tally affected by such an irregularity could be over 3 million and the council could, at the request of the candidates, re-count the affected ballot boxes, "it has yet to be determined whether the possible change in the tally is decisive in the election results," reported Khabaronline (http://www.khabaronline.ir/news-11204.aspx)

[jman3000]

At this point it's pretty much a universal truth that the vote was fraudulent. Only Ahmadinejad and the Ayatollah are claiming otherwise. If it wasn't for the regime constantly insulting the intelligence of the Iranian people it might not have been this bad... but they did what has traditionally worked in the past and it's bit them in the ass.

That look into the number endings is interesting, but ultimately I think it's pretty spurious.

[Winehole23]

It's the curse of analysis to dog the steps of common sense and to tell us little more.

So yeah. Crunching the numbers from Iran's Interior Ministry ratifies the I told you so's, long after most people were already convinced.

[Winehole23]

Still, you don't hang the outlaw before the verdict, even though you may be sure of it in advance.

[MannyIsGod]

Fivethirtyeight.com had these guys beat by at least a weak on the analysis. Nate Silver is on the ball, and he came to the same rough conclusions. I thought I posted it here but maybe not.

[Winehole23]

Fivethirtyeight.com had these guys beat by at least a week on the analysis. Nate Silver is on the ball, and he came to the same rough conclusions. I thought I posted it here but maybe not.Depends on which guys you mean, but yeah, I found some evidence for that. I don't think anybody's posted Nate Silver's bits on it yet, so here's a mini-round up.

There's this (http://www.fivethirtyeight.com/2009/06/ayatollahs-flawed-logic.html):



But this is simply not the case in Iran. All votes are counted are reported by the Interior Ministry. There is no other source of information. There are no election monitors. Nor does the fraud alleged involve any sort of physical process (e.g. stuffing ballot boxes). It is simply a matter of changing numbers on a spreadsheet. Under these conditions, it is essentially no more difficult to steal a thousand votes than one, a million than a thousand, or 11 million than one million.

This (http://www.fivethirtyeight.com/2009/06/karroubis-unlucky-7s.html), which echoes the WaPo piece upstream and jman's skepsis:



The reason I'm holding back from fully endorsing this is because it's not clear to me whether this particular distribution should indeed obey Benford's Law in the first place.


It's not clear to me whether the voting units that Iran's Interior Ministry reported on behave more like towns, in which case we might expect the voting distinctions to obey Benford's Law, or more like precincts, in which case we probably wouldn't. The way the units are described to me in the spreadsheet I'm working from are "city/county", which implies that sufficiently large cities are treated as their own units, whereas smaller ones -- it looks to me like perhaps those that have fewer than about 15,000 people -- have their results aggregated at some level resembling American counties. If there are these sorts of artificial constraints placed on the size of the reporting units, we might expect some anomalies from a Benford's Law perspective.

This (http://www.fivethirtyeight.com/2009/06/ahmadinejads-rural-votes.html) foreshadows the Chatham House conclusion about rural tallies:



So it's not exactly correct to say that Ahmadinejad's strength was in rural areas. What we certainly can say, however, is that almost all of the improvements that Ahmadinejad made over his 2005 totals came in rural areas. What was once a weakness of his turned into another strength.

This means that at least one of two things must be true. Either the urban-rural dynamics of Iran have changed significantly over the last four years -- at least insofar as it they affected perceptions of a candidate like Ahmadinejad. Or, alternatively, the election was rigged, and those who rigged it for some reason decided that rural votes were easier to steal.

Silver's multiple regression (http://www.fivethirtyeight.com/2009/06/if-he-did-it.html) analysis:



But with that disclaimer, at least one result is rather surprising. Namely, Ahmadinejad appeared to pick up most of the vote from Mehdi Karroubi, who is routinely described (http://www.voanews.com/english/2009-06-10-voa72.cfm) as the most liberal of the candidates. This is in spite of the fact that Karroubi himself was on the ballot this year; he appeared to retain only about 5 percent of his own vote.

Renard already detected (http://www.fivethirtyeight.com/2009/06/iran-does-have-some-fishy-numbers.html) this pattern, so I am not really telling you anything new. But it seems to me to be the key to explaining the Iranian election -- whether it was legitimate or whether it was rigged. Ahmadinejad won all of the provinces that Karroubi won in 2005, and his cumulative share of the vote in these 11 provinces was 66 percent, exceeding his overall total. In the province where Karroubi did best in 2005, his home province of Lorestan, Ahmadinejad got some 71 percent of the vote.

Now, most of those provinces where Karrobui did well in 2005 are rural, and it's possible that the rural tilt toward conservative candidates was greater in 2009 than in 2005. As I mentioned earlier (http://www.fivethirtyeight.com/2009/06/iranian-election-results-by-province.html), Ahmadinejad actually didn't do that badly in Iran's most urban province, Tehran, in 2005, but he performed relatively poorly there last week. If Ahmadinejad won the election, he did it by winning over these rural Karrobui voters. And if he stole it, those were the votes he stole or intimidated.

A secondary factor in Ahmadinejad's purported success was his ability to capture most of the vote from Akbar Hashemi Rafsanjani, a moderate (by Iranian standards) whom Ahmadinejad defeated in the 2005 run-off. Rafsanjani won three provinces in the first round in 2005: Kerman, Zanjan and Gilan. Ahmadinejad ostensibly won all three, grabbing 78 percent of the vote in Kerman, 77 pertcent Zanjan and 68 percent in Gilan. Kerman, indeed -- Rafsanjani's best province in 2005 -- was Ahmadinejad's best in 2009.

This (http://www.fivethirtyeight.com/2009/06/iranian-election-results-by-province.html) foreshadows Chatham House because Silver got info from a researcher. at St. Andrews.

This entry debunks (http://www.fivethirtyeight.com/2009/06/statistical-evidence-does-not-prove.html) Tehran Bureau's statistical analysis. That analysis has been scrubbed (http://tehranbureau.com/2009/06/13/faulty-election-data/) from the TB website.