Those who have been following the debate on the evidence for vote rigging should find this article interesting as it is critical of the analysis of Chatham House which has been at the forefront of those providing evidence for vote rigging. Of course most of the media never bother to look at so sophisticated a source as Chatham House, they can do with sound bites from the protesters and supporters or simplistic bits such as that concerning voting of over 100 per cent in some areas. As if the Iranian authorities themselves would admit this if it showed in itself that there were vote rigging!
This is from AsiaTimes
Jun 26, 2009
COMMENT Crunching the numbers
By Kaveh L Afrasiabi
A few days ago, just as the "color" movement's ferocious struggle to overturn the results of the 10th Iranian presidential elections was fading, it received a new lease of life via the publication of a British study  that casts serious doubt on the official results that saw President Mahmud Ahmadinejad re-elected. "Preliminary Analysis of the Voting Figures in Iran's 2009 Presidential Election" was published by Chatham House and the Institute of Iranian Studies, University of St Andrews, and edited by Iranian political scientist Professor Ali Ansari, director, Institute of Iranian Studies. The report has received a lavish reception in the Western media as a "sweeping condemnation" of the June 12 election results, by
virtue of repeatedly using such terms as "implausible" and "highly doubtful" in reference to aspects of the returned numbers from the nearly 40 million votes that were cast. The report identifies the "massive increase from 2005" as one of a "number of aspects" of the election as being "problematic". The authors question that the incumbent president could win 7 million more votes than he received the last time. Yet they overlook that his votes were extremely close to his voting percentage in 2005. One of the problems could be that the main author has no background in quantitative analysis as he is a qualitative political scientist. Compare this with another political scientist, US statistician Professor Walter Mebane, a leading expert on election fraud, who has made a similar statistical analysis of the Iranian election. He concluded that there is "no solid evidence of fraud". Another US statistic guru, Nate Silver, has concluded that the voting result was "valid based on statistical analysis". According to Mebane, who compared 366 district results with those for the 2005 elections, the "substantial core" of posted results are in line with the basic statistical trends. One of Mebane's conclusions is that "Ahmadinejad tended to do worst in towns where the turnout surged the most". Ansari reaches the opposite conclusion in making the same comparisons. Mebane has made the observation that "a model can never prove fraud - it can identify places where there may be fraud". Ansari's report presents his charts and figures as definitive statements on the election result. This raises questions over the timing of Ansari's report, in light of allegations by Iran of British meddling in the post-election turmoil in Iran sparked by supporters of the losing candidates. Iranian Foreign Minister Manouchehr Mottaki said on Wednesday Iran may downgrade ties with Britain, accusing London of meddling. The announcement came a day after British Prime Minister Gordon Brown said two Iranian diplomats had been expelled in a tit-for-tat move after Tehran ordered two British diplomats to leave. The Ansari study appears to want by the sheer force of its charts and figures to establish beyond doubt the fact of an election fraud, even though after two weeks the disgruntled candidates have failed to provide any tangible evidence. This despite the fact that the leading losing candidate, Mir Hossein Mousavi, had some 40,676 observers at the ballot boxes, and none has provided a formal complaint. At the outset of the study, the authors cite the abnormality of two areas having excess votes of more than 100%. Iran's Guardians Council, which has oversight of the elections, has identified some 50 towns which had more votes cast than their registered voters. The council points out that in some areas, such as Shemiranat, Mousavi won and that most of the towns are in the Caspian Sea resort, meaning the discrepancy could be attributable to heavy summer tourism combined with the result of a bureaucratic glitch with the Census Bureau. Having found no evidence of "major irregularity", the council has all but rejected the idea of annulling the votes. The Chatham House study says Iranians voted according to ethnic identities, claiming that this has been the case with the Azeris in all past elections. Yet in the 2005 elections, an Azeri candidate, Mehr Alizadeh, received only 28% of the votes in the province of East Azerbaijan. A weakness of the report is that that despite the lack of specific rural voting data in previous elections, remedied this year for the first time, the study claims privileged knowledge that neither in 2005 nor in 2009 did Ahmadinejad carry rural Iran. This year, Ahmadinejad did better than his reformist rivals in the "deprived" provinces of Chahar Mahal, South Khorasan and Kerman. Absent in the study is any reference to related works and findings, such as pre-election opinion sampling by pollsters Ballen and Doherty, who found that Ahmadinejad would win by a two to one margin and that only 16% of Azeris would vote for Mousavi. "Election results in Iran may reflect the will of Iranian people," they have written in the Washington Post.