Read Story Here
President Obama has stepped up his tough talk against the Iranian government’s violent acts toward the dissent of their citizens. On the same day Professor Walter Mebane Jr. has elevated the argument claiming widespread fraud in this latest election by dropping a helpful dose scientific analysis.
Mebane from the Department of Statistics of the University of Michigan is the latest in a long line of reputable pollsters and political scientists to claim the results of the latest Iranian election are rife with fraud and inconsistencies. Yesterday a 24-page document was released detailing what Mebane believes to be is strong evidence towards a major election swindle.
News flash to all! Mahmoud Ahmadinejad declared election victor and sitting Iranian president is probably not the fair and rightful winner.
In his document Mebane, with translation assistance from Dr. Richard Beam, carefully dissects several key data irregularities. It is complex stuff to be certain and a full reading of the PDF/text can be located at following the link: http://www-personal.umich.edu/%7Ewmebane/note22jun2009.pdf. Mebane’s overall conclusion however is a simple one, suggesting that data obtained gives “moderately strong support for a diagnosis that the 2009 election was affected by significant fraud.”
Statistical distribution is questioned, regressions are defined and outliers are highlighted. Being as how this is Iran with a limited amount of data observable to outsiders, Mebane’s task is a particularly difficult one. Mebane locates dozens of outliers and his regression models point toward extremely odd results coming from votes cast for three of the four candidates. If in fact a mass conspiracy is being orchestrated by the government of Iran the flaw in their cover-up plan might be how poorly the recent data purported correlates with that of the country's 2005 Presidential Election.
It seems fairly logical to assume, for instance, that areas of Iran that heavily supported Ahmadinejad in 2005 would, in the instance of a close election or landslide victory for the incumbent President, continue to produce big numbers for him in 2009. Furthermore there are issues with the dispersal of votes and overall voter turnout. The numbers for 2009 were significantly higher than four years ago in part because of a boycott by many in that election. Certain voting areas in Iran that were supportive of Ahmadinejad in 2005 would have to make up a smaller percentage of the overall pie chart than in 2009.
Pairing variables between both the 2005 and 2009 election totals shows several alarming discrepancies however. Mebane has created several statistical models many of which point to implausible “outlier” scenarios. Outliers are like red flag data that break from natural numerical fluidity to distort overall totals. Mebane’s models “wholly fail to describe” these outliers in 60 of 81 vote counts for Ahmadinejad, and in most instances the Iranian President did much better than the models would predict.
Mebane from the Department of Statistics of the University of Michigan is the latest in a long line of reputable pollsters and political scientists to claim the results of the latest Iranian election are rife with fraud and inconsistencies. Yesterday a 24-page document was released detailing what Mebane believes to be is strong evidence towards a major election swindle.
News flash to all! Mahmoud Ahmadinejad declared election victor and sitting Iranian president is probably not the fair and rightful winner.
In his document Mebane, with translation assistance from Dr. Richard Beam, carefully dissects several key data irregularities. It is complex stuff to be certain and a full reading of the PDF/text can be located at following the link: http://www-personal.umich.edu/%7Ewmebane/note22jun2009.pdf. Mebane’s overall conclusion however is a simple one, suggesting that data obtained gives “moderately strong support for a diagnosis that the 2009 election was affected by significant fraud.”
Statistical distribution is questioned, regressions are defined and outliers are highlighted. Being as how this is Iran with a limited amount of data observable to outsiders, Mebane’s task is a particularly difficult one. Mebane locates dozens of outliers and his regression models point toward extremely odd results coming from votes cast for three of the four candidates. If in fact a mass conspiracy is being orchestrated by the government of Iran the flaw in their cover-up plan might be how poorly the recent data purported correlates with that of the country's 2005 Presidential Election.
It seems fairly logical to assume, for instance, that areas of Iran that heavily supported Ahmadinejad in 2005 would, in the instance of a close election or landslide victory for the incumbent President, continue to produce big numbers for him in 2009. Furthermore there are issues with the dispersal of votes and overall voter turnout. The numbers for 2009 were significantly higher than four years ago in part because of a boycott by many in that election. Certain voting areas in Iran that were supportive of Ahmadinejad in 2005 would have to make up a smaller percentage of the overall pie chart than in 2009.
Pairing variables between both the 2005 and 2009 election totals shows several alarming discrepancies however. Mebane has created several statistical models many of which point to implausible “outlier” scenarios. Outliers are like red flag data that break from natural numerical fluidity to distort overall totals. Mebane’s models “wholly fail to describe” these outliers in 60 of 81 vote counts for Ahmadinejad, and in most instances the Iranian President did much better than the models would predict.
More than half of the 320 towns included in this part of the analysis exhibit
vote totals for Ahmadinejad that are not well described by the natural politic
process…these departures from the model much more often represent additions
rather than declines in the vote reported for Ahmadinejad. Correspondingly the
poorly modeled observations much more often represent declines than additions in
the votes reported for Mousavi – Walter Mebane, Jr. / University of
Michigan
With data available for thirty provinces Mebane can claim to the best of his knowledge that the 2009 election was in fact affected by significant instances of fraud. That said he does take into account that much of the possible fraud detected by his statistical models ride on oddities not only by the major parties but lower vote receiving candidates such as Mehdi Karoubi and Mohsen Razaei.
Outliers are a regular part of the polling process. The abundance of such can be seen even in instances such as our most recent Presidential election in 2008. Using data obtained from recent and similar American election cycles in 2000 and 2004 Barack Obama’s performance in traditionally strong Republican states like Virginia, North Carolina and Indiana would cause any pollster to pause at first glance. It should also be considered that unlike the normal process in this country there wasn’t exactly decent pre-election or sufficient exit polling to gather from in Iran.
Still Mebane can claim with reasonable confidence that while the data sets point to justified voting totals for Mousavi, they appear to show a range of distortions for Ahmadinejad and the two minor candidates. “The problem with the 2009 Iranian election” as Mebane states in closing, “is that the serious questions that have been raised are unlikely to receive satisfactory answers. Transparency is utterly lacking in this case. There is little reason to believe the official results announced in that election accurately reflect the intentions of the voters who went to the polls.”
No comments:
Post a Comment