Monday, June 15, 2009

Early Statistical Analysis Casts Doubts on Results

Renard Sexton at FiveThirtyEight, has posted an early statistical analysis of the election results, noting that "given the polling data in the run-up to the balloting, and the historical trend away from electoral domination in the first round by one candidate, this very fishy regional data tends to strongly support" the assertions of voting irregularities.

Specifically, Sexton notes that the historical trend in Iran is that the higher the turnout, the closer the election ("decreasing first round victor percentages"). On another issue that has been widely noted by commentators, Sexton asserts, that the provincial returns were highly unlikely to occur; comparing Karroubi and Ahmadinejad's provincial results from 2005 and 2009, Ahmadinejad beat his opponent by a factor of 10 in provinces Karoubi won handily in 2005. Thus, "Not only did Ahmadinejad beat Karroubi in his base of support, he crushed him beyond all recognition." Sexton notes that while his analysis is not definitive proof of wrongdoing, it is highly suggestive of suspect results.

No comments: