Comments on: Statistical evidence for Iranian election fraud? http://election.princeton.edu/2009/06/18/flash-statistical-evidence-for-iranian-election-fraud/ A first draft of electoral history Thu, 12 Nov 2009 02:35:25 -0500 http://wordpress.org/?v=2.8.6 hourly 1 By: Kuchikomi Prediction Analyses 2009 Japan Lower House Elections | achikule! http://election.princeton.edu/2009/06/18/flash-statistical-evidence-for-iranian-election-fraud/comment-page-1/#comment-3212 Kuchikomi Prediction Analyses 2009 Japan Lower House Elections | achikule! Tue, 06 Oct 2009 09:31:23 +0000 http://election.princeton.edu/?p=3086#comment-3212 [...] Another reason we should care about predictions is for data checking. Recently, Nate Silver of FiveThirtyEight accused Strategic Vision (a polling company) of having some shoddy, suspicious polling (Nate has huge balls doing this alone). Predictions can be applied to different datasets, and it usually outputs a general trend. When one results in a departure from the trend, it gives some evidence to look into information rigging. This came helpful in the recent Iranian election. [...] [...] Another reason we should care about predictions is for data checking. Recently, Nate Silver of FiveThirtyEight accused Strategic Vision (a polling company) of having some shoddy, suspicious polling (Nate has huge balls doing this alone). Predictions can be applied to different datasets, and it usually outputs a general trend. When one results in a departure from the trend, it gives some evidence to look into information rigging. This came helpful in the recent Iranian election. [...]

]]> By: The Devil Is in the Digits - Statistical analysis of the Iranian election results http://election.princeton.edu/2009/06/18/flash-statistical-evidence-for-iranian-election-fraud/comment-page-1/#comment-3177 The Devil Is in the Digits - Statistical analysis of the Iranian election results Tue, 23 Jun 2009 01:15:05 +0000 http://election.princeton.edu/?p=3086#comment-3177 By: C L O S E R » Blog Archive » Closing the week 26 - Featuring Iran http://election.princeton.edu/2009/06/18/flash-statistical-evidence-for-iranian-election-fraud/comment-page-1/#comment-3170 C L O S E R » Blog Archive » Closing the week 26 - Featuring Iran Sun, 21 Jun 2009 01:35:17 +0000 http://election.princeton.edu/?p=3086#comment-3170 [...] Statistical evidence for Iranian election fraud? Prof. Walter Mebane at the University of Michigan is knowledgeable about such analysis and is applying the methods to the data from the recent Iranian election. In addition to intrinsic peculiarities such as Benford’s law, he is also using 2005 election data as a baseline to help discover unexpected anomalies. [...] [...] Statistical evidence for Iranian election fraud? Prof. Walter Mebane at the University of Michigan is knowledgeable about such analysis and is applying the methods to the data from the recent Iranian election. In addition to intrinsic peculiarities such as Benford’s law, he is also using 2005 election data as a baseline to help discover unexpected anomalies. [...]

]]> By: The BRAD BLOG : 'Daily Voting News' for June 19, 2009 http://election.princeton.edu/2009/06/18/flash-statistical-evidence-for-iranian-election-fraud/comment-page-1/#comment-3167 The BRAD BLOG : 'Daily Voting News' for June 19, 2009 Fri, 19 Jun 2009 23:18:41 +0000 http://election.princeton.edu/?p=3086#comment-3167 [...] Iran: Statistical evidence for Iranian election fraud? http://election.princeto...-iranian-election-fraud/ [...] [...] Iran: Statistical evidence for Iranian election fraud? http://election.princeto…-iranian-election-fraud/ [...]

]]> By: Sam Wang http://election.princeton.edu/2009/06/18/flash-statistical-evidence-for-iranian-election-fraud/comment-page-1/#comment-3164 Sam Wang Fri, 19 Jun 2009 13:26:18 +0000 http://election.princeton.edu/?p=3086#comment-3164 Note that applying Benford's law for the first digit might be a problem depending on how districts are set up. If they are designed to be of approximately equal size then there might be a bias. For example, in the report from Poland (I'll blog about this later) the preponderance of leading 7's could be explained this way. This is why Mebane is using the second-digit Benford's law, a subtler but probably more reliable test. Note that applying Benford’s law for the first digit might be a problem depending on how districts are set up. If they are designed to be of approximately equal size then there might be a bias. For example, in the report from Poland (I’ll blog about this later) the preponderance of leading 7’s could be explained this way.

This is why Mebane is using the second-digit Benford’s law, a subtler but probably more reliable test.

]]> By: David Shor http://election.princeton.edu/2009/06/18/flash-statistical-evidence-for-iranian-election-fraud/comment-page-1/#comment-3161 David Shor Fri, 19 Jun 2009 08:28:27 +0000 http://election.princeton.edu/?p=3086#comment-3161 Hossein, Once he conditioned the results on 2005 first round results that became available to him, he found discrepancies that made him change his conclusion. Hossein,

Once he conditioned the results on 2005 first round results that became available to him, he found discrepancies that made him change his conclusion.

]]> By: Hossein http://election.princeton.edu/2009/06/18/flash-statistical-evidence-for-iranian-election-fraud/comment-page-1/#comment-3155 Hossein Thu, 18 Jun 2009 21:28:41 +0000 http://election.princeton.edu/?p=3086#comment-3155 Hi, Could you please check this out? Prof. Mebane said before that result conform Benford's law, and in the new edited version, he says that it seems to be a fraud. Hi,
Could you please check this out? Prof. Mebane said before that result conform Benford’s law, and in the new edited version, he says that it seems to be a fraud.

]]>