MATH: PROVES ELECTION FRAUD
Jun. 24th, 2009 03:12 pm![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
defrogI haven’t had much to say about the post-election furor in Iran, if only because what’s going on there isn’t all that radically different from elections we’ve seen in Kenya, Zimbabwe, East Timor, Cambodia, Florida and other countries where either the results are suspicious or the opposition refuses to accept the idea that they lost, because that’s Simply Not Possible. And no one really cared about those (except possibly Florida), though granted, we didn’t have Twitter then.
Anyway, that’s not to say that the Iranian election results aren’t suspicious. But when it’s yr word against the Guardian Council, how do you prove that votes were rigged?
With statistical mathematics!
According to a couple of PhD candidates at Columbia U, the biggest clue of election fraud in Iran is the official vote counts from different provinces – specifically, the last two digits of each result.
At the root of this is the fact that humans are generally bad at making up numbers.
Which means that when you crunch the statistical likelihood of vote counts arriving at the exact numbers they did, the results better fit the pattern of someone making up numbers.
DISCLAIMER: I barely passed algebra in high school. So don’t ask me if any of it’s true. But it’s an interesting idea.
Add it up,
This is dF
Anyway, that’s not to say that the Iranian election results aren’t suspicious. But when it’s yr word against the Guardian Council, how do you prove that votes were rigged?
With statistical mathematics!
According to a couple of PhD candidates at Columbia U, the biggest clue of election fraud in Iran is the official vote counts from different provinces – specifically, the last two digits of each result.
At the root of this is the fact that humans are generally 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 ....
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.
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.
Which means that when you crunch the statistical likelihood of vote counts arriving at the exact numbers they did, the results better fit the pattern of someone making up numbers.
DISCLAIMER: I barely passed algebra in high school. So don’t ask me if any of it’s true. But it’s an interesting idea.
Add it up,
This is dF
![[identity profile]](https://www.dreamwidth.org/img/silk/identity/openid.png){kind=small}
no subject
on 2009-06-24 09:59 am (UTC)I did well in maths, and I took just enough statistics to be able to identify when a proper statistical analysis has been performed or when a hack journo has just muddled the numbers around to show a story that they want, which is what you see in the vast majority of non-scientific journalism all across the world.
The article you linked looks like one of the few that has a proper mathematical backing. While it's hard to know for sure without seeing exactly what data they crunched or how they crunched it, it sounds like they did a proper and rigorous analysis. The .005 probability cited at the bottom of the article is very much the standard used to determine statistical significance.
In other words: yes, this is very good evidence that the election was rigged.
-- JF