A huge quantity of hot air has been expended arguing the details of the most recent Russian elections process. A host of groups, notably including several European elections monitoring groups, have reported widespread irregularities in the elections process and have called for investigations. The Russian elections board, for their own part, has ruled that any irregularities were minor and were neither severe or wide-spread enough to call into question the outcome of the election.
Any number of people could continue to argue this issue until they were quite blue in the face. Sometimes, though, all it really takes is a picture. It’s just a simple graph, showing the relationship between the number of people who voted in the Russian election (the reported percentage of voters who participated in each jurisdiction) with the party their vote favored (the reported percentage of voters who cast their ballot for each party in the jurisdiction).
Although the graph doesn’t reveal any great insights into the mechanics, it does serve as a compelling argument for the validity of claims that there were huge irregularities in the election. This distribution of data does not represent a fair and free election. Even worse, in my mind, is that it seems to show a situation in which the people perpetrating the fraud couldn’t even be bothered to hide it.
The troubling thing about the distribution of the data on the graph is that, generally speaking, the higher the reported voter turnout in a given district, the greater the percentage of voters who cast their ballot for United Russia (the party lead by Vladimir Putin). Think about this for a moment. It’s not that when more people voted a larger number of them voted for Putin. This would be expected and normal. Instead, when more people voted, a greater proportion of them voted for Putin.
This is troubling, because in any “normal” distribution of these data points, the percentage of votes cast for a given party would not change as the number of voters increased. If half of the people in the country favor Party A, and half favor Party B, you’d expect them to each get half of the votes. If more voters turn out in one district, both parties still get roughly half of the (larger number of) votes, because changing the number of participants does not have any effect on the relative popularity of the parties.
This becomes even more troubling when one discovers that, for example, Chechnya reported a staggering 90% voter turnout. This is a region that has been engaged in a simmering, at times deadly conflict with the Russian central authorities for years. Exit polls in Chechnya, conducted by outside organizations, reported that real voter turnout was vanishingly small. In spite of this, the region’s electoral board reported massive voter turnout with an overwhelming majority of voters favoring the very political regime that so many of their fellow Chechen’s have dedicated their lives and sacrificed their blood to oppose.
As with any statistical analysis, this one remains open to a huge amount of interpretation. I have no doubt that an impressive supply of further hot air will be dedicated to the task of evaluating the merits of any particular conclusion based on this data. If you are interested, I highly encourage you to investigate this information more fully. A good place to start might be this interpretation, published by the same person who produced the graph. There are certainly others available, and I don’t mean to suggest that this one in particular is definitively more correct than any other.
What’s most frustrating to me, I think, is that in spite of evidence as compelling as this that whatever the details, something is amiss with the election results, the story has already vanished from the headlines. A review of the BBC’s world news doesn’t mention the election at all, even with even the U.S. State Department calling for investigation into the results. Most people, including, as far as I can tell, most Russians, have shrugged their shoulders and turned their attention to other things.
This whole state of affairs worries and saddens me, and I wish there was something constructive I could do. Instead, I will have to content myself with playing with the data. As much as I’d like to say that it is critically important that voter fraud be carried out in such a way as to make it difficult to detect, it would seem that even the simple measures of covering your own tracks are a needless nod towards respectability.
-posted by Mark