The Lie of voter Fraud

Politifact did an interesting, though complicated piece about the lie of voter fraud.  You can read it here.

A synapses of it goes as follows:

• The 2010 census had 98,213 people over the age of 18 in Wood County, Ohio.
• The 2012 Voter Registrar lists 106,258 people as of Sept 17.  (Up to 108,014 as of the Politifact published date).
• A bunch of fools are insinuating that 106,258 people voted for Obama in Wood County in the 2012 election, being an obvious example of voter fraud.  This is an outright lie.
• The truth is that only 80,433 people are listed as 'active voters', and of those only 62,338 voted this November 6th.  
• Of the people that voted, only 31,596 voted for Obama.  The other 28,997 voted for Romney.

So, the idiots implying that there was massive, voter fraud are clearly trying to deceive people with the wrong numbers.  But let's talk about the right numbers.

Why are there 8,045 more people on the Voter Registrar list than in the census?

There are lots of reasons this happens.

1. Two years passed, and people aged.  Someone that was 16 in 2010 would not count in that 98,213 number. 
2. Some people moved into the county.
3. People that died did not get immediately removed from the list.
4. People that moved did not re-registrar to vote, removing themselves from the list.
5. College students may tell the Census that they 'live' with their parents, but register to vote in the the county they go to school.  (They may think that their parent's student financing need them to be claimed as a resident)

First, which way should we be cautious about.  Should we try to eliminate everyone we can from the voter registration list- preventing fraud?  Or should we be extra careful not to violate someone's voting rights by eliminate them unless we know that they are no longer an eligible voter?  Well, as there are practically no convictions or even arrests for voter fraud, we should obviously protect their voting rights - just as we protect our rights to own a gun or practice religion.   If on the other hand you can show significant numbers of arrests and convictions for voter fraud, then I could see changing those rules

College students are hard to measure, as are moves.  But birhts and deaths are public records easily looked at. 

In the US, about 0.839% of  people die every year.  (Source)

The birth rate is about 1.375% (4 million births/291 million people)

Not counting moves, that means there is a net increase of 0.536% of the population each year (on average.

So over two years, the actual number of people in the country probably grew from 98,213 to 99,268

That means that the Voter Registrar has an extra 6,990 people on it.Out of 99,268 that comes to an error rate of about 7%.

7% is fairly high number of non-people listed on a registrar.  But lets look at the number of people listed as 'active' but did not vote this election.

That would be 62338/80433 = 77.5% of the active voters voted, leaving 22.5% of 'active voters' not actually voting in one of the most contentious elections ever.

Given that almost 1/4 of people did not bother to vote, is it that surprising that 7% have moved away and did not re-registrar?  Or died and not been removed?

More importantly, those people SHOULD NOT have been removed during the past 9 months.  They should be removed NOW.   After the election.  Go through the 106,258-80,433 = 25,825 inactive voters and any that did not vote this year, becoming active again, telling them they will be removed from the voter lists and will not be able to vote again until they re-register.

That's how you do it - after the important election, not before.  That gives them two years to re-register for the congressional elections if we were mistaken, four years for the presidential one.