Colorado Counties Have More Voters Than People … Not So Much
A review of voter registration data for ten counties in Colorado details a pattern of voter bloat inflating registration rolls to numbers larger than the total voting age population. Using publicly available voter data and comparing it to U.S. Census records reveals the ten counties having a total registration ranging between 104 to 140 percent of the respective populations. …
…All ten counties investigated by Media Trackers reported voter turnout greater than the national average. Nine out of ten also showed voter turnout well above the Colorado average. Mineral and San Juan counties, which have voter registration numbers of 126 percent and 112 percent respectively, had voter turnout of 96 and 83 percent respectively. …
–Aron Gardner, Sep 4 2012
Colorado Counties Have More Voters Than People
The above article was written well before the election but was presented to me after Romney’s defeat by someone trying to make the case that this shows that Colorado is experiencing voter fraud.
So I decided to look at the numbers.
The web page for the Colorado Secretary of State has links to the voter registration data and the recent election results The latter is apparently outsourced to a commercial operation and, as of the time of this post, was last updated on Nov 08 2012.
For population data, I used the US Census Annual Estimates of the Resident Population by Age, Sex, Race, and Hispanic Origin for Counties: April 1, 2010 to July 1, 2011. Unfortunately, that data set uses an age group that groups 18 and 19 year-olds with teens too young to vote. My population estimates data begin at age 20 and are just a bit low without those 18 and 19 year olds. I cross-checked the ‘high’ counties determined below with the Census Quick Facts and determined that ‘bit low’ for the four counties of interest is about 2%.
Registered voters come in two categories: active and inactive. Associating this with the census population estimates for 2011 was relatively straightforward, as it was for the 2012 election ballot count. The only processing gottcha to be careful of is that the county names are sorted slightly different for counties with a space in the name.
I actually found 29 Colorado counties with total voter registrations (active and inactive) above 100% of the adult (20+) US Census estimates for 2011. This dropped to only three Colorado counties with voter registrations above 100% and one county with 95% when looking only at active voter registrations. One Colorado county had more ballots cast then the US Census adult population estimate.
Media Trackers compared voter participation to national and state averages. I found it more illuminating to display them as charted above. The chart above is for voter participation and the chart for active registrations (see below) looks very similar. The four “high” outliers are labeled, as are two “low” outliers. The state average voter participation (green, 67.7%) and the county median participation (blue, 69.0%) are also charted. None of the distributions of active registration, total registration, and voter participation are ‘standard normal distributions’ per the R Shapiro-Wilk test.
The four high outliers are very low population mountain counties. Three of the counties have less than a thousand adults by the 2011 census estimate. If we take the sum of the votes in the four high outliers counties (votes as tallied) and subtract out the “expected votes” based on median county voter participation by population (county population * median county participation rate), the “unexpected vote” count is only 1358 votes. Barack Obama won the state vote by 117,276 votes. Possible irregularities in the four high outliers could not have swung the state in favor of Obama.
So why do we have these four outliers if not for fraud? Responses to the original Media Trackers included possible under-counting by the US Census due to high transient populations (gaming, resorts, snow birds).
While Media Trackers focused on “too high” results, there are also two low outliers in agricultural plains counties. I wonder if those are related to a high population of non-citizen residents who are not eligible to register or vote.
I will also note that the Colorado voter participation numbers are based on the total registered voters. So while I think Media Trackers are puffing up their claims of voter registration anomalies by including inactive registrations in their calculations, that is the category of voter registration used by Colorado for their voter participation statistics.
In conclusion, there is no way that the anomalies explored above, even if fully exploited by pro-Democratic fraud, could have swung the election in favor of Obama. However. the fact that 29 of 64 counties show total registered voter rolls higher than the census estimates for adult populations in those counties could be an indication that the rules regarding the retention of inactive registrations might be too lax or that the execution of those rules to purge inactive registations are not being diligently followed. Bottom line, Media Tracker’s analysis seems inflated, but there may be a valid underlying issue regarding retention of inactive voter registrations.