Your browser (Internet Explorer 6) is out of date. It has known security flaws and may not display all features of this and other websites. Learn how to update your browser.
X
Post

Colorado: Comparison of Registered Voter Counts to Census Voting Age Population

By Earl F Glynn | Franklin Center

A comparison of US Census voting age population data in Colorado to voter registration data shows 17 of the 64 Colorado counties have more than 100% of the voting age population registered to vote.

Five of these 17 counties have had 100+% registration rates compared to census population for all available years of voter registration data (2004 to 2012).

An additional 9 counties have voter registration rates of 95% or higher.

Charts by county for the years 2000 to 2012 show how counties are maintaining their voter lists.

Many Colorado counties have inactive voter rates as high as 40%, but the definition of “inactive voter” in Colorado may be broader than in some nearby states.

There is higher voter fraud potential in areas of unmanaged voter lists and areas of high voter list bloat measured by inactive voters.


Overview

Yearly US Census population estimates for Colorado counties were extracted for the years 2000 to 2011.

Voter registration totals were extracted from the Colorado Secretary of State’s web site from election results reported for primary and general elections from 2000 to 2012.

Active and inactive voter registration counts by county were obtained from a file of Colorado voter registration data from Feb. 2012.

The census and voter registration data were plotted for easy visual comparison of any particular year or trends over time.


Results: County Charts

View the PDF with a chart for each Colorado County comparing US Census Voting Age Population to Registered Voter Counts [65 pages: 64 counties, state total]

(click on graphic to view PDF)

General comments about all charts:

  • The black line represents the total county population from the Census 2000 and the Census 2010 enumeration and census estimates for other years.
  • The blue line represents the census voting age population counts or estimates, those 18 years or older.
  • The voter registration points in red represent total registration values (Active + Inactive) from the Secretary of State’s online reports.
  • The points in green represent only the Active voters. The difference between Active and Total registration, the Inactive voters, is a measure of voter bloat.
  • Total and active voter counts from a file of Colorado voters are shown by a black-filled circle.

NOTE: US Census estimates are not as accurate for smaller counties on a relative basis.  See Limitations in US Census Voting Age Population.


Selected Colorado Counties

Mineral County:  Huge decrease in 2010 US Census population.  More voters than population now?

(click on image to enlarge)

What is the explanation for the huge decrease in US Census total population in Mineral County?  Why were the Census estimates not accurate for so many years?  Which is more correct, the census estimates or the voter registration numbers?

Based on total voters, Mineral County is the third smallest county in Colorado.


San Juan County:  More voters than census voting age population?

(click on image to enlarge)

Why has the voter registration for San Juan County always been more than 100% of census voting age population from 2004 to present?  (The red “+” marks are always above the blue line.)

Other counties with more voters than census population from 2004 to present include Cheyenne, Gilpin, Hinsdale, and Jackson.

San Juan County has the fewest number of voters of all Colorado counties with only 640 on Aug. 1.  Among this group Gilpin is the largest with 4,916 voters.


Ouray County:  Increasing number of voters while population values decline?

(click on image to enlarge)

Why in 2008 did the total registration for Ouray County start rising at a time the US Census said the total and voting age population was dropping?


Denver County:  Colorado’s most populous county

(click on image to enlarge)

Why did the registration as a percentage of voting age population in Denver County increase from roughly 80% to 90% between 2008 and 2009?

Colorado counties with more than 200,000 total registered voters from Aug. 1, 2012 data:

Two of these large counties, Boulder and Douglas, have a very high percentage of voting age population registered to vote.


Details of Technical Approach

Let’s first review the needed raw data files and then discuss the processing to create the charts above.

Raw Data: Colorado Voter Registration Statistics

Colorado Secretary of State Scott Gessler publishes end-of-month voter registration statistics online with counts of active and inactive voters, but the format of the files has changed over the years.

The original goal was to collect quarterly data, which was available from 2009 through 2012, but not always in earlier years.

From Sept. 2008 through 2012 the voter registration statistics can be downloaded in an Excel 2003 format, but the worksheet name changed four times.

Voter statistics files from May 2006 through March 2008 were put online in a Word 2003 .doc format.  The tables in these documents were copied and pasted into Excel file to be in a format that could be read by a computer program.

The Colorado voter statistics files from June 2004 through February 2006 were in a .txt text format, and could be read without change by a computer program.  However, the county names were not in alphabetical order within the file.

Active voters counts from 2002 were found in the 2002 Abstract of Votes Casts, a PDF file, but was not used since inactive voter counts were not also available.

[Note to Colorado Secretary of State: Standardization of file format and naming conventions would make analysis of the data easier.]

The script combine-text.R read the six text files from 2004-2006 and produced two summary files by tokenizing the files and selecting certain columns.

The script combine-ODBC.R read 7 Excel files converted from Word documents and 17 Excel files containing Colorado voter registration data, and combined the results from the combine-text.R script to create two files with all available Colorado voter registration data:

The scripts are somewhat rigid for the one-time conversion of the 30 voter registration data files into two summary files to be used to plot the charts.

Raw Data: Colorado Voter Registration File

A Colorado-Voters.R script read a text file of 3.4 million Colorado voters from earlier this year and created a cross-tabulation of “Active” and “Inactive” voters using the COUNTY and VOTER_STATUS fields.

Almost 4000 voters without a county field were ignored and a “Total” column was added using Excel to create the file Colorado-Active-Inactive-2012-03-19.csv.


Raw Data: US Census Data for Colorado

The links below are to the raw online Colorado census data at the US Census site.

Annual Estimates of the Resident Population by Selected Age Groups and Sex for Colorado Counties:

The technical article US Census Voting Age Population gives details of extracting US Census data for all states, including Nebraska, and plotting charts of that data.

Updated US Census county population estimates for July 1, 2012 are expected in April 2013.


R Program

After pre-processing the raw data described above, we can create charts showing US Census voting age population in Colorado and the number of registered voters by county.

Input files

Census files:

Voter registration summary files:

Processing Script: Colorado-Voting-Age-Population-and-Registered-Voters.R

After reading voter registration data into data.frames (voters.total, voters.active, voter.file) the script called function get.voting.age.census.data to read Colorado census data.

The census data was passed to the plot.county.population.charts function for plotting along with a “callback” function, plot.colorado.registration.

The plot.county.population.charts function plotted the census data and then used the plot.colorado.registration “callback” to plot the registration data on the same chart.

Output files


Analysis / Discussion

Census Continuity

This subsection discusses two types of census time series discontinuity:  (1) adjustments made to census figures to be the base for yearly census estimates, and (2) unexpected increases or decreases between census estimates in 2009 and the census numbers from 2010.

Census adjustment.  Each chart shows a time series of 16 points with US Census population estimates or census enumeration values for total population and voting age population.

All data for each county and the state totals are in the file Colorado-Voting-Age-Population.csv.  In the file, YEAR=1 (4/1/2000) marks data for the 2000 census figures and YEAR=13 (4/1/2010)  marks data for the 2010 census figures.

For an unknown reason the census figures for Broomfield County are missing for both the 2000 and 2010 census (the YEAR=1 and YEAR=13 values).  Perhaps the reason is related to Broomfield becoming a new Colorado County in late 2001.  The chart for Broomfield county shows breaks in the population lines because of the missing data.

In the file, YEAR = 2 (4/1/2000) and YEAR = 14 (4/1/2010) are labeled by the census as the “estimates base” used for all the subsequent yearly July 1 estimates.  Quite often the “estimates base” is the same as the census figures, but the adjustment made in the Colorado statewide totals can be seen as the small discontinuity in 2010 in the statewide chart:

An adjustment from 2010 census figure to 2010 estimates base for the State of Colorado can be seen as the discontinuity in the chart.

Change from 2009 estimate to 2010 census.  Normally the US Census estimates from 2009 are fairly consistent with the actual US Census counts from 2010, but sometimes there are unexpected increases or decreases in population as a result of the census.

The middle half of Colorado counties saw changes ranging from a decline of -1.65% to an increase of 3.35% from the 2009 population estimate to the actual 2010 census.

A review of the 2009 population estimates to 2010 census figures shows these large increases and decreases:

  • Increases:  San Juan (25.9%), Kiowa (12.9%), Costilla (11.9%)
  • Declines:  Mineral (21.9%), Saguache (13.9%), Huerfano (11.2%)

A boxplot of the percentage change suggests increases or decreases of roughly 10% or greater are statistical outliers.

Voter Bloat (inactive voters)

On August 1, 2012 Colorado had 3.5 million total voters with 2.3 million active voters.

The Colorado voter bloat, the number of inactive voters, was 1.2 million voters or about 35% of all voters.

This percentage of inactive voters seems high compared to nearby states of Kansas, Nebraska, or Missouri that I’ve previously analyzed, but the comparison is not so easy because of different definitions of inactive voters.

According to the 2010 Statutory Overview Report from the US Election Assistance Commission, Colorado marks a voter as “inactive” for “no response to confirmation notice or correspondence returned as undeliverable.”

But the definition of “inactive” voter in Colorado may be broader.  A KKCO 11 News story says there are two ways a voter may be marked “inactive” in Colorado:

  • Mail to a voter is returned to the election authority, or
  • A vote fails to vote in a general election.

According to that news story a Colorado voter may be marked as “inactive” if the last time a vote was cast was 2010.  This can be a problem for those who lasted voted in the 2008 presidential election and are now marked “inactive” for not voting in 2010.

Colorado counties with the lowest percentage of inactive voters are:

  • Hinsdale, 13.9%
  • San Juan, 15.6%
  • Kowa, 21.3%

Counties with more than 40% inactive voters:

  • Summit, 45.3%
  • Gunnison, 44.7%
  • Denver, 44.2%
  • Lake, 42.3%
  • Adams, 41.7%
  • Moffat, 41.5%
  • San Miguel, 40.8%
  • Pitkin, 40.3%

The voter bloat can be seen in the charts as the difference between total registration and active voter registration.

Amber McReynolds, Director of Election, Denver Clerk and Recorder’s Office, in a July 25, 2012 letter to Secretary of State Scott Gessler suggests the Colorado General Assembly needs to revisit the issue:

“the state’s inactive statutory scheme is outdated, inefficient, and very expensive to administer. We hope to see the General Assembly consider a change to the Colorado Revised Statutes next spring, which is the appropriate venue for this issue to be examined and resolved.”

While it’s tempting to compare Colorado’s 1.2 million inactive voters (35%) to Missouri’s 436,000 inactive voters (10.7%), with different definitions of “inactive voter” in both states this would be an apples and oranges comparison.  Unless states have similar definitions of “inactive voters” meaningful comparisons cannot be made easily.

Maintenance of Voter Lists (comparison to census)

U.S. federal law (Help America Vote Act of 2002) limits how state and local officials may remove names from voting rolls. But even with hands tied by the federal government, there is wide diversity in how county clerks maintain voting lists.

The US Census voting age population estimates give an upper bound on what the actual voter registration should be in a state or county. When lists are not maintained by the county clerks and election officials the “bloat” can grow over time resulting in more voters registered than the census says are of voting age.

Observations from the county charts yield these estimates of registration as a percent of voting age population in Colorado counties:

  • High Registration:  Mineral (130%), Ouray (120%), Gilpin (110%), Hinsdale (110%), Jackson (110%), San Juan (110%), Summit (105%), Archuleta (102%), Clear Creek (102%), Elbert (102%), Grand (102%), San Miguel (102%), Teller (102%), Cheyenne (100%),  Dolores (100%), Gunnison (100%), Pitkin (100%),  Custer (98%), Routt (98%), Boulder (97%), Douglas (97%), Baca (96%), Kiowa (95%), Montezuma (95%), Park (95%), Sedgwick (95%)
  • Low Registration:  Crowley (40%), Bent (55%), Logan (65%), Lincoln (70%)

Questions:

  • What is the explanation for the huge decrease in US Census total population in Mineral County?  If true, the current voter registration is more than 100% of the total population of all ages.
  • Why in 2008 did the total registration for Ouray County start rising at a time the US Census said the total and voting age population was dropping?
  • What’s the explanation for the huge population increase in San Juan County after the 2010 census?  Is voter registration really over 110%?
  • Why has the voter registration for Cheyenne, Gilpin, Hinsdale, Jackson, San Juan Counties always been more than 100% of census voting age population from 2004 to present?
  • Why did the registration as a percentage of voting age population in Denver County increase from roughly 80% to 90% between 2008 and 2009?

Overall about 85% of Colorado’s voting age population is registered to vote.


Conclusions

Seventeen of Colorado’s 64 counties have voter registration rates 100% or more of the US Census voting age population.  Another 9 counties have voter registration rates 95% or more of their voting age population.

Five counties have had over 100% voter registration rates compared to census population in all reporting years, 2004-2012.

Colorado’s 1.2 million inactive voters seems to be a very high number, but the state’s definition of an inactive voter only missing the last general election may be overly broad.

Unlike some states where the address and district assignments of an inactive voter may be in doubt, many of Colorado’s inactive voters may have valid addresses and simply decide not to vote or respond to mailings from the election offices.  This may be a matter for discussion in the Colorado General Assembly in a future session.


References

  • 2010 Statutory Overview Report, US Election Assistance Commission, May 2011.  [Gives definitions of active and inactive voters by state.  See PDF pages 16-18]

efg

Contact Info: Twitter: @WatchdogLabs, Facebook: http://www.facebook.com/WatchdogLabs

Leave a comment  

name*

email*

website

Submit comment