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

Ohio: 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 Ohio to voter registration data shows two counties, Wood and Lawrence, have more registered voters than perhaps they should.

Numerically Wood and Lawrence have more than 100% of census voting age population registered to vote.

An additional eight Ohio counties have 95% or more of their voting age population registered to vote.

Holmes County on the other extreme only has about 60% of its voting age population registered.

Charts by county for the years 2000 to 2012 show how counties are maintaining their voter lists based on the balance between census data and voter registration data.  There is higher voter fraud potential in areas of unmanaged voter lists and areas of high voter list bloat.

Information presented here includes statistics from the Sept. 23, 2012 Ohio voter registration file from the Ohio Secretary of State.


Overview

Yearly US Census population estimates for the 88 Ohio counties were extracted for the years 2000 through 2011.

Voter registration totals were extracted from the Ohio Secretary of State’s web site for the years 2000 through 2012 both from previous election statistics and online voter registration files collected at four points in time in 2011 and 2012.

Only total registration counts were available.  Separate counts of “active” and “inactive” voters were not available.

The census and voter registration data were plotted for easy visual comparison of any particular year. Trends over time can also be studied using the charts.


Results: County Charts

View the PDF with a chart for each Ohio County comparing US Census Voting Age Population to Registered Voter Counts [89 pages: 88 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 red “+” marks represent total voter registration numbers as reported by the Secretary of State at times of Ohio elections
  • The black-filled circles are voter registration numbers derived from voter files obtained from the Ohio Secretary of State.

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


Results for Selected Ohio Counties

Wood County: Nearly 110% registration of US Census voting age population — and still climbing

(click on image to enlarge)

One possible explanation for the Wood County chart may be high voter registration drive efforts (is that true?) at Bowling Green State University that has about 85% of its 17,000 students from Ohio at the same time the US Census counts many of the students at their home addresses outside Wood County.

This counting combination would result with a lower county age population line from the census data and a higher set of voter registration points.  I welcome feedback on how this can be confirmed.

Wood County’s registration chart seems to show differences by presidential election cycles:  2000-2004 (low historical registration), 2004-2008 (increasing registration to over 100% by 2008), 2008-2012 (above 100% with an uptick in 2012).  Does anyone have an explanation for this?


Lawrence County:  another Ohio county with 100+% voter registration when compared to census voting age

(click on image to enlarge)

Lawrence County’s registration dropped to under 90% in 2011 and then jumped to over 100% in 2012.  What changed in such a short period of time?

Why the recent large swing in voter registration numbers in the voter file from 2011 to 2012?


Holmes County:  Lowest voter registration county

(click on image to enlarge)

What explains the low voter registration in Holmes County?

Is the low voter registration related to the large Amish population– the largest Amish community in the world?

Wikipedia says about 36% in Holmes County speak either Pennsylvania German or German at home and another 7% speak Dutch.

Other counties with relatively low registration rates include Madison (home to several prison facilities) and Fayette.


Cuyahoga County:  Ohio’s most populous county

(click on image to enlarge)

CuyahogaCounty’s registration was over 100% from 2004 to 2008 and peaked near 110% in 2008 — there was a slow population decrease during this time of increased voter registration. In 2009 and 2010 voter registration was near 100% and then dropped to 90% this year.


Ohio statewide

(click on image to enlarge)

Ohio statewide registration peaked just short of 95% for the 2008 presidential election.


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: Ohio Voter Registration Statistics

The Ohio Secretary of State publishes Election Results & Data on its web site.  On each page of “Official Results” for any year there are “Voter Turnout” reports that show the breakdown of voters by county.  For example, this page shows the total registered voters by county at the time of the March 6, 2012 primary election.

I was tempted to “screen scrape” these online HTML tables but after study I found too many small differences.  Copying and pasting the HTML tables to Excel and cleaning them up manually was faster for me than attempting to deduce all the needed screen scraping rules.

The Ohio Secretary of State badly needs to set and follow naming conventions for files.  This page shows all the different naming conventions used for the Voter Turnout reports from 2000 to 2012.

The most recent turnout report can be downloaded as a CSV file, so future files will be easier to process.

The resulting Excel files with the turnout reports from 2000 to 2012 can be viewed here.

The Combine-OH-XLSX.R script used the RODBC package to read all the Excel files and create a single file, Ohio-Voter-Registration-2000-2012.csv, for processing described below.

Raw Data: Ohio Voter Registration File

Ohio provides a unique opportunity to study voter registration files since the files are online and free.  I have observed the files are every updated Sunday morning — at least for the last few weeks.

Statewide voter files can be downloaded from this page, or can be downloaded for particular districts, usually by candidates for public office.

A separate article, Technical Analysis of Ohio Voter Registration Files, describes R scripts used to analyze the voter data.  One of the files created is a file, 02-COUNTY_NUMBER.csv, which gives the frequency counts for the COUNTY_NUMBER data field.

The Combine-OH-Voter-File-County-Counts.R script combined such county frequency counts from these dates:  2011-07-07, 2012-07-09, 2012-09-02, and 2012-09-24.  The resulting combined file, Ohio-Voter-File-County-Counts.csv, was created for use in the processing described below.


Raw Data: US Census Data for Ohio

The links below are to the raw online Ohio census data at the US Census site. Annual Estimates of the Resident Population by Selected Age Groups and Sex for Ohio Counties:

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

Census estimates between the 2000 census and 2010 census appear to be fairly accurate for most Ohio counties.

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 Ohio and the number of registered voters by county.

Input files

Census files:

Voter registration summary files:

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

After reading voter registration data into data.frames (registered.dates and file.data) the script called function get.voting.age.census.data to read Ohio census data.

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

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

Output files


Observations about County Charts

  • Wood and Lawrence Counties are the only Ohio counties above 100% registration of US Census voting age population with Wood approaching 110% and Lawrence just above 100%.
  • Wood County’s registration chart seems to show differences by presidential election cycles:  2000-2004 (low), 2004-2008 (increasing to over 100% by 2008), 2008-2012 (above 100% with an uptick in 2012).
  • Lawrence County’s registration dropped to under 90% in 2011 and then jumped to over 100% in 2012.  What changed in such a short period of time?  Why the recent large swing in voter registration numbers in the voter file?
  • Other High Registration CountiesDelaware (95%), Fairfield (95%), Geauga (95%), Greene (95%), Henry (95%), Jackson (96%), Mercer (97%), Van Wert (96%).
  • Low Registration CountiesHolmes (60%), Madison (70%), Fayette (72%).  What are the explanations for the low percentages?
  • Cuyahoga County’s registration was over 100% from 2004 to 2008 and peaked near 110% in 2008 — there was a slow population decrease during this time of increased voter registration. In 2009 and 2010 voter registration was near 100% and then dropped to 90% this year.
  • Delaware, Fairfield, Monroe, Richland Counties were  about 100% registered at the time of the 2008 general election.
  • Franklin County was slightly above 100% registered at the time of the 2004 general election.
  • Gallia County was about 100% from 2004-2008 but has been about 90% before and after.
  • Registration in Mercer County was above 100% for much of 2003 to 2006.
  • Morrow County was near 100% 2008-2011 and dropped to about 94% this year.
  • Putnam County: 94% now, but has been consistently 95+% for much of last decade.
  • Shelby County: Voter registration has been steadily increasing since about 2005 faster than voting age population has been increasing.

References


efg

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

  • [...] Ohio: Comparison of Registered Voter Counts to Census Voting Age Population [...]

  • [...] Ohio: Comparison of Registered Voter Counts to Census Voting Age Population [...]

  • Interesting analysis, Earl, and thanks for going into depth into how you did it.

    What came to mind immediately that could account for some of the discrepancies that you find would be a pattern of absentee voter behavior. If, for instance, a county had a lot of residents that lived there but voted somewhere else as absentee, that would skew the results.

    I don’t know how you’d account for this, though – without going county by county and seeing where they issued absentee ballots to.

    Edward Vielmetti

    October 5, 2012

  • Thanks for the suggestion. I’m not sure how to get that info in Ohio without special open record requests. (In some other states, like Colorado, the voter history file gives info about voting method and that analysis could be done easily.)

    I plan to research the Ohio data more after the Nov. election, especially Wood County’s high registration. At least part of the explanation is “bloat” caused by college students who voted in college a number of years ago but have moved on. They’re still registered even though they are not voting anymore. There were several BGSU addresses with a large number of voters at a single address, a number of whom last voted in 2004, for example.

    efg

    October 5, 2012

  • [...] Ohio: Comparison of Registered Voter Counts to Census Voting Age Population [...]

  • [...] Ohio: Comparison of Registered Voter Counts to Census Voting Age Population, Watchdog Labs, Sept. 28, 2012. [...]

Leave a comment  

name*

email*

website

Submit comment