Research Sheds Light on Impact and Bias of Voter Purging in Michigan

Key modern examples include Georgia, where 100,000 voters were purged prior to the 2016 presidential primary, which impacted the 2018 gubernatorial race, and North Carolina purged 10% of its total voters after the 2016 election. There have also been cases in Michigan where outside groups tried to purge voter rolls in places like Detroit.

In these cases, the intent or result was that it influenced who participated in the election because many living people were purged erroneously, so the purpose was not just to maintain accurate voter rolls.

Method of Data Examination
In Michigan, the Bureau of Elections maintains a qualified voter file and voter records are canceled for the reasons noted earlier. Michigan uses the Electronic Registration Information Center, also known as ERIC, to uphold integrity by maintaining roles and notifying voters if they are not registered.

For their study, Sadler and Wojciechowski analyzed the just over 175,000 voters purged from the qualified voter file from Nov. 4, 2014, to Nov. 5, 2018, which equated to 2.5% of the 6.7 million voters in Michigan during that time.

They then brought in data considering four associated factors: total population density, racial composition, socioeconomic status and population residing in urban areas. From this data they established the percentage of voters purged according to their area-based racial community, socioeconomic status and 2020 election results.

“There are legitimate reasons why voter rolls periodically need to be purged, as people move and die over time and we need to remove them from the voter rolls,” Wojciechowski said. “However, when we included community mortality rates and the rates at which people were moving in and out of communities, communities with a greater proportion of Black residents still experienced increased levels of voter purging.”

Key Findings and Results
From the analysis, two key findings emerged. First, there was a clear urban-rural divide where cities tended to have higher purge rates than suburban or rural areas. For these cities, there was a strong correlation between the percentage of Black voters and the percentage of individuals voting for the Democratic nominee for president. During this period, there was a Republican legislature majority in Michigan.

Second, a regional pattern was observed where border areas had higher purging rates, namely at the southern part of the lower peninsula and the western part of the upper peninsula of the state.

The characteristics of places where voters were purged included locations where 17% of the population were Black, the average income was just under $50,000, life expectancy was just under 70 years, and the neighborhoods, purged had an average, were made up of 55% Democratic voters.

Since the data was on population, it is only known that they live in neighborhoods with those aggregate characteristics. That said, there were correlations between purges and more Democratic areas, which have more population density, lower life expectancy and more Black residents.

Recommended Next Steps
Outside the key findings, Sadler and Wojciechowski underscore how a majority party can disproportionately purge voters of the other party by using similar geographic techniques (i.e., focusing on certain parts of the state where they know opposing voters may live), including with higher voter purging rates during election years.

Additionally, there are also public health implications in how health and socioeconomic status can be tied to political power and voter disenfranchisement.

“Support for social programs known to make people healthier and happier and save taxpayer money should not be a political issue. Unfortunately, too much in public health has become politicized, as we saw with COVID-19,” Sadler said. “It matters who we elect to hold office because there is currently an imbalance in the value given to health and economic experts between the two major political parties.”

Importantly, this data only covered one four-year period and the reasons why people moved out of state were not known.

Sadler and Wojciechowski suggest that a review of data purging should be done in all states to determine how the party in power is operating and if discriminatory practices are in place. This would be a good first step, as well as states making sure data is accessible and does not come with costs or additional barriers.

States can also implement new election laws such as early voting periods and same-day registration that encourage people to exercise their franchise to vote, both of which have now gone into effect in Michigan.

Learn more about the structure of Michigan’s election system here.

Jack Harrison is Public Relations Coordinator, University Communications, Michigan State University. Kim Ward is Senior Public Relations Strategist at Michigan State University. The article was originally posted to the website of Michigan State University.