Case Study: COURIER’s Model Boosts Voter Turnout

COURIER
4 min readJun 6, 2022

COURIER and its local newsrooms in Arizona, Florida, Iowa, Michigan, North Carolina, Pennsylvania,Virginia, and Wisconsin were founded on the belief that fact-first, value-based, trusted local journalism can help strengthen civic engagement and democracy — boosting low-turnout voter participation in elections.

At COURIER, not only did we build our network of newsrooms on that belief, but we set out to prove it.

Spoiler: We did.

Results from a randomized-control-treatment (RCT) experiment of COURIER’s model show that its unique method of boosting news on social media had a positive, statistically significant impact on voter turnout in the 2021 Virginia state election.

The experiment tested the hypothesis that boosting factual, local coverage of the Virginia 2021 gubernatorial race and candidates, and localized coverage of federal policy solutions to passive news consumers will increase their voter participation in state elections. A pool of nearly 1 million voters was separated into three groups. Treatment Group A received boosted coverage on federal policy, Group B coverage on candidates, and no boosted coverage to a control group.

When comparing turnout rates, the pooled treatment group showed a statistically significant positive bump in voter participation in the election over the control group. Additionally, the actual impact of the boosted coverage on the voters who saw them is likely larger than suggested by this analysis. This is because the analysis metrics are based on the full size of the treatment group (n), rather than the number of people within the treatment group who use Facebook and saw the ads.

Let’s break it down further.

To prove that boosting news to low information voters on social media will increase their participation in elections, we wanted to run an experiment in advance of the Virginia 2021 elections.

For this experiment, we set out to test the effects of two distinct tracks of news coverage produced by COURIER’s Virginia newsroom, the Dogwood in the lead-up to the November elections: one focused on informing our audience about the two distinct gubernatorial candidates and where they stand on key issues that matter to Virginians, and the second focused on localizing the impacts of federal legislation passed in Washington.

Our hypothesis was that boosting these two tracks of coverage to segments of our audience that is made up of less politically engaged, passive news consumers across the Commonwealth would increase their voter participation in state elections, compared to a control group of the same audience who would not receive Dogwood’s boosted news.

We tested this in a treatment-control experiment in which close to 1 million eligible voters were put into one of three roughly equal groups — one that would be exposed solely to Dogwood’s boosted federal policy coverage, one exposed solely to its coverage of gubernatorial candidates and their positions, and the last a control group who would not be delivered any Dogwood coverage through this targeted paid digital media program.

The test results were clear: Boosted news has a statistically significant positive effect in the pooled treatment groups (federal + state groups) over the control group (no news).

Diving deeper into the results, we found that, while exposure to federal coverage was more effective than candidate coverage in boosting voter turnout, the candidate coverage did have a positive effect as well. We estimate that the cost per net vote for the federal news treatment group was ~$131, and that cost for the candidate coverage treatment group was ~$227. These results are an encouraging validation of Courier’s model given that the spend per voter was about 10X lower than the 2020 measurement project.

This experiment is just another positive proof-point that demonstrates COURIER’s unique model of boosting factual, local, civic journalism to low turnout voters online can measurably increase their participation in elections.

Why this matters.

The results of this experiment are yet another in a long list of validations for the model that COURIER and its outlets were founded upon.

Building scalable, local media infrastructure that can break through disinformation echo chambers and ensure less engaged voters are reached with quality, factual, and relevant information where they spend their time online can be the difference maker in elections that are won or lost at the margins. COURIER turns just three years old this June so just imagine for a second what these results might look like when our newsrooms are more widely known and trusted by even more voters.

--

--