Political Pivots | Excel

Ohio politics is hot. Several redistricting maps have been adopted, then ruled unconstitutional, then resubmitted, then implemented, then challenged again. The former swing state went for Donald Trump in 2016 and in 2020, then overwhelmingly approved an amendment to the state constitution to enshrine abortion rights and legalized recreational marijuana in 2023. Its five distinct regions comprise a microcosm of American politics.

Here, I used Excel to analyze and visualize data from 900,000 Ohio voters.

• There were decreases in both Hard GOP and Lean GOP voters from 2016 to 2020
• Among low turnout voters, the Designated Market Areas with the lowest approval rating for Biden are Lima and Parkersburg
• Energy Affordability is the most consistent concern across Congressional Districts

The dataset was provided for an interview assessment, so I'm not able to share it, but am assuming it was just a test set. After cleaning and formatting, there were 60 columns of demographics, party affiliation, issues of concern, and election participation. After removing duplicates, there were 899,958 rows, each representing a voter.

The first visualization is a slopegraph illustrating the decrease in GOP voters from 2016 to 2020.

The second visualization is a dynamic map with hyperlinked dropdown to select level of Turnout. It was built from a scatterplit of the average BidenApproval column (calculated by PivotTable) vs Designated Market Area location using an Ohio county map as a background image. The locations were plotted in a table with approximated x and y coordinates.

The final visualization is a heatmap highlighting issues of most concern by Congressional District. It was constructed with the count sums of each issue by District and then I applied color scale conditional formatting.

Yes, PowerBI and Tableau produce snappier visualizations, but they also make it easier to make mistakes. I think of Excel as foundational. Kind of like knowing how to change your own oil or install your own printer driver- if you can do it in Excel, you can do it anywhere. Plus, it was a fun challenge.