As we seek to better understand how the Dobbs decision has reshaped this midterm election, I’m excited to announce the first analysis of perhaps the most critically important post-Dobbs election dataset. Earlier this week, our team at TargetSmart compiled the voter history data from the Kansas election, detailing exactly which registered voters participated. This dataset provides the first look at how pro-choice voters prevailed by such a wide margin, and I’m excited to share that with you.
Starting with gender – women accounted for an astounding 56% of all ballots cast. Looking at the previous three general elections (a better benchmark than previous primaries, in that total turnout in this election surpassed that of every midterm general election in Kansas other than ’18), women had a previous high-water mark of 53.5% of ballots cast in 2016. In 2018 women made up 52.5% of general election turnout. This would seem to answer a question many have asked over the past few weeks, paraphrasing, “it’s good that women are registering to vote at a higher rate, but will they turn out?” The 12-point gender gap in turnout in Kansas is likely a record for the state, for any election.
Age was another critical component to the Kansas victory. Conventional wisdom says that young people don’t vote in midterms, and they certainly don’t vote in primaries. Turning that conventional wisdom on its head, voters under the age of 30 comprised over 14% of ballots cast, exceeding the youth vote share for each of the past three general elections in Kansas. For reference, more voters under the age of 30 voted in this election than voted in the 2018 general election.
Voters of color were a key component of the pro-choice victory in Kansas last month. Latino voters, though a small share of the electorate overall, are growing in importance in Kansas elections. Latinos accounted for 3.8% of turnout in this election, up from 2.3% of votes cast in the 2020 presidential general election. In fact, more Latinos voted in this August primary election than have voted in any election in Kansas, ever! And while AAPI voters are a small share of the electorate, they also surged in turnout, with this election producing a record number of AAPI voters as well.
We can assess turnout by party at this point as well. Republicans, unsurprisingly, accounted for the largest share of the electorate, at 51% of ballots cast. Somewhat surprisingly, this Republican turnout share exceeded the 2020 general election benchmark of 50.4% of all voters. That said, Democrats exceeded their 2020 electorate share by an even wider margin, accounting for 28.1% of votes in the 2020 general election, as compared to 32.1% in this election.
This turnout data by party makes it clear that a significant portion of Republicans voted “no” on the amendment, supporting the pro-choice position. In fact, even if you assume that every single Democratic and unaffiliated voter cast a “no” vote, that would still mean that 1 in 5 Republicans joined them. In reality, considering crossover voting, it is likely that it was closer to 1 in 4, even approaching 1 in 3 Republicans joined the pro-choice coalition.
As we consider this turnout data, where women, younger voters, and Latinos all exceeded historic precedents, we shouldn’t be surprised that the lone public survey on this amendment badly missed the wide margin of victory for pro-choice voters. In assessing the sample for the poll, they badly underpredicted the youth vote (12% of the sample, 21% of turnout) and women (8-point gender gap in the sample, 12 points in actuality). The pollster did not release information regarding weighting by race.
I’ll close with my key takeaways from this dataset:
- We knew that women were highly engaged by the voter registration data we are analyzing across the country. The Kansas data proves that engagement isn’t stopping at registration, but rather continuing to the ballot box.
- The Kansas coalition of women, younger voters, voters of color, independents, and even Republicans shows the depth and breadth of the movement motivated by the Dobbs decision.
- Polls that don’t consider the notion of younger voters, women, and voters of color accounting for a larger share of the electorate than they have in any recent election may miss the eventual election outcome by a wide margin.