Ranked Choice Voting and Racial Group Turnout Methodological Flaws Skew Recent Study on RCV

by Andrew Douglas // Published November 2, 2015

In the recent journal article “Writing the Rules to Rank the Candidates: Examining the Impact of Instant Runoff Voting on Racial Group Turnout in San Francisco Elections,” Professor Jason McDaniel of San Francisco State University argues that the adoption of ranked choice voting (RCV) in San Francisco mayoral elections has been associated with lower turnout among white voters and African Americans, as well as younger voters and voters with lower levels of educational attainment.

Because these findings run counter to much of the existing research on RCV, it is necessary to give this new study careful consideration. We find that omissions and inconsistencies in the construction of the study’s model are likely to have erroneously shifted attribution for declines in turnout among certain demographic groups to RCV, as more likely explanations are left unaccounted for. The issues include: 1) the omission of electoral competition as a control variable, 2) the decision to code the 2011 election, which featured a popular incumbent candidate, as an "open seat" race, and 3) an oversimplified view of "in-group" candidates and racialized politics. These methodological issues not only cloud the study's analysis, they also do so in a way that consistently disadvantages ranked choice voting, as they shift blame for declining turnout of whites and African Americans to RCV, while leaving more likely explanations unaccounted for.

It should be emphasized that ranked choice voting, just like every other voting method, is not perfect. The means through which RCV is administered in San Francisco should be improved. But, to dismiss RCV based on a single, flawed study is a mistake. We urge Professor McDaniel and other scholars to continue to study the effects of RCV, and to engage with the more nuanced operation of RCV in San Francisco.