Regarding Mark Simon’s April 16 column on district elections: See this study of over 7,000 cities (jstor.org/stable/25193833?seq=1) which found that districts somewhat improve representation for African American men; and that’s it. No improvement for other minorities, or for women, who are generally evenly distributed — you can’t draw a “women’s district.” In an at-large election, if 51% of the voters like candidates A and B and 49% like C and D, A and B win. Half of voters are out of luck. With districts, you can still end up with A beating C in district 1, and B beating D in district 2.
To improve representation, not in the tokenistic sense, but maximizing the people whose votes elect a councilmember, what you want is Proportional Representation. The Center for Election Science, a non-profit think-tank, advocates Proportional Approval Voting (electionscience.org/voting-methods/getting-proportional-with-approval-voting). The ballot is simple — vote up or down on each candidate. This can be implemented using existing machines, for free: just list each candidate like a ballot measure. Tabulation requires a little math, but it’s easy enough to work out on a blackboard. It’s based on a design from Thomas Jefferson that was used in the first apportionment of House seats.
In a proportional system, every vote has equal weight. If broadly adopted, in addition to improving racial diversity, we would see greater ideological diversity. I believe on principle that Silicon Valley Republicans and Libertarians and Central Valley Democrats and Greens, all deserve to be heard.