Michigan Independent Citizens Redistricting Commission (MICRC) Memo #3 — Recommendations for Managing and Analyzing Public Input in Future Rounds of Michigan Redistricting
This memo provides a comprehensive analysis of the CLOSUP team’s work in assisting the Michigan Independent Citizens Redistricting Commission (MICRC) in the analysis of public comments submitted during the 2024 Senate Mapping Redistricting process. The goal of this memo is to detail the public comment process and provide recommendations for managing and analyzing public input for future iterations of the CLOSUP team and the MICRC.
Key findings
• The CLOSUP team analyzed public comments for the Michigan Independent Citizens Redistricting Commission (MICRC) during the May and June 2024 state senate redistricting process. The team prepared two memos for the MICRC:
» Memo 1 presented relevant communities of interest (COIs) in Metro Detroit
» Memo 2 analyzed the “net-favorability” of the 12 draft Senate maps
• The four-person team (three analysts and a liaison to the commission) had two steps for each memo: (1) comment collection/aggregation and (2) comment analysis
» Comment collection required aggregating and sorting comments from the MICRC’s two online portals and from meeting transcripts
» Comment analysis required “coding” the comments into common themes, analyzing their frequency, and presenting relevant takeaways and data trends
• Most effective comment analysis strategies
» Solicit Word document versions of meeting transcripts from MICRC
» Divide comments for review, flagging unclear comments for secondary review
» Create unique IDs for individual commenters
» Use AI early with human oversight for a more consistent means of coding comments
» Meet weekly to discuss trends and update codebook as necessary
• Future recommendations for MICRC and CLOSUP team
» Adhere to thematic, bottom-up COI and public input philosophy
» Build and update a flexible and responsive codebook for public comments
» Leverage artificial intelligence with human oversight for demanding datasets