Applying Mapper to Chicago 2015 and 2019 Mayoral Data

This is a general repo for code and data related to a TDA project studying the 2015 and 2019 Chicago mayoral elections using the Mapper algorithm as implemented by Kepler Mapper with our own fork MGGG Kepler Mapper. Other project-related code is in the MGGG Summer 2023 repo. in directories beginning with the prefix tda-

Atlas for directories

data/

• acs_tables/ all of the ACS 2018 detailed tables aggregated on blockgroups in Cook County used for this code
• precincts_shapefile/ shapefile for Chicago's 2012 precincts, (source)
• tl_2010_17031_bg10/, tl_2010_17031_tabblock10/ 2010 census blockgroup and block shapefiles for Cook County
• 10decPL_chicago_blocks_raw.csv tables P1, P2, P3, P4 and H1 from the 2010 PL redistricting data on blocks in Cook County
• 10decPL_chicago_blocks_summary.csv summarized race computed categories for population over 18 computed from above data. See ~/summarize_pl.py for the computation.
• 10decPL_chicago_blocks_summary_pct.csv same but as percents of VAP
• acs_dataset_bgs.csv several variables computed from acs_tables on blockgroups in Cook County. See ~/notebooks/get_new_acs_dataset.ipynb for details
• Chicago_mayoral.csv dataset with demographic and mayoral election columns with little known provenance
• chicago_mayoral_with_ro_pct.csv above with runoff columns as percents of total vote
• enriched_chicago_mayoral.csv above with normalized geographic variables + first round vote normalized so that Garcia vote percent and Rahm vote percent sum to 1
• chicago_mayoral_acs_counts.csv above joined with variables from acs_dataset_bgs.csv after being disaggregated onto blocks then reaggregated onto precincts
• chicago_mayoral_acs_counts.csv above with acs columns normalized in per case ways
• chicago_2015_mayoral_runoff.csv clean election results from the chicago election commission for verifying the correctness of Chicago_mayoral.csv

notebooks/

• initial_exploratory.ipynb originial exploratory notebook for applying Mapper to Chicago mayoral election data + race. Also includes first attempt at map based vizualization
• get_new_acs_dataset.ipynb code using data/acs_tables for computing summary variables, disaggregating them to blocks, aggregating them back up to precincts and joining them onto existing data
• aggregate_pl_to_precincts.ipynb code to aggregate PL redistricting race data onto precincts
• baby_mapper_mapper.ipynb extracted map viz from original exploratory notebook to take first step to making it modular
• baby_mapper_mapper_distrib/ directory with everything needed to run baby_mapper_mapper notebook

Root dir

• mapper_boilerplate.py mildly structured example script for our mapper pipeline
• tmap_boilerplate.py attempt at doing the above for tmap library, in particular to use their SAFE computations