Nicar2026_skills_in_codex_claude majority-minority-change

This skill should be used when users need to analyze county-level racial composition change between two Census snapshots and identify where places crossed a majority-minority threshold.

install

source · Clone the upstream repo

git clone https://github.com/amkessler/nicar2026_skills_in_codex_claude

Claude Code · Install into ~/.claude/skills/

T=$(mktemp -d) && git clone --depth=1 https://github.com/amkessler/nicar2026_skills_in_codex_claude "$T" && mkdir -p ~/.claude/skills && cp -r "$T/.codex/skills/majority-minority-change" ~/.claude/skills/amkessler-nicar2026-skills-in-codex-claude-majority-minority-change-dcf8c6 && rm -rf "$T"

manifest: .codex/skills/majority-minority-change/SKILL.md

source content

Majority Minority Change

Use this skill to compare county demographics across two years and flag threshold crossings.

Requirements

R with
```
dplyr
```
,
```
readr
```
,
```
stringr
```
, and
```
jsonlite
```
Two local CSV files with county identifiers and required race/population fields

Bundled Census CSV snapshots for immediate use:

skills/majority-minority-change/data/county_race_acs5_2010.csv

skills/majority-minority-change/data/county_race_acs5_2020.csv

No API key required

Required Input Columns

Join keys available in both files:
- ```
county_fips
```
  (preferred), or
- both
```
state
```
  and
```
county
```
Metrics in both files:
- ```
total_population
```
- ```
non_hispanic_white
```

Standard Workflow

Confirm start/end files and year labels.

```
--state
```
accepts full names or USPS abbreviations (for example,
```
Georgia
```
or
```
GA
```
).

Default bundled files:

start:

skills/majority-minority-change/data/county_race_acs5_2010.csv

end:

skills/majority-minority-change/data/county_race_acs5_2020.csv

Review
```
skills/majority-minority-change/references/DATA_SCHEMA.md
```
for exact derived/output field names before downstream filtering or selecting columns.
- Expected output naming includes
```
nonwhite_share_<year_label>
```
  and
```
delta_nonwhite_share_pp
```
  .

Run script with optional state filter.

Example:

Rscript skills/majority-minority-change/scripts/analyze_majority_minority_change.R --input-start skills/majority-minority-change/data/county_race_acs5_2010.csv --input-end skills/majority-minority-change/data/county_race_acs5_2020.csv --start-label 2010 --end-label 2020

Review counties with largest percentage-point shifts and threshold crossings.

Output

Return:

non-white share in each year
percentage-point change
```
crossed_to_majority_minority
```
```
crossed_out_of_majority_minority
```