114 docs in four sections. Each section opens with an intro and
runs in clusters: a chapter about a kind of data, then the
briefs that use it. Generated from the doc files on 22 September 2026 by
_lib/make-index.py — re-run it after adding a doc rather than
editing this page. Click a column heading to sort (click a third time to
restore reading order); type to filter. 109 of 114 docs state a source in
their build script, and 83 commit a data/raw/ capture of that
source exactly as it arrives.
| Type | Doc | Tags | Updated | Data source |
|---|---|---|---|---|
| Front matter | ||||
| intro | introductionWhat data exists about American democracy, and what it says about the country | 2026-09-153e | — | |
| I. Data About the Population | ||||
| intro | part-1-census-bureauOne agency, three instruments, and the geography everything else is published on | 2026-09-153e | The corpus itself, which is why nothing here is downloaded. | |
| I.1 The Census and Its Products | ||||
| chapter | census-sourcerawThe one thing nothing else can do, and what it cost to keep doing it | 2026-09-153e | U.S. Census Bureau, 2020 Census P.L. 94-171 Redistricting Data… | |
| brief | census-accessrawOne question, a shelf of published answers, and how to say which one you mean | 2026-09-153e | https://www2.census.gov/programs-surveys/decennial/2020/data/">https://www2.census.gov/programs-surveys/decennial/2020/data/ | |
| brief | census-apirawThree ways into the Census Bureau, and what the catalog reveals about who the data is for | 2026-09-153e | https://api.census.gov/data/2020/dec/pl/variables.json">https://api.census.gov/data/2020/dec/pl/variables.json | |
| brief | census-coveragerawThe census grades itself, state by state, and mostly returns a blank | 2026-09-153e | U.S. Census Bureau, 2020 Post-Enumeration Survey estimation reports… | |
| brief | census-racerawTwo questions about one person on the 2020 census form, and the two boxes people use to answer neither | 2026-09-153e | https://api.census.gov/data/key_signup.html;">https://api.census.gov/data/key_signup.html; | |
| I.2 The Decennial Census | ||||
| chapter | census-decennialrawSeven questions asked once a decade, and the people it does not find | 2026-09-153e | 1. THE FILE ITSELF. U.S. Census Bureau, 2020 Census P.L. 94-171… | |
| brief | apportionmentrawEqual proportions, the seats the 2020 census moved, and a margin smaller than the instrument | 2026-09-153e | https://www.census.gov/data/tables/2020/dec/2020-apportionment-data.html">https://www.census.gov/data/tables/2020/dec/2020-apportionment-data.html | |
| brief | regional-shiftrawNobody proposed it, nobody voted on it, and no single decade announced it | 2026-09-183e | https://www2.census.gov/programs-surveys/decennial/2020/data/apportio… | |
| brief | demographicsrawWhat the 2020 census says about race, at eight scales, and why only the smallest one describes anyone | 2026-09-153e | Everything below is keyless and public. No Census API is used and no… | |
| brief | overplottingrawEvery populated census block in Georgia, and the 88% of them you cannot see | 2026-09-153e | Derived from other chapters in this book: areal-units | |
| I.3 The American Community Survey | ||||
| chapter | census-acsrawThree windows onto the same country, one of them closed, and a margin beside every number that has one | 2026-09-153e | U.S. Census Bureau, American Community Survey summary file, table… | |
| brief | migrationrawOne question on an American survey form, the map it draws, and how much of that map is not there | 2026-09-153e | https://www2.census.gov/programs-surveys/demo/tables/geographic-mobil… | |
| brief | chordrawEverybody moved, almost nobody left: a year of state-to-state flows in one circle, and the grouping the circle demands | 2026-09-153e | Derived from other chapters in this book: apportionment, migration,… | |
| brief | section-203Section 203 of the Voting Rights Act, and the survey that decides who gets a ballot they can read | 2026-09-153e | https://www.census.gov/programs-surveys/decennial-census/about/voting… | |
| I.4 Population Estimates | ||||
| chapter | census-peprawThe number you get when nobody counted, and the arithmetic that produces it | 2026-09-153e | U.S. Census Bureau, Population Estimates Program, county totals with… | |
| I.5 Census Geography | ||||
| chapter | census-geographyrawWhy so much data will not join | 2026-09-153e | Census Bureau Gazetteer Files, 2024.… | |
| brief | areal-unitsrawOne county's census blocks, and what moving the boxes does to the share | 2026-09-153e | derived/ga_block_race.csv built by build-block-race.R in this… | |
| brief | zip-codesrawA routing key for mail trucks, and everything it gets asked to stand for | 2026-09-153e | GeoNames, Postal Codes United States… | |
| brief | zero-population-blocksrawBuilding a national map of the census blocks that hold no people | 2026-09-153e | https://www2.census.gov/geo/tiger/TIGER2020/TABBLOCK20/">https://www2.census.gov/geo/tiger/TIGER2020/TABBLOCK20/ | |
| II. Survey Data | ||||
| intro | part-2-surveysAsking people, when nobody is obliged to answer | 2026-09-153e | The corpus itself, which is why nothing here is downloaded. | |
| II.1 What a Survey Can Establish | ||||
| chapter | surveys-sourcerawWhat no sample size will buy | 2026-09-153e | American National Election Studies, Time Series Cumulative Data File… | |
| brief | survey-accessrawFree, public, and unreachable — six survey archives, two ordinary programs, and a wall that judges the program rather than the person | 2026-09-153e | raw/probe-2026-08-13.csv, frozen, made by probe.py. Six addresses --… | |
| brief | age-structurerawWho is old enough to vote, who says they voted, and sixty years of the gap between the two | 2026-09-153e | https://www2.census.gov/programs-surveys/cps/tables/time-series/">https://www2.census.gov/programs-surveys/cps/tables/time-series/ | |
| II.2 Weighting | ||||
| chapter | ces-classrawThe Cooperative Election Study, the weights it applies to itself, and what a room can measure | 2026-09-153e | Cooperative Election Study Common Content, 2024 (Data Release 2),… | |
| brief | ces-statesSixty thousand people, and sixty-eight Vermonters: what the largest political survey in America buys, where it runs out, and the difference between being precise and being right | 2026-09-153e | Cooperative Election Study Common Content, 2024 (Data Release 2),… | |
| II.3 Political Polls | ||||
| brief | poll-weightingrawOne poll, five answers — 867 Florida voters, the weights that turned them into a result, and the winner that changes when you turn the dial | 2026-09-153e | TheUpshot/2016-upshot-siena-polls on GitHub, file… | |
| brief | poll-simulationEverything that goes wrong outside the sampling | 2026-09-153e | — | |
| II.4 Public Opinion | ||||
| chapter | anesrawSeventy-six years of survey answers, and three columns that run backwards | 2026-09-153e | American National Election Studies, Time Series Cumulative Data File… | |
| brief | party-idrawOne coding decision, and the most repeated number in American politics moves by a factor of five | 2026-09-153e | American National Election Studies, Time Series Cumulative Data File… | |
| brief | gender-gaprawFifty-two years of party identification, the one number everybody quotes, and the two numbers underneath it | 2026-09-153e | General Social Survey cumulative file, NORC at the University of… | |
| brief | ideologyrawA fifth of Americans decline to place themselves, and the survey invited them to | 2026-09-153e | American National Election Studies, Time Series Cumulative Data File… | |
| brief | partisan-economyrawOne economy, one question, and an answer that changes sign every time the White House changes party | 2026-09-153e | American National Election Studies, Time Series Cumulative Data File… | |
| brief | abortion-opinionrawThe most divisive issue in American politics, and whether opinion has actually moved | 2026-09-153e | American National Election Studies, Time Series Cumulative Data File… | |
| brief | gss-confidenceFifty-one years of one question, and the summary that does not survive it | 2026-09-153e | General Social Survey cumulative file, NORC at the University of… | |
| brief | thermometersNobody started liking their own side more | 2026-09-153e | American National Election Studies, Time Series Cumulative Data… | |
| brief | perception-gaprawEveryone guesses forty | 2026-09-153e | https://dataverse.harvard.edu/api/access/datafile/">https://dataverse.harvard.edu/api/access/datafile/ | |
| III. Administrative Data | ||||
| intro | admin-records-sourceRecords kept to run a system, about people who never asked to be described | 2026-09-153e | Nothing is downloaded. Every figure is read from the derived tables… | |
| III.1 Election Returns | ||||
| chapter | returns-sourcerawComplete, official, and permanently silent about who did it | 2026-09-153e | Everything here is read from files the corpus already built, so the… | |
| chapter | levels-of-aggregationWhich rung of the ladder answers which question | 2026-09-153e | — | |
| brief | county-returnsrawWhat it costs to assemble the file no agency publishes, and the half of the country where the answer changes | 2026-09-153e | Two files that answer the same question and were assembled in… | |
| brief | panhandle-claimrawWhat the Panhandle did: a claim about race and a Republican primary, tested four ways | 2026-09-153e | Florida Department of State, Division of Elections. Four files, all… | |
| brief | electoral-mapFifty-one rows, two inequalities, and the difference between where the votes are and where they matter | 2026-09-153e | https://github.com/jaytimm/PresElectionResults/raw/master/data/pres_b… | |
| brief | mappingrawOne election, three thousand counties, and the difference between coloring land and counting people | 2026-09-153e | https://www2.census.gov/geo/tiger/GENZ2024/shp/cb_2024_us_county_5m.zip">https://www2.census.gov/geo/tiger/GENZ2024/shp/cb_2024_us_county_5m.zip | |
| brief | wind-maprawEvery county's move from 2020 to 2024, drawn as three thousand arrows | 2026-09-153e | , STATED PLAINLY | |
| brief | sparklinesrawA table that is also fifty-one charts | 2026-09-153e | Derived from other chapters in this book: historical-campaigns | |
| brief | historical-campaignsForty-one elections in one file, and the question of whether the country was always divided this evenly | 2026-09-153e | https://github.com/jaytimm/PresElectionResults">https://github.com/jaytimm/PresElectionResults | |
| brief | distributionsrawFour hundred House districts drawn four ways, and an average that describes almost nothing | 2026-09-153e | Derived from other chapters in this book: house-competition | |
| brief | crossoverrawOne election against the next, one office against another, and the sliver of the map that was genuinely in play | 2026-09-153e | Derived from other chapters in this book: 06-putting-data-together,… | |
| brief | nationalizationrawOne election, not fifty | 2026-09-153e | https://voteview.com/static/data/out/members/HSall_members.csv">https://voteview.com/static/data/out/members/HSall_members.csv | |
| brief | bellwetherrawWhy a county that always picks the winner predicts nothing | 2026-09-153e | https://doi.org/10.7910/DVN/DGUMFI">https://doi.org/10.7910/DVN/DGUMFI | |
| brief | vote-targetingNorth Carolina, and the assumptions inside the arithmetic | 2026-09-153e | NCSBE publishes precinct-level results for every general election:… | |
| brief | ga-precinct-returnsrawThe state's own precinct file, the vote-method columns, and what counting five million ballots twice actually changed | 2026-09-153e | — | |
| brief | precinct-geographyPrecinct names and precinct lines drift independently, so a precinct that swung may be a precinct that moved | 2026-09-153e | Derived from other chapters in this book: ga-precinct-returns | |
| brief | cast-vote-recordsrawA congressional election recounted from the ballots themselves, and what the winner's majority is a majority of | 2026-09-153e | CVR… | |
| brief | residual-votesrawA punch card from Palm Beach County, and the one subtraction no agency performs | 2026-09-153e | https://www2.census.gov/geo/tiger/TIGER2020/COUNTY/tl_2020_us_county.zip">https://www2.census.gov/geo/tiger/TIGER2020/COUNTY/tl_2020_us_county.zip | |
| brief | eavsA survey the country's election offices fill in about themselves, and the empty box that decides who can be compared | 2026-09-153e | U.S. Election Assistance Commission, Election Administration and… | |
| III.2 Rosters of Officeholders | ||||
| chapter | rosters-sourceA roster is not a return, and the gap between them is 2,344 people | 2026-09-153e | Nothing is downloaded. Every figure is read from the derived tables… | |
| brief | house-competitionrawOne name on the ballot, and the seventy years of House elections that made it ordinary | 2026-09-153e | Gary C. Jacobson's House elections dataset, 1946-2014, held locally:… | |
| brief | incumbency-advantagerawA vote-share premium that rose for thirty years, peaked in the 1980s, and has nearly vanished since | 2026-09-223e | MIT Election Data and Science Lab, 2017, "U.S. House 1976-2024",… | |
| brief | midterm-lossA very reliable rule, and what it still cannot tell you | 2026-09-153e | Office of the Historian, U.S. House of Representatives, "Party… | |
| brief | retirementsrawTwo centuries of members leaving the House, and the one thing the record will not say | 2026-09-153e | https://voteview.com/static/data/out/members/HSall_members.csv">https://voteview.com/static/data/out/members/HSall_members.csv | |
| brief | primary-defeatsThe election that is supposed to substitute for the one nobody contests | 2026-09-153e | Derived from other chapters in this book: retirements | |
| brief | primary-positionsrawWhat candidates would not say, and the code that records the silence | 2026-09-153e | Brookings Institution, The Primaries Project. Compiled by Elaine C.… | |
| brief | nomination-anchorsrawThe dataset that was never released, and a column that exists in none of its tables | 2026-09-153e | Bawn, K., Brown, K., Ocampo, A. X., Patterson, S. Jr., Ray, J. L.… | |
| brief | nomination-rulesThe rule that picks the nominee, found without reading a single statute | 2026-09-153e | Federal Election Commission, *Federal Elections*, the biennial… | |
| brief | careersrawTwo-thirds of the newest careers have not finished, and the arithmetic that counts them anyway | 2026-09-153e | https://voteview.com/">https://voteview.com/ | |
| III.3 Voter Files | ||||
| chapter | voter-files-sourcerawThe only instrument that can tell you whether a survey is lying | 2026-09-153e | Georgia: Houston County registration extract 71754 and the state's… | |
| brief | voter-file-accessrawPublic by law, and you still cannot have it: fifty-one official pages, one downloadable statewide list | 2026-09-153e | Two things, both frozen in raw/, both dated 2026-08-13: | |
| brief | voter-filesrawOne county's file, the codes that decide which ballot you are handed, and the elections it has already forgotten | 2026-09-153e | , AND THEY ARE TWO. Houston County, GA registration extract 71754,… | |
| brief | false-matchesTwo people who have never met, one match key, and a purge rule built on the assumption that such a thing cannot happen | 2026-09-153e | 1. NEW JERSEY -- the primary source, and the one that carries a FULL… | |
| brief | disenfranchisementrawFelony disenfranchisement, state by state, and a category that outlives the sentence | 2026-09-153e | Uggen, Larson, Shannon & Stewart, "Locked Out 2024: Four Million… | |
| brief | gotvField experiments priced the mobilization tactics campaigns buy, and the price list does not rank the way anyone assumes | 2026-09-153e | Donald P. Green and Alan S. Gerber, "Get Out the Vote: How to… | |
| III.4 Roll Calls and Courts | ||||
| chapter | rollcalls-sourcerawEvery vote since 1789, and no note of what most of them were about | 2026-09-153e | Voteview, HSall_rollcalls.csv -- every recorded vote in the House… | |
| brief | dw-nominaterawThe widest gap since Reconstruction, and a chamber nobody redistricts | 2026-09-153e | https://voteview.com/">https://voteview.com/ | |
| brief | scdbNine justices, forty-five agreement rates, and a left–right line nobody put in | 2026-09-153e | Supreme Court Database, Washington University, 2024 release 01,… | |
| brief | oral-argumentrawSix defensible answers to who talks the most, and one number that turns out to be about a rule | 2026-09-153e | Oyez, a project of Cornell's Legal Information Institute, Justia and… | |
| brief | officeholder-agerawThe answer depends on which row of the life table you read | 2026-09-163e | https://voteview.com/static/data/out/members/HSall_members.csv">https://voteview.com/static/data/out/members/HSall_members.csv | |
| III.5 Campaign Finance | ||||
| chapter | finance-sourcerawThe only source in this book written by the people it is about | 2026-09-153e | Read from files the corpus already built, so the provenance is a… | |
| brief | campaign-financerawWhere a campaign's money comes from, and where the record stops naming it | 2026-09-153e | https://www.fec.gov/files/bulk-downloads/2024/weball24.zip">https://www.fec.gov/files/bulk-downloads/2024/weball24.zip | |
| brief | pie-radarrawTwo charts that hide their own data | 2026-09-153e | Derived from other chapters in this book: campaign-finance | |
| brief | independent-expendituresrawA junk filing in the federal record of political spending, and the difference between wrong and plausibly wrong | 2026-09-153e | FEC bulk downloads, independent expenditures, 2024 cycle:… | |
| brief | lobbyingrawA quarterly lobbying disclosure, and the two things the form was built not to record | 2026-09-153e | United States Senate / Clerk of the House, Lobbying Disclosure Act… | |
| brief | finance-networkrawThe FEC publishes edges, and leaves you to decide what the nodes are | 2026-09-153e | https://www.fec.gov/files/bulk-downloads/2024/independent_expenditure… | |
| III.6 Media and Attention | ||||
| brief | media-attentionA year of Wikipedia pageviews, and what a country looking something up does not tell you | 2026-09-153e | https://wikimedia.org/api/rest_v1/#/Pageviews%20data">https://wikimedia.org/api/rest_v1/#/Pageviews%20data | |
| brief | media-ideologyraw375,979 shared links, 184 news domains, and the one politician holding up the right-hand end of the scale | 2026-09-153e | `PolShares`, the demonstration dataset shipped with the R package… | |
| brief | streamgraphrawTwelve series that do not share a scale | 2026-09-153e | Derived from other chapters in this book: media-attention | |
| brief | campaign-visitsrawEight months of one campaign's travel log, and the half of America that never appears in it | 2026-09-153e | https://interactives.ap.org/embeds/ziEfd/54/">https://interactives.ap.org/embeds/ziEfd/54/ | |
| brief | follower-countsrawThree platforms, one question, and three different Congresses | 2026-09-153e | raw/followers-2026-08-15.csv, frozen, made by collect.mjs. One row… | |
| III.7 Records of Ordinary People | ||||
| brief | policingTraffic stops in one city, and a disparity that changes sign when you change the denominator | 2026-09-153e | U.S. Census Bureau, American Community Survey 5-year estimates, via… | |
| brief | jury-selectionPeremptory strikes in one Mississippi district, and a pattern no single trial can show | 2026-09-153e | APM Reports, "In the Dark" season 2, jury data.… | |
| brief | redliningrawA form filled in on the edge of Cleveland in 1939, and the ground it still describes | 2026-09-153e | HOLC areas, national… | |
| brief | namesrawThe field every record about a person carries, and the three jobs it is asked to do | 2026-09-153e | U.S. Census Bureau, "Frequently Occurring First Names in the 2020… | |
| IV. Putting Data Together | ||||
| intro | part-6-putting-data-togetherWhere the interesting questions and nearly all the errors live | 2026-09-153e | The corpus itself, which is why nothing here is downloaded. | |
| IV.1 Joining Files | ||||
| chapter | data-sourcesrawCounty FIPS, the join that never failed — and what that actually proves | 2026-09-153e | https://github.com/tonmcg/US_County_Level_Election_Results_08-24">https://github.com/tonmcg/US_County_Level_Election_Results_08-24 | |
| IV.2 Survey Meets Record | ||||
| brief | validated-turnoutCounted ballots against what people said they did, 1964-2024 | 2026-09-153e | 1. U.S. Census Bureau, Current Population Survey, Voting and… | |
| IV.3 Race Inference | ||||
| brief | surnamesrawA federal table of names, a county, and one line of arithmetic that decides who counts as a Black voter | 2026-09-153e | https://www2.census.gov/topics/genealogy/2010surnames/names.zip">https://www2.census.gov/topics/genealogy/2010surnames/names.zip | |
| brief | bisg-checkrawBISG checked against the truth, on forty-six thousand Georgia voters | 2026-09-153e | Houston County, GA voter registration and the geocoded block… | |
| brief | uncertaintyrawOne surname model scored 49,771 times, and a tight interval around a number that describes nobody | 2026-09-153e | Derived from other chapters in this book: bisg-check | |
| brief | rpvA Georgia county where the inference can be checked against the ballots | 2026-09-153e | Houston County, GA voter registration (extract 71754), the state's… | |
| IV.4 Districts | ||||
| brief | neutral-mapsrawFive Thousand Maps Nobody Drew | 2026-09-153e | The ALARM Project's 50-State Redistricting Simulations, Harvard… | |
| brief | mid-decaderawThe Same Votes, Redrawn: Texas in 2025, Florida in 2026 | 2026-09-153e | https://www2.census.gov/geo/docs/reference/cenpop2020/">https://www2.census.gov/geo/docs/reference/cenpop2020/ | |
| brief | redistrictingrawThree bias statistics on one column of numbers, and the states where they disagree | 2026-09-153e | https://github.com/jaytimm/PresElectionResults">https://github.com/jaytimm/PresElectionResults | |
| brief | vote-dilutionTwo legal tests, run on the census blocks of one Georgia county | 2026-09-153e | ---------------------------------------------- | |
| brief | sweet-spotrawSix defensible answers, and what happens between packing and cracking | 2026-09-153e | https://www2.census.gov/programs-surveys/decennial/rdo/datasets/">https://www2.census.gov/programs-surveys/decennial/rdo/datasets/ | |
| IV.5 Forecasting | ||||
| brief | models-markets746 resolved prediction markets, and what it takes to grade a forecast | 2026-09-153e | https://gamma-api.polymarket.com/markets?closed=true&order=volumeNum">https://gamma-api.polymarket.com/markets?closed=true&order=volumeNum | |
| brief | seat-forecastrawA working model of November 2026, and the one number in it that nobody measured | 2026-09-153e | 1. Wikipedia, "2026 United States House of Representatives… | |
| brief | senate-2026rawThirty-five races, rated in August, and a number you cannot revise in November | 2026-09-153e | https://en.wikipedia.org/w/api.php?action=parse&page=">https://en.wikipedia.org/w/api.php?action=parse&page= | |
| brief | election-nightThe 2026 Senate map, and how to tell a real shift from the order of the count | 2026-09-153e | https://unitedstates.github.io/congress-legislators/legislators-curre… | |
| IV.6 Proxies and Indices | ||||
| brief | turnout-denominatorrawOne numerator, two denominators, and a thirty-year decline that is not in the data | 2026-09-153e | https://www.jstor.org/stable/3117725">https://www.jstor.org/stable/3117725 | |
| brief | cost-of-votingrawFifty-six election laws compressed into one number, and the decisions that decide what it says | 2026-09-153e | https://costofvotingindex.com">https://costofvotingindex.com | |
| brief | whole-foods-cracker-barrelrawThe metric replicates to within a point, and then does not survive being asked about people | 2026-09-153e | OpenStreetMap contributors, United States objects tagged with the… | |
| brief | rank-sizerawThe same curve, six times — and how much you leave out before it becomes a law | 2026-09-153e | — | |