Skip to content

Six national surveys · 1972–2024

Chat with fifty years of American survey data.

Ask a question and get an answer computed from six of the country's longest-running surveys, with the source attached. Turn it into an essay, a brief or a post. Every number is checked against the data before it's published.

In 2024, said abortion should be legal for any reason — up from 39.5% in 2006.

Every number links to its source, years, sample size and margin of error.

66 essays so far, all built this way.

Every number comes from a real calculation.

Ask anything the surveys cover. ctzn.pub runs the numbers on the original data and shows its work.

Ask

Ask a question in plain English. You get the answer, its margin of error and a citation.

Write

Turn answers into essays and policy briefs, with slide decks coming soon. Every number is checked again before anything is published.

Cite

Every finding has its own permanent link, with APA and BibTeX ready to copy.

From raw data to citation

The same four steps for every number on the site.

Snapshot

A fixed copy of the source data, so a result reads the same next year as it does today.

Recipe

The calculation, using each survey’s own weights and rules.

Claim

The result, with its source and method, at a permanent link.

Citation

APA or BibTeX in one click.

Checked against published research

We reproduce published findings, then carry them forward to the latest survey years.

4 studies

Published findings from academic journals and federal agencies, reproduced here.

16 of 16

Figures from a federal agency’s 2005–2015 table, matched to the decimal.

0.005 points

The largest gap between our results and an independent analysis, across 289 estimates.

A statistician reviews every analysis

Before we publish results from a new source, a statistician reviews the full plan: which weights, which groups, and where a series changes over time. If the plan changes, it gets reviewed again. Nothing runs until it's approved.

Built for libraries

What libraries look for in a reference tool:

  • Click any number to see where it came from
  • Caveats shown right beside the results
  • Citations that keep working, guaranteed in writing
  • Questions are never used to train models
  • Built on public data, free to cite and share

For AI and research tools

Every essay is also published as Markdown and structured data, so research tools and AI assistants can read the findings and cite the sources.

  • Sourced
    Every figure lists its source and survey years.
  • Structured
    Markdown and structured data for every essay. APIs coming soon.
  • Current
    New essays appear as soon as they’re published.