---
name: poq-tables-figures
description: Use when building tables and figures for a Public Opinion Quarterly (POQ) manuscript so exhibits are self-contained, accessible, and show survey-based uncertainty correctly. POQ word caps count text and notes but exclude figures and tables, so exhibits carry weight without costing the budget — but they must report design-based uncertainty and weighting. Designs exhibits; it does not run the analysis.
---

# Tables & Figures (poq-tables-figures)

Exhibits are where an expert reviewer checks whether the survey result is real. At POQ the word caps
count **text and notes but exclude figures, tables, references, and appendices** — so exhibits do not
eat your budget, but they must be honest about **design-based uncertainty, weighting, and the base N**
behind every estimate.

## When to trigger

- Designing the main results table/figure or a key descriptive/trend exhibit
- Deciding what belongs in the article vs. an appendix
- A reviewer found an exhibit unclear, mislabeled, or missing its base/weighting note
- Presenting opinion trends, subgroup differences, or experimental treatment effects

## Principles

1. **Self-contained.** A reader should understand each exhibit from its title, labels, and note alone —
   state the **base N (weighted/unweighted), the population, the question wording (or short form), the
   mode, and the uncertainty measure** (CI / margin of error).
2. **Show survey uncertainty.** Report confidence intervals or margins of error from **design-based**
   estimates; never plot a point estimate with no uncertainty. Note if estimates are weighted.
3. **Figures over dense tables for effects.** Trend lines with uncertainty bands, coefficient/forest
   plots with design-based SEs, conjoint AMCE plots, predicted-probability plots communicate magnitude
   and uncertainty better than a wall of coefficients.
4. **Accessible.** Colorblind-safe palettes; legible in grayscale; no chartjunk, no 3D. Reviewers must
   parse it quickly.
5. **Reproducible.** Each exhibit is generated by the master script; numbers match the deposited
   replication package exactly (POQ re-runs it — see `poq-transparency-and-data-policy`).

## POQ-specific exhibits
- Opinion **trend lines** with uncertainty bands; mark wording/mode breaks in the series.
- **Weighted vs. unweighted** overlays where weighting matters to the read.
- Conjoint AMCE plots; survey-experiment treatment-effect plots with design-based intervals.
- A **disclosure-friendly methods table** summarizing population, mode, dates, N, and response rate
  (links to Appendix A — see `poq-survey-design-and-measurement`).

## Anti-patterns

- A point estimate or trend line with no interval / margin of error
- Tables that omit the base N or hide whether estimates are weighted
- Exhibits that require the prose to be intelligible (not self-contained)
- Color-only encoding that fails in grayscale or for colorblind readers
- Exhibit numbers/values that don't match the deposited code output


## Exhibit pass for Public Opinion Quarterly

Run this as a concrete capability pass. First lock the public-opinion construct, sampling frame, mode effects, weighting/nonresponse plan, and trend or causal interpretation; then test whether the manuscript addresses survey and public-opinion reviewers who inspect measurement, sampling, mode, nonresponse, and inference about attitudes or behavior.

- **Primary move:** For every table or figure, state the object, sample/case base, uncertainty display, and one sentence the exhibit proves for this venue.
- **Decision ledger:** return `claim / evidence / blocker / next edit` rows so the next pass can patch the manuscript directly.
- **Sibling comparison:** compare against Political Analysis for methods-first work, Journal of Politics for political-science theory, Communication Research for media-effects framing; if the neighboring outlet has the stronger audience claim, recommend re-routing before polishing.
- **Verification floor:** before submission-ready advice, re-open `resources/official-source-map.md` for volatile rules and name the one unresolved fact that could change the recommendation.

## Output format

```
【Main exhibit】what it shows + why a figure/table
【Self-contained?】title + labels + note + base N + wording/mode? [Y/N]
【Uncertainty shown?】CI / MOE from design-based estimates? [Y/N]
【Weighting noted?】weighted vs unweighted clear? [Y/N]
【Accessible?】grayscale-legible + colorblind-safe? [Y/N]
【Reproducible?】generated by master script, matches package? [Y/N]
【Next】poq-writing-style
```

## Supplementary resources

- [`../../resources/external_tools.md`](../../resources/external_tools.md) — plotting tools for trends, AMCE, and design-based intervals
- [`../../resources/official-source-map.md`](../../resources/official-source-map.md) — word-count rule (excludes figures/tables/appendices)