Tables & Figures (pubar-tables-figures)

Exhibits are where an expert reviewer checks whether the result is real — and where a practitioner reads the magnitude that drives your Evidence for Practice. At PAR the word count excludes tables, figures, charts, and appendices (检索于 2026-06；以官网为准), so the constraint is clarity, not word budget: every exhibit must communicate a magnitude with its uncertainty, fast.

When to trigger

• Designing the main results table/figure or a key descriptive exhibit
• Deciding what belongs in the article vs. an online appendix/supplement
• A reviewer found an exhibit unclear, mislabeled, or non-self-contained
• Translating a coefficient into something a public manager can read

Principles

1. Self-contained. A reader should understand each exhibit from its title, axis/column labels, and note alone. State units, sample, N, the estimator, and what the estimate is.
2. Figures over dense tables for effects. Coefficient/forest plots, marginal-effects and predicted-probability plots, event-study and RD plots communicate magnitude and uncertainty better than a wall of coefficients. Show intervals — a practitioner needs the effect size, not stars.
3. Accessible. Colorblind-safe palettes; legible in grayscale; no chartjunk, no 3D, no needless color. Reviewers and practitioner readers must parse it quickly.
4. Main text vs. supplement. Keep the few exhibits that carry the argument in the article; move balance tables, full specifications, and robustness grids to the online supplement.
5. Reproducible. Each exhibit is generated by the master script; numbers match the deposited materials exactly (TOP transparency — see pubar-transparency-and-data).

PA-specific exhibits

• Event-study plots around a reform to show pre-trends and dynamics of an administrative change.
• Predicted-probability / marginal-effects plots translating a model into managerial terms ("an agency at the 75th percentile of red tape is X points less likely to…").
• Maps for cross-jurisdiction variation; network diagrams for collaborative-governance structure.
• For qualitative/mixed work: process timelines, evidence tables linking claims to sources.

Execution bridge (StatsPAI / Stata MCP)

Generate exhibits from the fitted result, not by retyping numbers. Full map: execution-with-mcp. PAR is public administration — survey/observational and some experimental work; identification + clustered/multilevel inference, magnitude for practice.

• Tables: etable (multi-model) or did_summary_to_latex straight from the result_id.
• Figures: plot_from_result / enhanced_event_study_plot / event_study_table — axis units and the SE/clustering note baked in.
• Every note names the estimator + clustering and states the magnitude in interpretable units.

See a full fitted-result → exhibit chain in the JF execution walkthrough.

Anti-patterns

• Tables that require the prose to be intelligible (not self-contained)
• Reporting significance stars with no effect size or interval (practitioners can't act on it)
• Cramming every robustness check into the main text (use the supplement)
• Color-only encoding that fails in grayscale or for colorblind readers
• Exhibit numbers/values that don't match the deposited code output

Output format

【Main exhibit】what it shows + why a figure/table
【Self-contained?】title + labels + note + N/units/estimator present? [Y/N]
【Magnitude legible to a manager?】effect size + interval shown? [Y/N]
【Accessible?】grayscale-legible + colorblind-safe? [Y/N]
【Article vs supplement】split decided
【Reproducible?】generated by master script, matches package? [Y/N]
【Next】pubar-writing-style

Referee-pushback patterns and the PAR fix

• "I can't read the magnitude — the table is all stars." → Replace stars-only cells with effect sizes and intervals; add a marginal-effects or predicted-probability plot so a practitioner sees the size.
• "The exhibit isn't self-contained." → Put the sample, N, units, estimator, and what the estimate is into the title and note, so the figure stands alone without the prose.
• "This figure fails in grayscale / for colorblind readers." → Switch to a colorblind-safe palette and encode with shape/linetype, not color alone; check the grayscale print.
• "The main text is buried under robustness tables." → Keep the few exhibits that carry the argument in the article and move balance/robustness grids to the online supplement.
• "Numbers don't match the deposited code." → Regenerate every exhibit from the master script so the printed values and the deposited materials are identical (see pubar-transparency-and-data).

Calibration anchors (hedged)

• A PAR exhibit serves two readers: an expert checking whether the result is real, and a practitioner reading the magnitude that drives the Evidence for Practice. Design for both.
• Because the word count excludes tables, figures, charts, and appendices (检索于 2026-06；以官网为准), use the supplement freely for secondary exhibits — but keep the main argument to a few decisive ones.
• Confirm the current figure/table formatting and file-type requirements on the journal's author page; Wiley production specs evolve.

Supplementary resources

• ../../resources/external_tools.md — plotting/mapping/network packages
• ../../resources/official-source-map.md — word-count rule (excludes tables/figures/appendices)