jcp-tables-figures
GitHub为JCP论文构建符合APA规范的表格与图表，包括条件均值表、中介/调节路径图及交互图。强调展示效应量、置信区间及简单效应，确保图表独立可读并支持多研究一致性，辅助审稿人理解机制。
触发场景
安装
npx skills add brycewang-stanford/Awesome-Journal-Skills --skill jcp-tables-figures -g -y
SKILL.md
Frontmatter
{
"name": "jcp-tables-figures",
"description": "Use when building the exhibits for a Journal of Consumer Psychology (JCP) manuscript — condition-means tables, mediation\/moderation path figures, interaction plots, and APA-formatted statistical reporting that makes the process visible. Builds the exhibits; it does not run the analysis (jcp-data-analysis)."
}
Tables & Figures (jcp-tables-figures)
When to trigger
- Your means tables show stars but not the effect sizes/CIs a reviewer needs
- A mediation result is buried in prose instead of shown as a path diagram
- An interaction is described but there is no plot of the simple effects
- Tables/figures do not follow APA format and the copyeditor will bounce them
- Exhibits across studies are inconsistent, so the multi-study chain is hard to follow
Exhibits at JCP must make the process legible
In a JCP paper the exhibits exist to let a reader see the mechanism, not just the effect. A condition-means table that shows cells without the contrast that tests the hypothesis wastes space; a mediation figure that shows the indirect path with its coefficient and CI earns its place. Build exhibits in APA Style (7th edition; 检索于 2026-06；以官网为准) — table titles and notes, figure captions, decimals and statistics formatted per APA — because JCP formats manuscripts to APA and non-conforming exhibits create friction at every stage.
What each exhibit type should carry
| Exhibit | Must show | Common failure |
|---|---|---|
| Condition means | M, SD (or SE), n per cell; the focal contrast | a sea of numbers with no highlighted comparison |
| Mediation path figure | a→b paths with coefficients, the indirect effect + bootstrap CI | only c/c′ shown; CI omitted |
| Moderation / interaction plot | the interaction with probed simple effects (spotlight values) | bar chart with no simple-effect annotation |
| Moderated mediation | conditional indirect effects by moderator level + index | a single mediation figure that hides the moderation |
| Multi-study summary | a study-by-study row: design, N, key result, effect size | inconsistent metrics across studies |
A path diagram for the focal study and an interaction plot for the key moderation are usually the two highest-value figures; they carry the process argument visually.
Reporting conventions (APA + rigor era)
- Report effect sizes and CIs in tables, not asterisks alone; readers must judge magnitude.
- Keep decimal places consistent and follow APA for statistics (e.g., italic test letters, leading zeros per APA convention).
- Note Ns before and after exclusions in the table or its note.
- Each exhibit must be interpretable standalone — title, note defining abbreviations, and units.
- Do not duplicate the same result in a table and a figure; pick the form that shows the mechanism best.
Execution bridge (StatsPAI / Stata MCP)
Generate exhibits from the fitted result, not by retyping numbers (the usual source of
body-vs-appendix drift). Full map: execution-with-mcp. JCP is experimental consumer psychology; randomization inference, mediation done right (mediate, not naive controlling-away), and family-wise corrections matter most.
- Tables:
etable(multi-model columns) ordid_summary_to_latexstraight from theresult_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 effect size in interpretable units.
See a full fitted-result → exhibit chain in the JF execution walkthrough.
Checklist
- Every table/figure earns its place by making the effect or process clearer than prose would
- Mediation shown as a path figure with coefficients and the bootstrap CI
- Interaction shown with probed simple effects (spotlight/floodlight annotated)
- Effect sizes and CIs present; not asterisk-only
- APA formatting for tables, figures, notes, and statistics
- Ns (before/after exclusions) reported; abbreviations defined in notes
- Metrics consistent across the multi-study exhibits so the chain reads cleanly
Anti-patterns
- Asterisk theater: stars instead of effect sizes and CIs
- Buried mediation: the indirect effect described in text with no figure and no CI
- Naked interaction: an interaction plot with no simple-effect probing shown
- Decorative figures: a chart that repeats a table and adds nothing about the mechanism
- Non-APA exhibits: formatting that the JCP copyeditor will return
- Inconsistent multi-study tables: different metrics per study, hiding convergence
Output format
【Key exhibits】which table(s)/figure(s) carry the effect and the process
【Mediation figure】paths + coefficients + bootstrap CI shown? [Y/N]
【Interaction plot】simple effects probed/annotated? [Y/N]
【Reporting】effect sizes + CIs (not asterisk-only)? APA format? [Y/N]
【Multi-study table】consistent metrics across studies? [Y/N]
【Next skill】jcp-writing-style
版本历史
- 1839142 当前 2026-07-05 13:29
