jjstatsplot

Publication-ready statistical plots for clinicopathological papers

What it does

jjstatsplot is the group’s plotting front-end. It wraps ggstatsplot and ggplot2 in jamovi-friendly dialogs and produces plots that follow a consistent visual style across every paper.

Plot types

Between-groups comparisons — box / violin / dot plots with embedded test statistics and effect sizes.
Correlation scatters — with regression, CI, and statistics annotation.
Categorical plots — bar / mosaic with chi-square / Fisher’s results embedded.
Paired / repeated measures — slope plots, within-subject comparisons.
Forest plots — for meta-analyses and multi-subgroup summaries.
Kaplan–Meier — via jsurvival, using the same visual style.

When to use

You need a plot that’s ready to paste into a manuscript without hand-edits.
You want a consistent visual identity across all group papers.
You need statistics embedded directly in the plot (as reviewers increasingly expect).

Repos

jjstatsplot — main module.
ggstatsplot — upstream building block.
vijPlots — auxiliary plot helpers.

Quick start in jamovi

Analyses → ClinicoPath → Plots → Between-groups.
Select the numeric dependent variable and the grouping variable.
Plot appears with the appropriate test (t-test, ANOVA, Kruskal–Wallis) and effect size.
Copy out as PNG or SVG.

Quick start in R

library(jjstatsplot)
between_plot(
  data = mydata,
  y = "ki67",
  x = "grade"
)

Pitfalls

Default tests are auto-selected. If the module picks Kruskal–Wallis because of non-normality, don’t override to t-test without a defensible reason — the footnote already documents the choice.
Colors. The default palette is colorblind-safe. Please don’t swap it for the journal’s “house colors” manually — use the theme argument if the journal requires specific colors.
Effect sizes. Always report the effect size the module computes (Hedges’ g, omega-squared, Cramér’s V). It’s embedded in the plot; don’t strip it out.