Skip to contents

Radar Chart

Source: R/plt_radar.R
plt_radar.Rd

Visualise multivariate effect sizes (e.g. from stat_cohen()) as a radar chart. Supports base R (fmsb) or ggplot (ggiraphExtra) output modes.

Usage

plt_radar(
  res,
  ref_group = NULL,
  palette = NULL,
  output = c("base", "gg"),
  facet = FALSE,
  min_max = NULL,
  n_breaks = 6
)

Arguments

res

Data frame from stat_cohen(): first column is Variable, remaining columns are numeric values per group.

ref_group

Character, reference group for sorting variables. Default NULL uses the first group column.

palette

Colour palette. Character vector or palette name. Default uses pal_lancet.

output

Output mode: "base" (default, uses fmsb) or "gg" (uses ggiraphExtra). The gg mode automatically facets by group.

facet

Logical, facet each group separately (base mode only). Default FALSE.

min_max

Numeric vector c(min, max) for axis limits. Default NULL auto-detects from data.

n_breaks

Integer, number of axis breaks. Default 6.

Value

For output = "gg", a ggplot object. For output = "base", returns invisible(res) after printing.

Note

Requires fmsb for base mode, ggiraphExtra + ggh4x for gg mode.

See also

Other plot: PlotButterfly(), PlotButterfly2(), PlotRankCor(), plt_cat(), plt_cohen(), plt_con(), plt_dist(), plt_sankey(), plt_upset()

Examples

# Using iris Cohen's d
res <- stat_cohen(iris, "Species",
                  c("Sepal.Length","Sepal.Width","Petal.Length","Petal.Width"))

# Base radar chart
plt_radar(res)


# Base radar, faceted per group
plt_radar(res, facet = TRUE)

# ggplot radar: overlay (default, all groups on one chart)
plt_radar(res, output = "gg")

# ggplot radar: faceted (one panel per group)
plt_radar(res, output = "gg", facet = TRUE)

# ggplot radar with custom palette
plt_radar(res, output = "gg", palette = "Set2")

# Custom axis range (base mode)
plt_radar(res, min_max = c(-1, 1))


# Sort by specific group
plt_radar(res, ref_group = "versicolor")

# Simulated data
df <- data.frame(
  group = factor(sample(c("A","B","C"), 120, TRUE)),
  v1 = rnorm(120), v2 = rnorm(120), v3 = rnorm(120),
  v4 = rnorm(120), v5 = rnorm(120)
)
res2 <- stat_cohen(df, "group", paste0("v", 1:5))
plt_radar(res2, output = "gg")                         # overlay
plt_radar(res2, output = "gg", facet = TRUE)            # faceted
plt_radar(res2, output = "gg", palette = "Paired")