Skip to contents

Shape palette from Cleveland "Elements of Graphing Data" (discrete).

Source: R/shapes.R
cleveland_shape_pal.Rd

Shape palettes for overlapping and non-overlapping points.

Usage

cleveland_shape_pal(overlap = TRUE, unicode = FALSE)

Arguments

overlap

logical Use the scale for overlapping points?

unicode

If TRUE, return pch codes derived from Unicode glyphs, as this palette did before ggthemes 6.1.0. Glyph shapes are drawn by the device font, so they render as blank boxes in a font without coverage; the default returns base pch codes, which every font can draw.

Note

In the Elements of Graphing Data, W.S. Cleveland suggests two shape palettes for scatter plots: one for overlapping data and another for non-overlapping data. The symbols for overlapping data rely on pattern discrimination, while the symbols for non-overlapping data vary the amount of fill.

Following Tremmel (1995), the circle with a vertical line is replaced by an encircled plus sign.

cleveland_shape_pal(overlap = TRUE) supports four values on either branch.

cleveland_shape_pal(overlap = FALSE) supports three values by default and five with unicode = TRUE. Its five source symbols encode fill fraction, which base pch cannot express, so the two partially filled circles are dropped rather than approximated by a different shape. To encode a proportion, map alpha or fill instead; to restore the five glyphs, use unicode = TRUE with a font that covers Mathematical Operators, such as STIX Two Text.

The truncation is arguably an improvement. Tremmel (1995) Experiment 2 tested exactly this five-symbol set and found the fill-graded circles the worst performers measured, with the encircled plus and encircled dot the slowest pair and the one producing the most errors. The three shapes that survive are the better-separating subset.

References

Cleveland WS. The Elements of Graphing Data. Revised Edition. Hobart Press, Summit, NJ, 1994, pp. 154-164, 234-239.

Tremmel, Lothar, (1995) "The Visual Separability of Plotting Symbols in Scatterplots", Journal of Computational and Graphical Statistics, https://www.jstor.org/stable/1390760

See also

Other shapes: circlefill_shape_pal(), scale_shape_circlefill(), scale_shape_cleveland(), scale_shape_tremmel(), tremmel_shape_pal()

Examples

###   (discrete).

if (FALSE) { # \dontrun{
library("ggplot2")
p <- ggplot(mtcars) +
     geom_point(aes(x = wt, y = mpg, shape = factor(gear))) +
     facet_wrap(~am) +
     theme_bw()
# overlapping symbol palette
p + scale_shape_cleveland()
# non-overlapping symbol palette
p + scale_shape_cleveland(overlap = FALSE)
} # }