Exploratory plots for SAD validation

Produces the exploratory figures of a conventional SAD validation study: estimates against actual severity, Cb against r, boxplots of Lin's statistics, errors along the severity range, error densities, the standard deviation among raters along the severity range and the gain in accuracy relative to unaided estimates.

Usage

sad_plots(x, ccc = sad_ccc(x), baseline = NULL, palette = NULL)

Arguments

x: An object returned by sad_data().
ccc: Statistics per rater, by default sad_ccc(x).
baseline: Optional name of the method used as baseline (for example "UN"); when given, the plot of gain in \(\rho_c\) against baseline accuracy is added.
palette: Optional named vector of colours, one per condition.

Value

A named list of ggplot objects: scatter, cb_vs_r, box_ccc, error, error_density (requires ggridges), sd_vs_actual and, with baseline, gain_vs_baseline.

Examples

x <- sad_data(SADRustSoybean, wide = c(UN = "unaided", SAD = "aided"), group = "sad")
p <- sad_plots(x, baseline = "UN")
p$scatter