Plot symptom selection against the chance baseline

Draws the per-item selection frequencies from symptom_selection together with the frequency expected by chance. The baseline is drawn item by item rather than as a single reference line, because under the hierarchical rule it differs by cluster: a flat line would make two-item clusters look preferred when they are simply required.

Usage

plot_symptom_selection(
  selection,
  type = c("frequency", "enrichment"),
  weighted = FALSE,
  symptom_labels = NULL,
  bar_colour = "#4292c6",
  baseline_colour = "#252525"
)

Arguments

selection

A symptom_selection result.

type

Character. "frequency" (default) draws the selection frequency of each item with its own chance baseline marked; "enrichment" draws the ratio of the two, with a reference line at 1.

weighted

Logical. If TRUE, use the stability weighted columns (pi_freq, pi_enrichment), which requires symptom_selection to have been given a bootstrap_stability result. Default FALSE.

symptom_labels

Optional character vector of length 20 for the axis. Defaults to the label column of selection, or item numbers.

bar_colour

Fill colour used when the items have no cluster labels.

baseline_colour

Colour of the chance-baseline markers.

Value

A ggplot object.

See also

symptom_selection, plot_symptom_frequency.

Examples

# \donttest{
ptsd_data <- rename_ptsd_columns(simulated_ptsd[1:120, ],
                                 id_col = c("patient_id", "age", "sex"))
fit <- score_all_combinations(ptsd_data, n_symptoms = 4, n_required = 3,
                              show_progress = FALSE)
selection <- symptom_selection(fit)
plot_symptom_selection(selection)

# }