bar_plot: Get a bar plot for flagging percentage overall and stratified by provider sizes

Description

Generate a bar plot for flagging percentage.

Usage

bar_plot(
  flag_df,
  group_num = 4,
  bar_colors = c("#66c2a5", "#fc8d62", "#8da0cb"),
  bar_width = 0.7,
  label_color = "black",
  label_size = 4
)

Value

A ggplot object representing the bar chart of flagging results.

Arguments

flag_df: a data frame from test function containing the flag of each provider.
group_num: number of groups into which providers are divided based on their sample sizes. The default is 4.
bar_colors: a vector of colors used to fill the bars representing the categories. The default is c("#66c2a5", "#fc8d62", "#8da0cb").
bar_width: width of the bars in the bar chart. The default is 0.7.
label_color: color of the text labels inside the bars. The default is "black".
label_size: size of the text labels inside the bars. The default is 4.

Details

This function generates a bar chart to visualize the percentage of flagging results based on provider sizes. The input data frame test_df must be the output from package pprof's test function. Providers are grouped into a specified number of groups (group_num) based on their sample sizes, where the number of providers are approximately equal across groups. An additional "overall" group is included to show the flagging results across all providers.

Examples

Run this code

data(ExampleDataLinear)
outcome <- ExampleDataLinear$Y
covar <- ExampleDataLinear$Z
ProvID <- ExampleDataLinear$ProvID
fit_linear <- linear_fe(Y = outcome, Z = covar, ProvID = ProvID)
test_linear <- test(fit_linear)
bar_plot(test_linear)

data(ExampleDataBinary)
fit_logis <- logis_fe(Y = ExampleDataBinary$Y,
                      Z = ExampleDataBinary$Z,
                      ProvID = ExampleDataBinary$ProvID, message = FALSE)
test_logis <- test(fit_logis)
bar_plot(test_logis)