Inspect categorical variables — InspectCategoricalSummary • SciDataReportR

Provides a SciDataReportR-native categorical inspection summary and plot inspired by inspectdf::inspect_cat(), created because inspectdf is archived and no longer available on CRAN.

Usage

InspectCategoricalSummary(
  Data,
  Variables = NULL,
  Codebook = NULL,
  IncludeMissing = TRUE,
  MissingLabel = "(Missing)",
  RetainLabels = TRUE,
  SortLevelsBy = c("Frequency", "Value", "None"),
  SortVariablesBy = c("Input", "Label", "MissingPercent", "UniqueLevels", "TotalN"),
  Descending = TRUE,
  MaxLevels = 30,
  Plot = TRUE,
  PlotType = c("bar", "lollipop"),
  FacetScales = c("free_y", "fixed"),
  UsePercent = TRUE,
  LabelBars = TRUE,
  WrapLabels = 35,
  BaseSize = 11
)

Arguments

Data: A data frame.
Variables: Optional character vector of categorical variables to summarize. If NULL, character, factor, logical, labelled, and haven-labelled columns are detected automatically.
Codebook: Optional data frame with Variable and Label columns.
IncludeMissing: Logical. If TRUE, missing values are included as a level and in percentages.
MissingLabel: Character label to display for missing values.
RetainLabels: Logical. If TRUE, use variable labels from Codebook and value labels from labelled variables when available.
SortLevelsBy: One of "Frequency", "Value", or "None".
SortVariablesBy: One of "Input", "Label", "MissingPercent", "UniqueLevels", or "TotalN".
Descending: Logical. If TRUE, sort selected summaries in descending order.
MaxLevels: Positive integer giving the maximum number of levels to display per variable in the plot before collapsing lower-ranked levels to "(Other)".
Plot: Logical. If TRUE, return a ggplot object.
PlotType: One of "bar" or "lollipop".
FacetScales: One of "free_y" or "fixed".
UsePercent: Logical. If TRUE, plot percentages; otherwise plot counts.
LabelBars: Logical. If TRUE, add readable labels to bars or points.
WrapLabels: Integer number of characters used to wrap displayed level and facet labels.
BaseSize: Base font size passed to theme_minimal().

Value

A named list with Summary, a tibble containing categorical counts and percentages, and Plot, a ggplot object when Plot = TRUE or NULL when Plot = FALSE.

References

Rushworth A. inspectdf: Inspection, comparison and visualisation of data frames. Formerly available on CRAN; archived package.

Examples

df <- data.frame(
  group = factor(c("A", "B", "A", NA), levels = c("A", "B", "C")),
  flag = c(TRUE, FALSE, TRUE, NA),
  text = c("low", "high", "low", "mid")
)

result <- InspectCategoricalSummary(df, Plot = FALSE)
result$Summary
#> # A tibble: 9 × 13
#>   Variable Label Level     LevelLabel     N Percent Missing TotalN NonMissingN
#>   <chr>    <chr> <chr>     <chr>      <int>   <dbl> <lgl>    <int>       <int>
#> 1 group    group A         A              2    0.5  FALSE        4           3
#> 2 group    group B         B              1    0.25 FALSE        4           3
#> 3 group    group (Missing) (Missing)      1    0.25 TRUE         4           3
#> 4 flag     flag  TRUE      TRUE           2    0.5  FALSE        4           3
#> 5 flag     flag  FALSE     FALSE          1    0.25 FALSE        4           3
#> 6 flag     flag  (Missing) (Missing)      1    0.25 TRUE         4           3
#> 7 text     text  low       low            2    0.5  FALSE        4           4
#> 8 text     text  high      high           1    0.25 FALSE        4           4
#> 9 text     text  mid       mid            1    0.25 FALSE        4           4
#> # ℹ 4 more variables: MissingN <int>, MissingPercent <dbl>, UniqueLevels <int>,
#> #   VariableClass <chr>

plot_result <- InspectCategoricalSummary(df, Variables = "group", Plot = TRUE)
plot_result$Plot