peek_neat {neatStats}

Cursory Summaries and Plots per Group

Description

Cursory summaries and plots per group.

Usage

peek_neat(
  dat,
  values,
  group_by = NULL,
  filt = NULL,
  sep = ", ",
  collapse = NULL,
  f_print = FALSE,
  f_plot = NULL,
  iqr_times = 3,
  round_to = 4,
  group_n = TRUE,
  ...
)

Arguments

dat

Data frame (or name of data frame as string).

values

String, or vector of strings: the name(s) of the column(s) in the dat data frame, containing the vector(s) of values.

group_by

String, or vector of strings: the name(s) of the column(s) in the dat data frame, containing the vector(s) of factors by which the statistics are grouped.

filt

An expression to filter, by column values, the entire dat data frame before performing the aggregation. The expression should use column names alone; see Examples.

sep

String (comma by default) for separating group names.

collapse

Decides how to handle multiple columns of values. If NULL (default), displays each column of values as separate groups. Alternatively, any function can be given using which the columns are collapsed into a single column. For example, if mean is given for this parameter, a single column will be calculated based on the means of all given values columns. (NA is always ignored.)

f_print

Printing function; see details.

f_plot

Plotting function; see details. (Provide string to skip plotting.)

iqr_times

The multiplication of IQR to calculate Tukey's fences, when using default f_print function (set to TRUE); see Details. The default is 3 to spot Tukey's "far outliers" (e.g. by comparing the fences with min and max values). (Note that the usual fences, e.g. for box plots, use 1.5).

round_to

Number of significant fractional digits to round to, when using default f_print function (set to TRUE).

group_n

Logical. If TRUE, adds sample sizes (n) per group to plots when using default f_plot (NULL).

...

Any arguments to be passed to the f_plot function.

Details

If set to TRUE, prints to console the following data (per group): mean; 95 median; quantile_1st and quantile_3rd (first and third quantiles); "Tukey's fences" as fence_low and fence_upp; minimum and maximum values (min, max); number of NAs (na). Tukey's fences are the upper and lower limits with distances of X times the IQR from the actual IQR, where X is specified via the iqr_times parameter. Returns (invisibly) the same values, unrounded, via a data frame. If alternative f_print is given, prints whatever value is returned from the given function (and attempts, if possible, to create a data frame).

Creates and displays box plot(s) (per group) by default, along with overlayed violin plot (densities proportionate to sample sizes). If alternative f_plot is given, the first argument will be the values per group, and all plots will be arranged into a single plot and displayed together. To skip plotting, just give any character as argument (e.g. "none" or just "").

Value

Data frame with the printed values (if possible).

Examples

data("mtcars") # load base R example dataset

# overall info for wt (Weight)
peek_neat(mtcars, 'wt', f_print = TRUE)
#'
# now groupped by cyl (Number of cylinders)
peek_neat(mtcars, 'wt', group_by = 'cyl')

# grouped by cyl and gear
peek_neat(mtcars,
          'wt',
          group_by = c('cyl', 'gear'),
          f_print = TRUE)

# filter to only have cyl larger than  4
peek_neat(mtcars, 'wt', group_by = 'cyl', filt = cyl > 4)

# without plots
peek_neat(mtcars,
          'wt',
          group_by = 'cyl',
          f_plot = "",
          f_print = TRUE)

# with histogramms etc, using plot_neat
peek_neat(mtcars, 'wt', group_by = 'cyl', f_plot = plot_neat)

# with Q-Q plots, via ggpubr
peek_neat(mtcars,
          'wt',
          group_by = 'cyl',
          f_plot = ggpubr::ggqqplot)

# skewness and kurtosis data via psych
## Not run: 
info_df = peek_neat(
    mtcars,
    'wt',
    group_by = 'cyl',
    f_print = psych::describe,
    f_plot = ""
)
info_df # contains all data returns by psych::describe

## End(Not run)