`R/describe.R`

`describe_all.Rd`

This function extends `{describe}`

by applying to it all
columns of the specified class(es) in a data frame using functional
programming tools from the purrr package (e.g. `map`

).
To obtain a summary of a single variable in a data frame use
`describe`

instead.

```
describe_all(
data,
...,
class = "all",
digits = 3,
type = 2,
na.rm = TRUE,
sep = "_",
output = c("dt", "tibble")
)
```

- data
A data frame or tibble.

- ...
This special argument accepts any number of unquoted grouping variable names (also present in the data source) to use for subsetting, separated by commas, e.g.

`group_var1, group_var2`

. Also accepts a character vector of column names or index numbers, e.g. c("group_var1", "group_var2") or c(1, 2), but not a mixture of formats in the same call. If no column names are specified, all columns will be used.- class
The variable classes in data that you would like summaries for. Either "all" for all classes, or a character vector indicating which combinations of output classes you want. Specifying a subset will save time since summaries are only processed as needed. Options include "d" for dates, "f" for factors, "c" for character, "l" for logical, and "n" for numeric. If only a single class is requested or present in the data after excluding specified grouping variables, a data frame will be returned, otherwise you'll get a list of data frames (1 per summary class). If the only chosen class of variables is not detected in the input data an error will be returned that the class argument needs to be respecified.

- digits
This determines the number of digits used for rounding of numeric outputs.

- type
For numeric and integer vectors this determines the type of skewness and kurtosis calculations to perform. See

`skewness`

or`skew`

and`kurtosis`

or`kurtosi`

for details.- na.rm
This determines whether missing values (NAs) should be removed before attempting to calculate summary statistics.

- sep
A character string to use to separate unique values from their counts ("_" by default). Only applicable to factors and character vectors.

- output
Output type for each class of variables. dt" for data.table or "tibble" for tibble.

The output varies as a function of the class of input data/y, referred to as "y" below. Each output type is grouped together in a data frame and returned as a named item of a list, unless there is only one output type, in which case the data frame is returned directly.

**For all input variables, the following are returned (part 1):**

- cases
the total number of cases

- n
number of complete cases

- na
the number of missing values

- p_na
the proportion of total cases with missing values

In addition to part 1, these measures are provided for **dates**:

- n_unique
the total number of unique values or levels of y. For dates this tells you how many time points there are

- start
the earliest or minimum date in y

- end
the latest or maximum date in y

In addition to part 1, these measures are provided for **factors**:

- n_unique
the total number of unique values or levels of y

- ordered
a logical indicating whether or not y is ordinal

- counts_tb
the counts of the top and bottom unique values of y in order of decreasing frequency formatted as "value_count". If there are more than 4 unique values of y, only the top 2 and bottom 2 unique values are shown separated by "...". To get counts for all unique values use

`counts`

or`counts_tb`

instead.

In addition to part 1, these measures are provided for **character/string** vectors:

- n_unique
the total number of unique values or levels of y

- min_chars
the minimum number of characters in the values of y

- max_chars
the maximum number of characters in the values of y

- counts_tb
the counts of the top and bottom unique values of y in order of decreasing frequency formatted as "value_count". If there are more than 4 unique values of y, only the top 2 and bottom 2 unique values are shown separated by "...". To get counts for all unique values use

`counts`

or`counts_tb`

instead.

In addition to part 1, these measures are provided for **logical** vectors:

- n_TRUE
the total number of y values that are TRUE

- n_FALSE
the total number of y values that are FALSE

- p_TRUE
the proportion of y values that are TRUE

In addition to part 1, these measures are provided for **numeric** variables:

- mean
the mean of y

- sd
the standard deviation of y

- se
the standard error of the mean of y

- p0
the 0th percentile (the minimum) of y

- p25
the 25th percentile of y

- p50
the 50th percentile (the median) of y

- p75
the 25th percentile of y

- p100
the 100th percentile (the maximum) of y

- skew
the skewness of the distribution of y

- kurt
the kurtosis of the distribution of y

```
describe_all(mtcars)
#> variable cases n na p_na mean sd se p0 p25 p50
#> 1: mpg 32 32 0 0 20.091 6.027 1.065 10.400 15.425 19.200
#> 2: cyl 32 32 0 0 6.188 1.786 0.316 4.000 4.000 6.000
#> 3: disp 32 32 0 0 230.722 123.939 21.909 71.100 120.825 196.300
#> 4: hp 32 32 0 0 146.688 68.563 12.120 52.000 96.500 123.000
#> 5: drat 32 32 0 0 3.597 0.535 0.095 2.760 3.080 3.695
#> 6: wt 32 32 0 0 3.217 0.978 0.173 1.513 2.581 3.325
#> 7: qsec 32 32 0 0 17.849 1.787 0.316 14.500 16.892 17.710
#> 8: vs 32 32 0 0 0.438 0.504 0.089 0.000 0.000 0.000
#> 9: am 32 32 0 0 0.406 0.499 0.088 0.000 0.000 0.000
#> 10: gear 32 32 0 0 3.688 0.738 0.130 3.000 3.000 4.000
#> 11: carb 32 32 0 0 2.812 1.615 0.286 1.000 2.000 2.000
#> p75 p100 skew kurt
#> 1: 22.80 33.900 0.672 -0.022
#> 2: 8.00 8.000 -0.192 -1.763
#> 3: 326.00 472.000 0.420 -1.068
#> 4: 180.00 335.000 0.799 0.275
#> 5: 3.92 4.930 0.293 -0.450
#> 6: 3.61 5.424 0.466 0.417
#> 7: 18.90 22.900 0.406 0.865
#> 8: 1.00 1.000 0.265 -2.063
#> 9: 1.00 1.000 0.401 -1.967
#> 10: 4.00 5.000 0.582 -0.895
#> 11: 4.00 8.000 1.157 2.020
if (FALSE) {
describe_all(pdata) #all summary types in a list
#numeric summary only
describe_all(pdata, high_low, output = "dt", class = "n")
#numeric and logical summaries only
describe_all(pdata, high_low, output = "dt", class = c("n", "l"))
}
```