Raw numeric summaries (n, missingness, mean, SD, median, quartiles, IQR, range) of numeric variables, overall or by one or more grouping columns, as a long tibble for programmatic filtering. No formatting or rounding: use [summary_table()] or [tbl_table_one()] for presentation.
Usage
calc_grouped_summary(
data,
vars = NULL,
by = NULL,
stats = c("n", "n_missing", "pct_missing", "mean", "sd", "median", "q1", "q3", "iqr",
"min", "max")
)Arguments
- data
A data frame.
- vars
Numeric columns to summarise; default all numeric columns that are not in `by`.
- by
Grouping columns (character vector) or `NULL` for the whole data.
- stats
Statistics to return, any of `"n"`, `"n_missing"`, `"pct_missing"`, `"mean"`, `"sd"`, `"median"`, `"q1"`, `"q3"`, `"iqr"`, `"min"`, `"max"` (default all). `n` counts non-missing values; `pct_missing` is on a 0 to 100 scale.
Value
A tibble with the `by` columns, `variable` and the requested statistics, one row per group and variable (a missing `by` value forms its own group). Statistics that cannot be computed (an all-missing group, the SD of one value) are `NA`.
Examples
calc_grouped_summary(mtcars, vars = c("mpg", "hp"), by = "cyl")
#> # A tibble: 6 × 13
#> cyl variable n n_missing pct_missing mean sd median q1 q3
#> <dbl> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 4 mpg 11 0 0 26.7 4.51 26 22.8 30.4
#> 2 6 mpg 7 0 0 19.7 1.45 19.7 18.6 21
#> 3 8 mpg 14 0 0 15.1 2.56 15.2 14.4 16.2
#> 4 4 hp 11 0 0 82.6 20.9 91 65.5 96
#> 5 6 hp 7 0 0 122. 24.3 110 110 123
#> 6 8 hp 14 0 0 209. 51.0 192. 176. 241.
#> # ℹ 3 more variables: iqr <dbl>, min <dbl>, max <dbl>
calc_grouped_summary(airquality, stats = c("n", "pct_missing", "median"))
#> # A tibble: 6 × 4
#> variable n pct_missing median
#> <chr> <int> <dbl> <dbl>
#> 1 Ozone 116 24.2 31.5
#> 2 Solar.R 146 4.58 205
#> 3 Wind 153 0 9.7
#> 4 Temp 153 0 79
#> 5 Month 153 0 7
#> 6 Day 153 0 16