Counts the number of observations for a variable and calculates the corresponding percentages. Optionally converts the grouping variable to a factor using labels defined in the configuration object.
Usage
sum_up(dt, var, config, category = FALSE, digits = 1)
summarise(dt, var, config, category = FALSE, digits = 1)
summarize(dt, var, config, category = FALSE, digits = 1)Arguments
- dt
A
data.tableordata.frame. If adata.frameis supplied, it is converted internally to adata.table.- var
Variable to sum_up. Unquoted column name.
- config
Configuration list containing
value_map.- category
Logical. If
TRUE, convert the grouping variable to a factor using labels fromconfig$value_map.- digits
Number of decimal places used when rounding percentages.
Value
A data.table containing:
- var
The grouping variable.
- n
Number of observations.
- pct
Percentage of observations.
Details
The function expects a configuration object containing a
value_map element, where each variable name corresponds to a lookup
table with columns num_value and chr_value.
If category = TRUE and the variable is of class
haven_labelled, labels are extracted using
haven::as_factor(levels = "labels").
Otherwise, labels are taken from
config$value_map[[var_name]], using:
num_valueas factor levelschr_valueas factor labels
Examples
library(data.table)
#>
#> Attaching package: ‘data.table’
#> The following object is masked from ‘package:base’:
#>
#> %notin%
dt <- data.table(
kjonn = c(1, 2, 1, 2, 1, 1, 2, 2, 1, 2)
)
cfg <- list(
value_map = list(
kjonn = data.table(
num_value = c(1, 2),
chr_value = c("Mann", "Kvinne")
)
)
)
sum_up(dt, kjonn, cfg)
#> kjonn n pct
#> <num> <int> <num>
#> 1: 1 5 50
#> 2: 2 5 50
sum_up(dt, kjonn, cfg, category = TRUE)
#> kjonn n pct
#> <fctr> <int> <num>
#> 1: Mann 5 50
#> 2: Kvinne 5 50
