Get selected data The function leverage `dplyr::filter` to select the data
Examples
data <- data.frame(
x = seq(0, 9),
y = seq(10, 19),
mdv = c(1, 1, rep(0, 8)),
groups = rep(c("A", "B"), 5)
)
# Select all the data
getSelectedData(data, DataSelectionKeys$ALL)
#> x y mdv groups
#> 1 0 10 1 A
#> 2 1 11 1 B
#> 3 2 12 0 A
#> 4 3 13 0 B
#> 5 4 14 0 A
#> 6 5 15 0 B
#> 7 6 16 0 A
#> 8 7 17 0 B
#> 9 8 18 0 A
#> 10 9 19 0 B
# Select no data
getSelectedData(data, DataSelectionKeys$NONE)
#> [1] x y mdv groups
#> <0 rows> (or 0-length row.names)
# Select data from group A
getSelectedData(data, "groups %in% 'A'")
#> x y mdv groups
#> 1 0 10 1 A
#> 2 2 12 0 A
#> 3 4 14 0 A
#> 4 6 16 0 A
#> 5 8 18 0 A
# Remove missing dependent variable (mdv)
getSelectedData(data, "mdv == 0")
#> x y mdv groups
#> 1 2 12 0 A
#> 2 3 13 0 B
#> 3 4 14 0 A
#> 4 5 15 0 B
#> 5 6 16 0 A
#> 6 7 17 0 B
#> 7 8 18 0 A
#> 8 9 19 0 B