I would like to find a way to implement a series of piped functions through an lapply statement and generate multiple databases as a result. Here is a sample data set:
# the data
d <- tibble(
categorical = c("a", "d", "b", "c", "a", "b", "d", "c"),
var_1 = c(0, 0, 1, 1, 1, 0, 1, 0),
var_2 = c(0, 1, 0, 0, 0, 0 ,1, 1),
var_3 = c(0, 0, 1, 1, 1, 1, 1, 1),
var_4 = c(0, 1, 0, 1, 0, 0, 0, 0)
)
Here is the outcome I want:
$var_1
a b c d
1 1 1 1
$var_2
a b c d
0 0 1 2
$var_3
a b c d
1 2 2 1
$var_4
a b c d
0 0 1 1
I can recreate each list element individually with ease. Here is my sample code with dplyr:
d %>%
filter(var_1 == 1) %>%
group_by(categorical, var_1) %>%
summarise(n = n()) %>%
select(-var_1) %>%
rename("var_1" = "n") %>%
ungroup() %>%
spread(categorical, var_1)
# A tibble: 1 x 4
a b c d
<int> <int> <int> <int>
1 1 1 1 1
But, I want to automate the process across all columns and create an object that contains each row of information as a list.
Here is where I started:
lapply(d[,2:5], function (x) d %>%
filter(x == 1) %>%
group_by(categorical, x) %>%
summarise(n = n()) %>%
select(-x) %>%
rename("x" = "n") %>%
ungroup() %>%
spread(categorical, x))
Any help would be much appreciated!
We can gather into 'long' format, then do a group_split and spread it back after getting the sum of 'val' grouped by 'categorical'
library(tidyverse)
gather(d, key, val, -categorical) %>%
split(.$key) %>%
map(~ .x %>%
group_by(categorical) %>%
summarise(val = sum(val)) %>%
spread(categorical, val))
#$var_1
# A tibble: 1 x 4
# a b c d
# <dbl> <dbl> <dbl> <dbl>
#1 1 1 1 1
#$var_2
# A tibble: 1 x 4
# a b c d
# <dbl> <dbl> <dbl> <dbl>
#1 0 0 1 2
#$var_3
# A tibble: 1 x 4
# a b c d
# <dbl> <dbl> <dbl> <dbl>
#1 1 2 2 1
#$var_4
# A tibble: 1 x 4
# a b c d
# <dbl> <dbl> <dbl> <dbl>
#1 0 0 1 1
Or another option is to loop through the columns except the first one, and then do the group_by sum and spread to 'wide' format
map(names(d)[-1], ~
d %>%
group_by(categorical) %>%
summarise(n = sum(!! rlang::sym(.x))) %>%
spread(categorical, n))