pivot_longer the ouput of summarize_each with mean and sd functions

Question

I am trying to summarize all columns in my dataset (which has many columns, although the rep example bellow only has 2), getting mean and sd for each variable. I want the output to be in long format.

#Example dataset
d <- iris %>% select(Sepal.Length,Sepal.Width)
names(d) <- c("SepalLength","SepalWidth")

#Summarizing and trying to make it long
s <- d %>% summarize_each( list(mean=mean,sd=sd) )  # has summar stats, but they are in wide format

# trying to pivot.
s %>% pivot_longer( ??? what do I put here ???)

I tried a few variations (ex: pivot_longer(names_to = "key", values_to = "value")) but always get an error message.

Answer 1

We can use select_helpers inside

library(dplyr)
library(tidyr)
s %>%
   pivot_longer(everything())
# A tibble: 4 x 2
#  name             value
#  <chr>            <dbl>
#1 SepalLength_mean 5.84 
#2 SepalWidth_mean  3.06 
#3 SepalLength_sd   0.828
#4 SepalWidth_sd    0.436

---

Or if we need 'SepalLength', 'SepalWidth' as two columns

s %>% 
   pivot_longer(cols = everything(), 
       names_to = c(".value", "statistic"), names_sep="_")
# A tibble: 2 x 3
#  statistic SepalLength SepalWidth
#  <chr>           <dbl>      <dbl>
#1 mean            5.84       3.06 
#2 sd              0.828      0.436

---

Or if we need the 'mean', 'sd' as two columns

s %>%
    pivot_longer(cols = everything(), 
       names_to = c("colNames", ".value"), names_sep="_")
# A tibble: 2 x 3
#  colNames     mean    sd
#  <chr>       <dbl> <dbl>
#1 SepalLength  5.84 0.828
#2 SepalWidth   3.06 0.436

---

Or using gather

s %>%
  gather