Output of the dplyr summarize() fundtion

Question

Is there a convenient way to have dplyr::summarize_all() output the results in a more readable format without having to manually rearrange it after the fact?

Ultimately, I'd like to be able to port the output of summarize more easily to tables in Word, etc.

I would like to avoid the work of doing something like what's below.

Thank you

Example:

library(dplyr)
library(e1071) # for skewness() function
# make fake data
a <- rnorm(20)
b <- rnorm(20)
c <- rnorm(20)
x <- rnorm(20)
y <- rnorm(20)
z <- rnorm(20)

# create dataframe
dat = data.frame(a, b, c, x, y, z)

# run summarize()
descriptives <- dat %>% summarize_all(funs(mean, sd, skewness))
descriptives

# rearrange descriptives
matrix(descriptives, 6, 3, 
    dimnames = list(c("a", "b", "c", "x", "y", "z"), 
    c("mean", "SD", "skew")), byrow = FALSE)

# RETURNS
#  mean       SD        skew       
#a 0.1533271  0.8106499 -0.02879986
#b -0.5117311 0.5608904 -0.2668225 
#c 0.1267941  0.8214882 -0.4260682 
#x 0.05337055 0.9817041 -0.1932566 
#y -0.1091145 0.9050062 -0.3409686 
#z -0.3195788 0.8833493 -0.6663437

You should be able to automate your re-arranging process using the `tidyr` package. — AntoniosK, Nov 16 '17 at 17:50

score 2 · Accepted Answer · answered Nov 16 '17 at 17:52

library(tidyr)
library(dplyr)
library(e1071) # for skewness() function
# make fake data
a <- rnorm(20)
b <- rnorm(20)
c <- rnorm(20)
x <- rnorm(20)
y <- rnorm(20)
z <- rnorm(20)

# create dataframe
dat = data.frame(a, b, c, x, y, z)

# run process 
dat %>% 
  summarize_all(funs(mean, sd, skewness)) %>%
  gather() %>%
  separate(key, c("var","type")) %>%
  spread(type, value)

#   var          mean        sd   skewness
# 1   a  0.0182792019 0.9098886 -0.3851676
# 2   b  0.0003444183 0.9815170  0.6032848
# 3   c -0.2724927337 0.9557808 -0.1961959
# 4   x -0.2679435647 0.6557561 -1.0111428
# 5   y -0.1951287997 0.8190830  0.5120989
# 6   z -0.0395147539 1.2758244  0.0464844

Thank you for your help, I'd provide an up vote if I could, but I do not have the reputation to do so. — Peter Miksza, Nov 16 '17 at 19:06

score 1 · Answer 2 · answered Nov 16 '17 at 17:51

you need to rearrange your data in a tidy format, Read up https://cran.r-project.org/web/packages/tidyr/vignettes/tidy-data.html I did a quick melt, using the reshape2 package. You can do this using tidyr and the equivalent function would be gather(). Here's a handy reference -http://tidyr.tidyverse.org/ my brain is still too used to reshape2, slowly training my self to use tidyr

library(reshape2)
library(dplyr)
library(e1071)
descriptives <-melt(dat) %>%
group_by(variable) %>%
summarize_all(funs(mean, sd, skewness))

Thank you for your help. I'd accept both if I could and up vote both as well, but I do not have the "reputation" to do so. — Peter Miksza, Nov 16 '17 at 19:05

Output of the dplyr summarize() fundtion

2 Answers2