[FIXED] How to make and organise a table of means and sds in R, according to different groupings

Issue

Here is an example of what my data looks like:

group  valueA1 valueA2 valueB1 valueB2
1       2       3       4       5
1       4       4       5       5
2       5       4       3       2

I want to create a table with summary statistics organised in a few different ways. Firstly, I need my results organised by group, then by whether it contains A or B, then by whether it contains 1 or 2 in the variable name.

I understand that I can use something like this;

Agroup <- mydata %>%
    select(contains("valueA"))%>%
    summarise_all(list(mean=mean, sd=sd), na.rm = TRUE)

This gives me the list of means and sd, however. How can I print these as a table?

Solution

mydata = data.frame(
  group = c(1, 1, 2),
  valueA1 = c(2, 4, 5),
  valueA2 = c(3, 4, 4),
  valueB1 = c(4, 5, 3),
  valueB2 = c(5, 5, 2)
)

Agroup <- mydata %>%
  select(contains("valueA")) %>%
  summarise_all(list(mean=mean, sd=sd), na.rm = TRUE)
Agroup

gives you:

  valueA1_mean valueA2_mean valueA1_sd valueA2_sd
1     3.666667     3.666667   1.527525  0.5773503

and

Agroup <- mydata %>%
  select(contains("1")) %>%
  summarise_all(list(mean=mean, sd=sd), na.rm = TRUE)
Agroup

gives you:

  valueA1_mean valueB1_mean valueA1_sd valueB1_sd
1     3.666667            4   1.527525          1

Same applies to contains("valueB") and "2".

Answered By - Marc

Answer Checked By - David Goodson (PHPFixing Volunteer)