CODEWITHSUNDEEP
r
DocProc
DocProc

Reputation: 103

Comparing the one value for a row to the mean of all rows for the same column

I have a large dataset that I would like to design the ability for individuals to select a single person for comparison and then to see how their data compares to the average of all of the other people. I'm trying to find the most efficient way possible and am hoping for some ideas.

I've tried creating two separate dataframes and then using lists to manually add the rows together so that I can graph them in ggplot. The extra manipulation, however also includes mutating the data.

library(tidyverse)

mtcars$car <- rownames(mtcars)

df <- mtcars %>% 
    filter(mtcars$car == 'Datsun 710') %>% 
    group_by(car) %>% 
    summarize(disp = first(disp)) 

df2 <- mtcars %>% 
  summarize(disp = mean(disp))

The expected output is as follows

group   disp
Datsun 710  108
All cars    230.7219

Upvotes: 1

Views: 277

Answers (1)

akrun
akrun

Reputation: 886938

With the output we got, it can be changed to

library(dplyr)
mtcars %>%
      summarise(car = "All cars", disp = mean(disp)) %>% 
      bind_rows(df, .)
# A tibble: 2 x 2
#  car         disp
#   <chr>      <dbl>
#1 Datsun 710  108 
#2 All cars    231.

It can also be done with add_row. Create the column of row names with rownames_to_column, select the columns of interest, add the row with add_row and filter the rows based on the values of 'car'

library(tibble)
mtcars %>%
   rownames_to_column('car') %>% 
   select(car, disp) %>% 
   add_row(car = "All cars", disp = mean(.$disp)) %>%
   filter(car %in% c('Datsun 710', 'All cars'))
#         car     disp
#1 Datsun 710 108.0000
#2   All cars 230.7219

Upvotes: 1

Related Questions