Add a column to janitor::tabyl output where you count certain combinations by row in R

user_14661461 0
Viewed 32

Using janitor::tabyl() I get a desired contingency table output. This is done to check whether certain combinations are present, that should NOT be present. Therefore, I would like to create an extra column called "number_of_errors" that counts the combinations (by row), that I input.

If I run a tabyl on the iris dataset, for instance:

   gapminder %>% 
      tabyl(year, continent)

output:

 year Africa Americas Asia Europe Oceania
 1952     52       25   33     30       2
 1957     52       25   33     30       2
 1962     52       25   33     30       2
 1967     52       25   33     30       2
 1972     52       25   33     30       2
 1977     52       25   33     30       2
 1982     52       25   33     30       2
 1987     52       25   33     30       2
 1992     52       25   33     30       2
 1997     52       25   33     30       2
 2002     52       25   33     30       2
 2007     52       25   33     30       2

And now I would like to create an extra column, that counted certain combinations, that I give it. For instance, let's say that the combination 1952 and Europe can not exists, nor can 1962 have the value Oceania NOR Asia. Therefore, I want to create a column called "number_of_errors" with this desired output:

year Africa Americas Asia Europe Oceania number_of_erros
 1952     52       25   33     30       2 30
 1957     52       25   33     30       2 0
 1962     52       25   33     30       2 35
 1967     52       25   33     30       2 0
 1972     52       25   33     30       2 0
 1977     52       25   33     30       2 0
 1982     52       25   33     30       2 0
 1987     52       25   33     30       2 0
 1992     52       25   33     30       2 0
 1997     52       25   33     30       2 0
 2002     52       25   33     30       2 0
 2007     52       25   33     30       2 0

I have tried with a mutate and case_when, but to no avail. I've got this far:

out <- gapminder %>% 
          tabyl(year, continent)

out %>% 
  mutate(number_of_errors = case_when(subset(out, year == "1952") == "Europe") ~ count(subset(out, year == "1952") == "Europe")),
subset(out, year == "1962) == "Asia" | subset(out, year == "1962") == "Oceania") ~ sum(count(subset(out, year == "1962) == "Asia", count(subset(out, year == "1962) == "Oceania"))

Not sure if this made sense. There must be something more elegant and easy that works

Thanks

janitor dplyr r
1 Answers

Add further conditions as needed:

library(dplyr)

out <- data.frame(
  year = c(1952,1957,1962,1967,1972,1977,1982,1987,1992,1997,2002,2007),
  Africa = c(52,52,52,52,52,52,52,52,52,52,52,52),
  Americas = c(25,25,25,25,25,25,25,25,25,25,25,25),
  Asia = c(33,33,33,33,33,33,33,33,33,33,33,33),
  Europe = c(30,30,30,30,30,30,30,30,30,30,30,30),
  Oceania = c(2,2,2,2,2,2,2,2,2,2,2,2)
) %>%
  dplyr::mutate(number_of_errors = dplyr::case_when(
    year == 1952 ~ Europe,
    year == 1962 ~ Asia + Oceania,
    TRUE ~ 0
  ))

out
   year Africa Americas Asia Europe Oceania number_of_errors
1  1952     52       25   33     30       2               30
2  1957     52       25   33     30       2                0
3  1962     52       25   33     30       2               35
4  1967     52       25   33     30       2                0
5  1972     52       25   33     30       2                0
6  1977     52       25   33     30       2                0
7  1982     52       25   33     30       2                0
8  1987     52       25   33     30       2                0
9  1992     52       25   33     30       2                0
10 1997     52       25   33     30       2                0
11 2002     52       25   33     30       2                0
12 2007     52       25   33     30       2                0
user_100433230
Related