R. How to calculate level based on number in one dataframe?-CodePudding

This is probably a super easy and basic question but I can't find answer. I have a data frame with some nums. And I need to add another column to that database with strings that would represent this data. What I mean, for example:

A header	Another header
10	Bad
20	Good
15	Bad
35	Good

"Bad" and "Good" is intervals, for example, "Bad" is 0-20 and "Good" is 20 . How do I do that with R? I think I need to use apply func.

CodePudding user response：

library(tidyverse)

data <- tibble(num = c(10, 20, 15, 35))
data
#> # A tibble: 4 × 1
#>     num
#>   <dbl>
#> 1    10
#> 2    20
#> 3    15
#> 4    35

data %>%
  mutate(result = ifelse(num <= 20, "Bad", "Good"))
#> # A tibble: 4 × 2
#>     num result
#>   <dbl> <chr> 
#> 1    10 Bad   
#> 2    20 Bad   
#> 3    15 Bad   
#> 4    35 Good

^{Created on 2022-05-10 by the reprex package (v2.0.0)}

Or using base R:

data <- data.frame(num = c(10, 20, 15, 35))
data$result <- ifelse(data$num <= 20, "Bad", "Good")

If you have more than two levels (e.g. low, medium, and high), you can also use the function cut. There is also the function dplyr::case_when if the rules are more complex.

CodePudding user response：

You can use mutate and case_when:

df %>%
  mutate(status = case_when(v1 <= 20 ~ "Bad",
                              v1 > 20 ~ "Good"))

Output:

  v1 status
1 10    Bad
2 20    Bad
3 15    Bad
4 35   Good

Data

df <- data.frame(v1 = c(10,20,15,35))

CodePudding user response：

You can use cut.

x <- c(10, 20, 15, 35)
cut(x, c(0,20, Inf), c("Bad", "Good"), right=FALSE)
#[1] Bad  Good Bad  Good

What can easy be extended to more levels of gradation.

cut(x, c(0, 12, 25, Inf), c("Bad", "Normal", "Good"), right=FALSE)
#[1] Bad    Normal Normal Good

Or subset a vector.

c("Bad", "Good")[1 (x>=20)]
#[1] "Bad"  "Good" "Bad"  "Good"

Or using findInterval.

c("Bad", "Normal", "Good")[1   findInterval(x, c(12, 25))]
#[1] "Bad"    "Normal" "Normal" "Good"