"binning" rows into ranges (dplyr/R)

1
votes

I am having a lot of difficulty trying to place rows from a dataset into "bins". For example, suppose I have a data frame "df" with "var1" and "var2" :

enter image description here

I want to create a new variable called "var3" that follows this logic (R code):

1) if var1 <5 and var2<5 .... then var3 = "a"
2) if var1 between (5,10) and var2 between (5,10) .... then var3 = "b"
3) if var1 > 10 and and var2>10 .... then var3 = "c"

From a previous question I posted (If statements with multiple ranges (R)), I tried the following logic:

library(dplyr)
df %>%
  mutate(var3 = case_when(var1 < 5 & var2 < 5 ~ 'a', 
                          var1 > 5 & var1 < 10 & var2 > 5 & var2 < 10 ~ 'b', 
                          var1 >10 & var2 >10 ~ 'c'))

But when I inspect the df$var3, the logic does not seem to be correct (i.e. some entries for var3 do not have any values. note: the smallest possible value of var1 and va2 is 0).

Can someone please help me?

Thanks

UPDATE:

Sample dataset:

a <- rnorm(50,10,10)
b <- rnorm(50, 2,8)

var1 = abs(a)
var2 = abs(b)

df = data.frame(var1, var2)
2
rif-statementdplyrrangedata-manipulation
sure, I will update the question. thank you - stats555
please see the updates. thank you - stats555
Well, your picture are inconsistent with your logic (e.g., picture says var1 > 10 or var2 > 10 then 'c' but your text doesn't). - cuttlefish44
@cuttlefish44: i suspected this was probably the problem. i want what is in the picture - can you please help me the write the correct logic to match the picture? thank you for your help - stats555
Use set.seed to create random numbers. Tell us which exact line is not generating your expected output in the data i.e what it is giving now vs what it should actually be. - Ronak Shah

2 Answers

3
votes

If you want to use case_when,

library(dplyr)

## data make
set.seed(111)
df = data.frame(var1 = abs(rnorm(50,10,10)), var2 = abs(rnorm(50,2,8)))

## core
df <- df %>%
  mutate(var3 = case_when(var1 < 5 & var2 < 5 ~ 'a', 
                          var1 < 10 & var2 < 10 ~ 'b',
                          TRUE ~ 'c'))
## plot to check
with(df, plot(var1, var2, col = c(2:4)[as.numeric(as.factor(var3))], cex = 0.7))
abline(h = c(5, 10), v = c(5, 10), lty = 2)

enter image description here

2
votes

try this

library(dplyr)

set.seed(123)
df <- data.frame(var1 = round(runif(100)*20, 0),
                   var2 = round(runif(100)*20, 0))

df <- df %>% mutate(var3 = ifelse(var1 <= 5 & var2 <= 5, "a", ifelse(var1 <= 10 & var2 <= 10, "b", "c")))

to check

library(ggplot2)

df %>%
  ggplot() + geom_point(aes(x=var1, y= var2, color= var3))

enter image description here