Trouble with nested ifelse statement in R

Question

I'm trying to run a nested ifelse statement in R. Here's a look at the structure of my data using the glimpse() function from the tidyverse:

Rows: 22,104
Columns: 9
$ `Formation/Locality`    <chr> "Montmartre", "Montmartre", "Montmartre", "Fur", "Me...
$ Location                <chr> "Ile-de-France Region, France", "Ile-de-France Regio...
$ Environment             <chr> "terrestrial", "terrestrial", "terrestrial", "offsho...
$ `Palaeolongitude(N/-S)` <dbl> 47.4, 47.4, 47.4, 52.3, 46.9, 42.9, 47.5, 46.9, 46.2...
$ `Palaeolatitude(E/-W)`  <dbl> 1.6, 1.6, 1.6, 5.4, 4.8, 1.9, -5.2, 4.8, -93.6, -111...
$ TaxonomicLevel          <chr> "Order", "Order", "Order", "Order", "Order", "Order"...
$ TaxonomicName           <chr> "Upupiformes", "Upupiformes", "Upupiformes", "Trogon...
$ MinMax                  <chr> "MaxMa", "MaxMa", "MaxMa", "MaxMa", "MaxMa", "MaxMa"...
$ Age                     <dbl> 37.2, 37.2, 37.2, 55.8, 48.6, 37.2, 48.6, 48.6, 55.8...

I'm trying to get R to look at the Age column and if the value is within a certain range, for it to put the geological age name into a new column called AgeName. If the value is not within the range, I want it to move on to the next age range and so on and so forth. Here's my code so far:

pbdb_tidyish$AgeName <- ifelse(56>=pbdb_tidyish$Age&&47.8<pbdb_tidyish$Age,
                               "Ypresian",
                               ifelse(47.8>=pbdb_tidyish$Age&&41.2<pbdb_tidyish$Age,
                                      "Lutetian",
                                      ifelse(41.2>=pbdb_tidyish$Age&&37.8<pbdb_tidyish$Age,
                                             "Bartonian",
                                             ifelse(37.8>=pbdb_tidyish$Age&&33.9<=pbdb_tidyish$Age,
                                                    "Priabonian",NA))))

When I run this code, it creates the new column but fills the whole column with "Priabonian" so the dataset now looks like this:

Rows: 22,104
Columns: 10
$ `Formation/Locality`    <chr> "Montmartre", "Montmartre", "Montmartre", "Fur", "Me...
$ Location                <chr> "Ile-de-France Region, France", "Ile-de-France Regio...
$ Environment             <chr> "terrestrial", "terrestrial", "terrestrial", "offsho...
$ `Palaeolongitude(N/-S)` <dbl> 47.4, 47.4, 47.4, 52.3, 46.9, 42.9, 47.5, 46.9, 46.2...
$ `Palaeolatitude(E/-W)`  <dbl> 1.6, 1.6, 1.6, 5.4, 4.8, 1.9, -5.2, 4.8, -93.6, -111...
$ TaxonomicLevel          <chr> "Order", "Order", "Order", "Order", "Order", "Order"...
$ TaxonomicName           <chr> "Upupiformes", "Upupiformes", "Upupiformes", "Trogon...
$ MinMax                  <chr> "MaxMa", "MaxMa", "MaxMa", "MaxMa", "MaxMa", "MaxMa"...
$ Age                     <dbl> 37.2, 37.2, 37.2, 55.8, 48.6, 37.2, 48.6, 48.6, 55.8...
$ AgeName                 <chr> "Priabonian", "Priabonian", "Priabonian", "Priabonia...

Does anyone have any idea where I'm going wrong? I think it's just looking at the first Age value, running the ifelse statement then filling the whole column with the result of that as opposed to moving on to the next row.

Thanks,

Carolina

Why `56>=pbdb_tidyish$Age` and not `pbdb_tidyish$Age >= 56`? — Chris Ruehlemann, May 28 '20 at 14:28
I'd also strongly suggest using more spaces in your code - itwillbemucheasiertoreadanddebug. Many style guides recommend spaces between binary operators, e.g., `56>=pbdb_tidyish$Age&&47.8= pbdb_tidyish$Age && 47.8 < pbdb_tidyish$Age` — Gregor Thomas, May 28 '20 at 14:35

score 3 · Accepted Answer · answered May 28 '20 at 14:29

3

Without data it is unclear whether this is the only error but you should not use && here since it is not vectorized. That means, it only checks the value in the first row, returns either TRUE or FALSE, based on this single observation only, and recycles this value.

Use & instead.

For a comparison see this answer

answered May 28 '20 at 14:29

mnist

6,571
1
18
41

Yes that's all there was to it! Thank you! – Carolina Karoullas May 28 '20 at 14:38

score 1 · Answer 2 · answered May 28 '20 at 14:34

I think whenever you find yourself writing nested ifelse statements you should stop and ask yourself whether there might be a better way to achieve what you're trying to do. For example, the following single function call does what you are trying to achieve, and is easier to understand and maintain:

cut(pdb$tidyish, breaks = c(33.9, 37.8, 41.2, 47.8, 56),
    labels = c("Priabonian", "Bartonian", "Lutetian", "Ypresian"))

score 1 · Answer 3 · answered May 28 '20 at 14:38

You're already using the tidyverse, you should make yourself familiar with case_when:

pbdb_tidyish <- pbdb_tidyish %>%
  mutate(AgeName = case_when(
    (Age >= 33.9 & Age <= 37.8) ~ 'Priabonian',
    (Age > 37.8 & Age <= 41.2) ~ 'Bartonian',
    (Age > 41.2 & Age <= 47.8) ~ 'Lutetian',
    (Age > 47.8 & Age <= 56) ~ 'Ypresian',
  ))

Trouble with nested ifelse statement in R

3 Answers3