Using a for-loop to change a column in a dataframe

Question

I am a beginning user of R and i have written a code that I believe could be shortened with a for loop. The problem is I cannot figure out how to write the loop.

I have a dataframe with the column 'TestGrade' with values like 'Grade 1' or 'Kindergarten'. I am trying to change that column to be only a numeric value. For example 'Kindergarten' would be changed to 0 and 'Grade 1' would be changed to 1. I will provide code below of a sample dataframe and also how I solved the problem without a loop.

Any guidance will be greatly appreciated!

##Sample Data
FirstInitial <- c("A", "D", "M", "C", "J", "S", "K", "L", "M", "K", "G", "B", "F")
LastInitial <- c("S", "M", "T", "M", "A", "B", "H", "M", "S", "W", "L", "Z", "P")
TestGrade <- c('Kindergarten', 'Grade 1','Grade 2', 'Grade 3','Grade 4', 'Grade 5', 'Grade 6','Grade 7','Grade 8', 'Grade 9', 'Grade 10', 'Grade 11','Grade 12')

df <- data.frame(FirstInitial, LastInitial, TestGrade)

##The codes current function
if(any(df$TestGrade == 'Kindergarten')){
  df$TestGrade <- gsub('Kindergarten', '0', df$TestGrade)
}
if(any(df$TestGrade == 'Grade 1')){
  df$TestGrade <- gsub('Grade 1', '1', df$TestGrade)
}
if(any(df$TestGrade == 'Grade 2')){
  df$TestGrade <- gsub('Grade 2', '2', df$TestGrade)
}
if(any(df$TestGrade == 'Grade 3')){
  df$TestGrade <- gsub('Grade 3', '3', df$TestGrade)
}
if(any(df$TestGrade == 'Grade 4')){
  df$TestGrade <- gsub('Grade 4', '4', df$TestGrade)
}
if(any(df$TestGrade == 'Grade 5')){
  df$TestGrade <- gsub('Grade 5', '5', df$TestGrade)
}

if(any(df$TestGrade == 'Grade 6')){
  df$TestGrade <- gsub('Grade 6', '6', df$TestGrade)
}
if(any(df$TestGrade == 'Grade 7')){
  df$TestGrade <- gsub('Grade 7', '7', df$TestGrade)
}
if(any(df$TestGrade == 'Grade 8')){
  df$TestGrade <- gsub('Grade 8', '8', df$TestGrade)
}
if(any(df$TestGrade == 'Grade 9')){
  df$TestGrade <- gsub('Grade 9', '9', df$TestGrade)
}
if(any(df$TestGrade == 'Grade 10')){
  df$TestGrade <- gsub('Grade 10', '10', df$TestGrade)
}
if(any(df$TestGrade == 'Grade 11')){
  df$TestGrade <- gsub('Grade 11', '11', df$TestGrade)
}
if(any(df$TestGrade == 'Grade 12')){
  df$TestGrade <- gsub('Grade 12', '12', df$TestGrade)
}

The tangential advice I repeatedly offer up is to use explicit loops as a LAST choice in R. The apply functions and related packages are designed for coding efficiency. — SteveM
– SteveM, Commented Mar 29, 2019 at 15:42

Ronak Shah · Accepted Answer · 2019-03-29 15:19:37Z

7

We can use ifelse, assign 0 for "Kindergarten" and remove "Grade" from others

as.numeric(ifelse(df$TestGrade == "Kindergarten", 0, 
          sub("Grade ", "", df$TestGrade)))

#[1]  0  1  2  3  4  5  6  7  8  9 10 11 12

answered Mar 29, 2019 at 15:19

Ronak Shah

391k20 gold badges173 silver badges237 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

divibisan · Accepted Answer · 2019-03-29 15:30:12Z

5

We can use case_when

library(dplyr)
library(readr)
df %>%
  mutate(TestGrade = case_when(as.character(TestGrade) == "Kindergarten"~ 0,
                               TRUE ~ parse_number(TestGrade)))

#   FirstInitial LastInitial TestGrade
#1             A           S         0
#2             D           M         1
#3             M           T         2
#4             C           M         3
#5             J           A         4
#6             S           B         5
#7             K           H         6
#8             L           M         7
#9             M           S         8
#10            K           W         9
#11            G           L        10
#12            B           Z        11
#13            F           P        12

edited Mar 29, 2019 at 15:30

divibisan

12.2k11 gold badges44 silver badges63 bronze badges

answered Mar 29, 2019 at 15:25

akrun

891k38 gold badges590 silver badges700 bronze badges

Comments

Gregor Thomas · Accepted Answer · 2019-03-29 15:34:28Z

First shortening: you don't need any if(any(...)). gsub is smart, it's like a find/replace. The command gsub('Grade 9', '9', df$TestGrade) will replace 'Grade 9' with '9', and won't touch anything else. So deleting all your if statements, we get:

df$TestGrade <- gsub('Kindergarten', '0', df$TestGrade)
df$TestGrade <- gsub('Grade 1', '1', df$TestGrade)
df$TestGrade <- gsub('Grade 2', '2', df$TestGrade)
df$TestGrade <- gsub('Grade 3', '3', df$TestGrade)
df$TestGrade <- gsub('Grade 4', '4', df$TestGrade)
df$TestGrade <- gsub('Grade 5', '5', df$TestGrade)
df$TestGrade <- gsub('Grade 6', '6', df$TestGrade)
df$TestGrade <- gsub('Grade 7', '7', df$TestGrade)
df$TestGrade <- gsub('Grade 8', '8', df$TestGrade)
df$TestGrade <- gsub('Grade 9', '9', df$TestGrade)
df$TestGrade <- gsub('Grade 10', '10', df$TestGrade)
df$TestGrade <- gsub('Grade 11', '11', df$TestGrade)
df$TestGrade <- gsub('Grade 12', '12', df$TestGrade)

Next improvement, we could do a loop. This is exactly equivalent to the code above, just less typing.

pattern = c("Kindergarten", paste("Grade", 1:12))
replacement = as.character(0:12)

for (i in seq_along(pattern)) {
  df$TestGrade <- gsub(pattern[i], replacement[i], df$TestGrade)
}

Even better, we could be cleverer, make kindergarten a special case and just delete "Grade " from everything else, as in Juian's and Ronak's answers. Another variation of that is this:

df$TestGrade = as.character(df$TestGrade) # needed only if it is a factor
df$TestGrade[df$TestGrade == "Kindergarten"] = 0
df$TestGrade = sub("Grade ", "", df$TestGrade)
df$TestGrade = as.numeric(df$TestGrade) # if needed

If we really want to be fancy, we could set fixed = TRUE inside sub(). This tells sub we want exact matches only, we're not trying to use regular expressions. This will make the code run faster, but unless you've got a lot of data, you won't notice a difference. If you have 100,000+ rows, this method will be quite fast:

# optimized
df$TestGrade = as.character(df$TestGrade) # needed only if it is a factor
df$TestGrade[df$TestGrade == "Kindergarten"] = 0
df$TestGrade = as.integer(sub("Grade ", "", df$TestGrade, fixed = TRUE))

nsinghphd · Accepted Answer · 2019-03-29 15:22:38Z

This can be done without the need for a for loop using two line code. I also suggest you add stringsAsFactors = F in your data.frame command before running these lines

df$TestGrade[df$TestGrade == "Kindergarten"] = 0
df$TestGrade <- gsub("Grade ", "", df$TestGrade)

> df
   FirstInitial LastInitial TestGrade
1             A           S         0
2             D           M         1
3             M           T         2
4             C           M         3
5             J           A         4
6             S           B         5
7             K           H         6
8             L           M         7
9             M           S         8
10            K           W         9
11            G           L        10
12            B           Z        11
13            F           P        12

Thomas · Accepted Answer · 2019-03-29 15:23:03Z

3

You can write a key and set the grades as a factor. This will work even if the format of the grades changes.

key <- c('Kindergarten',
         'Grade 1',
         'Grade 2',
         'Grade 3',
         'Grade 4',
         'Grade 5',
         'Grade 6',
         'Grade 7',
         'Grade 8',
         'Grade 9',
         'Grade 10',
         'Grade 11',
         'Grade 12')
dat <- c('Grade 3', 'Grade 5', 'Grade 2')
dat <- factor(dat, levels = key)
dat <- as.numeric(dat) - 1
dat

We subtract 1 at the end because the factors start at 1 and you wanted kindergarten set to 0.

answered Mar 29, 2019 at 15:23

Thomas

513 bronze badges

Comments

Julian_Hn · Accepted Answer · 2019-03-29 15:18:38Z

2

This solves your Problem here:

df$TestGrade <- sapply(df$TestGrade,function(el)
  {
  if(el == "Kindergarten") return(0)
  else return(as.numeric(sub("Grade ","",el)))
}

answered Mar 29, 2019 at 15:18

Julian_Hn

2,1411 gold badge10 silver badges19 bronze badges

Collectives™ on Stack Overflow

Using a for-loop to change a column in a dataframe

6 Answers 6

Comments

Comments

Comments

Comments

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

6 Answers 6

Comments

Comments

Comments

Comments

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related