Removing rows from dataframe that contains string in a particular column

Question

So I'm cleaning up a huge data file in R and an example is as shown:

ID       Score
1001       4
1002       2
1003       h
1004       v
1005       3

Because the class of Score column is "character", I want to use the as.numeric function to convert 4,20 and 30 to numeric values. But since the data is dirty (contains unreasonable strings like h, v), I get the message:

NAs introduced by coercion.

When i run the following:

as.numeric(df$Score)

So what i want to do now is to remove the rows in the dataframe that contains strings with letters so that i would obtain:

ID       Score
1001       4
1002       2
1005       3

Does this answer your question? Remove the rows that have non-numeric characters in one column in R — camille
– camille, Commented Feb 8, 2020 at 18:58

Ronak Shah · Accepted Answer · 2020-02-08 06:41:39Z

1

There are multiple ways you can do this :

Convert to numeric and remove NA values

subset(df, !is.na(as.numeric(Score)))

#    ID Score
#1 1001     4
#2 1002    20
#5 1005    30

Or with grepl find if there are any non-numeric characters in them and remove them

subset(df, !grepl('\\D', Score))

This can be done with grep as well.

df[grep('\\D', df$Score, invert = TRUE), ]

data

df <- structure(list(ID = 1001:1005, Score = c("4", "20", "h", "v", 
"30")), class = "data.frame", row.names = c(NA, -5L))

answered Feb 8, 2020 at 6:41

Ronak Shah

391k20 gold badges173 silver badges237 bronze badges

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

2 Comments

Electric Over a year ago

thanks, but the first solution still produces a message "In eval(e, x, parent.frame()) : NAs introduced by coercion". But the second one works without having the message generated.

Ronak Shah Over a year ago

@Astrox yes, because we are trying to change the character value to numeric there hence you get the warning message. In this case, it is safe to ignore it.

Taher A. Ghaleb · Accepted Answer · 2020-02-10 05:10:53Z

0

You may use the str_detect in the tidyverse package, as follows:

df[str_detect(df$Score, "\\d"),]

or

df %>% filter(str_detect(df$Score, "\\d"))

Both produce the output:

#    ID Score
#1 1001     4
#2 1002    20
#5 1005    30

Hope it helps.

answered Feb 10, 2020 at 5:10

Taher A. Ghaleb

5,2905 gold badges33 silver badges47 bronze badges

Collectives™ on Stack Overflow

Removing rows from dataframe that contains string in a particular column

2 Answers 2

2 Comments

Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

2 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related