I have a dataframe below:
Country Population
123491 9.9
2348 4.3
USA 10.1
Australia 9.1
And I want to remove the rows where the Country is invalid, for example 123491 and 2348. The class of Country is "factor".
> sapply(df, class)
Country Population
factor numeric
I want to get the following as a result:
Country Population
USA 10.1
Australia 9.1
You could look for numbers as part of the Country column and exclude those that contain numbers.
library(tidyverse)
Country <- factor(c("123491", "2348", "USA", "Australia"))
Population <- c(9.9, 4.3, 10.1, 9.1)
df <- data.frame(Country, Population)
df %>%
filter(!(str_detect(Country, "\\d")))
You could subset your data frame using grepl:
df[!grepl("^\\d+$", df$Country), ]
Country Population
3 USA 10.1
4 Australia 9.1
Data:
df <- data.frame(Country=c("123491", "2348", "USA", "Australia"),
Population=c(9.9, 4.3, 10.1, 9.1))
Note: If you want to reject a country based on having any number in it, then just use grepl with the pattern \d:
df[!grepl("\\d", df$Country), ]
It can be done by checking whether the value can be converted to numeric, as follows:
suppressWarnings(df[is.na(as.numeric(as.character(df$Country))),])
# Country Population
#3 USA 10.1
#4 Australia 9.1
So, if the conversion to numeric produced NA, that means the value is of
a character type. I used suppressWarnings so you don't get a warning when converting characters to numeric values.
Hope it helps.
