CODEWITHSUNDEEP
r
John
John

Reputation: 3

Excluding Observations that Include a Specific Value

I would like to exclude observations that include specific values. That is, the observations in my columns look like this: 501.512.518. These represent three different individuals per observation. Now I want to exclude all observations that include e.g. the individual 512

Is there a way to create sub-samples that can exclude observations which include the value 512 but do not equal or start with the value 512?

Upvotes: 0

Views: 1564

Answers (3)

daniel
daniel

Reputation: 1246

The answer provide by @steveb is great, but you may want a base R alternative:

Example with mtcars dataset

df <- data.frame(Group = rep(c('A', 'B', 'C', 'D'), 50),
                 Number = sample(500:600, 200, replace = T))

to_drop <- c(512)

df <- df[!(df$Number %in% to_drop),]

 > df
    Group Number
1       A    518
2       B    536
3       C    518
4       D    505
5       A    544
6       B    511
7       C    507

Upvotes: 0

Tal J. Levy
Tal J. Levy

Reputation: 598

This question lacks some details, still I hope this could help. I would used the grepl function.
In order to remove all rows in the dataset that (one of its) columns (denoted as col) contain (but don't start with) 512, do:

newDF <- oldDT[!grepl('.512', oldDF$col), ]

grepl('.512', oldDF$col) will return a logical vector with TRUE every-time the column "col" has the pattern ".512" in it.
The ! infront of it will negate it and thus remove those rows.
Hope it helps.

Upvotes: 1

steveb
steveb

Reputation: 5532

It would be helpful if you had a more specific example but making some assumptions, and using dplyr you could do something like the following:

exclusion_values <- c(501, 512, 518)
new_df <- old_df %>% filter(! col_of_interest %in% exclusion_values)

Is this what you are looking for ?

Upvotes: 0

Related Questions