How to delete a specific value based on the column before in R-CodePudding

I have a dataframe in R that has a list of variables. However, I have printed the last two columns which I am interested in. I want to delete all occurrences of NAs in the value column only when bs_Scores == bs_24.

bs_Scores   value
bs_0    16.7
bs_1    41.7
bs_12   33.3
bs_24   NA
bs_0    25
bs_1    41.7
bs_12   NA
bs_24   0
bs_0    16.7
bs_1    41.7
bs_12   16.7
bs_24   16.7
bs_0    NA

Essentially, I want to delete all the rows within bs_24 that have an NA in the value column while preserving all the NAs in bs_0, bs_1 and bs_12.

SO it should look like this

bs_Scores   value
bs_0    16.7
bs_1    41.7
bs_12   33.3
bs_0    25
bs_1    41.7
bs_12   NA
bs_24   0
bs_0    16.7
bs_1    41.7
bs_12   16.7
bs_24   16.7
bs_0    NA

Thank you!

CodePudding user response：

with subset using dplyr you can use the code below


library(dplyr)

df %>% subset(!is.na(value) & bs_Scores != "bs_24" )

CodePudding user response：

Using base R with subset

subset(df1, !((is.na(value) & bs_Scores == 'bs_24')|bs_Scores == ""))

-output

 bs_Scores value
1       bs_0  16.7
2       bs_1  41.7
3      bs_12  33.3
5       bs_0  25.0
6       bs_1  41.7
7      bs_12    NA
8      bs_24   0.0
9       bs_0  16.7
10      bs_1  41.7
11     bs_12  16.7
12     bs_24  16.7
13      bs_0    NA

data

df1 <- structure(list(bs_Scores = c("bs_0", "bs_1", "bs_12", "bs_24", 
"bs_0", "bs_1", "bs_12", "bs_24", "bs_0", "bs_1", "bs_12", "bs_24", 
"bs_0"), value = c(16.7, 41.7, 33.3, NA, 25, 41.7, NA, 0, 16.7, 
41.7, 16.7, 16.7, NA)), class = "data.frame", row.names = c(NA, 
-13L))