How to assign a new value in a column of a dataframe based on a value from a different column in R?-CodePudding

I have a dataframe

P1   P2  P3_1  P3_2
1    3    1     30
2    NA   NA    NA
2    NA   NA    NA
2    NA   NA    NA
1    1    1     10

I'm trying to set all the NA's to be 0 instead of NA's if P1 = 2.

The code I have is

df$P2 <- ifelse(df$P1 == 2, df$P2, 0)

Which returns this

P1   P2  P3_1  P3_2
1    0    1     30
2    NA   NA    NA
2    NA   NA    NA
2    NA   NA    NA
1    0   1     10

It seems that P2 gets changed to 0 when P1 = 1 instead. I know I could just make it so all NA's are 0 instead, but this trend is specific for when the value is 2. Any other NA's I wouldn't want to change to 0.

Where did I go wrong with the ifelse() syntax? Is there a better way to do this and include all three columns that I need?

CodePudding user response：

When we create a logical in ifelse, the 'yes' condition should be 0 and 'no' would be 'df$P2'. Or without the ifelse, it can be assigned directly as well

df$P2[with(df, P1 == 2 & is.na(P2))] <- 0

-output

> df
  P1 P2 P3_1 P3_2
1  1  3    1   30
2  2  0   NA   NA
3  2  0   NA   NA
4  2  0   NA   NA
5  1  1    1   10

data

df <- structure(list(P1 = c(1L, 2L, 2L, 2L, 1L), P2 = c(3, 0, 0, 0, 
1), P3_1 = c(1L, NA, NA, NA, 1L), P3_2 = c(30L, NA, NA, NA, 10L
)), row.names = c(NA, -5L), class = "data.frame")