I have some trouble creating new columns from existing columns.
What I want to achieve is:
1) For rows where treatment equals "mixed", I want to locate the NA and then;
2a) If the NA is under pun1 or pun2 then;
the integer of eva3 and eva4 must be converted/copied to eva_out1 and eva_out2, respectively,
2b) Vice versa, if NA is under pun3 or pun4, then;
eva1 equals to eva_out1, where eva2 also has its integer converted to eva_out2
We will never have eva_out3 because of the dyadic group composition.
At first I tried this:
df5$eva_out1 <- with(df5, ifelse(
(is.na(pun1) | is.na(pun2)) & treatment == "mixed",
df5$eva_out1 <- eva3,NA ))
Which gets most of the job done, but I can't run this multiple times, because other values would be overwritten by the other block of code that goes up for eva_out2.
Then I tried this:
df5$eva_out1 <- with(df5, ifelse(
(is.na(pun1) | is.na(pun2)) & treatment == "mixed",
df5$eva_out1 == eva2 & df5$eva_out2 == eva3, ifelse(
(is.na(pun3) | is.na(pun4)) & treatment == "mixed",
df5$eva_out1 == eva1 & df5$eva_out2 == eva2, NA )))
And this
if(df5$treatment == "mixed") {
} if ( is.na(pun1) | is.na(pun2) ) {
eva_out1 <- eva3 & eva_out2 <- eva4
} else if ( is.na(pun3) | is.na(pun4) ) {
eva_out1 <- eva1 & eva_out2 <- eva2
} else {
eva_out1 <- NULL & eva_out2 <- NULL
}
But both either spew errors or do not give the results I need. I've been looking at functions, but ifelse() seems more legit in this situation. Correct me if I'm wrong though.
First 12 rows of the data:
. UniqueSS subject group part round treatment pun1 pun2 pun3 pun4 eva1 eva2 eva3
1 11 1 1 punishment 0 homogenous NA 0 0 0 0 0 0
2 12 2 1 punishment 0 homogenous 0 NA 0 0 0 0 0
3 13 3 1 punishment 0 homogenous 0 0 NA 0 0 0 1
4 14 4 1 punishment 0 homogenous 0 0 1 NA 0 0 0
5 11 1 1 punishment 1 homogenous NA 0 0 0 0 0 0
6 12 2 1 punishment 1 homogenous 0 NA 0 0 0 0 0
7 13 3 1 punishment 1 homogenous 0 0 NA 0 0 0 0
8 14 4 1 punishment 1 homogenous 0 0 0 NA 0 0 0
9 11 1 1 punishment 2 homogenous NA 0 0 0 0 0 0
10 12 2 1 punishment 2 homogenous 0 NA 0 0 0 0 0
11 13 3 1 punishment 2 homogenous 0 0 NA 0 0 0 0
12 14 4 1 punishment 2 homogenous 0 0 0 NA 0 0 0
For the third row we would have columns eva_out1 == 0, eva_out2 == 0, because member pun4 is in the same category group as pun3 and thus may not be converted to another column.
Thanks in advance!
Not sure I fully understand but is this what you're after?
df5$eva_out_1 <- df5$eva_out_2 <- NA
cond1 <- df5$treatment == "mixed" & (is.na(df5$pun1) | is.na(df5$pun2))
df5$eva_out_1[cond1] <- df5$eva3[cond1]
df5$eva_out_2[cond1] <- df5$eva4[cond1]
cond2 <- df5$treatment == "mixed" & (is.na(df5$pun3) | is.na(df5$pun4))
df5$eva_out_1[cond2] <- df5$eva1[cond2]
df5$eva_out_2[cond2] <- df5$eva2[cond2]