Creating new table from existing table in R

Question

I have an existing data set table. For example, the table is called Table1

         V1   V2            V3 
1      S301  OR     1575.3078990  
2      S301 AND     1006.5031070  
3      S301  OR      938.3647756  
4      S302  OR     1106.0894270  
5      S302 AND     1239.9842820  
6      S302  OR     885.3624568 

I'd like to reorganise this table into a new one where column V2 is split into 'And' and ' Or' columns, with the values being the mean of all the 'And' for S301 in table 1 V2 in the 'And' column
and mean for all the 'OR' for S301 in table 1 V2 in the 'Or column'.

I've been struggling with this for so long so any help would be greatly appreciated! Thank you all so much in advance.

Answer 1

You said you have an existing "data set table". This is not a type in R. I'll assume you meant data.table.

library(data.table)
theDT <- data.table(matrix(c(rep("S301", 3), rep("S302", 3), 
                           c("OR", "AND", "OR", "OR", "AND", "OR", 
                             1575.3, 1006.5, 938.4, 1106.1, 1240, 885.4)), 
                           ncol = 3))
theDT$V3 <- as.numeric(theDT$V3)
> theDT
     V1  V2     V3
1: S301  OR 1575.3
2: S301 AND 1006.5
3: S301  OR  938.4
4: S302  OR 1106.1
5: S302 AND 1240.0
6: S302  OR  885.4

then your answer is

> theDT[, mean(V3), by = c("V1","V2")]
     V1  V2      V1
1: S301  OR 1256.85
2: S301 AND 1006.50
3: S302  OR  995.75
4: S302 AND 1240.00

Answer 2

To get the AND and OR columns separated, you could do this.

dat.1 <- aggregate(V3~V2+V1,data=dat, mean)
dat.2 <- reshape(m, direction='wide', idvar='V1', timevar='V2')
names(dat.2) <- c('ID', 'AND', 'OR')

dat.1 looks like this

   V2   V1        V3
1 AND S301 1006.5031
2  OR S301 1256.8363
3 AND S302 1239.9843
4  OR S302  995.7259

dat.2 will end up looking like this:

    ID      AND        OR
1 S301 1006.503 1256.8363
3 S302 1239.984  995.7259