Question

If I have a data frame like this, which can be think as the responses of a survey for different respondents

Data = 
ID Q1 Q2 Q3
1  A1 B2 C3
2  A2 B1 C2
3  A1 B2 C3
...

and I also have the following score tables:

Q1 <- (c("A1", 10, "A2", 20, ...))
Q2 <- (c("B1", 10, "B2", 20, ...))
Q3 <- (c("C1", 10, "C2", 20, ...))

which mean that, if I choose “A1” in Q1, I get 10 marks, if I choose “B2” in Q2, I get 20 marks more, and if I choose “C1” in Q3, That is also 10 marks, the total score I got would be:
10 (A1) + 20 (B2) + 10 (C1) = 40

And now for every row in the Data, I need to calculation the total score by summing the scores for each columns according to the score tables.

Can any one suggest me how I can use a simple code to do so?
For now I can only think of make the score tables in to a Xx2 Matrices, and use a lot of nested-if and for-loop.

Answer 1

df <- read.table(text="ID Q1 Q2 Q3
1  A1 B2 C3
2  A2 B1 C2
3  A1 B2 C3",header=TRUE, stringsAsFactors=FALSE)

scores <- list(
Q1 = c("A1", 10, "A2", 20, "A3", 30),
Q2 =c("B1", 30, "B2", 10, "B3", 20),
Q3 =c("C1", 10, "C2", 30, "C3", 20)
)

#lookup table
scores <- do.call("cbind",scores)

#match
m <- apply(df[,-1],2,match,table=scores)

#look up
points <- matrix(as.numeric(scores[m+1]),ncol=ncol(df)-1)

#sum for each ID
result <- cbind(df,Score=rowSums(points))

  ID Q1 Q2 Q3 Score
1  1 A1 B2 C3    40
2  2 A2 B1 C2    80
3  3 A1 B2 C3    40