Question

I am trying to merge two data of different size by ID. However, for the values that match, both data contain duplicated entries, i.e., there may be three ID #3 in Data A and three ID#3 in Data B. When I try to merge the data, the result is much larger than both data combined.

C<-merge(A,B,by="ID",all.x=T,sort=F)

I want to merge the two data by the ID column, such that the first ID #3 in B pairs with the first ID #3 in A, and so on.

Also, I want the row order of Data A to remain the same. The sort=FALSE wasn’t much helpful: It places all the matching rows at the top, and the unmatched rows at the bottom.

Thanks for your help!

Answer 1

Before merging, you’ll need to add to each data.frame a column whose value records the index of each observation within its own ID group.

## Example data
A <- data.frame(ID=c(1,1,1,2), ht=1:4)
B <- data.frame(ID=c(1,1,2,2), wt=3:6)

## Add column with number of each observation within ID
A <- transform(A, ID2=ave(ID, ID, FUN=seq_along))
B <- transform(B, ID2=ave(ID, ID, FUN=seq_along))

## Now carry out the merge
merge(A, B, all.x=TRUE, sort=FALSE)
#   ID ID2 ht wt
# 1  1   1  1  3
# 2  1   2  2  4
# 3  2   1  4  5
# 4  1   3  3 NA