How to produce an R count matrix

问题:

In R, I can return the count results using the specific column names I am interested in as an array as below.

require("plyr")
bevs <- data.frame(cbind(name = c("Bill", "Llib"), drink = c("coffee", "tea", "cocoa", "water"), cost = seq(1:8)))
count(bevs, c("name", "drink"))

# produces
  name  drink freq
1 Bill  cocoa    2
2 Bill coffee    2
3 Llib    tea    2
4 Llib  water    2

How can I get the count result of two specific column names in a matrix which has columns: all unique drinks, rows: all unique names and cells: freqs (like below)?

     cocoa  coffee tea water
Bill   2      2     0   0
Llib   0      0     2   2

P.S: Obviously, the solution does not need to use plyr.

回答1:

You want a contingency table, which you can create using table:

table(bevs[, c("name", "drink")])
#      drink
#name   cocoa coffee tea water
#  Bill     2      2   0     0
#  Llib     0      0   2     2