Creating a table with the numbers of times variables occur together in R

Question

I have a list of variables occurring across different studies. Each variable is coded with 1 if it is present in the given study and with 0 if it not, with the table looking like this:

        var1 var2 var3
study1   1    0    0
study1   1    0    1
study1   0    0    0    etc.

My objective is to create a table with the number of times each variable occurs together with each of the other variables, i.e. something like this:

        var1 var2 var3
var1     -    2    4
var2     2    -    1
var3     4    0    -    etc.

How can I do this in R?

I have tried to look for a guide or a similar question, but come up empty.

Thank you in advance for your help!

Your input dataset indicates that var1 and var3 are both present and not present in study1. How is that possible? — Limey, Mar 02 '21 at 00:46
I think this is what you are looking for - the cross product of a tabulation gives the co-occurrence matrix - https://stackoverflow.com/questions/19977596/how-do-i-calculate-the-co-occurrence-in-the-table — thelatemail, Mar 02 '21 at 03:38

score 0 · Answer 1 · answered Mar 02 '21 at 03:34

The following code calculates how many times each variable occurs with each of the other variables and puts them in a symmetric matrix. The diagonal entries are -1.

df=data.frame(var1=c(1,1,0), var2=c(1,0,0), var3=c(0,1,1))
df

  var1 var2 var3
1    1    1    0
2    1    0    1
3    0    0    1

library(dplyr)
mat=diag(-1, ncol(df))
for (i in 1:(ncol(df)-1)) {
  for (j in (i+1):ncol(df)) {
      num=filter(df, all_of(df[,i])==1 & all_of(df[,j])==1) %>% nrow()
      mat[i,j]=num
      mat[j,i]=num
  }
}
mat

     [,1] [,2] [,3]
[1,]   -1    1    1
[2,]    1   -1    0
[3,]    1    0   -1

Creating a table with the numbers of times variables occur together in R

1 Answers1