So I have a data set that represents countries' votes in the UN. I would like to see how certain countries correlate. The possible values are "Yes" "No" "Abstain" or "No Vote". Now at first I thought it was very simple to just see how the countries correlate, but then I thought a bit deeper about it and decided an "Abstain" and a "No vote" could be similar to a "No". Some countries, however, are rarely present.
My idea is to create a scale like this: "Yes"=2 "No Vote"=1 "Abstain"=-1 "No"=-2
I would also like to weigh "No votes" so that if a country has a high number of "No vote"s each one will count less when comparing to other countries' "Abstain" or No"s
I have a fair amount of math education, but very little formal statistics education so I apologize if this is blatantly obvious or poorly worded.