The problem with these sorts of calculations is that people are much more likely to cite research done by people they know, people in their own country and work

The problem with these sorts of calculations is that people are much more likely to cite research done by people they know, people in their own country and works in a language they understand. So it ends up skewed in the favour of english speaking countries and countries with large numbers of computer scientists.