Depending on the distribution of international / domestic students, this also sounds wild and in congruence with the other commenter.

If there is 10 out of 100 international students, and they do 30% of 10 cases, then 30% (!!) of international students cheat while only 7% of domestic students cheat - in conclusion, international students do most of the cheating.

Regardless of this statistical thought experiment, I don't know what to do of data like this. It is not really ethical to act on it.