So why not take being an immigrant "into account statistically"?
It seems like the anti-PC folk have some things they think we should be PC about and take into account (like being a man) while being an immigrant is fair game. It doesn't make any sense to blame crime on one demographic factor, and ignore the others.
Its a matter of anti-PC folk selectively ignoring evidence that doesn't fit their worldview.
As for male criminal citizens, they should be sent to jail. In any case, as I mentioned before, they are already citizens, so the state has a special obligation towards them.
The point is also not to consider every immigrant a criminal just because some of them are, but to deal with life in one's country. If for example you could devise a test that would tell with 100% certainty if an immigrant will become a good, law-abiding citizen, a lot of immigration discussions would become a lot easier.
Btw suppose your suggestion would be followed, and male citizens would be deported because they are such a high risk for crime. Who would do the deporting? The world feminist army? Or should men just pack up and leave by themselves, admitting their own tendency to become criminals?
It should be possible to statistically distinguish the factors of being poor and being poor+immigrant. Or education level+immigrant or degree choice+immigrant. Point is, focusing on immigrants demographically makes no sense when there are plenty of other demographic factors that are better indicators of criminality.