> have “increased risk of committing severe, impulsive, violent recidivist crimes”, and that the variant is more prevalent in prison populations than the genera

> have “increased risk of committing severe, impulsive, violent recidivist crimes”, and that the variant is more prevalent in prison populations than the general public If your goal is to show that the gene variant causes violence, you probably need to compare violent vs nonviolent prison populations, not prison populations vs general public. We already know that there are aspects of genetics that affect likelihood of being imprisoned, all else being equal.