Incidentally you can play some interesting games with these numbers to present different narratives. For instance, the percentage of murdered men killed by the

Incidentally you can play some interesting games with these numbers to present different narratives. For instance, the percentage of murdered men killed by their intimate partner was 6% while for women it was 34%, that's a huge difference. That's basically 5x as many! And that "five times higher" interpretation is emphasized at the top of the PDF as the alternate title. But if you instead ask, of the people who were murdered by an intimate partner, how many were women and how many were men? That breaks down to 61% vs 39%, still a substantial difference but far short of that evident "5x" from the other framing. Viewed that way, it's more like 1.5x. Lies, damn lies, and statistics.