comments formfirst comment lyingwithstatistics

Doing background research related to yesterday’s blog post on bullying, I ran across this CBS News headline from last November: “Homicides of transgender women in U.S. reach alarming high.”

The total at that point in 2015: 22 trans-women. How alarming is that? It depends on what kind of story you want to tell.

You could tell the story that homicides in this category had almost doubled compared to the previous year. That would be true. Sort of. For you could also tell the story that a typical trans-woman’s likelihood of making it through the year without being murdered sagged by just ten 7 ten-thousandths of 1 percent (0.0007%) last year.

That’s based on common estimates that the U.S transgender population numbers 1.4 million persons, and about three-quarters of them are male-to-female. But maybe the population of interest includes only those who are actually living as women; for not all of those million or so persons have come out in that fashion. I couldn’t find an estimate of that population. Based on an estimate of about 32,000 post-surgical trans-women in the U.S., however (persons born with male genitalia who have transitioned as fully as possible to the female form), the homicide-related safety factor plummeted a whopping 3 one-hundredths of 1 percent (0.03%).

There’s definitely something alarming here, and here’s what it is: each of these versions is true in its own way, but one of them misrepresents the actual risk to any person or group so seriously, it’s better seen as co-opting statistics to peddle a politically expedient yet misleading message.

CBS chose to tell that version anyway. Why not? It’s more spectacular. It draws more viewers. And it fits their politics.

(For the record — since someone is bound to accuse me of insensitivity — yes, I agree that any murder at all is too many murders.)

Image Credit(s): YouTube (Show It Me Through Poetry).