Tuesday, January 17, 2012

This Is Everything I Hate About Statistics

Or maybe everything I hate about the way these statistical studies are reported. An article about how the number of pedestrians killed while wearing headphones has tripled popped up on a number of the websites I follow. Usually, I'm happy just reading the comments from those sites because, after all, I go there because I like the writers, but no one was answering the questions I had about the article. So I went to the main cited source, which was equally devoid of the information I desired. I can only assume that the  AFP reporter is lazy and only got the gist of the study or that the researchers have no idea what they are doing.

I don't really care one way or another about the results of the study. I go running with headphones on, but I hate seeing cyclists and drivers using them (their inattentiveness tends to have greater ramifications, especially drivers). As for pedestrians, I feel no ill will, though I choose not to wear them while walking. What irks me about the article and maybe the study is that, given the information we have, it's meaningless. Let's see what's missing...

-- Since we have no data on how many pedestrians listen to headphones while walking around in 2004 versus 2011 (the dates cited in the article), we have no idea if there is a real increase in deaths. It could easily be exactly what's expected if the pedestrian-killed-while-listening-to-headphones:total-pedestrians-that-listen-to-headphones ratio is the same.

-- Similarly, the article claims that from 2004 to 2011 there were 116 deaths in this group, with sixteen in 2004 and forty-seven last year. That leaves fifty-three deaths from 2005 to 2010. We aren't given the dispersal of those deaths, but I'm fairly certain the numbers over that time period do not constitute a rising trend. Maybe we aren't given a chart tracking this data because 2011 looks like an anomaly (just speculating...).

-- Maybe the reason there are more accidents of this nature is not because more pedestrians started picking up there headphones along with their keys and wallet, but because there are more pedestrians period. The article states that in more than half of the incidents, the pedestrian was run down by a train. Perhaps, said pedestrian was running to catch the very train that caught him (the article makes it known that 2/3 of the victims were male and under thirty, so I'm comfortable saying "him"). And yes, I'm fully aware of the irony in citing statistics in a write-up decrying just that.

-- Hell, and where are the statistics on public transportation usage? Most of us don't live right off of a bus/subway line. We've got to get there somehow. Does the study differentiate between pedestrians and commuters?

-- Why does the study exclude cell phones when so many of them are used as mp3 players now, not to mention have any number of other distractions attached? Isn't distraction distraction regardless of device? Why not shift some blame to buses that feature adds on the side of them? Surely that's a distraction. The researchers are claiming one form of distraction is worse than another, whether they intend to or not.

-- And what are the numbers on non-headphoned pedestrians? Are they going up as well? Surely pedestrians in headphones aren't the only people getting run down out there.

This is the exact type of report designed to try to scare people into doing something. It's vague and on shaky ground. Whether that's the researcher's fault or the reporter's, I'll never know because I haven't been able to find a link to the study. I get the sense that Dr. Richard Lichenstein has a child to whom he's trying to prove a point.

1 comment:

  1. So easily one bad driver could have plowed into a crowd with headphones on, and that explains the rise? Also: I don't think something in the double digits is really 'shocking' enough to cause concern. Talk to me with it's 50,000.

    (I usually wear headphones while biking, I keep the volume at a level that I am still able to adequately hear my surroundings. I also wear headphones while running, but with the same stipulation.)

    ReplyDelete