It’s commonly believed in many countries domestic violence spikes following a loss in a major sports event. It’s easy to see why “facts” like this spread easily: they seem to stand to reason (we tend to be upset when our favourite team loses, and most people are more prone to violence when they’re upset), as well as being an implicit criticism of our obsession with sports. The problem is, these “facts” are often false.
In the US a few years ago a story did the news circuit about domestic violence increasing following Superbowl Sunday. While it’s a good story, on examination the facts seem to have been largely manufactured, and the contention is not supported by academic evidence.
In 2003 a similar story did the rounds here, originating from a study commissioned by the National Collective of Independent Women’s Refuges, and already the idea is being dragged out again as pundits of all colours weigh in on the effect of the All Blacks’ loss on the national psyche.
I haven’t been able to track down a copy of the NCIWR report, so I can’t comment on its relevance, but anecdotal evidence alone that domestic violence goes up after an All Blacks game isn’t enough. This is because All Blacks games tend to happen on the weekend, and it’s possible that domestic violence always goes up on weekends anyway. Moreover, domestic violence will fluctuate randomly from week to week, so to suggest that the All Blacks are having an effect on domestic violence, we need to be able to prove that the any increase in incidents (after adjusting day of the week, and any other known factors) is more than can be explained by random variation.
The fact is that most of the data that we see is influenced by many factors, so show that a one-off occurrence has an effect, we need to make the necessary adjustments for other known influences.
In the same way, the most recent Molesworth and Featherston newsletter goes out on a limb when it claims: “New Zealand has been dumped out of the world cup four times now. Each time, our economy has accelerated in the following quarter. We have won the world cup once – in 1987, immediately before the stockmarket crashed.”
While they’re being flippant, we can’t automatically conclude that the All Blacks are helping the New Zealand economy because the World Cup tends to be held right before Christmas, when the economy peaks anyway.
It’s often said that “there are lies, damned lies, and statistics”. Statistics aren’t bad in and of themselves, but they can be bad when they’re used wrongly. I’d prefer it if we acknowledged that there are “lies, damned lies, and bad statistics”.
Update: The Dom doesn’t fail to deliver: http://www.stuff.co.nz/4229264a10.html