MetaEzra reader MP writes in to protest last week's coverage of the fact that Cornell's suicide rate is lower than the national average: You cite CU suicide rates over 21 years. The big 10 study is over 10. Apples, Oranges. And you leave off this gem: "The University of North Carolina at Chapel Hill’s Suicide Prevention Task Force 2004 report indicated that over a five year period seven UNC students committed suicide. This is consistent with the prevalence rates noted above for this size campus." I haven't compared the size of UNC to CU...let's assume they're comparable for a moment. This would create a problem for the idea that the narrative (not the facts) are the problem re:Cornell suicides. Wouldn't you agree? You can normalize and rationalize out to however many student suicide hours per weekend you want. My perception of reality is that people are worried about their friends and family... and that this is not a numbers game. People rightly perceive[d] an escalating number of suicides.I'd politely suggest that you should really know better than to 1) quote the ithaca journal for...anything. 2) use statistics in an argument.
Until proven otherwise, the statistics provided by Susan Murphy (as quoted by the Ithaca Journal) reflect responsible journalism.
But let's see. UNC has a campus of 30,000 students. 7 suicides over 5 years yields a rate of 1.4 a year. Let's normalize that to a rate per a hundred thousand students years... That gets us to 4.6 suicides per a hundred thousand student years.
If that number seems vaguely familiar, it's because that's Cornell's long-term student suicide rate prior to the tragic cluster of suicides this spring.
Taking a step back, from the numbers, however, nobody's arguing about whether or not there was an escalation of suicides this spring. There was a tragic clustering of suicides that could have precipitated even more tragedies had the administration not have reacted as strongly as it did. The cluster that we experienced was simply awful, but that doesn't mean that Cornell itself has a suicide problem. Suicide clusters can happen anytime and anyplace.
Statistics are great when they are used within context and with a scientific approach. However, when we're just quoting numbers at random and without any underlying approach to rationally compare things it's easy to get numbers to look like you want them to look. So yes, Cornell did have six suicides in one year, and UNC did have seven suicides over the course of five. But Cornell also has has six over the last four. You can slice and dice it a lot of different ways to conform to your world views.
But if we want to take a scientific approach to this, we should know that for rare events like suicides or murders, numbers are always quoted on a per 100,000 basis as that's the minimum sample size you need in order to make comparisons. A town of 10,000 with one murder in any given year (but none in the next nine) certainly doesn't have a higher long-term rate of murders than a city of 100,000 with one murder a year. See for example, this research piece by the CDC which quotes death rates nationwide on a per 100,000 basis.
So I rest my case.
People are rightfully worried about their friends and their loved ones at Cornell given the events of this spring. And one way to help friends and family is to help fight the myth that Cornell is a suicide school. Because, well, it's not. And we certainly don't want the myth to become a reality. It would be a sad day if Cornell started considering itself a suicide school (despite all evidence to the contrary) and it became more socially acceptable for Cornell students to take their own lives.
That would be the worst outcome of all.
