I posted this just a bit over a month ago on 12/11/14:
Newton: Will the Violence End?
On the anniversary of the Sandy Hook Elementary School mass murder, sources indicated that there were somewhere between 17 and 26 more school shootings.
Only two days later I posted:
And Yet Another School Shooting: 12/13/13
And now, a month and one day later:
Student opens fire at New Mexico school, injuring 2
What are we to make of this? I can state what most psychologists with a background in risk management know: such events are statistically rare. They are very difficult to predict.
We’re pretty much left with post-hoc analyses. And psych autopsies when the perpetrator was killed.
The concept of statistical rarity offers a sound argument when you examine the ratio of # of incidents to the total population.
How many students commit school shootings out of the population of students?
Viewed this way, a shooting may become but one additional statistic, lost after a few more breaking news stories emerge, buried in a sea of posts and tweets.
And we implement school safety procedures, enact new legislation, and attempt to identify and “treat” or “rehabilitate” high risk individuals. But the stats just don’t change. Sixteen or so shootings in a year don’t make much of a statistical dent overall.
Perhaps the variables are skewed.
Looking at total N as the universe of school shootings over a specified time period may be more helpful.
If, as a hypothetical, in a given geographic area there have been an average of 15 incidents a year over three years, and new safety features were installed and new safety procedures implemented at the beginning of the fourth year, and after two years the average number of incidents dropped to 13, a simple mathematical computation will indicate a 13% drop and that incidents are at 87% of what they were before.
Would that such analyses were this simple. Statistical computations are tough, levels of significance difficult to arrive at, let alone attributing the change to what you’d like to think was responsible for the change, as opposed to pure chance, one or more artifacts, or one or more active change ingredients not counted on or identified.
But this “feels good” statistics may offer hope and open up doors of investigation and change. There is something facially valid, for instance, about believing that steps taken by Long Island school districts that include “installation of security cameras, signing on with agencies that monitor buildings, hiring better-trained security guards, reviewing and modifying district security policies, and revamping visitor restrictions” make schools safer. See:
Absent evidence-based research, assumptions about improved safety are just that: assumptions. They must be put to the test.
Even something that looks like a “gold standard” may have an Achilles heel.
Fortified, bullet-proof cockpit doors are terrorist-proof. Aren’t they?
Think again. What happens when the door is opened so the flight attendant can bring food or beverages to the flight crew? Or the pilot has to use the bathroom? Infallible…not:
- Soon after 9/11, a new, fortified cockpit door was breached by a drunken passenger on an international flight.
- A few months later, an after-hours cleaning crew easily broke one off its hinges on a bet, by running a heavy snack cart into it, instantly gaining access to the cockpit.
The same article on school safety reported that:
Some experts have noted that the violence in Newtown was so random and occurred so rapidly that almost any security measures could have been defeated.
The admonishment, “They are not looking at the big picture that they need to be doing”, however, does not suggest doing nothing because no matter was is done, it too will be defeated.
The adage, “If you’re not part of the solution, you are part of the problem”, may be a neat slogan to motivate action fueled by passion.
But passion can take us down wrong alleys, lead to tunnel vision, reinforce stereotypes, and make us look for scapegoats.
Statistics are a funny thing. The same data looked through different lenses can lead to different interpretations with ensuing different actions that result in different consequences in the long run.
I think that looking at the universe of school shootings as a statistical base lends itself to viable solutions and interventions as long as caution is not thrown to the wind and scientific reason does not cave in to emotionally driven laxity.
If a school shooting is prevented, and we don’t know about it because a would-be perpetrator did not become a fait accompli, isn’t this a successful outcome?