School Shooting, Sandy Hook, CT

Another unspeakable tragedy. 20 children from 5 to 10 years old murdered by a gunman (one should say a monster) in their schoolrooms in a peaceful little town in Connecticut. Sandy Hook or Newtown was an idyllic place to live and raise children, well off for the most part, and nice people all around. We have friends who lived there until very recently. That idyll has been shattered.

Shortly after 9 a man dressed in army clothes, camo I guess, entered the school with two handguns and an assault rifle. He opened fire in the schoolrooms. 18 children were found dead at the scene and 2 more died at the hospital. 6 adults were also killed, including teachers and the principal. Another adult, the shooter's mother, was found murdered in her home.

The GCP has set and analysed several of these events, which though they result in a relatively small number of deaths compared to war and terrorist attacks, are deeply shocking to civilized people. We hear about it around the world, and each time we are again stunned. As President Obama said, "Our hearts are broken."

Robbie Parker, father of Emilie Parker, said through his tears, "Remember these beautiful children; keep them close to our hearts. Do not let their bright shining faces become extinguished. Let us do everything in our power to ensure their light will continue to shine brighter and brighter in all we do to remember them."

The GCP event was set for 9:00 to 15:00 local time (14:00-20:00 UTC), the usual 6 hour period. The result is Chisquare 21334.997 on 21600 for p = 0.899 and Z = -1.277.

Another view exploring the same data shows that the improbability of the deviations being just chance fluctuations increases steadily as time passes. As more information was broadcast we understood more of the horror. If you will allow poetic license, this looks like a building scream of anquish.

It is important to keep in mind that we have only a tiny statistical effect, so that it is always hard to distinguish signal from noise. This means that every "success" might be largely driven by chance, and every "null" might include a real signal overwhelmed by noise. In the long run, a real effect can be identified only by patiently accumulating replications of similar analyses.