Shootings in Naval Yard, Washington
From NY Times:
WASHINGTON — A former Navy reservist killed at least 12 people on Monday in a mass shooting at a secure military facility that led the authorities to lock down part of the nation’s capital — even after the gunman was killed — in a hunt for two other armed men spotted by video cameras, officials said.
But by Monday evening, the federal authorities said they believed the shooting was the act of a lone gunman, identified as Aaron Alexis, 34, who was working for a military subcontractor.
The chaos at the facility, the Washington Navy Yard, started just after 8 a.m. Civilian employees described a scene of confusion as shots erupted through the hallways of the Naval Sea Systems Command headquarters, on the banks of the Anacostia River a few miles from the White House and about a half-mile from the Capitol.
“I heard three gunshots, pow, pow, pow, straight in a row,” said Patricia Ward, a logistics management specialist from Woodbridge, Va., who was in the cafeteria on the first floor when the shooting started. “About three seconds later, there were four more gunshots, and all of the people in the cafeteria were panicking, trying to figure out which way we were going to run out.”
Police officers who swarmed the military facility exchanged fire with Mr. Alexis, 34, a former naval reservist in Fort Worth. Police officers shot Mr. Alexis to death, law enforcement officials said, but not before a dozen people were killed and several others, including a city police officer, were wounded and taken to local hospitals.
The GCP event was set for 6 hours, beginning at 08:00 local time (12:00 GMT). The result was Chisquare 21528.044 on 21600 df, for p = 0.634 and Z = -0.343. This analysis was corrected after removing one bad egg.
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.