NEW YORK -- A cool-headed pilot maneuvered his crippled
jetliner over New York City and ditched it in the frigid
Hudson River on Thursday, and all 155 on board were pulled
to safety as the plane slowly sank. It was, the governor
said, "a miracle on the Hudson."
One victim suffered two broken legs, a paramedic said, but
there were no other reports of serious injuries.
US Airways Flight 1549, an Airbus A320 bound for Charlotte,
N.C., struck a flock of birds just after takeoff minutes
earlier at LaGuardia Airport, apparently disabling the
engines.
The pilot, identified as Chesley B. "Sully" Sullenberger III
of Danville, Calif., "was phenomenal," passenger Joe Hart
said. "He landed it --I tell you what, the impact wasn't a
whole lot more than a rear-end (collision). It threw you
into the seat ahead of you.
"Both engines cut out and he actually floated it into the
river," he said.
In a city still wounded from the aerial attack on the World
Trade Center, authorities were quick to assure the public
that terrorism wasn't involved.
The plane was submerged up to its windows in the river by
the time rescuers arrived, including Coast Guard vessels and
commuter ferries that happened to be nearby. Some passengers
waded in water up to their knees, standing on the wing of
the plane and waiting for help.
The GCP event was set for 3 to 7 pm, local time. The plane
took off from LaGuardia at 3:26, hit a flock of birds about
a minute later, and ditched into the Hudson river about
3:33. The result is Chisquare 14115.638 on 14400 df,
corresponding to a p-value of 0.954, and a Z of -1.683. The
cumulative deviation of the data shows a pretty steady
downward trend during the 4 hour period. It is
interesting to speculate (assuming the result is not mere
chance fluctuation) on whether this might reflect an
emotional response to a near-tragedy that turns out not to be
tragic.
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.
