The joke called TSA

"Steven P. McNicoll" wrote in message
nk.net...

wrote in message
...

His point was that if you profiled pilots instead of Arabs, you would
have had a greater proablility of a hit on a 9/11 hijacker.

Would we? What fraction of worldwide terrorism is perpetrated by Arabs
and what fraction is perpetrated by pilots?

You've got the statistics backwards. The probability of a hit on a hijacker
if you profile trait X depends on the fraction of X's that are hijackers,
not the fraction of hijackers that are X's.

And as I mentioned earlier, the fraction of pilots in the US who were among
the 9/11 hijackers is larger than the fraction of Muslims in the US who were
among the 9/11 hijackers. (Both fractions, of course, are miniscule.)

You replied, first, that none of the 9/11 hijackers were pilots. When I
documented the falsehood of that reply, you retorted that none of the
hijackers were 757/767 pilots.

The only apparent reason for you to segue to that non sequitur was to avoid
acknowledging that your first reply (which was at least germane) was
factually mistaken.

--Gary

wrote in message
...

Would we? What fraction of worldwide terrorism is perpetrated by Arabs
and
what fraction is perpetrated by pilots?


Wrong question.


There's nothing wrong with the question.



The point is, a greater percentage of the U.S. registered pilot
universe have committred terrorism against the United States, than
have the percentage of the Arab uninverse.


Show your figures.



Therefore you would have been more likely to get a terrorist hit by
sampling the American pilot universe than by sampling the Arab
universe.

The fact that more terrorism is perpetrated by Arabs than pilots is
irrelevant.


Actually, it's completely relevant.

wrote:

On Wed, 22 Dec 2004 04:15:53 GMT, "Steven P. McNicoll"
wrote:


"Gary Drescher" wrote in message
...

"FAA records show that four of the 19 hijackers-one aboard each
flight-possessed FAA certificates as qualified pilots."
--9/11 Commission
(http://www.9-11commission.gov/staff_...tatement_4.pdf)


FAA records indicate that none of the hijackers were qualified in the
aircraft they were flying.




Irrelevant.

His point was that if you profiled pilots instead of Arabs, you would
have had a greater proablility of a hit on a 9/11 hijacker.



And if you had profiled Muslims, you'd have had virtually a 1.0 probability.

Matt

On 12/20/04 3:47 PM, wrote the following:
On Mon, 20 Dec 2004 21:26:33 GMT, Mitty wrote:


Unfortunately, "effective" is also known as "racial profiling".

Another case where political correctness has subverted simple logic.

Even children don't go fishing in areas where there are no fish because it's a
waste of time and money.

However, to be "PC," it is national policy that we waste huge amounts of time
and money searching people who have a negligible chance of being hijackers.
Which, of course, reduces the time and money available to actually work on
catching potential highjackers. Net result: Less safety, higher cost.



depends on whose ox is being gored.

I suspect there would have been a hue and cry from young white
christians if we had singled them out as terror suspects after McVeigh
and Nichols bombed the federal building in Oklahoma City.


The point is that (1) if you have a valid way to subset your population that (2)
gives you a significantly increased probability of finding the bad guy because
you can afford to expend more screening effort on the subset -- then (3) you
should do it.

Said less abstractly: If there are places in the lake where fishing effort has
a significantly higher probability of producing fish, then those are the places
to go fishing.

To your point: Assuming the selection of the subset is valid from a statistical
standpoint, it is irrelevant whether the members of the subset "raise a hue and
cry" or not.

To your implication that selecting young white Christians is logical: A single
crime committed by two members of a huge class is probably not a valid reason to
select the class for extra screening.

The arguments in the posts below, especially those about screening young Muslim
males versus screening pilots, all implicitly accept the idea that we should be
trying to fish where the fish are. It can also be true that it is difficult to
figure out where they are. That does not make it any less desirable to do so.

"Matt Whiting" wrote in message
...
wrote:

His point was that if you profiled pilots instead of Arabs, you would
have had a greater proablility of a hit on a 9/11 hijacker.


And if you had profiled Muslims, you'd have had virtually a 1.0
probability.

No, you're addressing a different probability. CFII and I were talking about
the probability that a given passenger screened for trait X would turn out
to be one of the 9/11 hijackers. That probability is higher for X=pilot than
for X=Muslim (though miniscule in both cases).

You're addressing instead the probability that one of the 9/11 hijackers
exhibits trait X. That probability indeed equals 1.0 for X=Muslim, or
X=male, or X=non-Asian, and many more. Obviously, then, that probability is
not the right criterion for identifying an effective profiling trait.

--Gary

"Mitty" wrote in message
...
On 12/20/04 3:47 PM, wrote the following:
On Mon, 20 Dec 2004 21:26:33 GMT, Mitty wrote:


Unfortunately, "effective" is also known as "racial profiling".

Another case where political correctness has subverted simple logic.

Even children don't go fishing in areas where there are no fish because
it's a waste of time and money.

However, to be "PC," it is national policy that we waste huge amounts of
time and money searching people who have a negligible chance of being
hijackers. Which, of course, reduces the time and money available to
actually work on catching potential highjackers. Net result: Less
safety, higher cost.



depends on whose ox is being gored.

I suspect there would have been a hue and cry from young white
christians if we had singled them out as terror suspects after McVeigh
and Nichols bombed the federal building in Oklahoma City.


The point is that (1) if you have a valid way to subset your population
that (2) gives you a significantly increased probability of finding the
bad guy because you can afford to expend more screening effort on the
subset -- then (3) you should do it.

Said less abstractly: If there are places in the lake where fishing
effort has a significantly higher probability of producing fish, then
those are the places to go fishing.

To your point: Assuming the selection of the subset is valid from a
statistical standpoint, it is irrelevant whether the members of the subset
"raise a hue and cry" or not.

To your implication that selecting young white Christians is logical: A
single crime committed by two members of a huge class is probably not a
valid reason to select the class for extra screening.

The arguments in the posts below, especially those about screening young
Muslim males versus screening pilots, all implicitly accept the idea that
we should be trying to fish where the fish are. It can also be true that
it is difficult to figure out where they are. That does not make it any
less desirable to do so.

so profile towards Muslim pilots its not a case of either / or

Gary Drescher wrote:
"Matt Whiting" wrote in message
...

wrote:


His point was that if you profiled pilots instead of Arabs, you would
have had a greater proablility of a hit on a 9/11 hijacker.


And if you had profiled Muslims, you'd have had virtually a 1.0
probability.


No, you're addressing a different probability. CFII and I were talking about
the probability that a given passenger screened for trait X would turn out
to be one of the 9/11 hijackers. That probability is higher for X=pilot than
for X=Muslim (though miniscule in both cases).

I have no problem with airports screening pilots more severely than
other passengers in the wake of 9/11. I don't know that the probability
is higher for pilots than for Muslims, however, I'll bet that the
probability of male Muslim pilots is much higher than for any other
demographic and thus worthy of being singled out for much greater
scrutiny at airport security screenings.


Matt

so profile towards Muslim pilots its not a case of either / or

Not at all either/or.

There should be and probably are a number of methods ("profiles") for picking a
subset for more intensive review. Some are publicized, like buyers of one-way
tickets bought at the last minute. The point is that if Muslims or Muslim
pilots are logically determined to merit extra attention then it is stupid
political correctness to use the pejorative "racial profiling" and argue that it
should not be done. "Racial profiling" may be just as valuable as "type of
ticket buyer profiling."

You can also profile for groups which merit minimal attention. For example,
"Chinese grandmothers" would be a good candidate group where screening effort
could be minimized. This, too, is sensible "racial profiling." Personally, I
would argue that "fat old white guys" is another "racial profile" where
screening could logically be minimized. Then I could get through airports
faster. :-)

There are lots of reasons to use "racial profiling." In medical research, for
example, it is logical to use race when studying sickle cell anemia (blacks),
cystic fibrosis (northern Europeans), and Tay-Sachs disease (Jews).

It is unfortunate that the "political correctness police" have hijacked a
perfectly innocent phrase and made it pejorative.

On 12/22/04 7:16 PM, wrote the following:
On Thu, 23 Dec 2004 00:43:10 GMT, Mitty wrote:


It is unfortunate that the "political correctness police" have hijacked a
perfectly innocent phrase and made it pejorative.



It is always "perfectly innocent" when it is not your ox being gored.

It might seriously affect your point of view if you are the one who
gets subjected to a patdown, strip search, detailed luggage
examination and other interminable delays every time you travel, while
your fellow citizens pass freely through the gates, simply because of
the color of your skin. You would soon join the "politically
correct", I am sure.

IMHO you need to take some courses that teach logical thinking. It makes
absolutely no difference what the members of the selected subset think about the
screening if the selection criteria meaningfully increase the probability of
catching bad guys. If they do not, that is a different matter.

And, BTW, nowhere have I suggested or seen suggested that "skin color" is a good
screening criterion. It almost certainly is not. "Young Muslim males" may well
be, however.

For now, no doubt, you would be happy to see the metal detectors
labeled "dark-skinned" and "light-skinned", and we could all line up
at our respective stations, and those of us lucky enough to be
"light-skinned" beings could enjoy the express service.

Those lucky enough to be in low risk groups certainly will enjoy the express
service. People objectively concerned about safety should be pleased that
resources are not wasted on screening low risk groups just to be PC. For
example, I would be very pleased to see Chinese grandmothers passed with minimal
screening.

I am always amused by how one's view of "political correctness"
changes in direct proportion with the direction in which the dirty
end of the "political correctness" stick points.

I am glad that you find your own strawmen to be amusing.

This thread is waay too OT. I'm outa here.

"Mitty" wrote in message
...

There are lots of reasons to use "racial profiling." In medical research,
for example, it is logical to use race when studying sickle cell anemia
(blacks), cystic fibrosis (northern Europeans), and Tay-Sachs disease
(Jews).
and bull****itus (N Americans)