Unlike the knowledge (written) test, in the USA, the
PTS is comprehensive. At the examiner discretion, it
covers everything the FAA believes a pilot needs to
know in order to fly safely. This is my understanding.
One could certainly argue that the PTS is either vague or
incomplete. But by design it is supposed to be comprehensive.

So everything in it is testable, and that is comprehensive.
This doesn't mean everything is TESTED during a given
practical test, just TESTABLE.

The written test seems an example of what you point out, however.

****Are examiners the best solution?******

The FAA collected statistics for pass rates of pilots for
various certificates. They compared the pass rates of
pilots flying with an FAA ASI for a practical test vs. the
pass rates for Designated Pilot Examiners.

The pass rates for both glider initial and add-on ratings
for DPEs was around 90%.

Over the same period, the FAA ASI pass rate for all types of
glider tests was 100%.

What is going on here? Well, the sample sizes were significant
(at least 30+) so that can't be it. One major difference is that
if a DPE has a string of perhaps 20-30 passes, they get
"looked at" a little bit harder.

The FAA ASIs do not get "looked at" harder for passing 100%

In any case, there is some statistically significant inconsistency
in these results.

How about eliminating Designated Pilot Examiners altogether?
Although they certainly put a human face on the FAA, are they
entirely necessary? If glider DPEs are failing 10% of the
applicants, and the FAA during it's mandatory random
flight test checks thinks 100% are fine, then there seems
to be a statistically significant standardization problem.

What do you think? Does the DPE 90% FAA 100% pass rate
surprise you? Are you thinking maybe you have a better shot
going to the FAA instead of a DPE for your next glider
practical test?

http://acra.faa.gov/iacra

is the automated FAA application system.
It can check the numbers by some computer formula to see
if the application is correct. And it can match data to the
student pilot license and medical info already in the database.

Beyond that, a "proctor" could put a logger with ENL in the
aircraft. Noise approximates engine RPM, gives buffet or stall horn,
and/or can record the voice of the pilot "That's the impending stall."

So give the guy a logger, have a "proctor" verify the takeoff, and
have the applicant do the manuever series off a clipboard or
audio tape instruction.

Land, and upload the flight log to FAA. A computer blindly
checks the data, and you get a pass or fail instantly.

I have not figured out how to test for coordination yet.
How do you know if the pilot is coordinated? Maybe a
360 45deg with a fast reversal to another 45deg 360.
The reversal would show differently on the track log coordinated
or not, maybe.

This would certainly provide consistency enforcing
the mathematical standards. It wouldn't test whether the
pilot was sweating profusely or crying during parts of the test,
however...and those are things we sure wouldn't want to see
once they carry a passenger.

But this seems pretty straightforward to implement.

Even if the DPEs remained to do the oral exam part,
the flight part could be done at one's leisure.

Hmmm...loggers are sure an interesting new device I
didn't know anything about until recently. Maybe
the FAA doesn't know about them so much either.

In article ,
Nyal Williams wrote:
At 17:30 04 February 2005, wrote:


The idea that the FAA sets minimum standards, and of
course all
instructors will train to higher standards, sounds
great in theory.
However in the real world, a large portion of the instructors
teach
only what will actually be tested on the practical
test. By debriefing
their students after flight tests, they have learned
exactly what a
particular examiner will expect. This then allows
them to train their
students
for a flight test with that specific examiner, rather
than bothering to
train for a thorough test in accordance with the PTS.

A bit chopped out

................. Unfortunately for this pilot, his
training was
done at an operaton known for shopping around for easy
examiners.

M Eiler



This notion of teaching to the test has come up in
political discussions about education. Even our current
US president was drawn into this about 4 years ago
and suggested that 'teaching to the test -- is teaching.'

Consider that the classroom teacher would teach multiplication
by teaching only those examples on the statewide test
for proficiency. No student would learn the entire
table -- just a few of the 5's and 10's and two or
three of the 6's and 7's -- maybe none of the 8's and
none of the 4's.





--

------------+
Mark J. Boyd