AFR Time: 1 in 60 Navigation Question

[Captain Noodle]

Been a while: Can someone check my answer for me. Much appreciated.

You are heading 240deg M and find yourself 4nm off track in 22nm. You want to regain track within the next 10nm, what heading would you turn to to correct for the track error.

I got Track Error of 11 degrees.
Closing Angle of 24 degrees.

Therefore total correction is 35 degrees, makes new heading 205 deg M?

I used a spreadsheet to do calculations, I guess in the air one would round the 22nm down to 20 nm to make 1 in 60 calc easier.

Thanks for help. Hope I didn't bring back too many bad memories.

Noodle

[mad_jock]

Looks fine the only thing which might be wrong is which direction the off track point was to begin with.

MJ

[Captain Noodle]

Good point Madness.

Missed that in the original question. Should read 'left of track'. I just assumed it was as I drew the diagram, off track to the East (left). I was lucky.

Always read the question, isn't that what we were taught in any exam we have ever set, and never assume anything.

Thanks
Noodle

[This message has been edited by Captain Noodle (edited 19 June 2001).]

[126.9]

Good heavens. Now I won't sleep for a few months! Can you believe that you had all that stuff at your fingertips once? 1 in 60 rule = my chance of getting laid on a nightstop, and that's in years btw!!!

g'night!

[Tinstaafl]

That's a very bad question. It's asking you to utilise a technique/method outside the method's normal limits of use.

A 1:60 calculation is subject to a rapid increase in error as the TE or CA increases. This is normally controlled within reasonable limits by NOT using the method to calculate an angle when that angle (eg TE or CA) is greater than about 15 deg.

So, a reasonable magnitude of error in this questions for the TE, but outside any reasonable magnitude of error for the CA.

Hence poor question design.

[This message has been edited by Tinstaafl (edited 20 June 2001).]

[mad_jock]

Unfortunatly JAR thinks it is good enough up to 20deg. And some of the feed back question have answers like 18 deg etc.

Yes i know your right. But as us JAA wannabies have learnt there is the proper way then the JAR way, which is very dependent on which country wrote the subjects question bank.

MJ

[Tinstaafl]

Out of curiosity, is this a multi choice question? If so what are the other options?

Just want to get an idea of the tolerances being allowed for in the candidates calculation.

Using trig, the correct amount to adjust heading (assuming drift is constant before & after the heading change) is 32.1 deg so rounded = 32 deg

Using 1:60 the adjustment is 34.9 deg so rounded the answer is 35 deg.

If any option in the correct sense uses less than the margin for error (including those allowed for from mechanical inaccuracy in the whiz wheel) then this is an invalid question according to fundamental principles of question design.

NB: It may still be partially valid if the question specifically states the method of calculation to be used ie 1:60 however that still leaves the poor design regarding the misuse of the tested skill.

Even using 20 deg as the limit of the 1:60 method they STILL fall outside it with the required answer for CA.

[Captain Noodle]

Tinnie,

Question is as written, not multiple choice.

"You are heading 240 deg M and find yourself 4nm left of track in 22nm. You want to regain track within the next 10nm, what heading would you turn to, to correct for the track error."

I read with interest your comments on the validity of this question. I agree with the point that 1:60 should only be used for TE or CA up to 15 degrees. The text books I have read do suggest this. But I think it is the method they are trying to get across here. But also it is all these other points that you are raising, that I am sure the question is trying to get the student to think about, rather then just come up with the calculated answer. My way of thinking is that all questions should be open to these kinds of analysis. One learns more.

As for the tolerances allowed. I assume the tolerances allowed are similiar to the tolerances one would allow whilst doing this calculation on the 'fly' in the 'head' in the sky. Hence my comment about rounding the 22nm down to 20nm.

I will let you know the outcome.

Noodle



[This message has been edited by Captain Noodle (edited 21 June 2001).]

[Tinstaafl]

It's well accepted fundamental principle in question design that there should not be any confounding factors in the question that are irrelevant to the syllabus item being examined. Similarly there should not be any item in the question stem or response (multi choice) that contra-indicates the use of the desired method to be used.

Inappropriate use of a method is most definately an irrelevant confounding factor and a contra-indication to using the method.

I haven't even introduced other items that are required to develop a valid question / exam.

For example, to use a 'cut-off' score as a pass / fail indicator (pass / fail if the candidate's score is above / below a certain cut off point, say, 75 marks out of a possible 100), testing of the exam questions must be done against a representative sample of the intended recipients to determine bell curve scores.

If 100% of the target candidates fail the exam's question then obviously the exam is set at too high a level for that particular level of expected knowledge. If 100% of candidate pass then it is too easy. The goal is to find an acceptable pass/fail ratio.

I can't recall the exact ratio but it's something like 2 or so standard deviations. WWW can you recall from your Uni psychology days?

If an exam's questions haven't been validated this way then the exam has to be marked 'on the curve'. This is where x% will pass out of all who sat the exam, and the remainder will fail.

In this sort of marking everyone is compared against everyone else & those who are, say, below 2 standard deviations will fail, the rest pass.

Eventually this sort of exam will get enough data collected over time to allow a simple percentage score pass/fail to be assigned.

Question & exam development is a well researched part of psychology, with quite definite principles & practices.

As far as I can determine NONE of the JAA questions have been validated. If the exams were presented at any university they'd be laughed out of the door.

[Engineer]

Could you use the double the error rule of thumb here similiar to tracking an NDB

i.e approx 12deg off so turn right by 24deg

If I am wrong don't all shout at once please

[mad_jock]

Tinstaafl


Do you want me to send you the feed back questions from the exams i have passed?

I think you will have a bit of a shock what we have to put up with with the JAR exams

MJ

[Tinstaafl]

Engineer, that would work if you wished to regain track in the same distance as you've just flown to get off track.

This question specifies a different distance in which it required to regain track, compared to the distance flown to get off track so 'double the TE' won't work.

mad_jock

Why not? Send me a few for a laugh. I'm just glad all that malarky is over & done with for me.