In defense of the HNC exam

[Doug_Adam]

Rick

"Every answer to every question on an exam SHOULD be available somewhere in class materials or lectures. This has been true in every class that I have taken."

I don't believe this is a good idea. Depending on the level you are teaching, it is common, and I think good, practice to have the students do their own research and not have all answers available. I have found that this more commonly reflects real life experience. After all, your boss cannot ensure that every answer to every problem is available somewhere in his instructions to you.

In fact, perhaps this question was well thought out to produce the very effects that we observe?

Doug

 

[rsdoran]

Ok Allen and all, I am going to pull my hole card. My idea about a page with answers is not new...maybe a few of you will recall this thread:
http://www.plctalk.net/qanda/showthread.php?s=&threadid=769&highlight=christmas

BTW is the old Santa Claus thread still available...it was a wild tangent but did have alot of info.

 

[Rick Densing]

Rick
I don't believe this is a good idea. Depending on the level you are teaching, it is common, and I think good, practice to have the students do their own research and not have all answers available.
Doug

For graded research papers and essays, yes. Exams, no. An exam is looking for one right answer. Research can result in more than one.

If you send your students out to find answers, and then discuss the results in class, that is good. But now the research has been integrated into the course material.

 

[Thomas]

It seems that this is largely a quibble about course "padding". Back in an ed methods class I recall that one should teach what is CRITICAL, OBSERVABLE, and DISTIGUISHABLE. That is to say: Is IT important? Can you see/taste/smell... IT? and can you tell the difference between IT and something else?

But in the pseudo real world of education the syllabus gets padded. It's easier than adding meaningful content. This padding takes on a life of it's own and becomes institutionalized... and you cna't get rid of it because it's in the "authorized version" blah blah blah.

As the old saying goes:

Those who can, do
Those who can't, teach.
Those who can't teach, write the manuals.

 

[Doug_Adam]

Rick

Quote:
'For graded research papers and essays, yes. Exams, no. An exam is looking for one right answer. Research can result in more than one.

If you send your students out to find answers, and then discuss the results in class, that is good. But now the research has been integrated into the course material.'

I agree with you for exams.
Now, do you think the students are sneaking out of the exam rooms to post this stuff here?

Actually, I agree that the question is stupid, and was probably badly thought out. It takes a careful type of sadist to carefully think up these types of questions on purpose. A couple of my lecturers did this sort of stuff occasionally to provoke a response and force the students to realise that they are not in high school any more.

Doug

 

[Pierre]

Lord knows how I lack the basic respect which should normaly be shown to teachers. It comes from my judgment of all the crappy ones I've had in the past.

I've also had many very good ones. The thing is, they where not paid to teach but to do. They didn't know it but they where fantastic teachers.

This Ashley is identical to the many unfit ones. BUT the reason I place him in this category might not be exactly what you guys thinck it is.

I've browsed through the refered document and to me it looks OK.

If the students manage to learn a little of all that is listed there, I beleive they will have aquired a lot of the basics needed to be something like a Plant Engeneer or a Technical Rep for a major brand.

Lets face it, this course is not what's needed to become a good system integrator or automation specialist. There is no course you can take to be one.

You can only hope to slowly become a good one after many years in the field.

I remember how this escellent teacher tought (hum! this spelling must be wrong) us a perfect hunderstanding of a PID using only pieces of woods, wheels, bellows, valves and plastic tubings. No numbers, no equations, only perfect theory and application.

Ashley should defend the document he mentions. He should be able to come up with descent arguments on wy the fact of even briefly touching the surface of some subjects IS enough to give a learning engineer a good start.

School is not an end, its only a beginning.

I don't care what computers they use. A cheap micro-controler would suffice.

I don't care what software they have access to. DOS, QBasic and a parallel port would be more than enough.

The goal of any courses we take is to show us the path. To traine us on how to traine ourselfs. To help us learn how to learn.

When they ask a dumb question about the packaging of a PLC, its not important that they use archaic words, whats important is that a student realises that they do come in different packages and that some level of technologie is in direct relation we them package styles.

Hey, they even thinck that "Flexible Manufacturing Cell" really means something. In fact it does. Its just not always called this way.

I even learned something. Remember a whyle ago, a post asked about what is considered "Advance functions".

Now I know, reading the document I found that it starts with: less than; greater than... and more

As for the students posting silly questions. At least they have access to a PC. You'd be surprised how many EE students don't even have one when they begin there school.

There are no stupid question, the saying goes.

Are those students lazy. Not more than myself, I'm to lazy to answer them.

Do they get cheated of good education. No way. They just get out of this school what they have put in it themself.

 

[Rick Densing]

Now, do you think the students are sneaking out of the exam rooms to post this stuff here?

Close. A lot seem to be a few hours away from taking them.

 

[msinclair]

I like this debate. Thought I'd add my two cents:

For graded research papers and essays, yes. Exams, no. An exam is looking for one right answer. Research can result in more than one.

I see an inherent flaw with this statement - if research can result in more than one right answer (and it usually can), how can an exam which looks for just one answer be a valid indication of skill, ability, or knowledge in a subject?

In my mind, this implies that an exam is strictly a way to measure how well someone can memorize what the instructor has told them. This philosophy "inspires" students to ask questions like we see on this board - the exam isn't about learning the subject - just getting the grade. This is a large part of why I've always preferred the "learn-by-doing" or "project" based methods of instruction.



There was a great story (I believe about Niels Bohr) where a student absolutely infuriated his Physics instructor by refusing to answer with the method the instructor was looking for.

The question was something on the order of "Using a barometer, determine the height of a tall building . . ."

His answers ranged from: "Find the superintendant and ask if he'd accept the barometer in payment for the information about building height."

To: "Measure the barometer height. Place it against the wall of the stairwell, and turn it end over end while climbing the stairs. When you reach the top of the building multiply the barometer height by the number of flips."

His instructor gave him failing marks, but Bohr protested saying that he answered the questions completely and with answers that worked.

Through arbitration he agreed to retake the test using conventional physics answers and (of course) passed with flying colors . . .

 

[Rick Densing]

There is no flaw in my statement, as you actually demonstrated. You may find flaws in the way exams are administered, as Bohr did, buy my statement holds true under the current system.

In the first method, Bohr doesn't know that he can accurately determine the height of the building. The superintendent may not know, may have inaccurate information, or is deathly afraid of barometers. It would require research to find out.

Of course the right answer is to drop the barometer from the top and time how long it takes to land.

 

[msinclair]

Rick,

Of course the right answer is to drop the barometer from the top and time how long it takes to land.

Yup - that was another of his answers (which of course is Physics based so should have gotten full marks ). There was actually a series of other answers too - I'd have to dig through some old paperwork to find the whole list. If I find it, I'll post it.

There is no flaw in my statement, as you actually demonstrated. You may find flaws in the way exams are administered, as Bohr did, buy my statement holds true under the current system.

No offense intended - perhaps I misspoke. The flaw is not in the statement that you made, but rather in what it implies is the purpose of the exam. An exam should test knowledge of a subject. If there are multiple acceptable answers, and only one of those answers gets full marks, then the exam is not testing knowledge, but rather memorization ability.

What I was trying to point out was that if the exam is testing memorization ability (and not learning the subject) then we are encouraging students to ask for spoon fed answers so that they can get the grade - not the knowledge. That's why I think we see questions like: "I'm stuck - can you experts post the traffic light solution".

Marc

 

[Antipathy]

marc,

An exam should test knowledge of a subject. If there are multiple acceptable answers, and only one of those answers gets full marks, then the exam is not testing knowledge, but rather memorization ability

Surely if a student has learned the subject thoroughly, he/she will have enough knowledge to give the answers that the examiners are looking for.
This isn't memorization just for passing an exam. It's knowing the subject well enough to give the desired answer as opposed to any old answer - which may be correct but doesn't demonstrate the required understanding of the subject in order to obtain a pass mark.

As for this Bohr character;

His instructor gave him failing marks, but Bohr protested saying that he answered the questions completely and with answers that worked.
Through arbitration he agreed to retake the test using conventional physics answers and (of course) passed with flying colors . . .

I think he was extremely lucky to be given another chance. If he knew the subject well enough to give the correct and desired answer then what was he doing?... Taking the p**s methinks. Which is fair enough and even quite funny but I don't think I'd have let him retake if the decision had been mine.
Harsh? Yes, but he would have been taught a valuable lesson in appropriate behaviour for serious circumstances.

Anthony

 

[msinclair]

Anthony,

Surely if a student has learned the subject thoroughly, he/she will have enough knowledge to give the answers that the examiners are looking for.

I largely agree with you, but:

The flip side of this is someone who doesn't have a clue what is going on, but can memorize and repeat verbatim what the instructor says. In general, testing (unless project based, essay, etc) does not distinguish between these two cases. I have met plenty of folks with great credentials (on paper) that can't solve a problem to save their lives. Got the grade, didn't learn the subject.


[up on my soap box one last time . . .]

As far as knowing a subject well enough to give the "right" answer - yes - in many cases that is absolutely true. The downside of course is when it gets to the point of: "Give the answer that the instructor wants even if you don't agree with it." (or even if the answer is not relevant like many have suggested about the differences between unitary, modular, and rack mount PLC's for example)

As you probably guessed - I am biased against most tests with "one right answer" (I hide that pretty well don't I? ). The only way that kind of test can be "justified" is if the information is spoon fed to the students and they are not encouraged to explore and learn what "other answers" are out there (for fear of learning the "wrong" answer). This type of test encourages memorization, not learning.

The key exception to my "one right answer" bias is a well written multiple choice test where each of the choices could be easily mistaken for a valid answer by someone who doesn't know the subject intimately.

[. . . steps off soap box]


Marc


PS: Here's a link about Niels Bohr (it doesn't mention my story, but gives some decent info about him).

http://www-gap.dcs.st-and.ac.uk/~history/Mathematicians/Bohr_Niels.html

 

[kalle]

Niels Bohr

Here is the story (or legend):

http://www.mentors.ca/bohr.html

 

[Allen Nelson]

And here is the TRUE story:

SNOPES investigation:
http://www.snopes.com/college/exam/barometr.htm
(note that in the story that SNOPES has, Bohr is the student, not the arbitrator.

A reprint of the "Saturday Review" story - the supposed first-hand account.
http://www.lhup.edu/~dsimanek/angelpin.htm

---

Much of the discussion here has been on the concept of "How do you test the material". That's why I focused on the term (used by the HNC criteria) "EVALUATE".

The current method of evaluating is to not tell the students anything, and have them come up with the definitions of U,M & RM themselves.

What I proposed is to give them the definition, and then see if they can use those definition in real-life.

If you were to test them on "Write a flip-flop program" - there is no single right answer. But there are answers that work, and answers that don't. Some disciplines have a single correct answer (math & physics, usually, the barometer question notwithsanding). Other subjects don't (anything that uses the word "Best" - "best" is subjective - my "best" is not necessarily yours; also any test question with a "Why")

---

Looking at the HNC Assessment criteria, about the only question we haven't gotten here, that I recall, is evaluate methods of programming programmable logic controllers.

It's just a matter of time.

 

[Antipathy]

Marc,
I'm not a huge fan of 'one-right-answer' exams but I also dislike muliple choice exams. They are simply too open to lucky guesses. Especially semi educated guesses.

It appears that there are pros and cons with any method of examining a students credibility.
I haven't done a HNC course so I don't know if the grades are exclusively exam driven, but it wouldn't surprise me if they are. My preference would be for a portion of the grade to be based on continual assessment by the lecturers and a portion based on the exams. That way, the students can't mess about all term (semester) and then rely on helpful fora such as here to help them out.

As for this Bohr character. If this was true of him (questionable).
Alright, so he went on to become one of the worlds most brilliant mathematicians... I'd still've smacked his arse (corporal punishment was still OK back then, wasn't it?) and made him sit through another year. He should have known better than give a frivolous answer.
Perhaps if the exam was prefaced with a requirement for the answers to demonstrate an understanding of the subject matter, he would have thought twice.

Anthony