Job Discussion Forums

[basiltherat]

Hi there
I am in the early stages of writing new tests, both mid term and final, for the Headway series because the contents of the original, published tests have been gradually (over time) leaked to future test takers at the company where I train.
I am considering designing multiple choice tests but am reluctant to always do the standard 'only-one-right-answer' type because trainees can sometimes find the right one just by process of elimination or simply guessing where the odds can be as good as 3 to 1. So, I am thinking about designing the test where there might be 2 or 3 or no right answer from the options given. In order to get a mark, they must check all (or , as the case may be, none) of the correct options possible and avoid checking those options which would be incorrect.

Would you consider this too difficult for or unfair to trainees ? Appreciate your views.
regards
basil

PS i'm sure there is a name for this type of MC test but for the life of me I can't think what it is.

[Stephen Jones]

If the main thing you want to test is the students' ability to understand test instructions then go ahead with it. If you want to test anything else forget it.

I fail to see what you have against reaching the answer by a process of elimination. After all they need to have the knowledge to do the elimination in the first place.

[moonraven]

I don't see why you feel it's necessary to give multiple choice tests. Why not give one that allows them to do some writing? (It'll be much harder to cheat on that one, too.)

[basiltherat]

I was actually considering a writing component, too. That would be an additional section to the 60 or so in the multiple choice section.
regards
basil

[lajzar]

The whole point of a multiple choice test is that there is a single correct answer. This logically implies that it should be possibel to find teh answer by a process of elimination.

I believe the Cambridge PET and KET exams have a variety of multiple choice question (and other) formats. You could do worse than look there for inspiration.

[Nauczyciel]

Who's next to reply to this post? I'm waiting!

[merlin]

unfair, yes.
You're taking a well-known format and changing it - this is what Stephen was hinting at. More than half your students woudn't read the instructions and would fail as a result. So it would erally be one of those "gotcha" tests where you are testing the students' testing habits - do they read the directions or not.

It's a lot like the one where they ahve to fill in the form that says "please read all instructions before you enter any data" and then at the end of the form it says "Do not write anything on any paper. return all forms blank."

It is, however fair to have some answers like:
a)a cat
b)a dog
c)a lizzard
d)both a and b
e)both a and c
f)none of the above

[basiltherat]

some constructive views. thnx
regards
basil

[juststeven]

Mr. Basil,
Check with your administrators! If your students are used to easy 'multiple guess' questions and 'filling in the blank' you will have a problem. I'm sure you're a dedicated professional and want to test their knowledge. But, if too many fail your exam (or even one in some places in the ME) you KNOW whose head will roll! IMO education in the ME is a 'bubble of illusion' and the best you can do is try to help those few students that really want to learn.

[leeroy]

Basil - forget the tests. I suggest a kind of "battle-royale", where the students have to compete against each other in a grand tournament (perhaps in some kind of roman gladiatorial arena?) to win your favour.

If not - I suppose a more open form of assessment is doable - albeit inherently more subjective in terms of assessment than simple multiple choice questions.

Each Headway unit has certain lexical and grammatical objectives, right? I presume these will be (primarily) what you are assessing. If this is the case - then instead of explicitly referring to each (like "Fill the gaps with an appropriate preposition", or, "choose the correct verb from a, b, c and d") - simply ask the students to write about what they have learnt from the particular chapter(s) in question. Perhaps have a score from 1-3 for each language point they correctly recollect studying - the score depending on the depth and accuracy that they refer to it.

That's just an idea of course - I've never actually done it! To be honest, I loath formal testing...

[Glenski]