How ACER calculates UMAT scores

Hey all,

I'm new to all this UMAT stuff. Can anyone explain how the scoring goes? Like is it 2 points for one question?- and then do you just calculate the percentage for that section?

Thanks

ACER does not release their methods of calculating scores. The 'percentage' or the percentile for a certain section is supposed to represent where your raw score lies in the entire UMAT cohort (e.g. 95th percentile means that your score was in the top 5% of test-takers). Hope that helped.

Edit: You should really look through the thread before asking your questions though.

Thanks for answering my question. I'm not really talking about the percentile - I understand you can't calculate that by yourself but I mean like the actual scores you get in each section? And I have looked in other threads - but can't seem to find answers.

Originally Posted by ilovethomjonnyedcolinphil

Thanks for answering my question. I'm not really talking about the percentile - I understand you can't calculate that by yourself but I mean like the actual scores you get in each section? And I have looked in other threads - but can't seem to find answers.

The actual scores, you can't calculate either.

The scores you will never know. It's all top secret, all you need to know if you may need 50 in each section for certain universities. And you will only know that when ou get your results

Item Response Theory (IRT).

I think they use that, some ****ed up shit.

Originally Posted by bobthebuilder94

Item Response Theory (IRT).

I think they use that, some ****ed up shit.

It's not ****ed up at all. The theory behind it's actually IMO a lot sounder. You get more marks for answering a difficult question than you do for answering an easier question. And the results are comparable between years, so UMAT scores can be used more than once.

Of course you don't think its ****ed up, you did good in it :P haha

Originally Posted by bobthebuilder94

Of course you don't think its ****ed up, you did good in it :P haha

*rolls eyes*
I don't think IRT is a bad system, no. As to the validity of the UMAT (and that's the argument that you can claim I have bias in), that's a different question.

But the way the answers you mark down are converted to scores is a lot better than it would be if you just scored it out of how many answers you got right.

IRT is based on well researched and well established statistical/assessment theory. It's a highly sophisticated and valid technique, and is very widely used. Most high school examinations (in Aus and NZ) now use it to some extent when marking. In no way is it unique to UMAT.