Scaling? (1 Viewer)

[deyveed]

Lets just say that almost everybody doing IT is very smart in computer stuff and they score really high even in the hard exams but they're crap at their other subjects.
This would make IT a low scaling subject right?
If i beat 90% of them, am i destined for some low scaling even though i beat heaps of smart people in a difficult subject?

 

[Lazarus]

Originally posted by deyveed
If i beat 90% of them, am i destined for some low scaling even though i beat heaps of smart people in a difficult subject?

No.

Two factors contribute to determine your scaled mark:

1) The scaling for the course (i.e. the scaled mean).
2) Your position in the course (e.g. the 90th percentile).

One can compensate for the other. Read the You and Your UAI guide published by the UAC.

 

[deyveed]

Yeah i understand but say that for IT, 90 percentile the scaling is still low because all the really smart computer nerd people do badly in their other subjects making IT an overall low scaling subject.

So even if i came 90 percentile, i'd still wont be getting enough scaling right?

 

[flyin']

deyveed, I think you're on the right track, which is why you have many people doing all the mainstream subjects like Maths Ext1, Ext2, Eng Ext1, Ext2, Phy, Chem, Econ, etc.

I agree that if you do a subject where the bulk of the people aren't comparatively good at other subjects (ie. just don't rank as well) and only well in that one subject, it's goin' to prove costly.

 

[Lazarus]

Originally posted by deyveed
So even if i came 90 percentile, i'd still wont be getting enough scaling right?

It depends on the course.

For Industrial Technology (using 2002 results), the 90th percentile results in a scaled mark of 29.8 - above average, but nothing special. The scaled mean is incredibly low (16.4), so you'd have to be right at the top to have a chance at a good scaled mark (39.7 in this case).

I feel that this wasn't the type of explanation you were after... so I'll stop here and take a different approach.

Originally posted by deyveed
If i beat 90% of them, am i destined for some low scaling even though i beat heaps of smart people in a difficult subject?

You wouldn't have beaten heaps of smart people in a difficult subject. The candidature is deemed to be of low ability (due to their lack of a good performance over all subjects), which is why the course scales low to begin with. And the difficulty of the subject is irrelevant.

Nope, that's not working either. You'll have to clarify your query.

 

[deyveed]

Thanks for the explanation Laz (sorry i wasn't clearer) but i think flyin' is a tad closer.

Originally posted by flyin'
I agree that if you do a subject where the bulk of the people aren't comparatively good at other subjects (ie. just don't rank as well) and only well in that one subject, it's goin' to prove costly.

Ok. In that case, how is it fair to the students in that subject?
Or is UAC assuming that if a person performs well in one subject, they should be performing just as well in their other subjects?

Anyone?

 

[walla]

I think your problem is that you are over-generalising WAY too much about all these unrecognised geniuses doing IT that can't do any other subjects. And in any case, if people can't perform in any but two units then there is simply no way that they can be deemed a high candidature and thus rewarded for their relative performance to each other in IT.

When it comes down to it, there is no fairer system than the one outlined by Laz unless you somehow "prove" that everyone in IT really is incredibly smart when it comes to computers and then somehow figure out how to make and exception for IT and distort the marks.

 

[rennidoodle]

The reason the scaling for IT is low is because there are 2 harder computer courses: IPT and SDD...

 

[Ragerunner]

Originally posted by rennidoodle
The reason the scaling for IT is low is because there are 2 harder computer courses: IPT and SDD...

That has nothing to do with it.

The reason it scales low because the candidture doing it isn't performing that well.