UAI and Percentile (1 Viewer)

[Plebeian]

Two questions:

1 - How does the Board compare you to your Year 10 cohort? Is this just an estimate of what the Year 10 people would have got, based on SC results or something? Or do they assume that people who left in Year 10 lie along the same distribution of good--> poor results as the Year 12 cohort?

2 - Table A7 of the 2003 UAC scaling report (here, pg64) shows the number of people on each UAI bracket of 1 (ie 99.95 to 99.00 or 84.95 to 84.00). The number seems to trend downwards as UAI decreases.
Some examples are 838 in 99 bracket, 794 in 91, 725 in 73 and 621 in 60.
Why is this?

 

[Survivor39]

1. Your uai is ranked against all your year 10 cohort, whether they have left, or not obtaining a UAI. The ones that had left, they are given a uai of 30 or less (making year 12 students who did the exam extremely difficult to obtain this rank, unless they get 0 for every exam). The bottom of the group. The ones that don't want a UAI will still be competing with you, except they won't 'see' their UAI.

2. No idea. It's a distribution thing I guess.

 

[Xayma]

They aren't given a UAI of 30 or less, they look at the SC results and the basic trend between SC results of those who completed the HSC and those who didn't.

Which I think is due to those differences, since a few of those 91's would drop out and alot more of the 60's.

 

[Lazarus]

1 - How does the Board compare you to your Year 10 cohort? Is this just an estimate of what the Year 10 people would have got, based on SC results or something? Or do they assume that people who left in Year 10 lie along the same distribution of good--> poor results as the Year 12 cohort?

This process is explained in the UAI Technical Report (different to the scaling reports). It is not an estimate, and does not involve making assumptions.

You can use the SC examinations to rank all students in the state comprising a particular age cohort - the HSC cohort for a particular year is only a subset of the total age cohort. Until recently, the SC only involved three compulsory exams, and so the maximum SC aggregate was 300.

Because the students who undertook the SC comprise the entire age cohort, including those who went on to do the HSC, the SC aggregates can be used to establish a set of equivalences between the total age cohort and the smaller HSC cohort.

For example, students scoring a SC aggregate of 240 out of 300 might be placed at the 90th percentile in the smaller HSC cohort but at the 94th percentile in the total age cohort.

This allows a TER of 90 (rank in the HSC cohort) to be transformed to a UAI of 94 (rank in the total age cohort).

Obviously, the SC aggregates of individual students don't play any role in this determination - the overall distributions of SC and HSC aggregates are used to determine the equivalences.

I'd strongly recommend reading the UAI Technical Report if you want further details on this.

2 - Table A7 of the 2003 UAC scaling report (here, pg64) shows the number of people on each UAI bracket of 1 (ie 99.95 to 99.00 or 84.95 to 84.00). The number seems to trend downwards as UAI decreases.
Some examples are 838 in 99 bracket, 794 in 91, 725 in 73 and 621 in 60.
Why is this?

The following graph (taken from p8 of the Technical Report) compares the SC aggregates of students who went on to complete Yr 12 and the HSC to those who finished school in Yr 10. Don't worry that it's from a few years ago - the distributions won't have changed significantly.


<center><img src="http://www.boredofstudies.org/other/uai_cohort.jpg"></center>


You can see that there is a high correlation between students who scored highly in the SC based on their SC aggregate and those who went on to complete Yr 12 and the HSC.

This means that when HSC students are ranked amongst their entire age cohort, they will be allocated the vast majority of the 'top' UAIs. As you move down the UAI rankings, more and more of them will be allocated to SC students rather than HSC students. This is why you see the general decrease in Table A7 of the 2003 Scaling Report.

 

[jm1234567890]

it also shows that the people get dumber as they get older

or the HSC was harder

 

[matt_f64]

it also shows that the people get dumber as they get older

or the HSC was harder

i think that a simple explanation for this is the abundance of 18th birthday parties in year 12.

 

[Plebeian]

Aha, it all makes sense ...thanks for replies (especially Lazarus)

 

[Manan]

Hi Everyone,

I came across this forum about a week ago, when researching extension2 english and have since joined. I just started year 12 and have got some great info from here. I have one quick question that I've asked so many people it's not funny...so please help

Is it better to do more than 10 units in the HSC, or does it not make a difference? In terms of scaling and UAI etc

Thanks a lot.

Cheers

 

[Plebeian]

This has been answered lots of times before, but:

Statistically people who do more units get better results. However, as the Board has often stated, it is very difficult to determine causality (ie. whether they're doing better because they do more units, or whether it's just that better students do more units).

IMO having a couple of units you can throw away if you bomb out in a test is good. If you have a full timetable, you are often more motivated to keep doing work as well - the plan that "I'll do 10 units and use the extra time to study" is probably fairly difficult to actually stick to. However, if you feel doing 12 units would be too much of an effort - especially if you're only doing 12 for the sake of it, and have picked up subjects you don't like / aren't good at it - then put the effort into ten.

I think there are other things that will make much more difference in the long run.

 

[Manan]

I see

That's what I thought-Im doing 11 units at the moment and have heaps of frees but that's cool....I do get some work done...I was just reading over some stats and noticed heaps of people who do extension 2 english etc do >10 units.

Anyways, thanks heaps.

Cheers