Study Score Prediction Question Thread

[brenden]

Nope, other way around. Highest exam performance goes to the 1st rank's SACs.

[abcdqdxD]

So rank 1's exam peformance has no effect on everybody else?

[brenden]

So rank 1's exam peformance has no effect on everybody else?

It can effect people, yes?

[spectroscopy]

What would you get for methods with
 
Ranked around 100/120ish in a select entry cohort , low c's on the exams

[brenden]

Lol, why on Earth are you asking this a year and three quarters before your Methods exam?!

Close to 25.

[spectroscopy]

i need to know if i can realistically get a 25 or not LMFAO

[brenden]

i need to know if i can realistically get a 25 or not LMFAO

This makes me want to cut babies.

[Mao]

Highest exam performance goes to the 1st rank's SACs.

I thought that was a myth. Highest exam mark is adjusted approximately to the 1st rank, but not exactly. The placement for latter ranks is even more complicated. It's got something to do with stretching one distribution (SAC) to resemble another distribution (exam) by fitting particular parameters. I never found the precise algorithm they used when I was looking into it.

Is there a publication that actually shows the algorithm?

EDIT: found the algorithm. For a discussion, see here: Re: Guide to how Study Scores and ATARs are calculated!

YOU MUST PROVIDE US WITH: Your ranking. The strength of your cohort. If you don't know the strength, tell us your school's median study score, search here http://www.bettereducation.com.au/results/vce.aspx.. and your hopeful exam(s) mark(s). Conversely, don't worry about your school or ranking and provide is with your SAC GAs after scaling (as has been done in the first reply below.)

I have no idea how this can be quantitatively useful in your calculation of SS estimates. Knowing the mean SS of a cohort cannot tell you what the spread of SS is. That only has predictive power if the student in question falls near the average.

[brenden]

I have no idea how this can be quantitatively useful in your calculation of SS estimates. Knowing the mean SS of a cohort cannot tell you what the spread of SS is. That only has predictive power if the student in question falls near the average.

Yeah I recognise it isn't mathematical, it's more a reasonable guess. I'd assume that the top 5 students at a school with a 33 median will be better than a school with a 28 median just by likelihood even if that assumption isn't necessarily mathematically correct. It's just a weak replacement for an estimate on strength.

[abcdqdxD]

Is it still possible to get a 50 in Further if I'm rank 2 or rank 3 out of ~100 students in a medium cohort provided I full-mark both exams?

[Will Sparks]

Is it still possible to get a 50 in Further if I'm rank 2 or rank 3 out of ~100 students in a medium cohort provided I full-mark both exams?

Well for other subjects, that'd be a 100% yes, but of course it's still possible in Further but you may have to hope that the others above you drop like one or two marks while you actually get full marks.

[abcdqdxD]

Ah cheers. Anyone know what a 9/9/9 in the exam will get in English assuming mid A+ SACs?

[Yacoubb]

Biology:

90/100 for SACS in a weak cohort and 106/110 for the exam. Rank 1.

[Will Sparks]

Ah cheers. Anyone know what a 9/9/9 in the exam will get in English assuming mid A+ SACs?

Should be at least 45 easily, I'm not exactly sure though, cause some schools have easy as shit SACs although 9/9/9 on your English exam is an amazing score.

Biology:

90/100 for SACS in a weak cohort and 106/110 for the exam. Rank 1.

Rank 1 so don't worry about your cohort.

106/110 sounds like an amazing score but I don't do bio, so it may be a less difficult score to get in other subjects (say like Physics or Chem).

I'd say 45+ easily though

[casbanjo]

Physics: 190/200 for SACs. Rank 1. Average cohort.
Exam: 80/90.