ATAR Notes: Forum

VCE Stuff => Victorian Education Discussion => Topic started by: Mellyboo on November 10, 2016, 12:39:50 am

Title: Subject scaling - smallish study
Post by: Mellyboo on November 10, 2016, 12:39:50 am
Hey all. I am doing a smallish study (usually less than 100 students) called Laboratory skills. I noticed that in 2013 the scale up was approximately +3.2 at 40 raw ss and in 2014 it was +2 at 40 raw ss. Now the weird thing is in 2015 it scaled DOWN -1 at 40 raw. I am really not understanding this trend since the exam difficulty from 2013-2015 stayed pretty much consistent. From what I understand, this meant that the students who achieved a raw study score of 30 in this subject must have done worse in their other subjects for this to scale down. Is this correct? And what could have caused such a dramatic +3.2 to -1 scale at 40 raw ss in just 3 years? Do you think someone noticed this and thought, hold on, thats too much of a scaleup /there was a calculation of scaleup error in the previous years or is everything completely computerised? Thanks all !

The data:
VCE VET Laboratory Skills   40 (raw study score)    39 (2015)   42 (2014)   43 (2013)
No scaling data before 2013 available.
Title: Re: Subject scaling - smallish study
Post by: Mellyboo on November 10, 2016, 12:44:58 pm
Bump: anyone?
Title: Re: Subject scaling - smallish study
Post by: Maths Forever on November 10, 2016, 03:27:51 pm
Hi Mellyboo,

I studied Laboratory Skills three years ago. You're right. The scaling is purely to do with the scores that students achieve in their other subjects (I think this is only based on the subjects that students have completed so far in their VCE (if they are in year 11), but I could be wrong).

This is very random, because last year only 29 students completed the subject. Before 2013, there were even less students, so no scaling data would have been accurate.

It is even harder to predict because in my year, the class was mostly comprised of year 11 students. Since a lot of these students only completed one or no other subjects in that year, this makes it even more unpredictable in terms of scaling.

Last year, I believe only 3 raw scores over 40 were given out, the highest of which was a 45. This is because a raw score over 40 is roughly 9% of the cohort (about 2.61 in a class of 29 students).

In short, it varies strongly from year to year and is not really based on exam difficulty at all.

Good luck for the exam next week!

Kind Regards,
Maths Forever
Title: Re: Subject scaling - smallish study
Post by: Mellyboo on November 11, 2016, 01:27:54 am
Thank you!! Hopefully those in my cohort do well in their other subjects then ! :D