Average mark for each year's exam

[Gentoo]

Hey guys, sorry if this has already been done but I did a quick calculation of the average for each year's exam for anybody curious (I've noticed some 'which years' exams were the hardest' questions have popped up on the boards which hopefully this can shed some objective light on (to the extent that the average mark represents 'difficulty' in the sense people use the term - it doesn't necessarily indicate the % of marks at the highest or lowest end of the spectrum for example, which is information we cannot glean from looking at the info given, only the average). I added up the averages for each question found in the exam report to obtain these (the averages for short answer questions are rounded to 1 decimal place though which *slightly* diminishes their accuracy - if you really wanted you could calculate the average mark for each question by calculating the mean manually for each individual question before adding them all up but that would take a very very long time).

2014
Exam 1: 23.2
Exam 2:

2013
Exam 1: 22.5
Exam 2: 38.97

2012
Exam 1: 22.5
Exam 2: 37.73

2011
Exam 1: 18.5
Exam 2: 38.67

2010
Exam 1: 21.8
Exam 2: 41* (median 41)

2009
Exam 1: 23.5
Exam 2: 44* (median 44)

2008
Exam 1: 23.7
Exam 2: 44* (median 45)

2007
Exam 1: 18.2
Exam 2: 39.5* (median 41)

2006
Exam 1: 22* (median 22)
Exam 2: 45.5* (median 46)

Notes: The * means that I've just gotten the average straight from the exam report which explicitly stated the mean mark (rounded to 1 or no decimal places) - VCAA stopped doing it years ago and didn't report it for some exam 1s even in the years they reported it for exam 2. I've also included the median for the years which VCAA reported it (which as I've mentioned before is impossible to calculate manually). Also, I haven't included the data from the Methods non-CAS exams from 2006-2009.

And if anybody wants to bet on the average mark for 2014 exam 2, now's your time.

[Floatzel98]

These exams are out of 40 and 80 right?

[Gentoo]

Yeah, exam 1 is out of 40 marks and exam 2 is out of 80 marks.

[EspoirTron]

Only a 2013er will understand the feels of reading time in the 2nd exam. Not only does the difficulty of the question count, but also the way in which it is worded.

[keltingmeith]

Only a 2013er will understand the feels of reading time in the 2nd exam. Not only does the difficulty of the question count, but also the way in which it is worded.

Supposedly 2012 was harder.
I demand a recount. >(

[dankfrank420]

Supposedly 2012 was harder.
I demand a recount. >(

Maybe you 2013 kiddies are just smarter. 

[Gentoo]

The scaling would indicate that. :p

I actually did recount 2012/2013 exam 2 just then - got the same for 2012 exam 2 but it appears I made an error initially for 2013 exam 2 and the average mark is actually higher - 38.97 (must have inputted a different MC answer for one of them). That is even higher than the 2011 one.

[keltingmeith]

The scaling would indicate that.

I actually did recount 2012/2013 exam 2 just then - got the same for 2012 exam 2 but it appears I made an error initially for 2013 exam 2 and the average mark is actually higher - 38.97 (must have inputted a different MC answer for one of them). That is even higher than the 2011 one.

LOL I was only joking about the recount, but that definitely shows me.

[Splash-Tackle-Flail]

So, wait, why did it scale so much in 2013 again? If the mean score is actually higher than usual? Is it standard deviation, or just people did better in other subjects? I don't see how it results in such a high scaling above 50.

[keltingmeith]

Scaling depends on the students other scores for that year, not the methods score in other years.

[Phy124]

The mean, standard deviation and median (by grade) are already listed for all three GAs (SACs, exam 1 & exam 2) in the grade distributions each year. They also show the "% of marks at the highest or lowest end of the spectrum" that you appear to be interested in. Here are the files for methods:

http://www.vcaa.vic.edu.au/Documents/statistics/2014/section3/vce_mathematical_methods_cas_ga14.pdf
http://www.vcaa.vic.edu.au/Documents/statistics/2013/section3/vce_mathematical_methods_cas_ga13.pdf
http://www.vcaa.vic.edu.au/Documents/statistics/2012/section3/vce_mathematical_methods_cas_ga12.pdf
http://www.vcaa.vic.edu.au/Documents/statistics/section2/vce_math_methods_cas_ga11.pdf
http://www.vcaa.vic.edu.au/Documents/statistics/section2/vce_math_methods_cas_ga10.pdf
http://www.vcaa.vic.edu.au/Documents/statistics/section2/vce_math_methods_cas_ga09.pdf
http://www.vcaa.vic.edu.au/Documents/statistics/section2/vce_math_methods_cas_ga08.pdf
http://www.vcaa.vic.edu.au/Documents/statistics/section2/vce_math_methods_cas_ga07.pdf
http://www.vcaa.vic.edu.au/Documents/statistics/section2/vce_math_methods_cas_ga06.pdf

Sorry if I've misinterpreted something here and this isn't what you won't, but I don't understand what it is you're doing

[Gentoo]

Oh wow, totally forgot about those LOL.

Are those the raw percentages/marks of the exams? Or have they already been adjusted (to make things easier for them statistically) at that point? I mean, for one thing, they're out of double the marks (which would indicate that? Or maybe not).

For instance, does 'A+' in the context of these reports even represent a raw score of 89-100%? Because every year after any VCAA exam for a given subject, people ask, "what's the A+ cutoff going to be for this year?" indicating that it changes from exam to exam and is NOT just 89% and above raw (it's 89%+ after adjustments). So could the mean/median in the report be subject to the same type of adjustment? Or are these figures all raw?

[Phy124]

Oh wow, totally forgot about those LOL.

Are those the raw percentages/marks of the exams? Or have they already been adjusted (to make things easier for them statistically) at that point? I mean, for one thing, they're out of double the marks (which would indicate that? Or maybe not).

For instance, does 'A+' in the context of these reports even represent a raw score of 89-100%? Because every year after any VCAA exam for a given subject, people ask, "what's the A+ cutoff going to be for this year?" indicating that it changes from exam to exam and is NOT just 89% and above raw (it's 89%+ after adjustments). So could the mean/median in the report be subject to the same type of adjustment? Or are these figures all raw?

All exams are marked by two different people so that is why exam 1 is out of 80 and exam 2 is out of 160.

The A+ cutoff is the lower mark in the range for A+, take for example the 2014 exam 1 where the A+ cutoff was 73/80 (or 36.5/40).

The cutoff does change from exam to exam because different exams have different difficulties, I'm not sure what you're referring to by adjustments. As far as I know students get a score on the exam, VCAA takes all the scores and then develops grade ranges based on the statistics that come from all these scores, then the student gets a grade range given to them from this.(Yes, they are "raw", AFAIK)

[Gentoo]

Yeah, so the act of getting an A+ isn't just a result of you getting 89%+ necessarily, rather how your score compares to what the totality of the cohort gets in their exams, otherwise the A+ cutoff would literally always be 72/80 and 144/160 (or whatever 89-100%, the supposed definition of an A+, corresponds to). So they're based on how the overall cohort scores and how their marks are distributed and are to this extent standardised, right? I think we're on the same page here lol. (sorry if my terminology is a bit sloppy)

Would the same process apply to the mean/median shown in those reports?