Ranking and Final Mark in Exam! (1 Viewer)

[chiefpasco]

Yo

He's a scenario I am in for the upcoming Bio exam.

I'm ranked 2nd, and the person ranked 1st is ahead of me by probably 7 marks from our school based assessments in Biology.

In the HSC exams, if I were to say, get a raw mark of 88 (then of course this will most likely allign to a B6), and the person who was ranked ahead of me got an raw exam mark of say, 88 as well. Will both of our final HSC Marks be the same? Or do they do voodoo bullshit and chop me down because my school based assessment had a gap between myself and this person.

churr

 

[DatAtarLyfe]

Mmm, i'm not too sure about how they account for the gap, but from what i know it seems like you'll both get the same. I think the gap is used if there are discrepancies i.e. someone stuffed up bad in external

Don't quote me though, i'm sure someone else can give a more definitive answer

 

[chiefpasco]

Mmm, i'm not too sure about how they account for the gap, but from what i know it seems like you'll both get the same. I think the gap is used if there are discrepancies i.e. someone stuffed up bad in external

Don't quote me though, i'm sure someone else can give a more definitive answer

i feel like that's the case too - i hope.

 

[aoc]

I think since you got the same HSC mark, your HSC mark would be the same, but his internal mark will be higher than yours

 

[KingOfActing]

That's incorrect, I'm afraid.

Your internal mark doesn't depend on other people individually, but on the cohort as a whole.

An example of a class of marks, in the form of Actual Internal Mark/Reported Internal Mark/External/HSC

95 - 88 - 88 || 88
87 - 84 - 88 ||86
82 - 82 - 80 ||81
73 - 76 - 75 ||76
60 - 63 - 63 ||63

Note this doesn't take scaling into account, but that's basically how it works. Your internal mark will be lower than first place, unless your cohort as a whole brings up the total marks by a lot.

 

[Zoinked]

Your internal mark will be anywhere from 6-8% below his most likely, as u said hes beating u by 7 marks, but we don't know if thats 7%, i.e his school mark is 100 and yours is 93, or he's beating u by 10%ish, i.e your average is 70%, his is 77%. I would say that say you both got 88 in the exam, that probably comes out to about 93 external. He would get 93 external and overall mark. You would get most likely (93+86)/2 which is 89.5 and rounds to 90. This is based off a lot of assumptions including if he is beating you by 7, what raw marks align to this year, etc.

 

[KingOfActing]

Your internal mark will be anywhere from 6-8% below his most likely, as u said hes beating u by 7 marks, but we don't know if thats 7%, i.e his school mark is 100 and yours is 93, or he's beating u by 10%ish, i.e your average is 70%, his is 77%. I would say that say you both got 88 in the exam, that probably comes out to about 93 external. He would get 93 external and overall mark. You would get most likely (93+86)/2 which is 89.5 and rounds to 90. This is based off a lot of assumptions including if he is beating you by 7, what raw marks align to this year, etc.

That's actually not quite right either, but close. Your relative difference to other people isn't preserved in moderation due it the moderation curve being quadratic, and not linear. Someone who is 50% behind first could go to being 10% behind first - it depends on the performance of the cohort as a whole.

 

[pikachu975]

That's actually not quite right either, but close. Your relative difference to other people isn't preserved in moderation due it the moderation curve being quadratic, and not linear. Someone who is 50% behind first could go to being 10% behind first - it depends on the performance of the cohort as a whole.

Yeah and most schools submit the marks really close on purpose, preserving ranks, to make sure there isn't a giant gap.

 

[trecex1]

Yeah and most schools submit the marks really close on purpose, preserving ranks, to make sure there isn't a giant gap.

What?? You realise the difference between the ranks matter a lot right? CHeck the example on the bostes website where they explain this, the relative gaps between people are preserved after moderation.
Otherwise you're saying schools only sending in the list of ranks without the marks would make zero difference

 

[trecex1]

That's actually not quite right either, but close. Your relative difference to other people isn't preserved in moderation due it the moderation curve being quadratic, and not linear. Someone who is 50% behind first could go to being 10% behind first - it depends on the performance of the cohort as a whole.

Can you link where your claim is from?
If you check http://www.boardofstudies.nsw.edu.au/hsc-results/moderation.html and scroll down to the 'relative difference (gaps)' they clearly say the relative gaps(after moderating) are similar to the relative gaps before moderation.

 

[A1P]

and the person who was ranked ahead of me got an raw exam mark of say, 88 as well. Will both of our final HSC Marks be the same?

Your HSC mark is half external half internal (after moderation). Same external mark but 7 internal marks behind you can't hope to get the same HSC mark. For that you would need to outscore 1st by around 7 external marks.

If this were between 2nd & 3rd your HSC mark would be 3-4 marks lower than that person's, since both are moderated up or down by roughly the same factor. Between 1st & 2nd difficult to predict because 1st gets the top exam mark for internal (which could be higher or lower than his internal) while 2nd gets moderated, could be up or down, by a large or small factor, depending on the cohort's external average vs internal average. Too many variables to predict.

 

[chiefpasco]

Your HSC mark is half external half internal (after moderation). Same external mark but 7 internal marks behind you can't hope to get the same HSC mark. For that you would need to outscore 1st by around 7 external marks.

If this were between 2nd & 3rd your HSC mark would be 3-4 marks lower than that person's, since both are moderated up or down by roughly the same factor. Between 1st & 2nd difficult to predict because 1st gets the top exam mark for internal (which could be higher or lower than his internal) while 2nd gets moderated, could be up or down, by a large or small factor, depending on the cohort's external average vs internal average. Too many variables to predict.

i just want a band 6 lol. 1st rank and myself are very strong in this subject, but the rest are pretty average - so basically, you're all telling me, my internal mark may be heavily moderated down? if my cohort behind me does poorly?

 

[A1P]

1st rank and myself are very strong in this subject, but the rest are pretty average - so basically, you're all telling me, my internal mark may be heavily moderated down? if my cohort behind me does poorly?

Say the cohort behind you does poorly in the external exam, but if they did even more poorly in the internal marks then they & you all get moderated UP. As said it depends on the cohort's external average vs internal average.

 

[chiefpasco]

cool, thanks for the info!!

 

[KingOfActing]

Can you link where your claim is from?
If you check http://www.boardofstudies.nsw.edu.au/hsc-results/moderation.html and scroll down to the 'relative difference (gaps)' they clearly say the relative gaps(after moderating) are similar to the relative gaps before moderation.

My source is the 20 page technical report on how moderation actually happens. I can pull it up later if you'd like. They're not wrong when they similar, but only in a non-technical sense. A "close" gap will remain "close" relative to the other gaps, but in reality it could become significantly bigger -- it's just that all the other gaps will become larger too.

 

[trecex1]

My source is the 20 page technical report on how moderation actually happens. I can pull it up later if you'd like. They're not wrong when they similar, but only in a non-technical sense. A "close" gap will remain "close" relative to the other gaps, but in reality it could become significantly bigger -- it's just that all the other gaps will become larger too.

Yeah please link it, can't find anything (do you mean the scaling report?).
And from how you worded your answer above, I'll take it you mean there are only a few exceptional cases the gap has significantly decreased/increased.
Though there's no way schools can simply alter all internal marks so the ranks are maintained, but the gaps decreased.

 

[KingOfActing]

Yeah please link it, can't find anything (do you mean the scaling report?).
And from how you worded your answer above, I'll take it you mean there are only a few exceptional cases the gap has significantly decreased/increased.
Though there's no way schools can simply alter all internal marks so the ranks are maintained, but the gaps decreased.

There you go! Of course, the cases where gaps change significantly are exceptional as they require a wildly different distribution of external marks to internal marks.

I think the important sentence is in the introduction: "The linear method also has the property of retaining the ratio of differences between pairs of Raw Assessments. The curved method does not do this. Within a localised region the curved method may approximate the linear method: to this extent the ratio of differences may approximately hold for adjacent pairs, but not for pairs widely separated on the Raw Assessments."

 

[pikachu975]

What?? You realise the difference between the ranks matter a lot right? CHeck the example on the bostes website where they explain this, the relative gaps between people are preserved after moderation.
Otherwise you're saying schools only sending in the list of ranks without the marks would make zero difference

I'm saying my school submits close ranks on purpose and a lot of schools do this

 

[KingOfActing]

I'm saying my school submits close ranks on purpose and a lot of schools do this

The ranks they submit don't actually matter though, because they'll get spread out.

My school does the opposite for some subjects - they submit first as "100" and last as "0" and spread everyone evenly in between, but it doesn't matter at all. So my "52" in English (which was really 70 something) is going to end up going to 80 something.

This is the exact goal of moderation - no school can "play" it to favour themselves. The only thing they can do, is make sure to not submit multiple people as first, because then what happens is if n people are tied first they get the average of the top n marks --- that way the mark can only really go down for first, so it's detrimental.