Secondary School Selection 2012

[ks2011]

Nebbermind:

But I thought the grades (A*, A etc) are allocated after the fitting into the bell curve and so the % should be roughly the same as the rest of the subjects. No meh?

Thats my understanding as well.

[Chenonceau]

Nebbermind:

But I thought the grades (A*, A etc) are allocated after the fitting into the bell curve and so the % should be roughly the same as the rest of the subjects. No meh?

It was a Minister who revealed that 80% of kids score A in Chinese. If this 80% is fact then the A are not fitted into bell curve. In essence, the A reflects raw score only. Uninformative. It would seem then that the A and B banding are stable 75 to 90 is A. No?

[Nebbermind]

The % for the rest of the subjects @ 43~44% seem to indicate that they are fitted into some models. But 80%...double!! Exempted from 'bell-curving'?

:scratchhead:

Some conspiracy theory? :scratchhead:

[wonderm]

Pen88n:

Getting a A or A* for Chinese does not guarantee the kid a high T-score for that subject. Imagine, if the mean is 80% and your kid score 81%, he will still get a A, but T-score for that subject is only 50.x even if it is a A. And yes, that means those with less than A will be scoring less than 50 for T-score for that subject. Sad and scary :nailbite: right???

You are right, A or A* does not represent high T-score. I also think even the A and A* marks are not fixed judging from the mistakes my ds told me about and their PSLE grades. At the end of the day, we just need to know T-score shows their relative position within their cohort, it doesn't really matter how hard are the papers. Don't be scared

[wonderm]

Chenonceau:

Nebbermind:

But I thought the grades (A*, A etc) are allocated after the fitting into the bell curve and so the % should be roughly the same as the rest of the subjects. No meh?

It was a Minister who revealed that 80% of kids score A in Chinese. If this 80% is fact then the A are not fitted into bell curve. In essence, the A reflects raw score only. Uninformative. It would seem then that the A and B banding are stable 75 to 90 is A. No?

I don't think the marks for A and A* are fixed. Despite the difference in difficulty levels of the papers over the years, the national % of A and A* for all subjects do not seem to change much. It is public information as it is shown on school websites every year.

http://www.nanhuapri.moe.edu.sg/cos/o.x?c=/wbn/pagetree&func=view&rid=32331

http://www.henryparkpri.moe.edu.sg/cos/o.x?c=/wbn/pagetree&func=view&rid=41039

Still, I don't know why Chinese has a much higher % of A and A*. In any case, T-score for each subject will still have a mean of 50.

[Chenonceau]

wonderm:

Chenonceau:

[quote=\"Nebbermind\"]But I thought the grades (A*, A etc) are allocated after the fitting into the bell curve and so the % should be roughly the same as the rest of the subjects. No meh?

It was a Minister who revealed that 80% of kids score A in Chinese. If this 80% is fact then the A are not fitted into bell curve. In essence, the A reflects raw score only. Uninformative. It would seem then that the A and B banding are stable 75 to 90 is A. No?

I don't think the marks for A and A* are fixed. Despite the difference in difficulty levels of the papers over the years, the national % of A and A* for all subjects do not seem to change much. It is public information as it is shown on school websites every year.

http://www.nanhuapri.moe.edu.sg/cos/o.x?c=/wbn/pagetree&func=view&rid=32331

http://www.henryparkpri.moe.edu.sg/cos/o.x?c=/wbn/pagetree&func=view&rid=41039

Still, I don't know why Chinese has a much higher % of A and A*. In any case, T-score for each subject will still have a mean of 50.[/quote]Yeah... I agree. Though still stumped by the proportion for Chinese.

[jtoh]

Maybe more students do better in Chinese compared to other subjects, hence the higher proportion of A/A*. But it also means that if you score an A (a low A particularly), your T-score will be dragged down lower compared to if you had scored a low A in the other three subjects.

[wonderm]

jtoh:

Maybe more students do better in Chinese compared to other subjects, hence the higher proportion of A/A*. But it also means that if you score an A (a low A particularly), your T-score will be dragged down lower compared to if you had scored a low A in the other three subjects.

I thought it is the other way round? For 2 bordreline A grades, the one with a smaller standard deviation will result in a higher T-score for that subject.

With more people scoring A and A*, the std deviation for that subject is smaller compared to another subject?

[Pen88n]

wonderm:

jtoh:

Maybe more students do better in Chinese compared to other subjects, hence the higher proportion of A/A*. But it also means that if you score an A (a low A particularly), your T-score will be dragged down lower compared to if you had scored a low A in the other three subjects.

I thought it is the other way round? For 2 bordreline A grades, the one with a smaller standard deviation will result in a higher T-score for that subject.

With more people scoring A and A*, the std deviation for that subject is smaller compared to another subject?

But the mean is much higher, which results in a small difference between the kid's score and the mean score. So even with a lower standard deviation, it still results in a lower T-score overall.

[wonderm]

Pen88n:

wonderm:

[quote=\"jtoh\"]Maybe more students do better in Chinese compared to other subjects, hence the higher proportion of A/A*. But it also means that if you score an A (a low A particularly), your T-score will be dragged down lower compared to if you had scored a low A in the other three subjects.

I thought it is the other way round? For 2 bordreline A grades, the one with a smaller standard deviation will result in a higher T-score for that subject.

With more people scoring A and A*, the std deviation for that subject is smaller compared to another subject?

But the mean is much higher, which results in a small difference between the kid's score and the mean score. So even with a lower standard deviation, it still results in a lower T-score overall.[/quote]Good point. But the fact that the mean is high means it is easier for the kid to score that particular score, right? I think we don't need to be overly worried about this stuff, just remember it is all about relative ranking.