Secondary School Selection 2012

[alng]

mummy so kiasu:

Same here. My son has lot of homework to do & we have to discourage to read his novel or he will have to go to bed at 12am.

janet_lee88:

My son has tons of homework to do...but still can't finish at 1145pm.

You mean your sons start to do homework from afternooon/evening till 11 plus 12 am? Wow, this is serious. I would suggest talking to the teachers. I talked to the teachers at the start of term 3 and they agreed not to assign homework on days where the students have supplementary classes. And they should allow more time (eg 2 days) to complete the homework. So far, I am happy with the homework assignment. Usually, the homework can be completed within 2-3 hours on non-supplementary-classes days.

[BeContented]

verykiasu2010:

BeContented:

Paper tough or not, isn't outcome the same? Vacancies remains. Bell curve also same. Child is graded against whole cohort, his position around same whether paper easy or difficult.....cos whole cohort move together. Or my thinking too simplified?

tough papers will \"benefit\" the high ability kids more as they can do more questions correctly and hence raw score is higher than peers, while the moderate performer will have lower raw score, and so in statistics, the standard deviation is higher, giving rise to higher t-score for the higher raw score, as the mean is lower plus higher standard deviation

when the paper is easy, both the moderate performers and high ability will have their raw scores bunched closely, as everyone could do similar easy questions; therefore the standard deviation is smaller and the t-score is lowered as the mean is higher than in a difficult paper

mean = t-score of 50, as a base, regardless what is the absolute number of the mean

as an example, if everyone scores 100% in the paper, mean = 100, standard deviation is zero because there is zero deviation from anyone, hence the t-score for that subject is 50.

if everyone score 100% in every of the 4 subjects, then everyone will get a t-score of 200 (4x50) for a perfect result in PSLE for everyone!

but in a cohort of 40plus thousand kids, it is impossible that everyone will score the same in every subject, hence the law of normal distribution sets in. because of this it is meaningless to compare the t-score of last year against the t-score of this year. Each year's t-score is a ranking exercise in itself. last year's 275 for easy paper could be having a much higher raw score than this this year's 275 of tough papers. hence it is observed that in a year of easy paper, the top psle score is lower while in year of tough papers, top score tend toward 28xhigh to 29xsomething

PSLE t-score is for ranking purpose and nothing else

Thanks for the clear explanation

Since each year's t-score is a ranking exercise in itself....the impact is really confined to students in that particular year right? Generally, the HA will already be getting better t-scores....so the number just get bigger with tougher paper. Other than the 'boosted t-score', any extra benefit? Will these higher t-score deprive others of any other things? If not, tough or easy paper again should not matter that much cos' the relative position will remain. PSLE t-score is almost useless once these kids enter Secondary school.

The only direct impact to me.....the higher the t-score, the more I can brag....haha :rotflmao: :rotflmao:

[verykiasu2010]

BeContented:

Since each year's t-score is a ranking exercise in itself....the impact is really confined to students in that particular year right? [YES] Generally, the HA will already be getting better t-scores....so the number just get bigger with tougher paper. Other than the 'boosted t-score', any extra benefit? [No, but more pride may be] Will these higher t-score deprive others of any other things? If not, tough or easy paper again should not matter that much cos' the relative position will remain. [YES] PSLE t-score is almost useless once these kids enter Secondary school. [YES, uesless]

The only direct impact to me.....the higher the t-score, the more I can brag....haha :rotflmao: :rotflmao:

you got it !

[jtoh]

BeContented:

PSLE t-score is almost useless once these kids enter Secondary school.


The only direct impact to me.....the higher the t-score, the more I can brag....haha :rotflmao: :rotflmao:

Yes! Bragging rights only.

Anyway I find that in secondary school, lower sec especially, those who are strong in languages and humanities will do better than those who are strong in math and science and weaker in languages. That's bec in lower sec you have English, Mother Tongue, History, Geog and Lit making up 5 subjects while Math and Science account for only 2 subjects. But that will change in upper sec when students get to choose their subject combinations.

[wonderm]

BeContented:

verykiasu2010:

[quote=\"BeContented\"]Paper tough or not, isn't outcome the same? Vacancies remains. Bell curve also same. Child is graded against whole cohort, his position around same whether paper easy or difficult.....cos whole cohort move together. Or my thinking too simplified?

tough papers will \"benefit\" the high ability kids more as they can do more questions correctly and hence raw score is higher than peers, while the moderate performer will have lower raw score, and so in statistics, the standard deviation is higher, giving rise to higher t-score for the higher raw score, as the mean is lower plus higher standard deviation

when the paper is easy, both the moderate performers and high ability will have their raw scores bunched closely, as everyone could do similar easy questions; therefore the standard deviation is smaller and the t-score is lowered as the mean is higher than in a difficult paper

mean = t-score of 50, as a base, regardless what is the absolute number of the mean

as an example, if everyone scores 100% in the paper, mean = 100, standard deviation is zero because there is zero deviation from anyone, hence the t-score for that subject is 50.

if everyone score 100% in every of the 4 subjects, then everyone will get a t-score of 200 (4x50) for a perfect result in PSLE for everyone!

but in a cohort of 40plus thousand kids, it is impossible that everyone will score the same in every subject, hence the law of normal distribution sets in. because of this it is meaningless to compare the t-score of last year against the t-score of this year. Each year's t-score is a ranking exercise in itself. last year's 275 for easy paper could be having a much higher raw score than this this year's 275 of tough papers. hence it is observed that in a year of easy paper, the top psle score is lower while in year of tough papers, top score tend toward 28xhigh to 29xsomething

PSLE t-score is for ranking purpose and nothing else

Thanks for the clear explanation

Since each year's t-score is a ranking exercise in itself....the impact is really confined to students in that particular year right? Generally, the HA will already be getting better t-scores....so the number just get bigger with tougher paper. Other than the 'boosted t-score', any extra benefit? Will these higher t-score deprive others of any other things? If not, tough or easy paper again should not matter that much cos' the relative position will remain. PSLE t-score is almost useless once these kids enter Secondary school.

The only direct impact to me.....the higher the t-score, the more I can brag....haha :rotflmao: :rotflmao:[/quote]I fully agree with VK2010's explanation. Just add on one minor implication, an easy paper favors those who are more mediculous and do not make careless mistakes. My ds1 used to be careless when he was in P6, so when he came out of his PSLE maths paper, and told me it was an easy paper and he finished it very fast, with more than 1 hour to spare. I almost fainted.

But such impact is very minor and should not affect the overall ranking too much. So you are right, I don't think it makes much difference whether the paper is easy or difficult.

[janet88]

If the paper is set ‘easy’, the child may let down his/her defense and be careless…a bit over-confident. But if paper is set difficult, child may also lose confidence as time is tight and yet not being able to move on smoothly.

[Oodles Mummy]

:goodpost:

verykiasu2010:

BeContented:

Paper tough or not, isn't outcome the same? Vacancies remains. Bell curve also same. Child is graded against whole cohort, his position around same whether paper easy or difficult.....cos whole cohort move together. Or my thinking too simplified?

tough papers will \"benefit\" the high ability kids more as they can do more questions correctly and hence raw score is higher than peers, while the moderate performer will have lower raw score, and so in statistics, the standard deviation is higher, giving rise to higher t-score for the higher raw score, as the mean is lower plus higher standard deviation

when the paper is easy, both the moderate performers and high ability will have their raw scores bunched closely, as everyone could do similar easy questions; therefore the standard deviation is smaller and the t-score is lowered as the mean is higher than in a difficult paper

mean = t-score of 50, as a base, regardless what is the absolute number of the mean

as an example, if everyone scores 100% in the paper, mean = 100, standard deviation is zero because there is zero deviation from anyone, hence the t-score for that subject is 50.

if everyone score 100% in every of the 4 subjects, then everyone will get a t-score of 200 (4x50) for a perfect result in PSLE for everyone!

but in a cohort of 40plus thousand kids, it is impossible that everyone will score the same in every subject, hence the law of normal distribution sets in. because of this it is meaningless to compare the t-score of last year against the t-score of this year. Each year's t-score is a ranking exercise in itself. last year's 275 for easy paper could be having a much higher raw score than this this year's 275 of tough papers. hence it is observed that in a year of easy paper, the top psle score is lower while in year of tough papers, top score tend toward 28xhigh to 29xsomething

PSLE t-score is for ranking purpose and nothing else

:goodpost:

[LindaYong]

Oodles Mummy:

:goodpost:

verykiasu2010:

[quote=\"BeContented\"]Paper tough or not, isn't outcome the same? Vacancies remains. Bell curve also same. Child is graded against whole cohort, his position around same whether paper easy or difficult.....cos whole cohort move together. Or my thinking too simplified?

tough papers will \"benefit\" the high ability kids more as they can do more questions correctly and hence raw score is higher than peers, while the moderate performer will have lower raw score, and so in statistics, the standard deviation is higher, giving rise to higher t-score for the higher raw score, as the mean is lower plus higher standard deviation

when the paper is easy, both the moderate performers and high ability will have their raw scores bunched closely, as everyone could do similar easy questions; therefore the standard deviation is smaller and the t-score is lowered as the mean is higher than in a difficult paper

mean = t-score of 50, as a base, regardless what is the absolute number of the mean

as an example, if everyone scores 100% in the paper, mean = 100, standard deviation is zero because there is zero deviation from anyone, hence the t-score for that subject is 50.

if everyone score 100% in every of the 4 subjects, then everyone will get a t-score of 200 (4x50) for a perfect result in PSLE for everyone!

but in a cohort of 40plus thousand kids, it is impossible that everyone will score the same in every subject, hence the law of normal distribution sets in. because of this it is meaningless to compare the t-score of last year against the t-score of this year. Each year's t-score is a ranking exercise in itself. last year's 275 for easy paper could be having a much higher raw score than this this year's 275 of tough papers. hence it is observed that in a year of easy paper, the top psle score is lower while in year of tough papers, top score tend toward 28xhigh to 29xsomething

PSLE t-score is for ranking purpose and nothing else

:goodpost:[/quote]I notice for the past years, the mean marks for MT is very high, about 80% of the them get A, thus could say that the mean is an A, or a high A. Whereas for the other 3 subjects the only about 40% of the student will get an A. Thus if your child score A for his MT, he could only get just a T-score of 50, whereas if he score A for the others 3 subjects, he will get a T-score higher than 50.

[jedamum]

LindaYong:

Oodles Mummy:

:goodpost: [quote=\"verykiasu2010\"]Paper tough or not, isn't outcome the same? Vacancies remains. Bell curve also same. Child is graded against whole cohort, his position around same whether paper easy or difficult.....cos whole cohort move together. Or my thinking ? simplified?

tough papers will \"benefit\" the high ability kids more as they can do more questions correctly and hence raw score is higher than peers, while the moderate performer will have lower raw score, and so in statistics, the standard deviation is higher, giving rise to higher t-score for the higher raw score, as the mean is lower plus higher standard deviation

when the paper is easy, both the moderate performers and high ability will have their raw scores bunched closely, as everyone could do similar easy questions; therefore the standard deviation is smaller and the t-score is lowered as the mean is higher than in a difficult pape

mean = t-score of 50, as a base, regardless what is the absolute number of the mean
IGIven
as an example, if everyone scores 100% in the paper, mean = 100, standard deviation is zero because there is zero deviation from anyone, hence the t-score for that subject is 50.

if everyone score 100% in every of the 4 subjects, then everyone will get a t-score of 200 (4x50) for a perfect result in PSLE for everyone!

but in a cohort of 40plus thousand kids, it is impossible that everyone will score the same in every subject, hence the law of normal distribution sets in. because of this it is meaningless to compare the t-score of last year against the t-score of this year. Each year's t-score is a ranking exercise in itself. last year's 275 for easy paper could be having a much higher raw score than this this year's 275 of tough papers. hence it is observed that in a year of easy paper, the top psle score is lower while in year of tough papers, top score tend toward 28xhigh to 29xsomething

PSLE t-score is for ranking purpose and nothing else

:goodpost:]

I notice for the past years, the mean marks for MT is very high, about 80% of the them get A, thus could say that the mean is an A, or a high A. Whereas for the other 3 subjects the only about 40% of the student will get an A. Thus if your child score A for his MT, he could only get just a T-score of 50, whereas if he score A for the others 3 subjects, he will get a T-score higher than 50.[/quote]Given limited resources, would you then advise a kid who is weak in MT to focus more on it, or to focus on getting A in another subject like science or math?

[PiggyLalala]

jedamum:

Given limited resources, would you then advise a kid who is weak in MT to focus more on it, or to focus on getting A in another subject like science or math?

Though the mean is high for chinese, I believe the standard deviation is low because the range of marks for chinese is not too big. So it is really very difficult to say which subject to focus more.

Luck plays an important part in any exam. The best scenario would be the subject that your child is good in is the difficult paper for that year so that while others can't solve the difficult sums, your child is able to solve; and that his weaker subject turns out to be a easy paper so that he, like the others, can solve most of the questions.

Btw, there is still time. You can still work on your ds chinese. I think reading chinese story books help. When they read more chinese, they can write better and maybe understand the comprehension passage better too.

:rahrah: :rahrah: :rahrah: all the best to our ds in their PSLE 2014.