Forecast COP for the top schools

Unlike entry to Primary Schools, admission into Secondary Schools is based on meritocracy. PSLE results are used as key admission criteria. Discuss everything related to PSLE and selection of Secondary Schools here.

Forecast COP for the top schools

Postby Hotspurs68 » Fri Dec 03, 2010 3:56 pm

Dear all,

I would be most grateful if you could help assist in forecasting the COP for ACS(I), RI, RGS, NYGH, MGS, SCGS and Dunmand High.

My estimates are as follows:

ACS(I) = 251
RI = 257
RGS = 258
NYGH = 262
MGS = 251
SCGS = 248
Dunman High = 250

What do you think?

Regards

Hotspurs68
OrangeBelt
OrangeBelt
 
Posts: 42
Joined: Tue Nov 30, 2010 7:46 pm
Total Likes: 0


Re: Forecast COP for the top schools

Postby livewith_vanilla » Fri Dec 03, 2010 7:25 pm

Hotspurs68 wrote:Dear all,

I would be most grateful if you could help assist in forecasting the COP for ACS(I), RI, RGS, NYGH, MGS, SCGS and Dunmand High.

My estimates are as follows:

ACS(I) = 251
RI = 257
RGS = 258
NYGH = 262
MGS = 251
SCGS = 248
Dunman High = 250

What do you think?

Regards


Unlikely for DHS to be at 250 .

livewith_vanilla
BlueBelt
BlueBelt
 
Posts: 415
Joined: Sat May 08, 2010 9:38 pm
Total Likes: 0


Re: Forecast COP for the top schools

Postby atutor2001 » Fri Dec 03, 2010 9:08 pm

Statistics is a measure of "normal" behaviour. Based on past statistics, the COP of all top schools will go up when the exam is easy.

If the COPs in this year for the top schools come down instead, the obvious statistical interpretation is that this cohort of students is different from previous years. That is, there is a bigger proportion of students who are not so strong in their studies. The top 5% did not "bunch together" like other years. Instead, their results spread out more when compared to past years.

The other possible interpretation is that last year's cohort is very good, getting very high COP despite the difficult paper. However, this is unlikely because last year's COPs are not very high as compared to other years.
Last edited by atutor2001 on Sat Dec 04, 2010 6:55 am, edited 1 time in total.

atutor2001
KiasuGrandMaster
KiasuGrandMaster
 
Posts: 1305
Joined: Thu Aug 27, 2009 11:08 pm
Total Likes: 6


Postby autolycus » Fri Dec 03, 2010 10:56 pm

In almost all cases the typical COP will not change much on average no matter whether the exam is harder or easier as long as the distribution remains normal. This is why, if the population remains about the same and the schools remain about the same, the COP will not vary by more than 1-2 points.

The only way it will not remain normal is if ALL the papers are very easy, so many students score full marks and the mean is high. Since it is not possible to score more than full marks, the distribution will be skewed high, as atutor2001 has said.

A drastic change in COP for a particular school can occur by specific market effects. For example, if school A suddenly launchs a highly desirable or attractive-looking programme B for elite students, then more top scorers will go there and the COP will rise. Or if school C suddenly has scandalous events with incompetent teachers, then nobody will want to go there and the COP will fall.

autolycus
BrownBelt
BrownBelt
 
Posts: 583
Joined: Mon Jan 04, 2010 9:11 pm
Total Likes: 1


Re: Forecast COP for the top schools

Postby Kiasu Friend » Sat Dec 04, 2010 8:10 pm

atutor2001 wrote: .......If the COPs in this year for the top schools come down instead, the obvious statistical interpretation is that this cohort of students is different from previous years. That is, there is a bigger proportion of students who are not so strong in their studies........


In my view, this year's cohort is no different from that of the previous years. Yet I too believe that the CoP will come down this year by at least 2 to 3 points when compared to last year. The reasons?

(1) This year the number of students appearing for PSLE is reported to be around 45,000. That is nearly 8% less than previous years (an unusual, yet significant fact). But the number of seats available in Secondary schools has not shrunk. This simply means that the competition will be less severe, leading to the fall of CoP.

(2) Most schools have reported an 'across-the-board' fall in scores. For the purpose of this analysis, I think it does not matter whether this fall in scores is due to the papers being easy or due to the cohort being less competent. All that matters is that an 'across-the-board' fall in scores will logically lead to a fall in CoP.

I hope you agree with this line of reasoning. Please post your comments.

Thanks to all.

Kiasu Friend
OrangeBelt
OrangeBelt
 
Posts: 53
Joined: Sun Aug 01, 2010 5:05 pm
Total Likes: 0



Postby autolycus » Sun Dec 05, 2010 1:09 am

If the distribution is the same but with a different mean, so that the fall in aggregate scores is indeed uniform across the board, then COP will drop but it won't matter because the lower COP would be as statistically common as the higher COP the previous year.

But since the population has been always about 42,000 and above, it is very likely a slightly skewed normal distribution, and the difficulty of entering any given school will remain very similar except where there are extreme circumstances for individual schools.

autolycus
BrownBelt
BrownBelt
 
Posts: 583
Joined: Mon Jan 04, 2010 9:11 pm
Total Likes: 1


Postby Muffins » Sun Dec 05, 2010 1:43 am

Hi guys, the COP for ACS(I)I last year was 251, and if this PSLE was tougher than the last year one, then the COP would be less....

:offtopic: 2 more posts to GMII!!!!!

Muffins
KiasuGrandMaster
KiasuGrandMaster
 
Posts: 2193
Joined: Sat Aug 22, 2009 10:28 am
Total Likes: 0


Postby Kiasu Friend » Sun Dec 05, 2010 8:36 am

autolycus wrote:If the distribution is the same but with a different mean, so that the fall in aggregate scores is indeed uniform across the board, then COP will drop but it won't matter because the lower COP would be as statistically common as the higher COP the previous year.

But since the population has been always about 42,000 and above, it is very likely a slightly skewed normal distribution, and the difficulty of entering any given school will remain very similar except where there are extreme circumstances for individual schools.


As usual, autolycus has given the perfect 'finishing touch'.

I agree with the conclusion - "the difficulty of entering any given school will remain very similar" although the COP is likely be lower for a majority of schools.

Kiasu Friend
OrangeBelt
OrangeBelt
 
Posts: 53
Joined: Sun Aug 01, 2010 5:05 pm
Total Likes: 0


Postby aqeqiq » Sun Dec 05, 2010 10:16 am

Found the following info in the SCGS Primary Homepage for affiliated students of SCGS:

COP for 2010 P6 students posted to S1 in 2011 - 210
COP for 2011 P6 students posted to S1 in 2012 - 220

Think the IP programme has caused the COP to increase but it will be from 2011 onwards for SCGS. I suppose it is to be expected. Question is from which year they will effect the increase.
aqeqiq
BlueBelt
BlueBelt
 
Posts: 271
Joined: Mon Oct 11, 2010 5:18 pm
Total Likes: 0


Re: Forecast COP for the top schools

Postby atutor2001 » Sun Dec 05, 2010 11:28 am

Kiasu Friend wrote:
atutor2001 wrote: .......If the COPs in this year for the top schools come down instead, the obvious statistical interpretation is that this cohort of students is different from previous years. That is, there is a bigger proportion of students who are not so strong in their studies........


In my view, this year's cohort is no different from that of the previous years. Yet I too believe that the CoP will come down this year by at least 2 to 3 points when compared to last year. The reasons?

(1) This year the number of students appearing for PSLE is reported to be around 45,000. That is nearly 8% less than previous years (an unusual, yet significant fact). But the number of seats available in Secondary schools has not shrunk. This simply means that the competition will be less severe, leading to the fall of CoP.

(2) Most schools have reported an 'across-the-board' fall in scores. For the purpose of this analysis, I think it does not matter whether this fall in scores is due to the papers being easy or due to the cohort being less competent. All that matters is that an 'across-the-board' fall in scores will logically lead to a fall in CoP.

I hope you agree with this line of reasoning. Please post your comments.

Thanks to all.


Hi Kiasu Friend

I agree with your analysis. However, just to clarify, my prediction for COP to go up is only for the TOP Schools i.e. RI, RGS, DH, NY, ... (because of skewing of the cure to the right) As for the rest of the schools, especially those with lower COP, I share your view.

Like what autolycus have said, whatever the COP may be, the level of competition for entry to the different schools will remain the same as any other years. However, the big difference is that the poor parents, armed with only the aggregate T-score and the COP for last year, are having a hard time trying to predict their kids chance of getting into their dream schools.

Perhaps MOE can be more open by also providing the "national position" besides the T-score. (Maybe not the exact position but just an indication e.g. "within top 500 or 1000, top 5000 etc). Then we know clearly where our kids stand. No need to feel bad that such a move is an over-emphasis on competition. Parents already know and accepted the fact that entry is by merit. IMHO, it is less stressful for parents to know the truth and put in a realistic option then cracking their head and praying for the best to happen. The same nightmare and frustration have been repeated year after year. I feel we should be open and mature enough to accept such ranking by "positions".

atutor2001
KiasuGrandMaster
KiasuGrandMaster
 
Posts: 1305
Joined: Thu Aug 27, 2009 11:08 pm
Total Likes: 6


Next

Return to Secondary Schools - Selection