<![CDATA[Bell Curve - To remove or not to remove?]]>I've come across a great blog and want to share it here:


http://hedgehogcomms.blogspot.sg/2012/10/piecing-together-psle-puzzle.html

The article has really made me to ponder over my views about PSLE results. As they are calculated based on the bell curve or t-score (your relative score), I'm more convinced that this is a fairer type of assessment since it takes into account other variables (e.g. difficulty of the papers). But witnessing so many children nowadays have been stressed, pressured and even gone hysterical, I wonder if it can help to de-stress the kids if the bell curve system is removed.

The blogger has pointed out:

\"Ranking fosters competition and discourages collaboration because it becomes a race where it's about outrunning your rivals rather than performing to the best of your ability. I'd rather score an A and have my schoolmates score Bs than score an A* along with everyone else.\"

I quite agree with him on this idea that this system is actually promoting an unhealthy competition.

For myself personally, I have been discouraged and depressed very much during my school years as despite how hard I kept trying, I seldom got the grades I wanted. I guess maybe a sense of achievement or a little of satisfaction could have made my school life better and more enjoyable!

]]>https://forum.kiasuparents.com/topic/42714/bell-curve-to-remove-or-not-to-removeRSS for NodeTue, 21 Apr 2026 07:54:39 GMTMon, 22 Oct 2012 16:13:03 GMT60<![CDATA[Reply to Bell Curve - To remove or not to remove? on Mon, 23 Sep 2013 17:20:10 GMT]]>The bell curve for national exams is here to stay and relevant.

]]>https://forum.kiasuparents.com/post/1101466https://forum.kiasuparents.com/post/1101466Mon, 23 Sep 2013 17:20:10 GMT<![CDATA[Reply to Bell Curve - To remove or not to remove? on Mon, 03 Dec 2012 05:09:56 GMT]]>

PreU_SGstudent:
Hi everyone! :scared:


I have been PSLE/ O levels/ A levels. To me, bell curve is essential to get into the place in higher institution. It is like a distinction of merit and your hardwork. My personal take is that, bell curve is used to differentiate A and B student and also C and D student. 

However, bell curve hits hard on A level students especially. In A levels, we are competing with the almightly. So therefore, it will be the survival of the fittest. 

Bell Curve on the other hand, helps in stronger students than the weakens which will hinder the gap even more. 

:hi5:

Yeah without the bell curve the whole of Singapore would score A1 for emath if the A1 was really at 75. .__. The bell curve pushes the A1 up to 90+ because of how easy the papers are.

]]>https://forum.kiasuparents.com/post/913222https://forum.kiasuparents.com/post/913222Mon, 03 Dec 2012 05:09:56 GMT<![CDATA[Reply to Bell Curve - To remove or not to remove? on Sat, 01 Dec 2012 05:42:16 GMT]]>Hi everyone! :scared:


I have been PSLE/ O levels/ A levels. To me, bell curve is essential to get into the place in higher institution. It is like a distinction of merit and your hardwork. My personal take is that, bell curve is used to differentiate A and B student and also C and D student. 

However, bell curve hits hard on A level students especially. In A levels, we are competing with the almightly. So therefore, it will be the survival of the fittest. 

Bell Curve on the other hand, helps in stronger students than the weakens which will hinder the gap even more. 

:hi5:

]]>https://forum.kiasuparents.com/post/912425https://forum.kiasuparents.com/post/912425Sat, 01 Dec 2012 05:42:16 GMT<![CDATA[Reply to Bell Curve - To remove or not to remove? on Fri, 30 Nov 2012 13:32:44 GMT]]>Dear AWSP; again appreciate your insight, Please keep posting and also make your point across to MOE, I believe you have a valid point! Since MOE is looking at reducing the stress level of PSLE, I am sure they will welcome any constructive suggestions.


On the side note, PSLE is one-and-all exam for majority students, except DSA CO /WL students, that is also a cruel reality to some students like my dd’s classmates. A few of them were constantly top of school, but their t score for PSLE were 25+, very unlikely to go to their dream schools. The luck factor during exam plays a part. So if MOE allows 40% P4-P6 academic performance and 60% PSLE score, that will greatly reduce the stress for one exam.

]]>https://forum.kiasuparents.com/post/912175https://forum.kiasuparents.com/post/912175Fri, 30 Nov 2012 13:32:44 GMT<![CDATA[Reply to Bell Curve - To remove or not to remove? on Thu, 29 Nov 2012 14:04:44 GMT]]>

1girl1boy:
:goodpost:

But I think your effort is falling on deaf ears. 😄
I suggest that we leave this forum since both you and I are done with PSLE.
Don't waste your effort.
Congrats for finishing PSLE as well. Couldn't resist the posting while waiting for the US markets to open.

]]>https://forum.kiasuparents.com/post/911526https://forum.kiasuparents.com/post/911526Thu, 29 Nov 2012 14:04:44 GMT<![CDATA[Reply to Bell Curve - To remove or not to remove? on Thu, 29 Nov 2012 13:56:30 GMT]]>

ruohoo97:

Thanks AWSP, thank you for detailed insight on Tscore, very enlightening indeen. The only thing, although sometimes distorted, Tscore is so far the best placement measurement to serve this ranking purpose. Unless MOE wants to another holistic approach for students to select secondary schools, visa verse, the bell curve will stay for a while.
I think the first step to solving the issue is the recognise the distortion effect it creates. I would say that this contributes to a lot of unnecessary uncertainty and we cannot discount the magnification effect and the distortion in the weighting between the subjects. As atutor2001's exercise also pointed out and as I highlighted in the switching of the sign for those students with a subject that is particularly weak, landing on the wrong side of the mean really penalise you if you are not with the rest of the crowd in a skewed distribution.
Most importantly, the whole thought process of the Tscore appears to be all about fitting the score into 0-75 ( I think the designers did a very conservative fitting and use the factor 10/(standard deviation). Why not just take the percentile rankings for each subject and add together? There will be no distortion and true equal weighting while maintaining the whole objective of ranking. But the formulation will not look so \"sexy\" and appears too simple. I reserve my critical comments on the designers of Tscore for making such a \"marketing\" move.

The simple method above may lose a very tiny bit of granularity but we have to be aware that the granularity from the present Tscore comes from distortion in weighting. However as with all detailed ranking, the method I just proposed may be too clinical and not palatable to most.

My own \"socialist\" leaning is to have grades with percentile ranking for the award of the grades and point system for the grades. However this will definitely not be granular enough for sorting into schools. I maintain my view that rhetoric aside, MOE has to bring the schools more or less in line in terms of standard. Or it has to increase the supply of places at the IP schools both at PSLE and at 0 levels. Only then parents and students have the \"true option\" of selecting the school to match their pace. The current allocation of places for O levels entrants into the IP schools is ridiculous because of the IP programs. Thats why everyone is \"stampeding\" at PSLE.

I will not claim moral highground and say parents are too fixated with IP school because we are all guilty of putting our kids into the best schools that they can get into. And we are all aware that the chances narrow sharply after PSLE which is why PSLE is a high staked exams.

]]>https://forum.kiasuparents.com/post/911522https://forum.kiasuparents.com/post/911522Thu, 29 Nov 2012 13:56:30 GMT<![CDATA[Reply to Bell Curve - To remove or not to remove? on Thu, 29 Nov 2012 12:07:03 GMT]]>

AWSP:


the whole exercise of t-score is just for the single purpose of sec school allocation by merit and nothing else. the guy ranked higher will choose his sec school first. that's it

t-score is not a measure of cleverness or ability (in a narrow sense) but it takes into account carelessness on the actual exam day

and it is totally meaningless to compare t-scores across different years / cohort...as there is zero meaning to rank one guy from this year against another guy from another year...

if everyone is given unlimited amount of time, everyone will score full marks and have the same score

the guy who \"invented\" the t-score formula is now the principal of NUSH

well, for those people who demand for \"transparency\" aren't they insinuating there is cover up ? after all the t-score formula is not a state secret, and the PSLE markers are their teachers
You missed my point. While I agree with you on that we may need a granular formula which is because of the competition for places into the IP school., my hypothesis is
a) the t-score is designed SOLELY for data fitting into nice channel of 0-100 ( or even less) and NOTHING ELSE.
b) we have over speculated on that a difficult paper equate to low SD etc. I can easily design a paper where everyone score 0-10 marks for a difficult paper(low SD) or everyone score 90-100 for an easy paper (low SD again). Anything in between is unpredictable and pure speculation. AND it is not in the design.
There is no need to name the architect fof T-score although I know who he is as you have known.
I am re-engineering the thought process in the design as a point of interest and not because I am crying foul.
c)By virtue of points a and b, there are unintended effects which will be interesting to study as well.

the t-score formula is NOT just for competition for IP places.

it is used for ranking for the whole cohort to determine who has the right to choose his school before the next guy, including the guy with t-score of 43

why are people so fixated with IP or the top 10%? the root cause is really the sour grapes attitude mixed with kiasu mentality

why can't people recognise and accept that normal distribution of talent and examination ability exist in the population ? what is wrong by accepting such truth ?

do you guys seriously expect everyone to be able to score the same top marks ? then the questions will be set even more difficult to differentiate the differing ability of every pupil...want more stress ?
a) I dont think the guys on this thread are sour grapes or fixated over the IP schools placing. For me, PSLE is over and this PSLE has no bearing on me or my kids as the youngest have DSAed into an IP school and is not taking part in the S1 posting. (her Tscore is sufficient to qualify her into the school any way)
b) the point of interest is that I am trying to get people to open their eyes and not take too many things for granted like
i) normal distribution - even in finance and economics, researchers are shifting out of normal distribution into fat tailed distributions and skewed distribution (power law distribution). Can we be so sure?
ii) are there some unintentional effects in the design of the T-score. Does it cause distortion?
iii) what is the thought process in the design of the T-score
I re-emphasize that this is just a discussion of interest.
There are other issues that kiasuparents should examine carefully:
a) are we on the right track with the type of testing in Science and Maths?
b) are we making a big detour doing models and heuristics in Maths which all the kids ditch in secondary school?
c) can pure rhetoric that all schools are the same solve the stress and competition issue without bringing the secondary schools more or less in line in terms of quality and facilities?
d) should we adjust the requirements in mother tongue so that it is not so \"punishing\" while we want to encourage or promote the learning.
T-score is but one small part of the broadbased issues that we are facing.
I am not one to accept all \"truths\" without examining the \"truths\" first.[/quote]

Thanks AWSP, thank you for detailed insight on Tscore, very enlightening indeen. The only thing, although sometimes distorted, Tscore is so far the best placement measurement to serve this ranking purpose. Unless MOE wants to another holistic approach for students to select secondary schools, visa verse, the bell curve will stay for a while.

]]>https://forum.kiasuparents.com/post/911437https://forum.kiasuparents.com/post/911437Thu, 29 Nov 2012 12:07:03 GMT<![CDATA[Reply to Bell Curve - To remove or not to remove? on Thu, 29 Nov 2012 10:25:55 GMT]]>

AWSP:



iii) what is the thought process in the design of the T-score
I re-emphasize that this is just a discussion of interest.
There are other issues that kiasuparents should examine carefully:

a) are we on the right track with the type of testing in Science and Maths?
b) are we making a big detour doing models and heuristics in Maths which all the kids ditch in secondary school?

c) can pure rhetoric that all schools are the same solve the stress and competition issue without bringing the secondary schools more or less in line in terms of quality and facilities?

d) should we adjust the requirements in mother tongue so that it is not so \"punishing\" while we want to encourage or promote the learning.

T-score is but one small part of the broadbased issues that we are facing.
I am not one to accept all \"truths\" without examining the \"truths\" first.
:goodpost:

]]>https://forum.kiasuparents.com/post/911397https://forum.kiasuparents.com/post/911397Thu, 29 Nov 2012 10:25:55 GMT<![CDATA[Reply to Bell Curve - To remove or not to remove? on Wed, 28 Nov 2012 13:18:49 GMT]]>:goodpost:

But I think your effort is falling on deaf ears. 😄
I suggest that we leave this forum since both you and I are done with PSLE.
Don't waste your effort.

AWSP:
I thought I could make the argument more rigorous in maths in addition to some of the numerical examples from atutor2001 since I am quite free this morning. Here is the working.
z= std deviation, x= raw score and y = mean.
Tscore= 50 + (10/z) *(x-y)

Taking derivative and holding x constant to assess the effect on Tscore.
d(Tscore) = -(10/(z*z)) (x-y)dz -(10/z)dy

The equation is in fact very 'rich' and gives a lot of insight.
The first term is the contribution of changes in standard deviation and the second term is contribution from changes in the mean. Both are negative signed meaning the increases in z and y results in decreases in Tscore.
The net effect would depend on z movement in relation to y movement. A simple measurement could be the skewness of the distribution.

Also atutor2001's number example highlights the effect of landing on the wrong side of the mean where x-y changes sign.

But the crucial point which I would like to highlight is that it is very easy to note that even if a person's percentile does not change, the cohort's mean and standard deviation changes can have quite a strong impact especially across subjects. AND it is not a \"moderation effect\" as we always thought.
Moderation effect is about percentile ranking using a bell curve where a person can still score an A despite getting 40 marks if he is in the say top 10%.

In a certain sense, I would agree with Tharman that the T-score is too granular.

I hope the above will be interesting enough to share. Sorry for the calculus if you couldnt understand.

]]>https://forum.kiasuparents.com/post/910705https://forum.kiasuparents.com/post/910705Wed, 28 Nov 2012 13:18:49 GMT<![CDATA[Reply to Bell Curve - To remove or not to remove? on Wed, 28 Nov 2012 04:09:41 GMT]]>

bupashu:
AWSP:

[quote=\"bupashu\"]
by raw scores, there will be many pupils with the same aggregate scores

hence the raw scores have to be transformed (hence the name t-score = transformed score) to take into account how \"easy\" or \"difficult\" of each subject (measured by mean & SD)...in order to rank the pupils for the purpose of sec one posting

the whole exercise of t-score is just for the single purpose of sec school allocation by merit and nothing else. the guy ranked higher will choose his sec school first. that's it

t-score is not a measure of cleverness or ability (in a narrow sense) but it takes into account carelessness on the actual exam day

and it is totally meaningless to compare t-scores across different years / cohort...as there is zero meaning to rank one guy from this year against another guy from another year...

if everyone is given unlimited amount of time, everyone will score full marks and have the same score

the guy who \"invented\" the t-score formula is now the principal of NUSH

well, for those people who demand for \"transparency\" aren't they insinuating there is cover up ? after all the t-score formula is not a state secret, and the PSLE markers are their teachers

You missed my point. While I agree with you on that we may need a granular formula which is because of the competition for places into the IP school., my hypothesis is
a) the t-score is designed SOLELY for data fitting into nice channel of 0-100 ( or even less) and NOTHING ELSE.
b) we have over speculated on that a difficult paper equate to low SD etc. I can easily design a paper where everyone score 0-10 marks for a difficult paper(low SD) or everyone score 90-100 for an easy paper (low SD again). Anything in between is unpredictable and pure speculation. AND it is not in the design.
There is no need to name the architect fof T-score although I know who he is as you have known.
I am re-engineering the thought process in the design as a point of interest and not because I am crying foul.
c)By virtue of points a and b, there are unintended effects which will be interesting to study as well.

the t-score formula is NOT just for competition for IP places.

it is used for ranking for the whole cohort to determine who has the right to choose his school before the next guy, including the guy with t-score of 43

why are people so fixated with IP or the top 10%? the root cause is really the sour grapes attitude mixed with kiasu mentality

why can't people recognise and accept that normal distribution of talent and examination ability exist in the population ? what is wrong by accepting such truth ?

do you guys seriously expect everyone to be able to score the same top marks ? then the questions will be set even more difficult to differentiate the differing ability of every pupil...want more stress ?[/quote]Your arguments are easy to comprehend, very sound and indeed commendable. I concur with your view 100%.

]]>https://forum.kiasuparents.com/post/910194https://forum.kiasuparents.com/post/910194Wed, 28 Nov 2012 04:09:41 GMT<![CDATA[Reply to Bell Curve - To remove or not to remove? on Wed, 28 Nov 2012 02:51:19 GMT]]>I thought I could make the argument more rigorous in maths in addition to some of the numerical examples from atutor2001 since I am quite free this morning. Here is the working.

z= std deviation, x= raw score and y = mean.
Tscore= 50 + (10/z) (x-y)

Taking derivative and holding x constant to assess the effect on Tscore.
d(Tscore) = -(10/(zz)) (x-y)dz -(10/z)dy

The equation is in fact very ‘rich’ and gives a lot of insight.
The first term is the contribution of changes in standard deviation and the second term is contribution from changes in the mean. Both are negative signed meaning the increases in z and y results in decreases in Tscore.
The net effect would depend on z movement in relation to y movement. A simple measurement could be the skewness of the distribution.

Also atutor2001’s number example highlights the effect of landing on the wrong side of the mean where x-y changes sign.

But the crucial point which I would like to highlight is that it is very easy to note that even if a person’s percentile does not change, the cohort’s mean and standard deviation changes can have quite a strong impact especially across subjects. AND it is not a "moderation effect" as we always thought.
Moderation effect is about percentile ranking using a bell curve where a person can still score an A despite getting 40 marks if he is in the say top 10%.

In a certain sense, I would agree with Tharman that the T-score is too granular.

I hope the above will be interesting enough to share. Sorry for the calculus if you couldnt understand.

]]>https://forum.kiasuparents.com/post/910086https://forum.kiasuparents.com/post/910086Wed, 28 Nov 2012 02:51:19 GMT<![CDATA[Reply to Bell Curve - To remove or not to remove? on Tue, 27 Nov 2012 14:04:24 GMT]]>

bupashu:
AWSP:

[quote=\"bupashu\"]
by raw scores, there will be many pupils with the same aggregate scores

hence the raw scores have to be transformed (hence the name t-score = transformed score) to take into account how \"easy\" or \"difficult\" of each subject (measured by mean & SD)...in order to rank the pupils for the purpose of sec one posting

the whole exercise of t-score is just for the single purpose of sec school allocation by merit and nothing else. the guy ranked higher will choose his sec school first. that's it

t-score is not a measure of cleverness or ability (in a narrow sense) but it takes into account carelessness on the actual exam day

and it is totally meaningless to compare t-scores across different years / cohort...as there is zero meaning to rank one guy from this year against another guy from another year...

if everyone is given unlimited amount of time, everyone will score full marks and have the same score

the guy who \"invented\" the t-score formula is now the principal of NUSH

well, for those people who demand for \"transparency\" aren't they insinuating there is cover up ? after all the t-score formula is not a state secret, and the PSLE markers are their teachers

You missed my point. While I agree with you on that we may need a granular formula which is because of the competition for places into the IP school., my hypothesis is
a) the t-score is designed SOLELY for data fitting into nice channel of 0-100 ( or even less) and NOTHING ELSE.
b) we have over speculated on that a difficult paper equate to low SD etc. I can easily design a paper where everyone score 0-10 marks for a difficult paper(low SD) or everyone score 90-100 for an easy paper (low SD again). Anything in between is unpredictable and pure speculation. AND it is not in the design.
There is no need to name the architect fof T-score although I know who he is as you have known.
I am re-engineering the thought process in the design as a point of interest and not because I am crying foul.
c)By virtue of points a and b, there are unintended effects which will be interesting to study as well.

the t-score formula is NOT just for competition for IP places.

it is used for ranking for the whole cohort to determine who has the right to choose his school before the next guy, including the guy with t-score of 43

why are people so fixated with IP or the top 10%? the root cause is really the sour grapes attitude mixed with kiasu mentality

why can't people recognise and accept that normal distribution of talent and examination ability exist in the population ? what is wrong by accepting such truth ?

do you guys seriously expect everyone to be able to score the same top marks ? then the questions will be set even more difficult to differentiate the differing ability of every pupil...want more stress ?[/quote]a) I dont think the guys on this thread are sour grapes or fixated over the IP schools placing. For me, PSLE is over and this PSLE has no bearing on me or my kids as the youngest have DSAed into an IP school and is not taking part in the S1 posting. (her Tscore is sufficient to qualify her into the school any way)
b) the point of interest is that I am trying to get people to open their eyes and not take too many things for granted like
i) normal distribution - even in finance and economics, researchers are shifting out of normal distribution into fat tailed distributions and skewed distribution (power law distribution). Can we be so sure?
ii) are there some unintentional effects in the design of the T-score. Does it cause distortion?
iii) what is the thought process in the design of the T-score
I re-emphasize that this is just a discussion of interest.
There are other issues that kiasuparents should examine carefully:
a) are we on the right track with the type of testing in Science and Maths?
b) are we making a big detour doing models and heuristics in Maths which all the kids ditch in secondary school?
c) can pure rhetoric that all schools are the same solve the stress and competition issue without bringing the secondary schools more or less in line in terms of quality and facilities?
d) should we adjust the requirements in mother tongue so that it is not so \"punishing\" while we want to encourage or promote the learning.
T-score is but one small part of the broadbased issues that we are facing.
I am not one to accept all \"truths\" without examining the \"truths\" first.

]]>https://forum.kiasuparents.com/post/909673https://forum.kiasuparents.com/post/909673Tue, 27 Nov 2012 14:04:24 GMT<![CDATA[Reply to Bell Curve - To remove or not to remove? on Tue, 27 Nov 2012 13:26:38 GMT]]>

AWSP:
bupashu:


by raw scores, there will be many pupils with the same aggregate scores

hence the raw scores have to be transformed (hence the name t-score = transformed score) to take into account how \"easy\" or \"difficult\" of each subject (measured by mean & SD)...in order to rank the pupils for the purpose of sec one posting

the whole exercise of t-score is just for the single purpose of sec school allocation by merit and nothing else. the guy ranked higher will choose his sec school first. that's it

t-score is not a measure of cleverness or ability (in a narrow sense) but it takes into account carelessness on the actual exam day

and it is totally meaningless to compare t-scores across different years / cohort...as there is zero meaning to rank one guy from this year against another guy from another year...

if everyone is given unlimited amount of time, everyone will score full marks and have the same score

the guy who \"invented\" the t-score formula is now the principal of NUSH

well, for those people who demand for \"transparency\" aren't they insinuating there is cover up ? after all the t-score formula is not a state secret, and the PSLE markers are their teachers

You missed my point. While I agree with you on that we may need a granular formula which is because of the competition for places into the IP school., my hypothesis is
a) the t-score is designed SOLELY for data fitting into nice channel of 0-100 ( or even less) and NOTHING ELSE.
b) we have over speculated on that a difficult paper equate to low SD etc. I can easily design a paper where everyone score 0-10 marks for a difficult paper(low SD) or everyone score 90-100 for an easy paper (low SD again). Anything in between is unpredictable and pure speculation. AND it is not in the design.
There is no need to name the architect fof T-score although I know who he is as you have known.
I am re-engineering the thought process in the design as a point of interest and not because I am crying foul.
c)By virtue of points a and b, there are unintended effects which will be interesting to study as well.

the t-score formula is NOT just for competition for IP places.

it is used for ranking for the whole cohort to determine who has the right to choose his school before the next guy, including the guy with t-score of 43

why are people so fixated with IP or the top 10%? the root cause is really the sour grapes attitude mixed with kiasu mentality

why can't people recognise and accept that normal distribution of talent and examination ability exist in the population ? what is wrong by accepting such truth ?

do you guys seriously expect everyone to be able to score the same top marks ? then the questions will be set even more difficult to differentiate the differing ability of every pupil...want more stress ?

]]>https://forum.kiasuparents.com/post/909650https://forum.kiasuparents.com/post/909650Tue, 27 Nov 2012 13:26:38 GMT<![CDATA[Reply to Bell Curve - To remove or not to remove? on Tue, 27 Nov 2012 13:21:03 GMT]]>

enoawng:
bupashu:


by raw scores, there will be many pupils with the same aggregate scores

hence the raw scores have to be transformed (hence the name t-score = transformed score) to take into account how \"easy\" or \"difficult\" of each subject (measured by mean & SD)...in order to rank the pupils for the purpose of sec one posting

the whole exercise of t-score is just for the single purpose of sec school allocation by merit and nothing else. the guy ranked higher will choose his sec school first. that's it

t-score is not a measure of cleverness or ability (in a narrow sense) but it takes into account carelessness on the actual exam day

and it is totally meaningless to compare t-scores across different years / cohort...as there is zero meaning to rank one guy from this year against another guy from another year...

if everyone is given unlimited amount of time, everyone will score full marks and have the same score

the guy who \"invented\" the t-score formula is now the principal of NUSH

well, for those people who demand for \"transparency\" aren't they insinuating there is cover up ? after all the t-score formula is not a state secret, and the PSLE markers are their teachers

thanks for a good summary for the need for t-score.

Do you know the reason(s) why raw scores cannot be published to the individual student in addition to the t-score?

ok if they give you the raw scores, are you going to collect all 45000 x 4 raw scores to compute the mean and the SD ?

it is meaningless

]]>https://forum.kiasuparents.com/post/909645https://forum.kiasuparents.com/post/909645Tue, 27 Nov 2012 13:21:03 GMT<![CDATA[Reply to Bell Curve - To remove or not to remove? on Tue, 27 Nov 2012 12:52:21 GMT]]>

bupashu:

by raw scores, there will be many pupils with the same aggregate scores

hence the raw scores have to be transformed (hence the name t-score = transformed score) to take into account how \"easy\" or \"difficult\" of each subject (measured by mean & SD)...in order to rank the pupils for the purpose of sec one posting

the whole exercise of t-score is just for the single purpose of sec school allocation by merit and nothing else. the guy ranked higher will choose his sec school first. that's it

t-score is not a measure of cleverness or ability (in a narrow sense) but it takes into account carelessness on the actual exam day

and it is totally meaningless to compare t-scores across different years / cohort...as there is zero meaning to rank one guy from this year against another guy from another year...

if everyone is given unlimited amount of time, everyone will score full marks and have the same score

the guy who \"invented\" the t-score formula is now the principal of NUSH

well, for those people who demand for \"transparency\" aren't they insinuating there is cover up ? after all the t-score formula is not a state secret, and the PSLE markers are their teachers
thanks for a good summary for the need for t-score.

Do you know the reason(s) why raw scores cannot be published to the individual student in addition to the t-score?

]]>https://forum.kiasuparents.com/post/909632https://forum.kiasuparents.com/post/909632Tue, 27 Nov 2012 12:52:21 GMT<![CDATA[Reply to Bell Curve - To remove or not to remove? on Tue, 27 Nov 2012 12:52:16 GMT]]>Hmm… need not name who the architect behind T-score, most of us aware….formula not a secret, it’s how and what values used to transformed

]]>https://forum.kiasuparents.com/post/909631https://forum.kiasuparents.com/post/909631Tue, 27 Nov 2012 12:52:16 GMT<![CDATA[Reply to Bell Curve - To remove or not to remove? on Tue, 27 Nov 2012 12:35:40 GMT]]>

bupashu:

by raw scores, there will be many pupils with the same aggregate scores

hence the raw scores have to be transformed (hence the name t-score = transformed score) to take into account how \"easy\" or \"difficult\" of each subject (measured by mean & SD)...in order to rank the pupils for the purpose of sec one posting

the whole exercise of t-score is just for the single purpose of sec school allocation by merit and nothing else. the guy ranked higher will choose his sec school first. that's it

t-score is not a measure of cleverness or ability (in a narrow sense) but it takes into account carelessness on the actual exam day

and it is totally meaningless to compare t-scores across different years / cohort...as there is zero meaning to rank one guy from this year against another guy from another year...

if everyone is given unlimited amount of time, everyone will score full marks and have the same score

the guy who \"invented\" the t-score formula is now the principal of NUSH

well, for those people who demand for \"transparency\" aren't they insinuating there is cover up ? after all the t-score formula is not a state secret, and the PSLE markers are their teachers
You missed my point. While I agree with you on that we may need a granular formula which is because of the competition for places into the IP school., my hypothesis is
a) the t-score is designed SOLELY for data fitting into nice channel of 0-100 ( or even less) and NOTHING ELSE.
b) we have over speculated on that a difficult paper equate to low SD etc. I can easily design a paper where everyone score 0-10 marks for a difficult paper(low SD) or everyone score 90-100 for an easy paper (low SD again). Anything in between is unpredictable and pure speculation. AND it is not in the design.
There is no need to name the architect fof T-score although I know who he is as you have known.
I am re-engineering the thought process in the design as a point of interest and not because I am crying foul.
c)By virtue of points a and b, there are unintended effects which will be interesting to study as well.

]]>https://forum.kiasuparents.com/post/909613https://forum.kiasuparents.com/post/909613Tue, 27 Nov 2012 12:35:40 GMT<![CDATA[Reply to Bell Curve - To remove or not to remove? on Tue, 27 Nov 2012 12:02:31 GMT]]>

AWSP:
The T-score formulation is interesting, its complication is probably unwarranted for. The point is not about trusting or not trusting it but to understand its effect first. That has sort of eluded us.

If you look at the formulation and try to understand the intuition. It looks weird. First, it tries to \"recenter\" the scores around the 50 mark by \"differencing the score and the mean and adding to 50.
Secondly, it also tries to rescale the difference by a factor of 10/(standard deviation). I can only guess that 10 appears to be some form of \"historical standard deviation\".
It may appear that the guys formulating it assume that the standard deviation will be wide for an easy paper and narrow for a difficult paper. I think that is a disputable assumption.
However my guess is that the guys are merely trying to fit the scores into a channel by rescaling it using a standard deviation. And that channel happens to be 0-100.
The more I look at it, the more likely that my guess is correct.
It is possibly something that only Singaporeans can create. Uniquely Singapore!
And we boast that we are more advanced than everyone else who use the good old percentile to grade!
PS I dont think it is our intent to take them(MOE) to court. We are just discussing this out of interest and for the fun of it.
by raw scores, there will be many pupils with the same aggregate scores

hence the raw scores have to be transformed (hence the name t-score = transformed score) to take into account how \"easy\" or \"difficult\" of each subject (measured by mean & SD)...in order to rank the pupils for the purpose of sec one posting

the whole exercise of t-score is just for the single purpose of sec school allocation by merit and nothing else. the guy ranked higher will choose his sec school first. that's it

t-score is not a measure of cleverness or ability (in a narrow sense) but it takes into account carelessness on the actual exam day

and it is totally meaningless to compare t-scores across different years / cohort...as there is zero meaning to rank one guy from this year against another guy from another year...

if everyone is given unlimited amount of time, everyone will score full marks and have the same score

the guy who \"invented\" the t-score formula is now the principal of NUSH

well, for those people who demand for \"transparency\" aren't they insinuating there is cover up ? after all the t-score formula is not a state secret, and the PSLE markers are their teachers

]]>https://forum.kiasuparents.com/post/909584https://forum.kiasuparents.com/post/909584Tue, 27 Nov 2012 12:02:31 GMT<![CDATA[Reply to Bell Curve - To remove or not to remove? on Tue, 27 Nov 2012 11:55:49 GMT]]>let me repost in simpler english:

x is the score; y is the mean and z is the standard deviation.
x-y = difference between mean and score.
50 + (x-y) = recentering a score around 50. if score is 70 and mean is 60,
the score is shifted to 60 ie 50 + (70-60).
50 + (10/z)*(x-y)= rescaling (x-y) by a factor of 10/Z and recentering to 50.

If we try to put intelligence into the rescaling factor, we should weigh the factor 10/Z lower for an easy paper and higher for a difficult paper. That means the standard deviation is higher for easy paper and lower for difficult paper. Is that TRUE always?

Or is it just a pure data fitting exercise, fitting the range of marks into a channel of 0-100?

]]>https://forum.kiasuparents.com/post/909581https://forum.kiasuparents.com/post/909581Tue, 27 Nov 2012 11:55:49 GMT<![CDATA[Reply to Bell Curve - To remove or not to remove? on Tue, 27 Nov 2012 10:43:06 GMT]]>

VALyap:
AWSP:

The T-score formulation formulation is interesting, its complication is probably unwarranted for. The point is not about trusting or not trusting it but to understand the effect first. That has sort of eluded us.

If you look at the formulation and try to understand the intuition. It looks weird. First, it tries to \"recenter\" the scores around the 50 mark by \"differencing the score and the mean and adding to 50.
Secondly, it also tries to rescale the difference by a factor of 10/(standard deviation). I can only guess that 10 appears to be some form of \"historical standard deviation\".
It may appear that the guys formulating it assume that the standard deviation will be wide for an easy paper and narrow for a difficult paper. I think that is a disputable assumption.
However my guess is that the guys are merely trying to fit the scores into a channel by rescaling it using a standard deviation. And that channel happens to be 0-100.
The more I look at it, the more likely that my guess is correct.
It is possibly something that only Singaporeans can create. Uniquely Singapore!
PS I dont intend to take them to court. I am just discussing this out of interest.


my guess is, in MOE Not more than 5 persons involved in T-score , processes, formulation and final approval including the minister… just a wild guess! given the fact it's like state secret….just the chief statistician and Directors of examination branch & minister etc..

It is extremely weird that the only consideration in the T-score is data fitting into 0-100 range. Reflects the state of affairs in our civil service, perhaps. Ngiam Dong Tow is right in his recent comments on the civil service perhaps.

]]>https://forum.kiasuparents.com/post/909521https://forum.kiasuparents.com/post/909521Tue, 27 Nov 2012 10:43:06 GMT<![CDATA[Reply to Bell Curve - To remove or not to remove? on Tue, 27 Nov 2012 10:33:40 GMT]]>

AWSP:
The T-score formulation formulation is interesting, its complication is probably unwarranted for. The point is not about trusting or not trusting it but to understand the effect first. That has sort of eluded us.

If you look at the formulation and try to understand the intuition. It looks weird. First, it tries to \"recenter\" the scores around the 50 mark by \"differencing the score and the mean and adding to 50.
Secondly, it also tries to rescale the difference by a factor of 10/(standard deviation). I can only guess that 10 appears to be some form of \"historical standard deviation\".
It may appear that the guys formulating it assume that the standard deviation will be wide for an easy paper and narrow for a difficult paper. I think that is a disputable assumption.
However my guess is that the guys are merely trying to fit the scores into a channel by rescaling it using a standard deviation. And that channel happens to be 0-100.
The more I look at it, the more likely that my guess is correct.
It is possibly something that only Singaporeans can create. Uniquely Singapore!
PS I dont intend to take them to court. I am just discussing this out of interest.

my guess is, in MOE Not more than 5 persons involved in T-score , processes, formulation and final approval including the minister… just a wild guess! given the fact it's like state secret….just the chief statistician and Directors of examination branch & minister etc..

]]>https://forum.kiasuparents.com/post/909512https://forum.kiasuparents.com/post/909512Tue, 27 Nov 2012 10:33:40 GMT<![CDATA[Reply to Bell Curve - To remove or not to remove? on Tue, 27 Nov 2012 10:21:17 GMT]]>The T-score formulation is interesting, its complication is probably unwarranted for. The point is not about trusting or not trusting it but to understand its effect first. That has sort of eluded us.

If you look at the formulation and try to understand the intuition. It looks weird. First, it tries to "recenter" the scores around the 50 mark by "differencing the score and the mean and adding to 50.
Secondly, it also tries to rescale the difference by a factor of 10/(standard deviation). I can only guess that 10 appears to be some form of "historical standard deviation".
It may appear that the guys formulating it assume that the standard deviation will be wide for an easy paper and narrow for a difficult paper. I think that is a disputable assumption.
However my guess is that the guys are merely trying to fit the scores into a channel by rescaling it using a standard deviation. And that channel happens to be 0-100.
The more I look at it, the more likely that my guess is correct.
It is possibly something that only Singaporeans can create. Uniquely Singapore!
And we boast that we are more advanced than everyone else who use the good old percentile to grade!
PS I dont think it is our intent to take them(MOE) to court. We are just discussing this out of interest and for the fun of it.

]]>https://forum.kiasuparents.com/post/909499https://forum.kiasuparents.com/post/909499Tue, 27 Nov 2012 10:21:17 GMT<![CDATA[Reply to Bell Curve - To remove or not to remove? on Tue, 27 Nov 2012 09:23:19 GMT]]>

enoawng:
Will_lim:



http://hedgehogcomms.blogspot.sg/2012/10/piecing-together-psle-puzzle.html

I don't want to debate on the pros and cons of the bell curve but the author is right in that moe should be more transparent to provide the raw scores of each subject, the average and the standard deviation.

The 12 year olds can work out the t-scores themselves.....

lots of parents already have problem & difficulties understanding the t-score formula....u think the students at P6 can handle statistics, mean, median & standard deviation and 45,000 x 4 raw scores ?

but if you don't trust the t-score given, you can always request to review the t-score from MOE, and if you are still not satisfied, go file a court order to order MOE to reveal all the stats and get an independent committee of inquiry to audit the t-score process and procedures, including the computer software codes for it

]]>https://forum.kiasuparents.com/post/909411https://forum.kiasuparents.com/post/909411Tue, 27 Nov 2012 09:23:19 GMT<![CDATA[Reply to Bell Curve - To remove or not to remove? on Mon, 26 Nov 2012 15:41:52 GMT]]>

Will_lim:


http://hedgehogcomms.blogspot.sg/2012/10/piecing-together-psle-puzzle.html
I don't want to debate on the pros and cons of the bell curve but the author is right in that moe should be more transparent to provide the raw scores of each subject, the average and the standard deviation.

The 12 year olds can work out the t-scores themselves.....

]]>https://forum.kiasuparents.com/post/908662https://forum.kiasuparents.com/post/908662Mon, 26 Nov 2012 15:41:52 GMT