* Eunoia JC (EJC)
-
thsheng99\" post_id=\"1900543\" time=\"1551425407\" user_id=\"37400:
Given that there are 25 %JAE, and if the 75% of the IP kids really performed above-expectation (say >=85 UAS), I can conclude that the JAE kids would have performed below expectation for the mean UAS to be 82.4.
Are you saying the rest of the IP JCs have no IP kids? Furthermore, there is still a substantial 25% that came in through JAE in EJC. Please look at the COP at 2017Grandyma\" post_id=\"1900412\" time=\"1551397436\" user_id=\"153793:
EJC value adding...you mean with 75% IP kids in this 2018 A level cohort ( its P6 intake that yr i think was COP 252-257 and higher) , you didnt expect this cohort to churn out Mean RP of 82.4 results regardless its pioneer batch or not. From info shared, RP 82.4 is not better than some other JCs.
Which single digit JCs are you refering that they didnt produce the so called ' value adding' results and that EJC has more 'value added' results??
Junior College\tArts\tScience / IB
Raffles Institution, RI\t4\t4
Hwa Chong Institution, HCI\t4\t4
Victoria Junior College, VJC\t7\t5
Anglo-Chinese School (Independent), ACSi\t–\t5
National Junior College, NJC\t7\t6
Nanyang Junior College, NYJC\t7\t6
Anglo-Chinese Junior College, ACJC\t8\t7
St. Joseph’s Institution, SJI\t–\t7
Eunoia Junior College, EJC\t9\t9
There are at least 8 schools with better COP than EJC. With a UAS of 82.4 is close to or similar to the likes of NJC and VJC ( who has COP of 5-7)
Furthermore, given that it is a pioneer year ( with no proven record) with presumably many teething problems and that it is operating from a temporary campus in remote Mt Sinai where transportation links are not that good, would you not think that they have been \"value-adding\"?
You are just being over-critical. I am glad that my grandma is more reasonable -
floppy\" post_id=\"1901081\" time=\"1551693609\" user_id=\"97579:
Mean must be = Median?
If you take principal's comment at face value, median > 85 (however, we also know 5 in 10 could also mean 45%Grandyma\" post_id=\"1901033\" time=\"1551679512\" user_id=\"153793:
Thank you all for answering my question.
So its right to say Mean can be higher or lower than Median.
Regard to NJC Mean, yes fair enough to say its Mean likely negative skewed as Principal didnt share much abt its cohort scores.
How about NYJC?
Principal shared \"More than 5 out of 10 in you got more than 85 pt\".
So can say its Median is slightly > 85.X?
Skewed towards?
).
If 50% of the students score between 85 and 90, an equal amount would have to score between 80 and 85 to have mean = median, or a better ratio for mean > median, which are very unlikely to be the case. Chances are, there will be a handful scoring in the 70s or lower. Hence, NYJC mean should also be in the low 80s. How low? Insufficient data to know.
Principal said ' MORE than 5 in 10' not 5 in 10
Dont understand yr login....hahaa but thanks. -
Grandyma\" post_id=\"1901033\" time=\"1551679512\" user_id=\"153793:
Seen no response for the question I posted.
Thank you all for answering my question.obm\" post_id=\"1900989\" time=\"1551669708\" user_id=\"162988:
It is likely to be left skewed distribution or negative skew for NJC, based on that pathetic bit of info provided by the school. There were 35% H2As and the Math & Science depts did pretty well according to one table compiled by a member in the Alevel 2018 megathread.
NJC's '1 in 2 scored As for H3' cannot be counted as it's not part of UAS. I hazard a guess its Mode is on the right side. But as to how flat the curve is, we need its full set of data to determine how big the margin is between its Mean and Median. For negative skews, the Median is higher than Mean, hence the preference by its Principal to present the Median.
https://postimg.cc/dkvrbBmH
So its right to say Mean can be higher or lower than Median.
Regard to NJC Mean, yes fair enough to say its Mean likely negative skewed as Principal didnt share much abt its cohort scores.
How about NYJC?
Principal shared \"More than 5 out of 10 in you got more than 85 pt\".
So can say its Median is slightly > 85.X?
Skewed towards?
Its ok I have gotten the answer. -
Grandyma\" post_id=\"1901151\" time=\"1551748738\" user_id=\"153793:
Mean must be = Median?
If you take principal's comment at face value, median > 85 (however, we also know 5 in 10 could also mean 45%floppy\" post_id=\"1901081\" time=\"1551693609\" user_id=\"97579:
[quote=Grandyma post_id=1901033 time=1551679512 user_id=153793]Thank you all for answering my question.
So its right to say Mean can be higher or lower than Median.
Regard to NJC Mean, yes fair enough to say its Mean likely negative skewed as Principal didnt share much abt its cohort scores.
How about NYJC?
Principal shared \"More than 5 out of 10 in you got more than 85 pt\".
So can say its Median is slightly > 85.X?
Skewed towards? ).
If 50% of the students score between 85 and 90, an equal amount would have to score between 80 and 85 to have mean = median, or a better ratio for mean > median, which are very unlikely to be the case. Chances are, there will be a handful scoring in the 70s or lower. Hence, NYJC mean should also be in the low 80s. How low? Insufficient data to know.
Principal said ' MORE than 5 in 10' not 5 in 10
Dont understand yr login....hahaa but thanks.[/quote]If you don’t understand, I cannot help you. As lee_yl says, it’s statistics 101. I understand a crash course in statistics can be overwhelming for some elderly folks. Perhaps your grandchild who has gotten an A in Maths (I hope) has a better way of explaining to you (I hope again). -
Grandyma\" post_id=\"1901151\" time=\"1551748738\" user_id=\"153793:
Mean must be = Median?
If you take principal's comment at face value, median > 85 (however, we also know 5 in 10 could also mean 45%floppy\" post_id=\"1901081\" time=\"1551693609\" user_id=\"97579:
[quote=Grandyma post_id=1901033 time=1551679512 user_id=153793]Thank you all for answering my question.
So its right to say Mean can be higher or lower than Median.
Regard to NJC Mean, yes fair enough to say its Mean likely negative skewed as Principal didnt share much abt its cohort scores.
How about NYJC?
Principal shared \"More than 5 out of 10 in you got more than 85 pt\".
So can say its Median is slightly > 85.X?
Skewed towards? ).
If 50% of the students score between 85 and 90, an equal amount would have to score between 80 and 85 to have mean = median, or a better ratio for mean > median, which are very unlikely to be the case. Chances are, there will be a handful scoring in the 70s or lower. Hence, NYJC mean should also be in the low 80s. How low? Insufficient data to know.
Principal said ' MORE than 5 in 10' not 5 in 10
Dont understand yr login...out of context....hahaa...its ok...but thanks.[/quote] -
ezke\" post_id=\"1901110\" time=\"1551708160\" user_id=\"138513:
It’s hard to draw that conclusion without further info.
Given that there are 25 %JAE, and if the 75% of the IP kids really performed above-expectation (say >=85 UAS), I can conclude that the JAE kids would have performed below expectation for the mean UAS to be 82.4.thsheng99\" post_id=\"1900543\" time=\"1551425407\" user_id=\"37400:
...
With a UAS of 82.4 is close to or similar to the likes of NJC and VJC ( who has COP of 5-7)
...
However, it’s clear that on the whole, EJC performance for the pioneer batch isn’t that far off from NJC / VJC (I suspect both will still be marginally better), and is on par or better than TJC. I don't believe in this 'value-added' nonsense. The result is consistent with the quality provided by EJC three feeder school. -
floppy\" post_id=\"1901154\" time=\"1551749346\" user_id=\"97579:
If you don’t understand, I cannot help you. As lee_yl says, it’s statistics 101. I understand a crash course in statistics can be overwhelming for some elderly folks. Perhaps your grandchild who has gotten an A in Maths (I hope) has a better way of explaining to you (I hope again).[/quote]Your low 80s RP is at which range?
Mean must be = Median?Grandyma\" post_id=\"1901151\" time=\"1551748738\" user_id=\"153793:
[quote=floppy post_id=1901081 time=1551693609 user_id=97579]
If you take principal's comment at face value, median > 85 (however, we also know 5 in 10 could also mean 45% ).
If 50% of the students score between 85 and 90, an equal amount would have to score between 80 and 85 to have mean = median, or a better ratio for mean > median, which are very unlikely to be the case. Chances are, there will be a handful scoring in the 70s or lower. Hence, NYJC mean should also be in the low 80s. How low? Insufficient data to know.
Principal said ' MORE than 5 in 10' not 5 in 10
Dont understand yr login....hahaa but thanks.
But I did prove that Mean can be higher or lower than Median which was initially believed by the rest that Mean is always < Median.
Nyjc More than 5 in 10 got More than 85 pts. So more top scorers saturated on the positive skewed. Yes, there will low 70s scorers but P has said more than 50% got 85 pts or more.
Median will be > 85.0.
Other JCs with RP 83.X or more will also have 70s or lower scorers in the cohort, right??
Whats yr take for NJC Mean with its Median at 84.4?? -
Grandyma\" post_id=\"1901272\" time=\"1551766969\" user_id=\"153793:
:?
Your low 80s RP is at which range?
But I did prove that Mean can be higher or lower than Median which was initially believed by the rest that Mean is always < Median.
Nyjc More than 5 in 10 got More than 85 pts. So more top scorers saturated on the positive skewed. Yes, there will low 70s scorers but P has said more than 50% got 85 pts or more.
Median will be > 85.0.
Other JCs with RP 83.X or more will also have 70s or lower scorers in the cohort, right??
Whats yr take for NJC Mean with its Median at 84.4??
You didn't prove anything nor is there a need for you to.
There are only 3 possible outcome in statistics:
1. Mean = Median
2. Mean > Median
3. Mean < Median
I don't think there's anyone out there who believe in a fourth outcome. Given the way rank points are being distributed, what most people here are saying (and believing, correctly if I may add) is that schools with median > 8X is more likely to have a mean < 8X than a mean > or = 8X.
With that, welcome once again to Statistics 101.
I don't recall anyone here have disputed that it isn't possible for NYJC median to be > 85. Given that NYJC COP for 2017 is 7 / 6 for Arts / Science, why shouldn't more than 50% of the students do well? :scratchhead: What most people have an issue with, is comparing mean with median, and obfuscating the two. As you have said, there will be students in NYJC scoring in the low 70s. Therefore, I won't be surprised if NYJC mean is also in the low 80s. Same goes for NJC (with a median of 84.4). My range of low 80s? 80 to 82.5 i.e. same bracket as EJC, TJC. That said, it's fair to assume that NYJC (and NJC) should be closer to the higher end of that range. Apart from the two colleges named R and H, this is actually not a bad result at all (and why EJC is rightfully proud of the school's performance). -
floppy\" post_id=\"1901287\" time=\"1551771716\" user_id=\"97579:
Previously, forummers here believed that only Mean is always lower than Median right, if Im not wrong? Idk who said this and was accepted by the rest.
:?Grandyma\" post_id=\"1901272\" time=\"1551766969\" user_id=\"153793:
Your low 80s RP is at which range?
But I did prove that Mean can be higher or lower than Median which was initially believed by the rest that Mean is always < Median.
Nyjc More than 5 in 10 got More than 85 pts. So more top scorers saturated on the positive skewed. Yes, there will low 70s scorers but P has said more than 50% got 85 pts or more.
Median will be > 85.0.
Other JCs with RP 83.X or more will also have 70s or lower scorers in the cohort, right??
Whats yr take for NJC Mean with its Median at 84.4??
You didn't prove anything nor is there a need for you to.
There are only 3 possible outcome:
1. Mean = Median
2. Mean > Median
3. Mean < Median
With that, welcome to Statistics 101.
I don't recall anyone here has disputed that it isn't possible for NYJC median to be > 85. Given that NYJC COP for 2017 is 7 / 6 for Arts / Science, why shouldn't more than 50% of the students do well? :scratchhead:
What most people have an issue with, is comparing mean with median, and obfuscating the two. As you have said, there will be students in NYJC scoring in the low 70s. Therefore, I won't be surprised if NYJC mean is in the low 80s. Same goes for NJC. My range of low 80s? 80 to 82.5 (i.e. same bracket as EJC, TJC
). Apart from the two colleges named R and H, this is actually not a bad result.
Regarding those JCs with RP 83 and above, they also have low 70s scorers, why then RP can be 83 pt above? You havent explained.
Njc didnt reveal abt its % cohort performance overall RP so you said Njc RP can be 80-82.5, ok loh but to be confirmed.
But Nyjc revealed More than 50% with 85 pt above, in Mathematical logic, its Mean cant be 4-5 pts lower than its Median of > 85 pts.
You biased lei.
If say 1-3 pts lower If we take it to be negative skewed, then Nyjc RP can be 82.5 the lowest to 84.X the highest. -
Grandyma\" post_id=\"1901291\" time=\"1551773900\" user_id=\"153793:
\"forummers here believed that only Mean is always lower than Median right\"
Previously, forummers here believed that only Mean is always lower than Median right, if Im not wrong? Idk who said this and was accepted by the rest.
Regarding those JCs with RP 83 and above, they also have low 70s scorers, why then RP can be 83 pt above? You havent explained.
Njc didnt reveal abt its % cohort performance overall RP so you said Njc RP can be 80-82.5, ok loh but to be confirmed.
But Nyjc revealed More than 50% with 85 pt above, in Mathematical logic, its Mean cant be 5 pts lower than its Median of > 85 pts.
You biased lei.
If say 1-3 pts lower If we take it to be negative skewed, then Nyjc RP can be 82.5 the lowest to 84.X the highest.
> This is mischaracterizing what forummers say.
> As far as I can recall, forummers understand that there are 3 possible outcomes and mean < median is the most likely outcome (ok, most forummers understand... there are some stubborn ones who
).
\"Regarding those JCs with RP 83 and above, they also have low 70s scorers, why then RP can be 83 pt above?\"
> Which are the JCs? I don't know.
> As to why? Because there are 3 possible outcomes (I'm also getting old, sound like a broken record ). When it is said that mean < median is the most likely outcome, it doesn't mean that it is the ONLY possible outcome.
> Without conjuring my Harry Potter self, I can tell you that RP > 83 is easier to achieve a 'mean > median' than for RP > 85 to do so (although both is quite unlikely, statistically speaking).
> A distribution like this:
70, 75, 82, 82, 82, 83, 88, 89, 90, 90, 90 will have a median of 83 and a mean of 83.7, i.e. mean > median.
\"But Nyjc revealed More than 50% with 85 pt above, in Mathematical logic, its Mean cant be 5 pts lower than its Median of > 85 pts.\"
> Ok, understandable why you get that impression. You quoted my first draft.
> This is what I added:
\"That said, it's fair to assume that NYJC (and NJC) should be closer to the higher end of that range.\"
> Based on what had been shared by NYJC, I don't think NYJC mean is anywhere near to 84.x. It's the data that has NOT been shared that often shines the brightest.
> And \"more than 50% with 85 pt above...\" is not mathematics. It's a statement, and a fairly useless one where statistics and mathematics are concerned. Based on mathematical logic, it's mean CAN absolutely be 5 points lower than median. However, based on statistical analysis, this is an extremely unlikely (but not impossible) scenario. Two different things.
> A distribution like this:
70, 70, 71, 71, 80, 85, 86, 86, 86, 87, 88 will have a median of 85 and a mean of 80.0, i.e mean < median by 5 points.
A bit more of these and we can move on to Statistics 201.
Hello! It looks like you're interested in this conversation, but you don't have an account yet.
Getting fed up of having to scroll through the same posts each visit? When you register for an account, you'll always come back to exactly where you were before, and choose to be notified of new replies (either via email, or push notification). You'll also be able to save bookmarks and upvote posts to show your appreciation to other community members.
With your input, this post could be even better 💗
Register Login
