This discussion is now closed.
Check out other Related discussions
- tompkins table inquiry for cambridge
- what oxford college?
- Corpus Christi
- hypothesis Testing with only > Mean side of z table
- Complete University Guide Table 2024
- BTEC Level 3 IT Unit 2 Exam May 2024
- Paris 2024 Paralympics: Para table tennis
- Paris 2024 Olympics: Table Tennis
- Cie igcse computer science paper 2
- Would someone get prosecuted
- Is King's College really good?
- BTEC IT Unit 2 January 2024 exam discussion
- I forgot to cite a table for my dissertation for my AP
- First choice Uni dropped by 23 places in league tables
- is kcl good?
- OCR A - A level chem - do we need to know history of the periodic table
- Paris 2024 Paralympics: Para Triathlon
- The Sports Hurt/heal game
- Do you get given a college time table(uk)?
- AQA NEA Computing - URGENT !!!
The Not-Norrington (Poohbear)Table
Despite its popularity, there are a number of issues with using the Norrington table:
1. It uses an arbitrary weighting of the scores, a semi-opaque weighted average that corresponds to nothing.
("Final year students score 5 for a first, 3 for a 2:1, 2 for a 2:2 and 1 for a 3rd" )
Why a 1 for a 3rd ? Why not a 0.1 or even a -1 ?
2. Getting a 2:2 or a 3 is the kiss of death for your career.
Having these death-of-career marks contribute to the score is counterintuitive and meaningless
3. What really matters is the percentage of Firsts.
At Oxford, almost everyone ( except those with deep, deep problems ) gets a First or a 2:1. Most of us can confidently expect to get at least a 2:1 .
4. What I want to know is what percentage of candidates in a college get Firsts. I can counter a minor college-culture problem by working hard. However, I do want to make sure that my college does not get one-third the number of Firsts that the average college gets. Countering that severe a cultural problem may well be beyond my merely human efforts.
5.I suspect that Edward Tufte would have called the Norrington Table "chart-junk" or "a corrupt technique in evidence presentation" for it hides the structure of the data and imposes unwarranted and not immediately obvious assumptions into the displayed data. http://www.edwardtufte.com/bboard/q-and-a-fetch-msg?msg_id=0001et&topic_id=1
In this new table, I rank colleges according to the percentage of firsts. Among other things, it reveals how close many colleges are to each other.
So, here follows the Poohbear Table
.
1. It uses an arbitrary weighting of the scores, a semi-opaque weighted average that corresponds to nothing.
("Final year students score 5 for a first, 3 for a 2:1, 2 for a 2:2 and 1 for a 3rd" )
Why a 1 for a 3rd ? Why not a 0.1 or even a -1 ?
2. Getting a 2:2 or a 3 is the kiss of death for your career.
Having these death-of-career marks contribute to the score is counterintuitive and meaningless
3. What really matters is the percentage of Firsts.
At Oxford, almost everyone ( except those with deep, deep problems ) gets a First or a 2:1. Most of us can confidently expect to get at least a 2:1 .
4. What I want to know is what percentage of candidates in a college get Firsts. I can counter a minor college-culture problem by working hard. However, I do want to make sure that my college does not get one-third the number of Firsts that the average college gets. Countering that severe a cultural problem may well be beyond my merely human efforts.
5.I suspect that Edward Tufte would have called the Norrington Table "chart-junk" or "a corrupt technique in evidence presentation" for it hides the structure of the data and imposes unwarranted and not immediately obvious assumptions into the displayed data. http://www.edwardtufte.com/bboard/q-and-a-fetch-msg?msg_id=0001et&topic_id=1
In this new table, I rank colleges according to the percentage of firsts. Among other things, it reveals how close many colleges are to each other.
So, here follows the Poohbear Table
.Scroll to see replies
Calculated using the 2005 data from http://www.ox.ac.uk/aboutoxford/facts/collegefigs.shtml
Based on Undergraduate Student Results record data as recorded at Trinity Term 2005 on 05.10.05, for the degrees of: B.A., B.F.A., B.Th., M.Biochem., M.Chem., M.Earth Sci., M.Eng., M.Math., M.Phys ( the latter M.* are undergrad degrees despite the names ! )
College percentage of Firsts achieved: Sorted by Rank
%Firsts Pooh-Rank
38 1 St John's College
37 2 Merton College
37 2 Balliol College
34 4 Magdalen College
32 5 University College
30 6 Hertford College
29 7 Wadham College
28 8 Christ Church
28 8 Keble College
27 10 Exeter College
27 10 St Anne's College
26 12 Jesus College
26 12 New College
24 14 Somerville College
24 14 Mansfield College
23 16 Corpus Christi College
23 16 Trinity College
23 16 St Catherine's College
23 16 Queen's College
22 20 Worcester College
21 21 Pembroke College
21 21 Brasenose College
20 23 St Peter's College
19 24 St Hilda's College
19 24 Lincoln College
19 24 St Edmund Hall
19 24 St Hugh's College
16 28 Oriel College
14 29 Lady Margaret Hall
4 30 Harris Manchester College
25 Overall Average
Based on Undergraduate Student Results record data as recorded at Trinity Term 2005 on 05.10.05, for the degrees of: B.A., B.F.A., B.Th., M.Biochem., M.Chem., M.Earth Sci., M.Eng., M.Math., M.Phys ( the latter M.* are undergrad degrees despite the names ! )
College percentage of Firsts achieved: Sorted by Rank
%Firsts Pooh-Rank
38 1 St John's College
37 2 Merton College
37 2 Balliol College
34 4 Magdalen College
32 5 University College
30 6 Hertford College
29 7 Wadham College
28 8 Christ Church
28 8 Keble College
27 10 Exeter College
27 10 St Anne's College
26 12 Jesus College
26 12 New College
24 14 Somerville College
24 14 Mansfield College
23 16 Corpus Christi College
23 16 Trinity College
23 16 St Catherine's College
23 16 Queen's College
22 20 Worcester College
21 21 Pembroke College
21 21 Brasenose College
20 23 St Peter's College
19 24 St Hilda's College
19 24 Lincoln College
19 24 St Edmund Hall
19 24 St Hugh's College
16 28 Oriel College
14 29 Lady Margaret Hall
4 30 Harris Manchester College
25 Overall Average
The thing is thought, it doesn't look that different.
I mean there aren't many colleges I can use as markers but I know we were 13th in the Norrington this year - and your table puts us @ 12
Merton always come very high, and they're 2
Hildas always comes near the bottom and they're 24th
I mean there aren't many colleges I can use as markers but I know we were 13th in the Norrington this year - and your table puts us @ 12
Merton always come very high, and they're 2
Hildas always comes near the bottom and they're 24th
Bekaboo
The thing is thought, it doesn't look that different.
I mean there aren't many colleges I can use as markers but I know we were 13th in the Norrington this year - and your table puts us @ 12
Merton always come very high, and they're 2
Hildas always comes near the bottom and they're 24th
I mean there aren't many colleges I can use as markers but I know we were 13th in the Norrington this year - and your table puts us @ 12
Merton always come very high, and they're 2
Hildas always comes near the bottom and they're 24th
It reveals that the differences in the percentages are minimal for most colleges. Hildas has nearly as many firsts as Brasenose ! ( 19 % for Hildas and 21% for Brasenose ) and is pretty close to Corpus Christi, St Caths and Trinity !
%Firsts
23 Corpus Christi College
23 Trinity College
23 St Catherine's College
21 Brasenose College
19 St Hilda's College
Thats really cool 
- good effort 
If nothing else its another table to cross compare with the Norrington table and as you say it highlights how close all the colleges are to each other. I think its a nice little indication that it doesnt really matter where you go... its more individual talent than anything else.
Plus Keble is 8th
- I cant complain 
Btw are these stats taken over one year or multiple years?
- good effort If nothing else its another table to cross compare with the Norrington table and as you say it highlights how close all the colleges are to each other. I think its a nice little indication that it doesnt really matter where you go... its more individual talent than anything else.
Plus Keble is 8th
- I cant complain Btw are these stats taken over one year or multiple years?
theoretically, that could still be flawed to an extent. what's a first? college X could get 38% of 1sts which all scored between 70-73% whilst college X could get 37% of 1sts which all scored between 72-84, with the majority falling at the later end of the bracket. just hypothesising.
silence
theoretically, that could still be flawed to an extent. what's a first? college X could get 38% of 1sts which all scored between 70-73% whilst college X could get 37% of 1sts which all scored between 72-84, with the majority falling at the later end of the bracket. just hypothesising.
True.
But I do not have access to the raw marks.
Otherwise, I could have done pretty box and whisker plots and histograms
I have a copy of R http://www.r-project.org/ and I enjoy playing with it.
And in any case, the final degree classification will just say First or 2:1
That still ignores the finals gap. Hilda's is never going to rise up the Norrington Table, or yours, because statistically, women get less firsts than men at Oxbridge. Your table is very well done, though.
Russell_G
although u say that a 2:2 or 3rd is the kiss of death, why not adapt your table to take this into account. atm a college that got 40% firsts, 60% 3rds would come top.
True.
But I cannot think of an intuitive score to assign to a 3rd.
Should it be a 0 ? or a -5 ? or something else altogether, like the I.E.E.E specification of NAN ("Not a Number" ) http://en.wikipedia.org/wiki/NaN
sanjiv
I'm just wondering - how "easy" is it to get a 3rd? Compared to a 2:2 or 2:1?
What do you have to do (or NOT do) to get a 3rd?
What do you have to do (or NOT do) to get a 3rd?
Only 1.3% "achieved" a 3rd ( 43 of 3195) and only 0.3% failed ( 9 of 3195 )
I guess that you have to have serious drug, alcohol or psychiatric problems
( I include pathological procrastination and laziness in this category ) to get a 3rd or to fail.
After all, if one is bright enough to get into Oxford in the first place, one would not perform dismally due to purely intellectual or academic reasons.
I suppose that probably depends on the individual. I know people who do next to no work and yet are getting really high marks, and are clearly aiming for firsts, and I know people who work the whole time and don't get the marks to show for it. The average Oxford student would not be looking at a third I don't imagine.
well, in order to tell what you should do with a 3rd, we really have to decide what the purpose of the norrington's table actually is.
1) A competition between finalist students in the way of sport
2) An actual evaluation of teaching quality, resources and all of the other stuff that goes into enabling a student to do well
If it's 1, make a 3rd 0, a pass -1 and a fail -2 and keep the others the way they are.
If it's 2, take anything at 3rd level or lower out of the equation. The reason is that every college obviously provides the teaching, resources etc. to allow every student to get a 2.2 (probably a 2.1). Those who get 3rds obviously have something wrong with them and don't really reflect any of the qualities that are looked for in reason 2 in each college.
1) A competition between finalist students in the way of sport
2) An actual evaluation of teaching quality, resources and all of the other stuff that goes into enabling a student to do well
If it's 1, make a 3rd 0, a pass -1 and a fail -2 and keep the others the way they are.
If it's 2, take anything at 3rd level or lower out of the equation. The reason is that every college obviously provides the teaching, resources etc. to allow every student to get a 2.2 (probably a 2.1). Those who get 3rds obviously have something wrong with them and don't really reflect any of the qualities that are looked for in reason 2 in each college.
zoraster
2) An actual evaluation of teaching quality, resources and all of the other stuff that goes into enabling a student to do well
2) An actual evaluation of teaching quality, resources and all of the other stuff that goes into enabling a student to do well
but then - should we take into account the "value added" teaching quality?
i.e. compensate for features such as the numbers of first choice applicants a given college has to select from in each subject, rather than waiting for pooled applicants, or say, the mean entrance/aptitude exam scores or A2 points of applicants at their point of entry?
i think things can easily become self-perpetuating, as people have mentioned..
INteresting, poohbear.
The reason why your table (of Firsts) is SIMILAR to Norington (though superior) is that only a tiny fraction get thirds, and APPROX the same % get 2.1 + 2.2 in each college.
One further point: I think you table would be further improved by getting % that got 1st + 2.1, because a 2.2 and a third are a waste of time. A 2.1 or a 1st is a requirement for (say) the Bar.
The reason why your table (of Firsts) is SIMILAR to Norington (though superior) is that only a tiny fraction get thirds, and APPROX the same % get 2.1 + 2.2 in each college.
One further point: I think you table would be further improved by getting % that got 1st + 2.1, because a 2.2 and a third are a waste of time. A 2.1 or a 1st is a requirement for (say) the Bar.
Related discussions
- tompkins table inquiry for cambridge
- what oxford college?
- Corpus Christi
- hypothesis Testing with only > Mean side of z table
- Complete University Guide Table 2024
- BTEC Level 3 IT Unit 2 Exam May 2024
- Paris 2024 Paralympics: Para table tennis
- Paris 2024 Olympics: Table Tennis
- Cie igcse computer science paper 2
- Would someone get prosecuted
- Is King's College really good?
- BTEC IT Unit 2 January 2024 exam discussion
- I forgot to cite a table for my dissertation for my AP
- First choice Uni dropped by 23 places in league tables
- is kcl good?
- OCR A - A level chem - do we need to know history of the periodic table
- Paris 2024 Paralympics: Para Triathlon
- The Sports Hurt/heal game
- Do you get given a college time table(uk)?
- AQA NEA Computing - URGENT !!!
Latest
Trending
Last reply 1 day ago
Oxbridge applicants for modern languages - 2025 entry (applying in 2024)University of Oxford
62
116
Last reply 1 month ago
Any Oxford Phil & Theo applicants heard back about interviews?
