Normalisation kills CBSE/ISC scorers: a devastaing example with actual data sources

This article was submitted by Chandrasekhar Balachandran, a candidate who appeared for KEAM 2011. This will provide an insight into how a CBSE student is subject to a shot in the back as per normalization and the data of mean and standard deviation released by the CEE website. This article uses the data exactly as per in the official documents released to arrive at the “after normalization” section. No guess work was done whatsoever. However, in case you find something here that is totally wrong, comment and let us know. We will publish your side of the story after verification.

To know the standard deviation and mean marks of different boards, refer the official KEAM 2011 Highlights. To know how we calculated the normalised marks and the shifts, read the official document on How Normalisation Works


A SAMPLE SCORE SHEET

“To a question at a subsequent question-and-answer session, he said normalisation would imply an inherent advantage for the Kerala stream in relation to other streams, including the CBSE and the ISC.” – The Hindu, 11 May 2011 

Consider a sample mark sheet of two students from different syllabi:

STREAM    ENTRANCE MARK    BOARD MARK   TOTAL

CBSE           320/960                     240/300             560/1260 

KERALA     280/960                     317/340             560/1260

WHY WE CHOSE SUCH A SCORE SHEET

First, let me explain why we put in a variation in board marks for the two students. CBSE Physics, Chemistry and Math exams have a maximum scorable mark of 100. The maximum marks for Physics and Chemistry for State syllabus is 120. Their Math paper, however is out of 100. So we chose a total score 317/340 which when converted into 300 marks will be approximately equal to 280. Thus total score of both students when the score is adjusted without normalization(entrance mark out of 960 and board marks out of 300after aforementioned conversion) will be approximately equal.

State student Board Marks conversion: (317/340)*300 = 280

To find total score: 280 (Entrance marks) + 280 (converted Board marks) = 560/1260

Let the marks of CBSE student be 90, 70 and 80 for Maths, Physics and Chemistry respectively. When subject to normalization, the data takes a swing.

AFTER NORMALISATION

(calculation done as per the data given in KEAM 2011 Highlights and How Normalisation Works, both released officially by CEE Kerala official website)

SCORE OF CBSE STUDENT

SUBJECT     INITIAL   NORMALISED   SHIFT

Maths           90            81.6719             -8.3218

Physics         70           78.4805              +8.4805

Chemistry    80           81.6614              +1.6614

TOTAL SCORE AFTER NORMALISATION: 241.8138/300

Converting this to 500 we have a final board mark score 403.023/500

Since the State result does not undergo normalization mathematics (because it is taken as the basis of comparison) we can directly convert the State student’s score into 500:

 (317/340) * 500 = 466.1765

Let us just take it as 466 to keep it simple (We are not that stingy to complain about each fourth decimal point, but be aware four decimal places are taken according to the CEE official documents).

THE CLIMAX OF THE ARTICLE

Now we can finally get to the point. Take a look at the table and see how conveniently this is a crown of thorns for the CBSE student.

STREAM    ENTRANCE MARK   BOARD MARK    TOTAL

CBSE            166/500                             403/500               569/1000   

KERALA        145/500                        466/500               611/1000

Now take another look: Those who had equal marks in 1260 are separated by 42 marks when it  came to 1000 after normalisation. Is it me or is this the way it is supposed to work: giving state sylabiian an advantage by putting the CBSE student at undue disadvantage?

OUR SUGGESTIONS

An exam when conducted should give each student a fair shot at it. Reforms should not be introduced so that a particular set of students are given a push to the top. If the government expects the stat students to fare better, I suggest they improve the faculty, conduct continuous valuation and improvise on the teaching methods. 

• Related post: Why I Think 50-50 And Normalisation Is Educational Blasphemy

Image Squashed tomato: Boaz Yiftach / FreeDigitalPhotos.net

Image 3D Pie Chart: digitalart / FreeDigitalPhotos.net

Image: renjith krishnan / FreeDigitalPhotos.net

Read more »

Sneak Peek: A devastating example using actual data references

Chandrasekhar Balachandran has come up with a devastating live example indicating clearly why normalisation hits CBSE and ISC students badly. The initial draft has been recieved. Currently his example calculations are undergoing mathematical cross-checking. The article will be published at noon at 2:00PM IST (or sooner depending on the verification).

Chandrasekhar explains clearly using sample marks of two students from different streams and his basic high-school Mathematics skills how hard hitting normalisation can be. If you have never understood why this is serious, or if you haven't yet been convinced, this article will be a mathematical side of the proof why we think this is system is bad.

Stay tuned everybody.

Image: jscreationzs / FreeDigitalPhotos.net

Read more »

Why I think 50-50 and Normalisation is Educational Blasphemy

This article is meant for one thing and one thing only. It aims to shed light onto a few yet-not-looked-into facts that are seemingly open loopholes in the "equal weightage" and "normalisation" systems that have been adopted in the implementation and publication of Kerala Engineering Entrance Exam of KEAM 2011. This will be an expose of the flaws in simple yet convincing words so that you don't have to complete a degree to just understand this whole aritcle. References have been given to all related articles some directly from KEAM website and to a few reliable news sources.

FIRST OF ALL, ITS NOT EQUAL WEIGHTAGE

For the preparation of Engineering Rank list, equal weightage was given to
the marks obtained in the Entrance Examination for Engineering (Paper I &        
Paper II put together) and the grade/marks obtained for Mathematics, Physics and
Chemistry/Computer Science/Biotechnology/Biology put together in the Second
year of the qualifying examination (Plus  Two or equivalent), after effecting the
normalization procedure as described in Clause 9.7.4 (b)(ii) of the Prospectus.  (Page 2- Highlights of KEAM 2011) see more  

No, its not equal weightage. Its sounds fine: entrance marks are taken out of 500 and boards out of 500. The end result will be out of 1000 where each of the exams were given 50-50 weightage, right? WRONG!

Think this: a typical exam would be of maximum marks say 100. We have three subjects here: Physics, Chemistry and Maths. So our total marks of boards will be out of 300. This value gets converted its equal corresponding value of a maximum marks of 500, according to the Prospectus. Lets see an example..

Suppose my score was 80 in Phy, 90 in Chem and 75 in Math after normalisation. My total now would be 245 out of 300. This will translate to 408.3333 out of 500 according to the rules.  The important thing here to notice is that my marks, now after rules implementation has INCREASED from 245 to 408.3333.

Now take a look at my sample entrance score.. Lets say I have a total score of 400 marks from my Paper I (Physics and Chemistry)  and Paper II  (Mathematics). Thats 400/960 which will translate into 208.3333. Here keep in mind my entrance score has DECREASED. 

Lets put it together: my board marks get increased when converted into a 500 marks score whereas my entrance score has decreased while both are maintained at a grand total of 1000. This means every one mark I score in entrance has less value than the one mark I scored in my board exam. Take a look:

Board exam 1 mark conversion:  (1 / 300 ) * 500 = 1.6667

Entrance exam 1 mark conversion: (1/960) * 500 = 0.5208

Clearly, though it is supposed to be equal weightage, my score for Entrance has less value than that of my Board exam (in fact, it has halved! ). Conclusion: Equal weightage does not exist. You can fool a layman by saying its both out of 500 and hence it must have equal value.. . In reality, a little 5th grade Math does not agree.

NOTE: Only in the case of OISCA Academy Senior High School, Japan and Tamil Nadu Board of Higher Secondary Education is the total greater than 500 (each subject being of max mark 200). In this case the marks of the students will decrease.


SECONDLY, NORMALISATION PROCESS IS UNJUST

In an article published in The Hindu a paragraph reads:

Dr. Krishnan said that any swing — either upward or downward — of scores in various streams would be evened out during the process of normalisation, as it would be applied equally to all candidates in that stream.

This is perfectly acceptable given data is collected from every board that has candidates from appearing for the exam. However, this was not the case. Frankly I'm impressed that they got so far. It shows they put in a lot of effort. CEE Kerala got data from boards such as United States Office of Education, Illinois and National University of Singapore. But a few were reported to be missed out (source will be provided on next update). These students' marks  has been compared to that of CBSE values. So the actual "normalisation" does not take place. If they had a tougher exam than the CBSE and their mean value was substantially lower (or higher) than that of CBSE (note here even one mark average is important as seen in the above case) justice is well.. not served. Even though its a minority case, they are students too. They wrote the exams too. They deserve the right normalisation too.

This implies the very reason for implementation of  normalisation procedure fails as these values taken- though 'standard', are on a false base.



THIRDLY, AFTER NORMALISATION AVERAGES WILL VARY FROM INITAL EXPECTATIONS

This I found out on the day of declaration of result. Its difficult to explain all the math involved here but here is what happens:

Person 1 has 80,90 and 60 in Phy, Chem and Math respectively.
Person 2 has 60, 60 and 70 in the exams.

Their total after converting to 500 without normalisation will be around 383 and  316 respectively.

But after normalisation based on the data, it was found that Person 1 has around 9 marks less (374) and Person 2 has slightly more lets say a 0.6 marks gain.

Notice here that Person 1 has lost a good 9 marks from normalisation. If you don't see that happening, you can take my word for it. It has happened.  Take a look at who else is complaining about this.

I can only say : "wierd".



The future of a lot of smart students is at stake. It is for us citizen to ensure that these kids get where they deserve to be. The Entrance exams were brought in so that there would be a fresh system free from previous errors for admission to colleges. Change with reason and well researched solutions is perfectly acceptable. But as of now there is none so significant to have the method revised. If the present government is convinced of the shortcomings of the newly adopted system, it should do away with the new system and publish the ranking based on the Entrance marks alone. This will not only create order in a world of chaos but will ensure that the in future more thought will be put in before any revolutionary plan is put to practice.


NOTE: If you disagree with the points above, state so in the comments section with a link to a reliable source or with convincing proof; I'll be only happy to be enlightened on my bad. Have anything to add? Feel free..
This article will be updated soon. Stay tuned.
Image: Fingers Pointing - Teerapun / FreeDigitalPhotos.net
Image: Tick jscreationzs / FreeDigitalPhotos.net

Read more »