Re: Very important CLAT Notification

[Shamnad]

Thanks Dipto,

Clatapult recommends ignoring the negative marking. Comments? Ramanuj?

http://clatapult.com/clat-2013-pattern-answer-every-question-ignore-the-negative-marking/

On 23-Dec-2012, at 3:23 PM, diptoshree basu wrote:

Dear Teamleaders

Please take note that CLAT 2013 will impose negative marking of 0.25 marks for each wrong answer.  Also, there will be static GK included in the general knowledge section. Please find the notification here:  http://www.clat.ac.in/patterns.php# 

Please make sure that each of the scholars are informed about this and are taught strategies on cracking papers which have negative marking system.

Thanks and Regards

Dipto

--

Diptoshree Basu

Executive Director,

IDIA

[Shamnad]

Thanks Hersh,

Can you respond specifically to Shouvik's concern that the probability hypothesis may not square out too well in a small subset of the whole paper (the portions for which the student does not really know the answer and cannot make a very calculated guess)...

Thanks,

SB

On 25-Dec-2012, at 11:24 AM, Hersh Sewak wrote:

Mathematically, Tanuj's argument is spot on. Presently, I will secure a benefit of 2.5 marks for marking the same options for 40 questions, ie, 6.25% or 0.06. Earlier it was 10 marks, ie, proportion of .25 (25%).

However, it will work more accurately with greater number of questions, that is, the chances to secure positive score will increase as the number of questions which have to be guess-answered increases. 

Example, I have 10 questions to guess answer, then simple rule of percentage will say, I ought to secure .6 marks (or to be more accurate somewhere around .25) However, the error margin is high enough, to also make the score zero or negative. Consider that out of the 10, only 2 were correct, then I score zero; and if 4 were correct, then I score 2.5.

However, if I have 60 questions which I have to guess answer, then the 6.25% will give more accurate score. 

Why? Because, as the aggregate increases, the answer begins to shift towards the expected mean, that is 6.25%

Conclusion: In all, Tanuj's point will work if the candidate is in danger of leaving/failing to answer a large chunk of the paper. It would however, yield unpredictable result if the number of such questions is less, that is, somewhere around 10-20.

All in all, it toughens the answering strategy for kids vis-a-vis their earlier counterparts. 

Hope it helps :)

H.

[sha...@gmail.com]

Tanuj surely can't be catering advice to only those that know a small percentage of answers. And if he is not so catering, then he is squarely wrong..

Sent from BlackBerry® on Airtel

---

From: shouvik kumar guha <shouvikk...@gmail.com>

Sender: idiatea...@googlegroups.com

Date: Tue, 25 Dec 2012 23:13:22 +0530

To: <idiatea...@googlegroups.com>

ReplyTo: idiatea...@googlegroups.com

Cc: diptoshree basu<dba...@gmail.com>; IDIA Gujarat<idia...@googlegroups.com>; <idiawb...@googlegroups.com>; <idia...@googlegroups.com>; <idiane...@googlegroups.com>; <idiao...@googlegroups.com>; Shishira Rudrappa<shis...@spirecapital.in>; since 1985 ...........Nimojjito ochenay<85.a...@gmail.com>

Subject: Re: Very important CLAT Notification

I think Hersh has already voiced my concern too, by saying that the strategy may work better the more the number of questions that the student is unsure about. The closer this number comes to the total number of questions, the more likely is for this strategy to bear fruit. However, for it to be a sureshot benefit, the two numbers need to be so close as to mean that the student does not know at least 60% or more of the questions, in which case, I don't think he can be said to be properly prepared for the exam anyway.
Shouvik.