The curtains on CAT 2013 were drawn on Nov 11th. However, the final word has not been said.
While CAT 2013 had all the trappings of a CAT exam, similar to those of the past three years, it left more questions than answers.
The original issues with an Online Aptitude Test conducted over just 20 odd days over 40 different sessions remain unsolved.
The changes in the difficulty level from one session to another were marked. If there were more questions of Algebra throughout all the sessions, the difficulty of some of these were beyond compare. You were left wondering if you were attempting a JEE exam.
Add to the fact that despite a gag over revealing questions (Copyright!), there was plenty of sharing of these questions. Some question (s) were repeated so often that the students who wrote the exam in the later dates were extremely pleased to find the question and gleefully mark the correct answer. This was undue advantage to students who had the network to gather the questions before they wrote their exam.
The famed network of the IITs, BITSians, NITs and some top Engineering and Degree colleges in the Metros had all the advantages.
The IIMs profess (in their oft repeated proclamations) that they want to get more Non-Engineers and bring more diversity to the MBA class.
If anything, the conduct of the exam and the contents of the exam have made the diversity a distant dream! Unless the IIMs give EXTRAAA marks for the non-engineers (esp. the BAs, BComs, BPharms) there is no way there is going to diversity at the campuses in the coming academic year.
Overall, the CAT paper (if one can call it a single paper!) was DIFFICULT. There were a lucky few who had easier sets of questions. However, there is no guarantee for these students (despite doing well) that they would get a higher percentile.
Some of the issues of CAT have remained unanswered for five years now.
- How would the raw scores be normalized to get Equated scores?
- Since NO two students are getting the same paper, then how is each question, and therefore the entire paper evaluated?
- Does this mean that the questions are categorized into difficulty levels and each students get equal number of easy, medium and difficult questions (assuming there are only three levels of difficulty)?
- Does that then mean that each question has a differential marking depending on the difficulty level of that question? What about the negative marking then? Would that also follow the same principle?
- Would accuracy be rewarded more if the scores were same for two different candidates?
- A little more analysis tells that it is impossible for each one to get exactly the same overall level of difficulty in the paper? If that were the case, how would normalization happen?
- With quite a few questions getting leaked in various private fora, there is little chance for CAT to be a level playing field. In fact, students from non-elite institutes and from tier-2/tier-3 cities may be at a distinct disadvantage?
Well, there are more unanswered questions than the number of questions in each section of CAT (pun, if there is one, intended)!
But, let’s still try to analyse and give some clarity, direction to students who have written CAT. Some assumptions have to be taken to be able to analyse and predict cutoffs to help students to plan their B School applications.
Assumption 1: That almost all students get similar level of difficulty and hence a student getting a score in any one slot is going to get a very similar percentile as another getting an identical score in any other slot.
Assumption 2: That there is no difference in any person writing the exam in the early part of the CAT window as compared to a person writing the exam in the later half of the CAT window. (A restatement of the above!)
CAT 2013 – Paper Analysis
Summary of the Paper!
CAT was difficult and getting around 90 marks out of 180 would catapult a student to get more than 98.5 percentile at the least!
In Section I, getting a score of 30 could fetch one a percentile around 88-91.
In Section II, getting a score of 30 could fetch a percentile around 83-85.
To get a 99 percentile, one had to score in the range of 42-45 in Section I and around 46-49 in Section II.
Section I – Quantitative Aptitude & Data Interpretation (30 Qns)
The pattern of the paper in terms of division of questions between QA and DI was by and large same. Most students got 21 Qns in QA and 9 in DI whereas some of the others got 20 Qns in QA and 10 in DI.
Geometry and Algebra were all over the place. Polynomials, Functions & Graphs, Inequalities, Quadratic Equations were given in plenty. The number of questions in Arithmetic was in the range of 3 to 5 questions (out of 21) to a majority of students. Some sprinkling of Trigonometry, Coordinate Geometry did stump a few students. Around 10 questions were almost always doable, whatever the difficulty level of the overall paper. There were sitters as well as very difficult questions in each and every slot (and for every student)
A good student would have attempted around 12-14 questions in around 40 minutes.
There were 3 sets of DI for almost all students. While the majority of students got all traditional DI sets – Tables, Bar Charts, Line Graphs, Pie Charts, Combination graphs, etc. there were a few who got a set of Venn Diagrams questions instead of a third set of DI. At least two of the three sets were doable and one set was either difficult or lengthy.
A good student would have attempted 6 to 8 questions in around 25 minutes.
VistaMind has done a survey on a vast number of people to put together the average number of attempts in Section-I and it was found that the average order of attempts was in the range of 15 to 17 questions. However, the top 10 percentile of the students had attempted in the range of 19-20 in the first section.
Students have proclaimed that the actual CAT paper resembled many of the Smart CATs and the students whose slot had more difficult questions opined that the GENIUS SmartCAT conducted by VistaMind was almost a replica of the real CAT!
The scores and therefore the percentiles would depend on the accuracy levels that one has in the paper.
Section II – Verbal Ability, Reading Comprehension, Logical Reasoning (30 Qns)
The pattern was similar to that of CAT 2012. There were about 10-12 questions in Verbal Ability, 9-10 questions in Reading Comprehension (3 RC passages) and 9-10 questions in Logical Reasoning (3 sets). As can be seen from the above pattern, some students probably were lucky to get 10 questions each in RC and LR.
The verbal ability area had hardly any surprises. Para Jumbles, Para Completion, FIBs, Grammar, Vocab Usage were the usual areas given to just about every student in every slot. Whatever the level of preparation, it is almost impossible to predict if one has done well in Verbal Ability area. Especially Para Jumbles and Para Completion (not to talk about Grammar) questions can be tricky and with the time pressure and CAT exam pressure, a perfectly innocuous question can appear draconian.
A good student would have attempts close to 7 or 8 questions in this part of the paper.
There were 3 sets of passages in the paper. Once again, there were at least two sets of passages and questions direct and not very difficult. However, one set of the three was unusually difficult and for some abstruse. Each of the RC sets was around of 500 – 700 words.
A good student would have attempted at least two sets and around 6 questions.
As per expectations, there were 3 sets of logical ability. Quite a few students got simple sets in Arrangements – Sequential as well as Circular, Distribution, Selection, Networks as well as numerical puzzles. Two sets out of the three were normally doable whereas the third set was difficult and time consuming.
A good student would have attempted 6 questions and would most probably have got all of them right! A very good student would have got all 9 correct.
Average Overall attempt in Section-II was in the range of 20-22. However a top 10 percentile of the students would have attempted closer to 23-25 questions.