CAT 2018 Slot 1 Analytics

There are numerous questions that are needed to be solved before you take the CAT. Mind you, these are not questions that are a part of the syllabus. These are questions that directly affect your CAT test-taking strategy. For example, "How many questions should I attempt to score a 99%ile in CAT?", "What should my ideal accuracy be in each section?", and even "How many easy and difficult questions to expect per section?".

Here, we will answer a couple of these questions using the power of Machine Learning. Post CAT 2018, we asked aspirants to fill in their number of attempts in our score calculator. This gave us the cleanest data that there could be regarding the attempts of the students in CAT 2018.

Therefore, lets dive in to the data and figure out some interesting insights from it.

Attempts and Accuracy for 90+%ile

The graphs given below give us an interesting perspective of what does it take to get a 90+%ile in CAT.

The first graph, on the left hand side, suggests what number of attempts led to what percentile in CAT 2018. Reading this graph is simple. For example, for someone with a 95%ile in CAT 2018, an attempt of roughly 58-60 questions out of 100 questions was required. Similarly, for a 99%ile, 70-72 attempts were required in CAT 2018.

The second graph, on the right hand side, in the image given above, suggests what sort of accuracy leads to what kind of a percentile. Effectively, this graph states that ..............

How the toppers managed attempts and accuracy in CAT 2018 Slot 1?

What our insights suggest here is something that is extremely interesting. The graph given below should tell you that there is no one and only one way to get a 99%ile in CAT. With the number of attempts and accuracy being the variables, there are many combinations of the attempts and accuracy that can get you to a 99%ile.

In the graph, the color of the pointer suggests the percentile. For example, the grey dot signifies a 99%ile, a yellow dot signifies 95%ile and so on.

The graph plotted between the No. of Attempts and Accuracy, suggests multiple combinations to get a particular percentile.

For example, in order to get a 99%ile, one could have attempted just 60 questions in the entire paper with an accuracy of 85-90%.

At the same time, 99%ile could also be achieved at a lower accuracy of around 70% but with a higher number of attempts, as high as 80 attempts. This should given hope to you no matter what your current performance level might be. Which means that if you are a student who generally has a lower accuracy in mocks, you now know that the number of attempts for you have to be at what level. Similarly, if you have a high level of accuracy in the mocks, then a lower number of attempts will also be fine.

CAT Analytics of CAT 2018 Slot 1 - VARC Section

Our data science team delved deeper into each section and found out analytics that have the potential to entirely change the way you take CAT.

Have a look at this graph on the right side which is for the VARC section of CAT 2018. As the legend suggests, the colors signify the level of difficulty of each question. Here you must know what our definition of level of difficulty is over here.

Easy - A question is marked in Green color, i.e. easy, if at least 75% of the aspirants who attempted that question got it correct

Medium - A question is marked in Yellow color, i.e. medium, if 50% to 75% of the aspirants who attempted that question got it correct

Difficult - A question is marked in Red color, i.e. difficult, if less than 50% of the aspirants who attempted that question got it correct

Here it can be seen that there are exactly 5 out of 10 questions of Verbal Ability that were difficult. This means, that less than 50% of the students who attempted the question got it correct. What should also be seen is that, all of these 5 questions were non-MCQ questions.

This simply tells you the fact that cracking the non-MCQ questions in VA (which are usually the Para Jumble questions) is extremely difficult. Therefore, your focus in the VARC section should be on capitalizing from the questions on Reading Comprehension. With 24 out of 34 questions being from RCs, it becomes important for you to master the RCs.

Yes, it is worth mentioning that while RCs are non-negotiable, it is the VA section that can make all the difference for someone who is willing to get a 99+%ile.

CAT Analytics of CAT 2018 Slot 1 - DILR Section

DILR section is the Achilles heel for many CAT aspirants. Therefore, in order to truly understand the section, lets have a look at the naked picture of the section.

Here, the same definition of Easy, Medium and Difficult applies as mentioned for the VARC section. The CAT 2018 Slot 1 DILR section had 8 sets which are depicted in the graph on the left.

As you can see, there are two sets that are extremely easy as all the four questions of this set are easy. These are Q. 35 to Q. 38 and Q. 47 to Q. 50. Apart from these, there are a few other easy questions sprinkled across the entire section. This suggests that while you should be able to spot and definitely solve the easy sets, you should also look for the easy questions in various other sets.

At the same time, you should definitely stay away from extremely difficult sets, like the last set of CAT 2018 Slot 1 DILR. Therefore, a student should have ideally attempted the following sets in this paper:

Definitely Attempt - Set 1, Set 4, anyone from Set 3 and Set 7

Not Attempt - Set 5, Set 8

CAT Analytics of CAT 2018 Slot 1 - QA Section

The analytics of the QA section, which is the last section in CAT, gave a good overview of what should be your attempt strategy.

As you can see from the graph, there was a good number of easy and medium questions in the paper. Slot 1 of CAT 2018 has 12 easy questions along with 16 medium level questions.

For any candidate, these 28 questions should be the questions that you should be looking to maximise your score from.

While that is what should be the ideal strategy for any candidate, sadly students tend to miss out on a lot of easy questions in the paper. The most interesting insight from this is the fact that the last 10 questions of QA had as many as 7 easy questions. However, if you see the attempt ratio for these last 10 questions you will see that few students have attempted these questions.