Analysing the English section of CLAT mocks

Taking a mock tells you your score. Analysing it tells you why, and only the "why" changes next week's score. Many candidates take mock after mock with flat results because they look at the score, glance at the answer key and move on. The same five mistakes then appear in every paper.

For English, a proper review takes about as long as the section itself. That is time well spent.

The review routine

Do this on the same day as the mock, while you still remember how you thought.

1. Record the numbers. Questions attempted, right, wrong, skipped, and the time you spent on the section.
2. Sort every question into one of four buckets (below). Include the ones you got right.
3. For each wrong or unsure answer, find the line in the passage that settles it. Write the line down.
4. Name the error using a fixed list, so patterns show up across mocks.
5. Re-solve the skipped questions untimed. Could you have got them with more time? That tells you whether the problem is speed or skill.
6. Check your passage order. Did you start with the passage that suited you best, or just the first one?
7. Write one lesson for the next week. One, not five.

The four buckets

The second bucket is the one most candidates ignore. A lucky right answer is a mistake waiting to happen in the real exam.

The error log

Keep one running log for English, across all mocks.

After five or six mocks, sort the log by the Error column. The two most frequent errors are your study plan for the next month.

A worked analysis

Suppose a candidate's mock English section looks like this. The numbers are an example, not a benchmark.

• Questions: 24. Attempted 21, right 15, wrong 6, skipped 3.
• Score: 15 − (6 × 0.25) = 15 − 1.5 = 13.5.
• Time: 31 minutes, against a plan of 25.

Sorting the six wrong answers:

Two more answers were right but unsure, both on inference.

What this tells the candidate:

• Inference is the real weakness: three wrong and two shaky, five of 24 questions.
• The EXCEPT and vocabulary errors are careless. Fixing just those two would have given 17 right and 4 wrong: 17 − 1 = 16, a gain of 2.5 marks with no new knowledge.
• The six extra minutes probably came from rereading on the inference questions, not from slow reading.

The one lesson for the week: apply the must-be-true test to every inference option and write the supporting line. Practice: ten inference questions a day for seven days, from the inference guide and sectional tests.

Looking at the missed question

Here is one of the three inference errors, re-created.

The city's new cycle lanes have been praised and attacked in equal measure. Shopkeepers on the main street say they have lost customers who used to park outside. The council's own counts, however, show more people passing the shops on foot and by bicycle than before the lanes were built. Whether those passers-by spend as much as drivers did is a question the counts cannot answer.

Which of the following can be inferred?

(a) The shopkeepers' sales have fallen.
(b) The council's counts do not settle whether the shops are better or worse off.
(c) Cyclists spend more than drivers.
(d) The cycle lanes should be removed.

The candidate chose (a). The correct answer is (b).

• (a) is what the shopkeepers "say". It is a reported claim, and the passage never confirms it. Treating it as fact is the probably-true trap.
• (b) follows from the last sentence: the counts show footfall, not spending.
• (c) is the opposite gap: the passage says this is unknown.
• (d) is an opinion the passage never takes.

Log entry: inference · chose (a) · correct (b) · reported claim treated as fact · "a question the counts cannot answer" · check who is making each claim.

Timed versus untimed accuracy

Once a fortnight, redo a past sectional without a timer. Compare the two accuracy figures.

Track the trend, not the score

A single mock can mislead. Passages vary in difficulty, and so do you. Track a rolling average.

Mock 3 looks like a setback, but the average of the four is (13.5 + 15 + 12.75 + 16.25) ÷ 4 = 57.5 ÷ 4 ≈ 14.4, and the trend is rising. Judge progress on the average of the last three or four mocks.

Practice: diagnose the error

Name the error type in each case: extreme option, outside knowledge, misread stem, reported view, time plan, or dictionary meaning.

1. The author wrote that "some regulation has gone too far". A candidate chose "The author opposes all regulation".
2. A candidate chose an option about a real government scheme because they knew it to be true. The passage never mentioned it.
3. The question asked which statement was NOT supported. The candidate chose a statement the passage clearly supported.
4. The passage said "critics argue that the tax is unfair" and then defended the tax. The candidate answered that the author finds the tax unfair.
5. A candidate spent eleven minutes on the first passage and guessed on all of the last passage.
6. Asked what "arrest" meant in "arrest the decline", a candidate chose "take into custody".

Answers

1. Extreme option. "Some" became "all".
2. Outside knowledge. True in the world, absent from the passage.
3. Misread stem. They answered the opposite question.
4. Reported view. The critics' view is not the author's.
5. Time plan. Set a limit per passage and move on when you reach it.
6. Dictionary meaning. In context, "arrest" means stop or halt.

What to do next

• After your next mock, sort every English question into the four buckets and log each error.
• Choose one lesson for the week and practise only that.
• Redo one sectional untimed this fortnight and compare the two accuracy figures.
• Use improving English accuracy for fixes to each error type.