SSC CHSL Tier 1 – Most Repeated Questions

The Tier 1 Blueprint and the Repeat Map

A declaration first, because you may have arrived here expecting something else. Every question in this book is original and newly written to the published SSC syllabus and to the exam's own question pattern. No past paper is reproduced, and nothing is taken from any other publisher's question bank. SSC does not maintain a public archive of CHSL question papers, so no honest book can hand you one — what it can hand you is the set of question types that recur cycle after cycle, and the skill to solve each on sight. That is what this chapter builds. It does two jobs. First, it makes the paper itself concrete: what the 60 minutes actually contain, what each question is worth, what a wrong answer really costs, and how to spend the clock. Second, it lays out the repeat map — the master table of recurring question archetypes across General Intelligence, General Awareness, Quantitative Aptitude and English, each banded by weight and pointed at the chapter that teaches it. Read this chapter with a pen. The marks arithmetic here is not decoration. In a paper where the difference between a selected and an unselected candidate is often a handful of marks, understanding that a wrong answer swings your score by 2.5 marks — not 0.5 — changes how you sit the exam. Scope note: this book covers Tier 1 only. Tier 2 has a different structure, including a Computer Knowledge module and a Skill or Typing Test, and none of that is taught here.

KarmSakha preparation topics covered

• The Tier 1 CBT structure: four parts of 25 questions, 2 marks each, 100 questions / 200 marks in one composite 60-minute sitting
• Negative marking of 0.50 per wrong answer, and the exact expected-value mathematics of when a guess pays
• Break-even guessing accuracy, and the expected value of a guess with 4, 3 and 2 surviving options
• Net-score formulas, and why accuracy beats raw volume of attempts at this marking scheme
• A minute-by-minute 60-minute allocation plan with the seconds-per-question rate each block implies
• Bilingual paper rules, the Group C post family Tier 1 feeds, and what is notice-controlled and must be verified on ssc.gov.in
• The honest methodology behind this book's repeat map, and why frequency is banded High / Medium / Low instead of counted
• The master table of recurring question archetypes across all four sections, each mapped to its teaching chapter

What the Tier 1 Paper Actually Is

SSC CHSL Tier 1 is a computer-based test, or CBT — you answer on a terminal, not on paper. It is one paper, not four papers. The 100 questions are loaded in four numbered parts and a single clock runs across the whole sitting. Every question is objective with four options. A correct answer earns 2 marks. A wrong answer costs 0.50 marks. A question you never touch costs nothing at all.

Notice the consequence of equal marking. Because every question in every part carries the same 2 marks, a General Awareness item you recognise in eight seconds pays exactly what a mensuration problem that eats two minutes pays. The paper does not reward difficulty. It rewards marks collected per minute. That single fact drives almost every strategic decision in this book.

There is no sectional timing inside Tier 1. The 60 minutes are composite — you may spend them in any proportion across the four parts, and you may jump between parts freely using the on-screen question palette. Free navigation is an advantage only if you have decided the plan before the clock starts. Candidates who improvise usually discover at minute 50 that they have 25 unread Quantitative Aptitude questions left.

Language and posts. Three of the four parts are set bilingually, in English and in Hindi. Part IV, English Language, is set only in English — testing English through a translation would defeat the purpose. Tier 1 feeds a family of Group C posts: Lower Division Clerk and Junior Secretariat Assistant, Data Entry Operator and Data Entry Operator Grade 'A', and, in cycles where the Department of Posts participates, Postal Assistant and Sorting Assistant. Which posts appear in any given cycle is decided by that year's notice, not by tradition.

The Arithmetic of Negative Marking

Most aspirants treat negative marking as a vague threat. Treat it as arithmetic instead, and it becomes a tool. Let p be your probability of getting a particular question right. Marking it gains 2 marks with probability p and loses 0.50 marks with probability (1 − p). The expected value of that mark, written EV, is:

EV = 2p − 0.50(1 − p) = 2p − 0.50 + 0.50p = 2.5p − 0.50

Set EV = 0 to find the break-even point. 2.5p = 0.50, so p = 0.50 / 2.5 = 0.20. Read that carefully: you only need to be right 20% of the time for marking an answer to be worth exactly as much as leaving it blank. Above 20% accuracy, marking gains you marks in the long run. Below it, you bleed.

Now apply it. A blind guess among four live options has p = 1/4 = 0.25, which is already above the 20% break-even, so EV = 2.5(0.25) − 0.50 = +0.125 marks. Eliminate one option and p = 1/3, giving EV = 2.5(0.3333) − 0.50 = +0.333 marks. Eliminate two and p = 1/2, giving EV = +0.75 marks. Every elimination roughly doubles the payoff.

Two honest cautions, because the mathematics alone is not the whole story. First, +0.125 marks is a tiny edge; it is real, but it is not worth 40 seconds of clock you could have spent on a question you would definitely get right. Guess blindly only in the last 60 seconds, on questions you were never going to attempt anyway. Second, EV describes the long-run average across many questions, not any single question — over 20 blind guesses you expect about 5 right and 15 wrong, which nets 10 − 7.5 = +2.5 marks, but any one guess is a coin toss.

The other number worth burning into memory is the swing. Turning a right answer into a wrong one does not cost you 0.50 marks. It costs the 2 you did not gain plus the 0.50 you did lose: a swing of 2.5 marks. One careless misread in Quant is worth more than one whole extra correct answer.

Accuracy Beats Volume — With Numbers

Coaching folklore says 'attempt more'. The marking scheme disagrees, and you can prove it in thirty seconds using Net = A(2.5a − 0.5).

Candidate P attempts 90 questions at 70% accuracy. That is 63 right and 27 wrong: Net = 2(63) − 0.5(27) = 126 − 13.5 = 112.5. Candidate Q attempts only 75 questions but at 88% accuracy. That is 66 right and 9 wrong: Net = 2(66) − 0.5(9) = 132 − 4.5 = 127.5. Q attempted 15 fewer questions and finished 15 marks ahead. The extra 15 attempts P made were worth, on average, less than nothing after the penalty absorbed them.

The break-even logic explains why. Marks come from the bracket (2.5a − 0.5). At a = 0.88 each attempt is worth 1.70 marks; at a = 0.70 each attempt is worth 1.25 marks. Volume multiplies whatever that bracket contains, so raising accuracy raises the value of every attempt you have already made, while raising volume only adds more attempts at the value you already had.

This is also why per-minute return, not difficulty, should decide what you touch. Suppose you have three minutes. Option A: crack one hard Quantitative Aptitude question you are confident of, for +2 marks — a rate of 0.67 marks per minute. Option B: clear nine General Awareness questions at 20 seconds each with 60% accuracy — that is 9 × (2.5 × 0.60 − 0.50) = 9 × 1.0 = 9 marks, a rate of 3.0 marks per minute, about four and a half times better. The hard question feels more valuable. It is not.

So the working rule is a three-pass sweep of the paper. First pass: answer only what you recognise instantly, in every section, and never linger. Second pass: return to questions that need work but have a clear method. Third pass: in the final minutes, use eliminations on whatever is left, because at 20% break-even almost any informed guess is positive. Mark-for-review is your friend on pass one and your enemy on pass three.

• Pass 1 (about 30 minutes): sight-readable questions only. If a question does not resolve in the time budget for its section, flag it and move. Do not re-read.
• Pass 2 (about 24 minutes): flagged questions with a known method — multi-step arithmetic, seating arrangements, cloze passages, comprehension.
• Pass 3 (about 3 minutes): eliminate and mark. Anything where you can strike out even one option is worth +0.33 marks on average.
• Never re-open a question you have already answered confidently. Re-checking a correct answer gains 0, but talking yourself out of it costs 2.5.

The 60-Minute Allocation Plan

Sixty minutes across 100 questions averages 36 seconds per question. That average is misleading, because the four sections are not equally expensive. General Awareness is pure recall — you either know it or you do not, and staring longer will not manufacture the fact. Quantitative Aptitude is computation, and computation takes time no matter how well you have prepared. So the budget should be deliberately unequal.

The plan below front-loads the cheap sections and protects the expensive one. Start with General Awareness because it converts knowledge into marks faster than anything else in the paper and because clearing it early gives you a psychological score floor. Take Reasoning next: it is the highest-accuracy section for most candidates once the archetypes are drilled. English third. Quantitative Aptitude last, with the largest block, because it is the only section where extra time reliably converts into extra marks.

Hold yourself to the per-question rate, not just the block total. If General Awareness is budgeted at 5 minutes, that is 300 seconds for 25 questions — 12 seconds each. At 12 seconds you either recall or you skip. Any GA question that has taken 30 seconds is a question you do not know, and the extra 18 seconds were stolen from Quant.

Practise with this clock, not just with the questions. Run at least one timed full paper each week and record your split per section. If you consistently finish Reasoning in 10 minutes rather than 13, move the spare 3 minutes into Quant permanently and rewrite your plan. The allocation is a starting hypothesis to be tuned against your own data, not a rule handed down.

How This Repeat Map Was Built — And What It Is Not

You deserve to know exactly how the frequency bands in the next table were derived, because a great deal of what circulates online on this subject is invented.