Reasoning

Verbal & Logical Reasoning — Series, Analogy, Classification & Coding-Decoding

Verbal reasoning forms the backbone of every Reasoning Ability section in SSC, Banking, RRB, Insurance and State PSC exams. The four sub-topics below — Number/Letter Series, Analogy, Classification (Odd-One-Out) and Coding-Decoding — together account for 8–12 marks per paper and are the easiest, fastest-scoring questions if you spot the pattern in the first 20 seconds.

Exam relevance: SSC CGL/CHSL Tier-I: 8–10 questions. RRB Group D / NTPC: 6–10 questions. IBPS PO / SBI PO Prelims: 4–6 questions. State PSC Reasoning: 5–8 questions.

In-Depth Tutorial

A formal, accessible 800–1000 word walkthrough of this topic, written for the serious aspirant. Switch to हिन्दी using the toggle on the right.

What Verbal and Logical Reasoning Really Tests

Verbal and logical reasoning is the part of the examination that tests how clearly and systematically a candidate can think when presented with information in the form of words, numbers, letters or short statements. Unlike subjects such as history or general science, this section does not assess what you have memorised; it measures how quickly your mind can recognise an underlying rule, generalise that rule, and apply it to a new situation. Every government recruitment body — from the Staff Selection Commission to the Institute of Banking Personnel Selection, the Railway Recruitment Boards and the Union Public Service Commission — includes verbal reasoning because the work culture in administrative, banking and railway services demands logical clarity under time pressure.

The questions in this segment are short, deceptively simple, and almost always solvable in twenty to thirty seconds once the rule is identified. The challenge is not difficulty in the academic sense; it is the discipline of remaining calm, reading the problem twice, and resisting the urge to compute before the rule has been understood. Most aspirants who score below average in reasoning do not fail because of weak intelligence — they fail because they begin solving before they have truly observed the structure of the question.

Number and Letter Series — The Art of Pattern Recognition

A series question presents a sequence of numbers or letters arranged according to a definite, hidden rule. The candidate must either find the next term, identify the missing term, or detect the term that breaks the pattern. The first habit to develop is mechanical: always write the first differences below the series before attempting anything else. If the differences are constant, the series is arithmetic. If the differences themselves form a pattern (such as 2, 4, 6, 8), the series is built on a second-level arithmetic rule. If the ratio of consecutive terms is constant, the series is geometric. If none of these apply, examine squares, cubes, prime numbers, and finally alternate-position rules where odd-indexed and even-indexed terms follow separate sequences.

For letter series, replace each letter with its alphabetical position number, identify the numeric pattern, and then translate the answer back into a letter. This single discipline reduces an entire family of questions to ordinary number-series problems. With about thirty solved examples, an average aspirant typically reaches an accuracy of ninety percent in this sub-topic.

Analogy — Reasoning by Relationship

An analogy question presents two pairs and asks the candidate to identify the relationship that ties the first pair together, then apply the same relationship to complete the second pair. The relationships are drawn from a stable list: synonyms and antonyms, worker and tool, cause and effect, raw material and finished product, country and capital, part and whole, category and example, and arithmetic operations such as squares, cubes and constant differences.

The professional approach is to test two or three candidate rules against the first pair before locking in an answer. Examiners deliberately design distractors that satisfy a superficial rule but fail when applied carefully. For instance, in the pair Doctor : Hospital, the relationship is workplace, not service-recipient — so the matching pair must be Teacher : School, not Teacher : Student. Recognising this discipline of multiple-rule testing is what separates a seventy-percent-accuracy candidate from a ninety-five-percent-accuracy candidate.

Classification — Spotting the Outlier

Classification questions, often called odd-one-out, present four or five terms and ask which one does not belong. The professional method is to identify the rule that the majority of the options satisfy and then mark the option that violates it. Common classifying rules include prime numbers versus composites, perfect squares versus non-squares, even versus odd, multiples of a particular integer, vowels versus consonants, leap-year-eligible numbers, and biological or geographical categories such as mammals, birds, mountain ranges or planets.

Two safeguards prevent careless errors. First, never finalise the answer after testing only one rule — always confirm that the chosen odd-one-out fails the rule that the other three satisfy. Second, if two rules seem to identify two different outliers, the examiner has built in a deeper classification, and the deeper rule is almost always the correct one.

Coding–Decoding — Translating Between Codes

Coding-decoding tests whether the candidate can apply a defined transformation rule consistently. The transformation may be a positional shift in the alphabet (such as +1, +2 or −3), a mirror image (where A is paired with Z, B with Y), a reversal of letter order within the word, or a substitution based on a key. The reliable method is to write the alphabet 1 to 26 across the top of the rough sheet at the start of the paper. Every coding-decoding question can then be solved by simple addition or subtraction of position numbers, which is faster and far less error-prone than counting letters mentally.

Once the rule has been confirmed on the example pair given in the question, the same rule must be applied character-by-character to the target word. With consistent practice, the candidate can solve a coding-decoding question in under twenty seconds.

How to Practise and What to Expect in the Exam

A disciplined preparation plan covers verbal reasoning in approximately three weeks. In the first week, the candidate should solve ten series, ten analogy, ten classification and ten coding-decoding questions every day, focusing on rule identification rather than speed. In the second week, the same volume of practice should be timed at thirty seconds per question. In the third week, full mixed sets of forty questions in twenty minutes should be attempted to simulate examination pressure.

On the actual examination day, the candidate should approach this section with the goal of attempting every question, because verbal reasoning is among the highest-yield, lowest-cost segments of the entire paper. A score of eight to ten correct answers out of ten in this segment forms a strong foundation that compensates for slower segments such as data interpretation or comprehension.

1Number & Letter Series

A series is a sequence of numbers or letters arranged according to a definite rule. The task is to find the next term, the missing term, or the wrong term that breaks the pattern.

Common patterns: (i) Arithmetic — constant difference (+d). (ii) Geometric — constant ratio (×r). (iii) Squares/Cubes (n², n³). (iv) Mixed/double-difference — first differences themselves form a pattern. (v) Alternate series — odd and even positions follow separate rules. (vi) Letter series based on alphabet positions (A=1, B=2, … Z=26) or skip patterns (+1, +2, +3, …). Always compute first differences first, then ratios, then squares/cubes.

Examples
  • Number series: 3, 7, 15, 31, 63, ? → each term ×2 +1 → next = 127.
  • Mixed series: 2, 6, 12, 20, 30, ? → diffs 4, 6, 8, 10 → next = 42.
  • Letter series: A, C, F, J, O, ? → skip +1, +2, +3, +4, +5 → next = U.
  • Alternate: 5, 8, 11, 16, 17, 24, ?, ? → odd terms +6 (5,11,17,23) & even terms +8 (8,16,24,32) → 23, 32.
Exam tip: If first differences are constant → AP. If ratios are constant → GP. If neither, look at squares, cubes, primes, or alternate-position rules.

2Analogy (Number, Letter, Word & Meaning-Based)

Analogy tests whether you can identify the relationship between a given pair and apply the same relationship to another pair.

Steps: (1) Decode the rule between the FIRST pair (synonym, antonym, part-whole, cause-effect, worker-tool, ±n, ×n, n², n³, alphabet-position shift). (2) Apply the SAME rule to the second pair. Common rules: worker-tool (Carpenter : Hammer), category-member (Bird : Sparrow), country-capital, raw-material → product, addition/subtraction of letters, mirror image of letters.

Examples
  • Word analogy: Doctor : Hospital :: Teacher : ? → School (workplace).
  • Number analogy: 4 : 64 :: 5 : ? → n³ → 125.
  • Letter analogy: AB : ZY :: CD : ? → mirror image (A↔Z, B↔Y) → XW.
  • Mixed: 7 : 50 :: 9 : ? → n² + 1 → 82.
Exam tip: Always test 2–3 possible rules on the first pair before locking one. Many MCQs include traps where two rules fit pair-1 but only one fits the answer pair.

3Classification (Odd-One-Out)

Classification asks you to identify the term that does NOT belong to the same group as the other three (or four).

Pick the rule that the majority of options satisfy. Common groupings: prime numbers, perfect squares/cubes, even/odd, multiples of n, vowels vs consonants, alphabet positions in AP/GP, country/capital pairs, fruits vs vegetables, mammals vs birds, planets, mountain ranges, leap-year-eligible numbers.

Examples
  • Numbers: 25, 36, 49, 50, 81 → all are perfect squares except 50. Odd = 50.
  • Letters: BD, FH, JL, NQ → all have a +2 letter gap except NQ (gap 3). Odd = NQ.
  • Words: Mango, Banana, Apple, Carrot → Carrot is a vegetable; rest are fruits.
  • Numbers: 11, 13, 17, 21, 23 → all primes except 21 (= 3×7). Odd = 21.
Exam tip: If 4 options share property X and 1 option shares property Y, that one is the odd one. Never lock the answer until you have verified the same property in all the other three.

4Coding-Decoding

Coding-decoding tests how a word/number is converted to a coded form using a fixed rule, and how to decode a new word using the same rule.

Common types: (i) Letter-shift coding (A→B, B→C, …): each letter shifted by a fixed number. (ii) Reverse-alphabet coding (A↔Z, B↔Y, …). (iii) Number-position coding (A=1, B=2 …). (iv) Substitution coding (real meaning replaced by another word). (v) Pattern-based number coding (sum/product of letter positions). Always identify the FIRST and LAST letters before applying the rule to all letters.

Examples
  • If CAT is coded as DBU, then DOG = ? → each letter +1 → EPH.
  • If A = 1, B = 2 …, then code for FACE = 6 + 1 + 3 + 5 = 15.
  • If MONDAY is coded as YADNOM, then FRIDAY = ? → reverse → YADIRF.
  • If 'sky' means 'red', 'red' means 'green', 'green' means 'blue', then the colour of grass is = 'red' (originally 'green', re-coded).
Exam tip: When several letters in a coded word repeat at the same position, the encoding is almost always positional (1↔1, 2↔2 …) — verify by mapping individual letters, not the full word.

Short Tricks & Shortcuts

Use these speed tricks in the exam. Each trick is followed by a worked example so you can verify the shortcut yourself.

Trick 1First-difference test for any series

Compute differences of consecutive terms. If constant → AP. If they form their own series → mixed/double-difference. If ratios are constant → GP. Otherwise try squares, cubes, primes.

Example: 2, 6, 12, 20, 30, ? — diffs 4,6,8,10, next diff 12 → 42.
Trick 2Letter-position shortcut (A=1 … Z=26)

Convert letters to positions instantly. To find a letter shifted by ±k, add/subtract from the position. If result > 26, subtract 26; if < 1, add 26 (cyclic).

Example: Letter 6 places after T (=20) → 20+6 = 26 → Z.
Trick 3Mirror-image letter pairs

Memorise: A↔Z, B↔Y, C↔X, D↔W, E↔V, F↔U, G↔T, H↔S, I↔R, J↔Q, K↔P, L↔O, M↔N. Sum of mirror-pair positions = 27.

Example: What is the mirror image of L? → 27 − 12 = 15 → O. ✓
Trick 4Coding by EJOTY rule

Memorise EJOTY (5, 10, 15, 20, 25) for the alphabet — it lets you locate any letter's position in 1–2 jumps.

Example: Position of W? → T = 20 → +3 → 23 → W. ✓
Trick 5Eliminate by 'majority property' in classification

If three options share property X (primes, even, squares, fruits) and one does not, that one is the odd one. Never decide on a single attribute alone — re-test all four against TWO properties.

Example: 11, 13, 17, 21, 23 — 11/13/17/23 are primes; 21 = 3×7 → odd one.

Quick Revision Facts

  • EJOTY: E=5, J=10, O=15, T=20, Y=25 — use to find any alphabet position fast.
  • Mirror-pair sum is always 27 (A+Z = 1+26 = 27).
  • AP: constant difference. GP: constant ratio. Always test these two first.
  • There are 26 letters, 5 vowels (A, E, I, O, U) and 21 consonants.

Frequently Asked Questions

Compute the differences between consecutive terms. The wrong term will produce a difference that breaks the otherwise regular pattern. Replace it with the value that fits the pattern.

Check the FIRST letter pair: if the encoded letter is close to the original (1–5 positions away), it is shift coding; if the sum of positions is 27, it is mirror coding.