E ExamMaster

SBI PO · General Intelligence & Reasoning

Dictionary Order & Word Building

Arrange words as in a dictionary; form one meaningful English word from given letters; count letter-pairs in a word that have as many letters between them as in the alphabet.

Three official CGL items share this topic, and all three treat letters as ordered objects rather than as meanings. Dictionary order (indexing) asks you to arrange words the way a printed dictionary does, or to say which of four or five words stands first or last. Letter-pair formation — the banking 'how many such pairs' stem — asks you to count pairs of letters inside one word that have as many letters between them in the word as in the English alphabet. Word building asks you to form one ordinary English word from a jumble, or from the first and last letters of given words. None of this is analogy: you are not naming a semantic relation. You are comparing positions. The marks sit in the first differing letter, in an equal gap, and in refusing a near-word.

  • SBI PO
  • Easy level
  • 5 concepts
  • 16 practice questions

1Dictionary order: first differing letter, then the shorter prefix

A dictionary does not sort words by how long they are, or by what they mean. It compares them letter by letter from the left, the same way you already compare two numbers digit by digit from the left. Hold one pair for the whole concept: NEW and NEWS. Write them one above the other and read down each column: N against N (tie), E against E (tie), W against W (tie). Those three letters do not decide anything. The decision waits for the first column that is not a tie.

Here the first three letters are identical and then NEW runs out of letters while NEWS still has S. The dictionary treats 'end of the word' as earlier than any leftover letter, so the shorter word that is a prefix of the longer one comes first: NEW before NEWS. The same prefix rule puts CAT before CATCH and before CATALOG. It is not that shorter words always win — CATCH is shorter than CATALOG and still comes after it, because after CAT the next letters are C versus A, and A precedes C. Length is consulted only when one word is exactly the leading part of the other.

So the whole method is two questions in a fixed order. First: at the first letter that differs, which letter comes earlier in A–Z? That word is earlier. Second, only if every letter of the shorter word matches: the shorter prefix comes first. Meaning ('news feels more complete than new') is not a key, and neither is counting letters and putting the long word last.

Figure. Stack NEW over NEWS and walk left to right. N, E and W tie; then NEW ends while NEWS still has S. The shorter prefix comes first.

Order of attack

  1. Stack the wordsWrite the two words one above the other, aligned on the first letter, so each column is one position.
  2. Walk left to rightSkip every tied column. The first column that differs decides: the earlier A–Z letter wins that pair.
  3. If one word endsWhen every letter of the shorter word has matched and the other word still has letters left, the shorter prefix comes first — NEW before NEWS, CAT before CATCH.
  4. Refuse length and meaningDo not put the longer word last as a habit, and do not ask which word 'feels more complete'. CATCH is shorter than CATALOG and still follows it, because C follows A.
NEW vs NEWS, column by column
PositionNEWNEWSDecision
1NNTie — keep walking
2EETie — keep walking
3WWTie — keep walking
4(end)SShorter prefix NEW comes first

NEW before NEWS

Which comes first in an English dictionary, NEW or NEWS?

  • Columns 1–3: N=N, E=E, W=Wtie through NEW
  • Column 4: NEW ends; NEWS has Sprefix rule
  • Shorter prefix precedes the longer wordNEW before NEWS

Pro tip. End-of-word beats leftover S. The same comparison puts CAT before CATCH.

In an English dictionary, CATCH and CATALOG share C-A-T. Which word comes first, and why?
  1. CATCH, because it is the shorter word
  2. CATALOG, because after CAT the next letter is A, and A precedes C
  3. CATCH, because C precedes A

After the shared CAT, CATALOG continues with A and CATCH with C. A precedes C, so CATALOG is earlier. Length would have picked CATCH and is the wrong key once the words are not a prefix pair.

2First and last in a list of four or five words

CGL almost never asks you to sort just two words. The live stem hands you four, five or six numbered words and asks either for the full dictionary order (as a permutation of those numbers) or for who stands first or last. The method does not change: you still compare letter by letter and you still apply the prefix rule. What changes is that you must not invent a second key — length, 'looks later', 'more syllables' — once more than two words are on the page.

Take the five-word CGL-style list Seniority, Senna, Sennet, Sennight, Sennit. All five open SEN. The first differing letter is in the fourth seat: Seniority has I, and the other four have N. I precedes N, so Seniority is first in the dictionary — and it is first even though it is the longest word in the list. Length would have put Senna first; that is the trap. Among the four SENN-words the next split is Senna (A) before Sennet (E) before the two SENNI-words; then Sennight (G) before Sennit (T). Last is Sennit.

The NEW family from the prefix concept is the same trap in different clothes. Sorted, NEW, NEWER, NEWEST, NEWS come in that order: NEW first (it is a prefix of the other three), then NEWER before NEWEST (R precedes S), and NEWS last because after NEW the leftover letter is S, which follows the E of NEWER and NEWEST. Students who sort by length put NEWS second (just after NEW) and NEWEST last. The dictionary does the opposite with NEWS.

Figure. After shared SEN, Seniority splits on I and the other four on N, so Seniority is first even though it is the longest. Among the SENN-words the order is Senna, Sennet, Sennight, Sennit — last is Sennit (T after G).

Order of attack

  1. Find the first splitWrite the shared opening letters once. The first seat where the list disagrees is the only seat that matters for first and last at that stage.
  2. Read that seat as A–ZThe earliest letter in that seat is the first word (I in Seniority). The latest letter, after you finish later splits, is the last word (T in Sennit).
  3. Split only the remaining groupOnce Seniority is peeled off, compare Senna / Sennet / Sennight / Sennit among themselves. Do not re-sort the whole list from scratch by length.
  4. Refuse a length rankingA longer word can be first (Seniority) and a shorter leftover can be last (NEWS after NEWER and NEWEST). Length is not a key.
Five-word CGL list after the shared SEN
WordAfter SENNext splitDictionary place
SeniorityIorityI precedes NFirst
SennaNaA precedes ESecond
SennetNetE precedes IThird
SennightNightG precedes TFourth
SennitNitT is last among SENNILast

Who is first, who is last

Among Seniority, Senna, Sennet, Sennight and Sennit, which word stands first in an English dictionary, and which stands last?

  • Shared opening SEN; fourth letter I vs NI precedes N, so Seniority is first
  • Among SENN-: A (Senna), E (Sennet), I (Sennight, Sennit)Senna then Sennet, then the SENNI pair
  • Sennight has G after SENNI; Sennit has TG precedes T, so Sennit is last

Pro tip. Seniority is the longest word and still first. Length is not the key; the I-versus-N split is.

Arrange NEW, NEWS, NEWER, NEWEST as in an English dictionary. Which word comes last?
  1. NEWEST, because it is the longest
  2. NEWS, because after NEW the leftover S follows the E of NEWER and NEWEST
  3. NEWS, because it is only one letter longer than NEW and so sits at the end

The order is NEW, NEWER, NEWEST, NEWS. After the shared NEW, E precedes S, so both NEWER and NEWEST come before NEWS. Length would put NEWS second and NEWEST last — both wrong.

3Letter pairs whose word-gap equals the alphabet-gap

The banking and CGL stem is almost always the same sentence: how many pairs of letters in a given word have as many letters between them in the word as in the English alphabet? 'Between' means the letters sitting strictly in the middle — not the two letters of the pair themselves. Number the alphabet once and for all: A = 1, B = 2, through Z = 26. Then a pair of letters at word seats i < j counts when the seat-gap equals the alphabet-gap: |j - i| = |\mathrm{pos}(L_j) - \mathrm{pos}(L_i)|. That is the same claim as 'letters between in the word = letters between in the alphabet'.

Hold PRIDE for the whole concept. Seats and positions: P = 16 at seat 1, R = 18 at 2, I = 9 at 3, D = 4 at 4, E = 5 at 5. D and E are next to each other in the word (0 letters between; seat-gap 1) and next to each other in the alphabet (0 letters between; position-gap 1). The gaps match, so DE is one pair. Direction does not have to run forward in the alphabet: a later seat may hold an earlier letter, and that backward pair still counts when the absolute gaps match. GOLDEN has such a backward pair, G to D.

The trap in PRIDE is the pair PR. P and R appear in alphabetical order, and coaching notes sometimes count them for that reason alone. They are adjacent in the word (0 between) but Q sits between them in the alphabet (1 between). Seat-gap 1, alphabet-gap 2 — not equal, so PR is not a pair. Alphabetical order inside the word is not the test. Equal gaps are the test. Scanning every pair in PRIDE leaves only DE. The count is 1.

Figure. In PRIDE, DE has 0 letters between in the word and 0 in the alphabet, so it counts. PR has 0 between in the word but 1 (Q) between in the alphabet. IE has 1 between in the word and 3 in the alphabet. The count is 1.

How it works

  1. Number both tracksWrite seats 1…n under the word and A = 1 … Z = 26 under each letter. PRIDE becomes 16, 18, 9, 4, 5.
  2. Test every pair, not only neighboursFor seats i < j, count the pair when |j-i| equals the absolute alphabet-position gap. Neighbours are the 0-between case; GOLDEN's G–D pair skips two letters and still counts.
  3. Count both directionsIf the later letter is earlier in the alphabet, still use the absolute gap. Backward pairs are legal.
  4. Reject order-only pairsP then R in PRIDE is alphabet order with unequal gaps. Do not count it.
PRIDE pair scan (the pairs students actually debate)
PairSeatsBetween in wordBetween in alphabetCount?
DE4 and 500 (D–E consecutive)Yes
PR1 and 201 (Q sits between)No — unequal gaps
IE3 and 51 (D)3 (H, G, F)No

How many pairs in PRIDE?

How many pairs of letters in PRIDE have as many letters between them in the word as in the English alphabet?

  • Positions: P=16, R=18, I=9, D=4, E=5 at seats 1–5five letters, ten candidate pairs
  • DE: seats |5-4|=1, alphabet |5-4|=1gaps match — 1 pair
  • PR: seats |2-1|=1, alphabet |18-16|=2unequal — reject (Q is between in the alphabet only)
  • Every remaining pair fails the equal-gap testcount = 1

Pro tip. PR is the mark-losing pair. Order in the alphabet without an equal gap is not a pair.

In PRIDE, why is PR not a counted letter-pair?
  1. P and R are not next to each other in the word
  2. They have 0 letters between them in the word but 1 letter (Q) between them in the alphabet, so the gaps are unequal
  3. Backward pairs are never counted, and R comes after P

PR is adjacent in the word (0 between) and has Q between in the alphabet. Equal-gap fails. They are neighbours, and the pair is forward, so the other two options are simply false.

4One meaningful English word from a jumble

A meaningful-word item gives you a short heap of letters — either a jumble such as RIFA, or named seats of a longer word — and asks for one ordinary English word that uses each given letter exactly once. 'Meaningful' here is a dictionary test, not a feeling: the answer must be a word you would find in an English dictionary, used with its ordinary sense. A personal name, a brand, or a string that merely looks word-shaped is not enough.

Take the jumble RIFA — the four letters R, I, F, A, each used once. Try the letters in the order they were handed to you: RIFA is not an English word. Shuffle them. FAIR is an ordinary English word (just, pale, a village market). FIRA looks like a name; RAIF and FRIA look word-shaped; none of those three is the dictionary word the stem wants. When the stem says 'only one meaningful word can be formed', you must also check that no second dictionary word uses the same letters. For R, I, F, A the one ordinary word is FAIR.

The trap is always a near-word sitting in the options — FIRA next to FAIR, TEAD next to DATE, COULD offered when the letters were supposed to make a different word, or a name offered as if it were an English noun. If two genuine dictionary words exist (EARTH and HEART from A, E, H, R, T), the stem will ask 'how many' or 'which of the following', not 'the' word. A stem that promises one word is promising that the near-word option is not in the dictionary.

Figure. The bag is R, I, F, A. Collection-order RIFA is not a word. FIRA is name-shaped. FAIR is the ordinary English dictionary word. RAIF is not.

Order of attack

  1. List the letters onceWrite the given letters (or the named seats of the longer word) as a bag: R, I, F, A. Each letter will be used once.
  2. Try collection order, then shuffleRIFA is not a word. Rearrange. Look for a common noun, verb or adjective — FAIR — not a name.
  3. Reject the near-wordFIRA, RAIF and FRIA are the options written to look finished. They are not English dictionary words.
  4. Confirm uniqueness when the stem claims itIf the stem says 'only one meaningful word', check that a second anagram (EARTH beside HEART) does not also qualify. For R, I, F, A only FAIR does.
RIFA trials
ArrangementDictionary word?Keep?
RIFA (as given)NoNo — collection order
FIRANo (name-shaped)No — near-word trap
FAIRYes — ordinary EnglishYes — the word
RAIF / FRIANoNo

Jumble RIFA

If it is possible to form only one meaningful English word from the letters R, I, F, A using each letter once, which word is it?

  • Bag of lettersR, I, F, A — each once
  • Collection order RIFA; name-shaped FIRAneither is an English dictionary word
  • Rearrangement FAIRordinary English word; unique among common anagrams

Pro tip. FIRA is the option that exists to steal the mark. Looking finished is not the same as being in the dictionary.

The letters R, I, F, A can form one ordinary English word. The answer is FAIR and not FIRA because
  1. FIRA uses the letters in a different order, so it is a different bag
  2. FAIR is an English dictionary word and FIRA is only name-shaped, not that word
  3. The stem forbids rearranging, so the letters must stay R-I-F-A

Both strings use the same four letters once. The test is 'ordinary English word', which FAIR passes and FIRA fails. The stem of this item type allows rearranging; collection order RIFA is not a word either.

5Build a word from named first and last letters

CGL word-building often does not hand you a jumble. It names seats on other words: the first letter of this word, the last letter of that word, sometimes both ends of each word in a short list. Your first job is extraction, not anagramming. Take only the letters the stem actually named. Using a middle letter, or taking every first letter when the stem asked for last letters, is a different bag and a wrong word.

Hold this stem: the first letter of TIGER, the last letter of PRIDE, the first letter of APPLE and the last letter of BEND. First of TIGER is T. Last of PRIDE is E — the same E that closed the DE pair in the pair-count concept, now reused as a building brick. First of APPLE is A. Last of BEND is D. The bag is T, E, A, D. In the order you picked them those letters spell TEAD, which is not an English word. Rearrange the bag: DATE is the ordinary English word. TEAD is the near-word trap for this flavour — collection order looking finished.

Two stem-readings stay distinct. If the stem says 'which meaningful word can be formed', you may rearrange, and TEAD is wrong. If a rare stem says the letters are placed sequentially and already form the word, you would keep the collection order — and then the answer would exist only when that order is already a word (as when the first letters of FAST, IRON, AREA, ROAD are taken and then rearranged to FAIR, or, if a stem wanted first letters kept in list order, something that already reads as a word). Default CGL reading: extract, then form one dictionary word from that bag, and refuse the collection-order near-word.

Figure. Extract only the named seats: first of TIGER is T, last of PRIDE is E, first of APPLE is A, last of BEND is D. Collection order TEAD is not a word; the ordinary English word is DATE.

Order of attack

  1. Extract, then stopRead each naming clause once: first of TIGER → T; last of PRIDE → E; first of APPLE → A; last of BEND → D. Do not pull I, G, R or any other letter.
  2. Write the bagT, E, A, D. Collection order TEAD is a note, not yet an answer.
  3. Form one dictionary wordUnless the stem forbids rearranging, shuffle the bag. DATE is the ordinary English word.
  4. Reject collection orderTEAD uses the right letters in the order you picked them and is still not a word. That is the near-word for this stem.
Named letters for the TIGER / PRIDE / APPLE / BEND stem
Source wordNamed seatLetter
TIGERfirstT
PRIDElastE
APPLEfirstA
BENDlastD

First of TIGER, last of PRIDE, first of APPLE, last of BEND

Those four named letters are taken. Which meaningful English word can be formed?

  • Extract: first TIGER, last PRIDE, first APPLE, last BENDT, E, A, D
  • Collection order TEADnot an English word
  • Rearrange T, E, A, DDATE

Pro tip. TEAD is FAIR's cousin: the right letters in pickup order. Rearrange unless the stem says not to.

The first letters of FAST, IRON, AREA and ROAD are taken. A meaningful English word formed from those letters is
  1. FIAR, the letters in list order
  2. FAIR, the ordinary English word from the bag F, I, A, R
  3. FAST, because it is one of the source words

The bag is F, I, A, R. FIAR is collection order and not a word; FAST uses letters the stem did not give you from the other three words. FAIR is the dictionary word from that bag.

Notes

  • Dictionary order is letter-by-letter from the left. The first position where two words differ decides which comes earlier; the earlier alphabet letter wins. Meaning and word-length are not sorting keys.
  • When one word is a prefix of the other (NEW and NEWS), the shorter word comes first — the dictionary treats 'end of word' as earlier than any leftover letter. The same rule picks who stands first or last in a list of four or five words.
  • Letter-pair formation: two letters in a word form a counted pair when the number of letters sitting between them in the word equals the number sitting between them in the English alphabet (both directions). In PRIDE the only such pair is DE, not PR.
  • Meaningful-word items give a jumble (or named letters of a longer word) and ask for one ordinary English word that uses each given letter once. A near-word or a name that is not in the dictionary is the trap.
  • Word-building from first and last letters: take only the letters the stem names — first of this word, last of that word — then form one English word. The letters in collection order are not automatically the answer.

Formulas

  • Dictionary compare: walk left to right; at the first index i where the letters differ, the word whose letter is earlier in A–Z comes first.
  • Prefix rule: if word X is exactly the leading letters of word Y, then X precedes Y (NEW before NEWS; CAT before CATCH).
  • Letter-pair test: letters at word positions i < j form a pair iff |j - i| = |\mathrm{pos}(L_j) - \mathrm{pos}(L_i)|, where \mathrm{pos}(A)=1, \ldots, \mathrm{pos}(Z)=26. Equivalently, letters-between in the word equals letters-between in the alphabet.
  • Alphabet positions: A=1, B=2, \ldots, Z=26. Gaps are absolute — a later letter in the word may be earlier in the alphabet (backward pairs count).
  • Word building: extract the named letters (first, last, or both, as the stem says), then find the one ordinary English word that uses each extracted letter once.

Exam traps & shortcuts

  • Stop at the first differing letter. Everything after that letter is noise for that pair of words.
  • Never sort a dictionary list by length. NEW, NEWER, NEWEST, NEWS ends with NEWS, not NEWEST, because S comes after E.
  • Write A = 1 under every letter of the given word before counting pairs. Then test every pair, not only neighbours — GOLDEN's G–D pair skips two letters.
  • If the letters in the order you picked them do not spell an English word, rearrange them. The collection-order string (TEAD, FIRA) is the near-word trap.

Reference tables

Use these in order on any two words, then peel a four- or five-word list by the same splits.

Dictionary comparison rules
SituationWhat decidesExample
First letters differEarlier A–Z letter winsMarble before Margin (B before G)
Opening letters tieWalk to the first differenceSenna before Sennet (A before E after SENN)
One word is a prefix of the otherShorter prefix comes firstNEW before NEWS; CAT before CATCH
Same opening, leftover letters differCompare the leftovers, not the lengthsNEWER before NEWS (E before S after NEW)

A pair counts only when the two gaps are equal. Number seats and alphabet positions before you start.

Letter-pair equal-gap test
CheckFormulaPRIDE verdict
Seat gap|j-i| for seats i<jDE: 1; PR: 1
Alphabet gap|\mathrm{pos}(L_j)-\mathrm{pos}(L_i)| with A=1DE: 1; PR: 2
Count the pair?The two gaps are equalDE yes; PR no
DirectionAbsolute gap; backward pairs countGOLDEN's G–D is backward and counts

Recap

Read only this the night before.

Dictionary
First differing letter decides. If one word is a prefix of the other, the shorter one comes first: NEW before NEWS.
First / last
Same rule on four or five words. Length is not a key: Seniority can be first; NEWS can be last after NEWER and NEWEST.
Letter pairs
Count a pair when letters-between in the word equals letters-between in the alphabet. PRIDE has one pair, DE. PR is the trap.
Meaningful word
Use each given letter once. The answer is an ordinary English dictionary word (FAIR), not a near-word (FIRA).
First / last letters
Extract only the named seats, then form one word. T, E, A, D → DATE, not TEAD.

Practise Dictionary Order & Word Building

Reading is free and needs no account. Practice, mocks and progress live in the app.

  • 16 exam-style questions on this topic, with explanations
  • A 5-question practice set that ends the chapter
  • Timed mocks scored with the real marking scheme
  • Readiness tracked per topic, kept on your device
Continue with Google — freeNo card, no trial. Works offline once installed.