Study guide
5.1 Part 4 Rules: Word Limits, Exact Stems & Dual-Point Scoring
Cambridge B2 First Exam
Track your progress as you read
Finished sections get a check mark.
1 Chapter 1: Exam Architecture, Cambridge English Scale & Administration
1.1 Exam Architecture & CEFR B2 Benchmark 1.2 The Cambridge English Scale, Subscale Scoring & Results Profiles 1.3 Delivery Formats: Computer-Based vs. Paper-Based Testing & Interface Strategy 1.4 Registration, Exam-Day Regulations, Permitted Materials & Timing Rules
2 Chapter 2: Reading & Use of English — Part 1: Multiple-Choice Cloze
2.1 Part 1 Mechanics, Lexical Precision & Collocations 2.2 Dependent Prepositions, Idiomatic Expressions & Fixed Phrases 2.3 Phrasal Verbs & Eliminating High-Frequency Distractors
3 Chapter 3: Reading & Use of English — Part 2: Open Cloze
3.1 Part 2 Grammar Focus: Determiners, Pronouns & Quantifiers 3.2 Relative Clauses, Auxiliary Verbs & Structural Linkers 3.3 Fixed Grammatical Patterns & Prepositional Phrases
4 Chapter 4: Reading & Use of English — Part 3: Word Formation
4.1 Part 3 Strategy: Identifying Parts of Speech & Sentence Context 4.2 High-Yield Prefixes: Negation, Reversal & Degree 4.3 High-Yield Suffixes: Nouns, Adjectives, Verbs & Adverbs 4.4 Internal Spelling Shifts, Irregular Stems & Compound Derivatives
5 Chapter 5: Reading & Use of English — Part 4: Key Word Transformations
5.1 Part 4 Rules: Word Limits, Exact Stems & Dual-Point Scoring 5.2 Conditional Structures, Wishes & Regrets 5.3 Passive Voice, Causative Structures & Reported Speech 5.4 Modals of Deduction, Speculation & Inverted/Comparative Structures
6 Chapter 6: Reading & Use of English — Parts 5–7: Reading Comprehension
6.1 Part 5: Multiple-Choice Reading — Gist, Detail, Attitude, Tone & Reference 6.2 Part 6: Gapped Text — Text Structure, Cohesion, Coherence & Reference Markers 6.3 Part 7: Multiple Matching — Scanning, Paraphrase, Implication & Selective Reading 6.4 Reading Paper Time Management & Answer Sheet Protocol
7 Chapter 7: Writing Paper — Part 1: Compulsory Discursive Essay
7.1 Cambridge Writing Assessment Framework: The Four Subscales Explained 7.2 Part 1 Compulsory Essay: Analyzing the Prompt & Generating the Third Point 7.3 Essay Structural Blueprint: Introduction, Balanced Body Paragraphs & Conclusion 7.4 Advanced Cohesive Devices, Complex Syntax & Lexical Range for B2 Essays
8 Chapter 8: Writing Paper — Part 2: Genre Writing
8.1 Articles: Engaging the Reader, Catchy Headlines & Descriptive Flair 8.2 Reviews: Structuring Evaluation, Balanced Critique & Recommendations 8.3 Formal & Informal Letters/Emails: Conventions, Purpose & Tone Modulation 8.4 Reports: Headings, Factual Presentation, Neutral Tone & Recommendations
9 Chapter 9: Listening Paper — Strategy & Parts 1–2
9.1 Listening Paper Architecture, Acoustic Traps & The Two-Pass Auditory Method 9.2 Part 1: Multiple Choice Short Extracts — Decoding Genre, Gist, Function & Context 9.3 Part 1: Identifying Speaker Feelings, Attitudes, Opinions & Agreement 9.4 Part 2: Sentence Completion — Monologue Prediction, Exact Dictation & Word Limits
10 Chapter 10: Listening Paper — Parts 3–4 & Transfer Protocol
10.1 Part 3: Multiple Matching — Strategy for Five Monologues & Theme Tracking 10.2 Part 3: Neutralizing Paraphrase Traps & Distractor Analysis in Monologues 10.3 Part 4: Long Interview / Discussion — Tracking Topic Shifts & Complex Attitudes 10.4 Final 5-Minute Transfer Period, Spelling Pitfalls & Answer Sheet Protocol
11 Chapter 11: Speaking Paper — Overview & Parts 1–2
11.1 Speaking Exam Overview, Test Room Dynamics & The Two Examiners 11.2 The Speaking Assessment Scale: Analytical Subscales vs. Global Achievement 11.3 Part 1: The Personal Interview — Expanding Responses with Reasons & Examples 11.4 Part 2: The Individual Long Turn — Comparing Photographs, Speculating & Managing the 1-Minute Clock
12 Chapter 12: Speaking Paper — Parts 3–4 & Final Exam Execution
12.1 Part 3: Collaborative Task — Discussion Phase, Spidergram Exploration & Negotiation 12.2 Part 3: Reaching a Decision & Productive Disagreement Techniques 12.3 Part 4: Discussion Phase — Broadening Themes, Abstract Reasoning & Justification 12.4 Final Preparation Blueprint: 8-Week Study Schedule, Mock Exams & Confidence Routine
5.1 Part 4 Rules: Word Limits, Exact Stems & Dual-Point Scoring
Key Takeaways Explain This Topic AI
Part 4 Key Word Transformations consists of 6 discrete items (numbered 25–30 plus Example 0), each featuring a complete lead-in sentence, an invariant key word in bold capitals, and a gapped second sentence requiring 2 to 5 words to complete an exact paraphrase.
Each transformation is evaluated out of 2 raw marks (12 marks total), contributing directly to the Reading and Use of English paper's Use of English subscale.
The Cambridge marking scheme splits every target answer into two separate scoring units [Mark 1 | Mark 2], allowing candidates to earn full credit (2 marks), partial credit (1 mark), or zero credit (0 marks).
The invariant key word must appear completely unaltered in the gap; modifying its tense, part of speech, number, or affixation results in an immediate award of 0 marks for the entire item.
Contractions count as two words (e.g., 'didn't' = 2 words), with the handbook's single exception that can't counts as cannot; writing full uncontracted forms is the safeguard against word-count disqualification.
Last updated: September 2026
5.1 Part 4 Rules: Word Limits, Exact Stems & Dual-Point Scoring
Quick Answer: Reading & Use of English Part 4 consists of 6 discrete items (Questions 25–30, plus Example 0). For each item, you are given a complete lead-in sentence, a single key word printed in bold capital letters, and a second sentence with a central gap. You must complete the gap using between two and five words (inclusive), one of which must be the key word completely unchanged , so that the second sentence has an identical meaning to the first. Each question is worth 2 raw marks (12 marks total), split into two independent scoring units [Mark 1 | Mark 2] . This means you can earn 2, 1, or 0 marks per item. Always write full, uncontracted words ( did not rather than didn't ) because contractions count as two words — the one handbook exception being can't , which counts as cannot , that is, one word.
Architecture and Mechanics of Part 4
Part 4 represents the pinnacle of grammatical and lexical manipulation on the Cambridge B2 First examination. While Parts 1, 2, and 3 test vocabulary in context, closed-class grammatical function words, and morphological word formation respectively, Part 4 synthesizes all these linguistic faculties into structural paraphrasing.
Key Structural Parameters
- Feature Official Specification
- Number of Items 6 test items (Questions 25–30) + 1 unnumbered Example (Item 0)
- Prompt Structure Complete lead-in sentence + Invariant key word (bold capitals) + Gapped target sentence
- Permitted Word Count Strictly between 2 and 5 words (inclusive of the key word)
- Raw Marks Available 2 marks per question (Maximum 12 raw marks)
- Paper Weighting Contributes directly to the Cambridge English Scale Use of English subscale
- Scoring Split Dual-point marking boundary [Mark 1 | Mark 2] awarding 0, 1, or 2 marks
Answer Format Written directly on the separate Answer Sheet in CAPITAL LETTERS (paper-based) or typed (computer-based)
In Part 4, candidates do not read an extended continuous passage. Instead, each question stands alone as an independent syntactic puzzle. The prompt requires candidates to express the exact semantic proposition of the lead-in sentence using completely different grammatical architecture and lexical collocations centered around the invariant stem.
The Strict Word Count Constraint (2 to 5 Words)
Cambridge regulations regarding word count are mathematically absolute. There is zero examiner discretion or leniency:
1 word: 0 marks (automatically disqualified for being under-length).
2, 3, 4, or 5 words: eligible for full or partial credit (1 or 2 marks).
6 or more words: 0 marks (automatically disqualified for being over-length, even if the phrasing is semantically flawless and natural).
The Contraction Trap
One of the most frequent causes of lost marks among intermediate candidates is the careless use of contracted verb forms. Under official Cambridge assessment criteria, contractions count as two words , reflecting their constituent lexical parts:
- didn't = did not = 2 words
- haven't = have not = 2 words
- she's = she is or she has = 2 words
- they'd = they would or they had = 2 words
- wasn't = was not = 2 words
- The One Documented Exception: Can't
The B2 First handbook spells this out for Parts 2 and 4: candidates "must count contracted words (with the exception of can't = cannot ) as the full form (e.g. didn't = two words did not )". So can't is the single contraction Cambridge treats as one word, because its uncontracted equivalent cannot is itself written as one word. Every other contraction counts as two. Even so, writing cannot rather than can't removes any doubt when you are operating near the 5-word boundary.
Hyphenated Words
A hyphenated compound that standard English orthography genuinely requires ( well-known , first-class , twenty-five ) is normally treated as one word . Cambridge does not publish a rule for hyphens in Part 4, so do not build an answer that only fits inside the 5-word limit because of a hyphen: rephrase so the count is unambiguous.
The Golden Exam Rule: NEVER use contractions in Part 4. Always write out full, uncontracted grammatical forms ( did not , would have , could not , is not ). Writing uncontracted forms prevents counting miscalculations, guarantees clarity on your answer sheet, and ensures your tally matches the official marking key.
The Invariant Key Word Rule
The word printed in bold capital letters between the two sentences is the key word (or stem). It is completely immutable:
Do NOT change the part of speech: Changing the noun DECISION into the verb decide results in an immediate score of zero marks .
Do NOT alter verb tense or inflection: Changing SEE into saw , seen , or seeing scores zero marks .
Do NOT add prefixes or suffixes: Changing LIKE into dislike or likely scores zero marks .
Do NOT change singular to plural or vice versa: Changing SUGGESTION into suggestions scores zero marks .
Do NOT alter spelling: Any misspelling of the key word invalidates the entire response.
If you modify the key word by even a single letter, your response receives 0 marks for both halves of the item , regardless of whether the rest of your sentence is grammatically sublime.
The Dual-Point Marking Scheme: Unlocking Partial Credit
Unlike Parts 1, 2, and 3—where each gap yields either 1 mark or 0 marks—Part 4 awards up to 2 marks per question . Cambridge assessment guidelines divide every official answer key into two distinct scoring segments separated by a vertical bar: [ Mark 1 ∣ Mark 2 ] \mathbf{[\text{Mark 1}\ |\ \text{Mark 2}]} [ Mark 1 ∣ Mark 2 ]
Segment 1 (Mark 1): Typically tests a specific lexical chunk, phrasal verb particle, dependent preposition, or fixed idiom.
Segment 2 (Mark 2): Typically tests grammatical agreement, auxiliary verb selection, clause connector, or non-finite verb pattern (gerund or infinitive).
Because each segment is marked independently, candidate performance on each item falls into one of three tiers:
2 marks (Full Credit): Both Segment 1 and Segment 2 are 100% grammatically and orthographically accurate.
1 mark (Partial Credit): One segment is entirely correct, while the other segment contains an error (such as a wrong preposition, incorrect tense, or missing particle).
0 marks (Zero Credit): Both segments are incorrect, the word count is violated (<2 or >5), or the invariant key word was modified.
Worked Exemplar: Deconstructing Cambridge Partial Credit
To understand how examiners assign marks in real testing conditions, analyze this standard B2 First item:
Lead-in Sentence: I haven't seen Mark for nearly five years.
Key Word: LAST
Target Sentence: It ____________________________________ I saw Mark.
Model Marking Split (worked example, not a published Cambridge key): is almost five years | since OR has been almost five years | since
- Mark 1: is almost five years / has been almost five years [1 mark]
- Mark 2: since [1 mark]
Let us evaluate five authentic student submissions against this benchmark:
Candidate A writes: is nearly five years since (5 words)
Segment 1 ( is nearly five years ): Completely correct semantic transfer and tense agreement. Awarded 1 mark.
Segment 2 ( since ): Completely correct temporal connective. Awarded 1 mark.
Total Score: 2 marks (Full credit).
Candidate B writes: is nearly five years when (5 words)
Segment 1 ( is nearly five years ): Correct. Awarded 1 mark.
Segment 2 ( when instead of since ): Incorrect temporal conjunction. Awarded 0 marks.
Total Score: 1 mark (Partial credit awarded for mastering the main clause structure!).
Candidate C writes: has been nearly five years that (6 words)
Total word count is 6 words, violating the strict 5-word ceiling.
Total Score: 0 marks (Disqualified automatically, regardless of segment accuracy).
Candidate D writes: was five years since (4 words)
Segment 1 ( was five years ): Incorrect (wrong copular tense was and omits the vital adverbial meaning of nearly/almost ). Awarded 0 marks.
Segment 2 ( since ): Correct temporal connector. Awarded 1 mark.
Total Score: 1 mark (Partial credit salvaged by supplying the correct linker!).
Candidate E writes: is nearly five years lasting (5 words)
Modified the invariant key word LAST into the participle lasting .
Total Score: 0 marks (Immediate invalidation of the entire item).
The 5-Step Transformation Solving Protocol
To eliminate erratic guessing and secure consistent 2-mark scores, apply this systematic 5-step operational protocol to every Part 4 question:
Step 1: Scan & Compare: Read the lead-in sentence and the target frame side by side. Mentally cross out or identify all information already expressed in the second sentence. Isolate the exact semantic kernel that is missing.
Step 2: Analyze the Key Word: Determine the part of speech of the key word. Activate its associated collocations, dependent prepositions, or grammatical patterns (e.g., if the key word is PREVENT , activate the structural template prevent + object + FROM + -ing ).
Step 3: Synthesize the Target Bridge: Construct the grammatical transformation around the key word. Ensure that the tense, aspect, and voice of the second sentence faithfully match the temporal reality of the prompt.
Step 4: Audit Word Count: Count the words in your draft answer individually on your fingers. Expand all contractions into full words ( did not , would rather ). Confirm that the total word count is strictly between 2 and 5 words inclusive.
Step 5: Meaning & Orthography Check: Read the completed second sentence aloud in your head. Does it mean EXACTLY the same as the first sentence? Is the spelling 100% correct? Did you keep the key word completely unchanged? If yes, transfer your answer in clear capital letters.