TOEIC Link Writing — Lexical Density and Information Packaging Control: How Content-Word Ratio Engineering Moves the Vocabulary Band from 23 to 28
Lexical density is the most teachable vocabulary discriminator on the TOEIC Link writing module that almost no candidate explicitly trains. The metric — the ratio of content words to total words in a passage — is a direct proxy for how much information each clause carries, and the scoring rubric rewards higher densities sharply at band 26 and above. Internal practice-corpus analysis indicates that candidates in the 23-to-25 band write at densities around 0.42 to 0.46, while candidates in the 27-to-30 band write at densities around 0.52 to 0.58. The four-to-twelve point gap is not vocabulary breadth — both groups know the words — it is the engineering of how the known words are packaged into clauses, and that engineering is closable through a four-week protocol.
The TOEIC Link writing module tests information-packaging control across four task types — email response, opinion essay, summary task, and integrated reading-writing task — and each task type rewards a distinct packaging signature that aligns with the task's rhetorical register. For broader context on the writing module, see the writing vocabulary precision and collocation discipline guide, the writing tone and register control guide, and the writing theme-rheme progression and topic continuity guide.
What lexical density actually measures
Lexical density is the ratio of content words (nouns, verbs, adjectives, adverbs) to total words (content words plus function words such as articles, prepositions, conjunctions, auxiliary verbs, and pronouns). A density of 0.50 means half the words in the passage carry semantic content; the other half carry grammatical structure.
Spoken English typically runs at 0.36 to 0.42. Conversational written English runs at 0.42 to 0.48. Professional written English runs at 0.50 to 0.58. Academic and policy writing runs at 0.55 to 0.65. The TOEIC Link writing module rewards candidates whose density signature matches the register of the task — professional written English for email response and opinion essay, academic register for summary task and integrated reading-writing task.
The metric is robust because it is hard to fake. A candidate cannot simply substitute longer words for shorter ones to raise density. Density rises only when the candidate restructures clauses to remove function-word scaffolding — converting prepositional phrases into compound nominals, converting auxiliary-verb constructions into single lexical verbs, converting subordinate clauses into participial phrases. Each restructure removes function words and concentrates meaning.
The three packaging mechanisms
Mechanism 1 — Nominalization
Nominalization converts a clause into a noun phrase, removing the auxiliary verb, the subject pronoun, and the conjunction that the clause required. Example transformation: "The committee decided that the proposal should be approved" (eleven words, six content) becomes "The committee approved the proposal" (five words, four content). Density rises from 0.55 to 0.80. The technique is the default for opinion-essay and summary-task writing where clause economy is rewarded sharply.
The pattern works best when the nominalized form is a single-word or two-word noun phrase. Three-word and longer nominalizations begin to introduce their own function-word scaffolding and the density gain reverses. Candidates should drill ten clause-to-noun-phrase conversions per day on professional source material until the conversion is automatic.
Mechanism 2 — Participial reduction
Participial reduction converts a finite subordinate clause into a non-finite participial phrase, removing the subordinator, the subject, and the auxiliary verb. Example transformation: "The proposal that was submitted by the operations team addresses the bottleneck" (twelve words, seven content) becomes "The proposal submitted by the operations team addresses the bottleneck" (ten words, seven content). Density rises from 0.58 to 0.70.
The pattern is the default for descriptive paragraphs where the candidate needs to attach modifying information to a noun phrase without breaking the main clause's rhythm. Participial reduction works for both present participles (active voice) and past participles (passive voice). The technique is mechanically simple and can be drilled to fluency in roughly two weeks.
Mechanism 3 — Compound nominal construction
Compound nominal construction takes two or three nouns and combines them into a single noun phrase that removes the prepositional connector. Example transformation: "The strategy for the integration of the channel partners" (nine words, four content) becomes "The channel-partner integration strategy" (four words, three content). Density rises from 0.44 to 0.75.
The pattern is the default for technical and business writing where compound nominals carry domain-specific meaning concisely. The technique is high-leverage but high-risk because over-compounded noun phrases ("channel-partner integration strategy implementation governance framework") become unreadable. Candidates should keep compound nominals to a maximum of four nouns and should always test readability aloud after construction.
The five over-packing failure modes
Failure 1 — Function-word starvation
The candidate removes so many function words that the prose becomes telegraphic and the grammatical relations between clauses become ambiguous. The output reads as note-form and the scoring rubric penalizes it under the clarity category. The remediation is to drill density-target writing where the candidate aims for a specific density range (0.52 to 0.55) rather than maximizing density.
Failure 2 — Compound nominal overstacking
The candidate constructs noun phrases of five or more compounded nouns that the reader cannot parse without re-reading. The output reads as jargon-dense and the scoring rubric penalizes it under the clarity category. The remediation is to drill compound nominal decomposition where the candidate takes existing over-compounded phrases and re-breaks them into two or three shorter noun phrases joined by prepositions.
Failure 3 — Nominalization-induced agency loss
The candidate nominalizes verbs so aggressively that the agent of the action disappears from the sentence. Example: "The decision was made" hides who decided. The output reads as evasive and the scoring rubric penalizes it under the precision category at band 27 and above. The remediation is to drill agent-restoration exercises that retrofit nominalized drafts to name the responsible actor.
Failure 4 — Register-density mismatch
The candidate writes at academic density (0.60+) for an email-response task that the rubric expects to be at professional density (0.50 to 0.55). The output reads as stiff and inappropriate to the genre, and the scoring rubric penalizes it under the register category. The remediation is to drill density-by-genre exercises that tag target densities to each task type before drafting.
Failure 5 — Participial reduction ambiguity
The candidate reduces a subordinate clause to a participial phrase that allows two possible attachment readings (which noun does the participle modify). The output reads as ambiguous and the scoring rubric penalizes it under the clarity category. The remediation is to drill participle-attachment audits that ask "which noun does this participle modify" for every reduced clause in the draft.
The four-week drill protocol
Week 1 — Density measurement on professional source material
The candidate spends the first week building density-recognition fluency on existing high-quality writing. The drill routine is to take ten paragraphs per day from professional sources (annual reports, policy memos, business correspondence), measure the density of each paragraph by counting content words and total words, and produce a density-distribution note that documents the typical density per genre. The week's output is a seventy-paragraph density-distribution corpus.
Week 2 — Single-mechanism transformation drills
The candidate spends the second week drilling each of the three mechanisms in isolation. The drill routine is to take five low-density paragraphs per day from the candidate's own previous writing and rewrite each paragraph three times, each time emphasizing one mechanism (nominalization, participial reduction, compound nominal). The week's output is a 75-paragraph transformation corpus that documents the candidate's command of each mechanism and the density delta each mechanism produces.
Week 3 — Density-target drafting
The candidate spends the third week drafting paragraphs to a specific density target. The drill routine is to take five topic sentences per day and write three paragraphs per topic sentence, each at a different density target (0.48, 0.54, 0.60). The week's output is a 75-paragraph density-controlled production corpus that demonstrates the candidate's ability to hit a specified density on demand.
Week 4 — Genre-matched packaging under timed conditions
The candidate spends the fourth week writing full task responses under timed conditions with explicit density targets matched to the task genre. The drill routine is to take ten TOEIC Link writing prompts per day (across all four task types), draft a full response under the time limit, measure the density of the response, and self-assess against the target density for that task type. The week's output is a 70-response timed production corpus that demonstrates target-matched packaging under exam conditions.
Density targets by TOEIC Link writing task type
Email response task
Target density: 0.50 to 0.54. The email-response task rewards professional written English with conversational accessibility. Densities above 0.55 read as stiff and inappropriate to the email genre; densities below 0.48 read as casual and underdeveloped. The mechanism mix should emphasize participial reduction and selective nominalization, with compound nominals reserved for genuinely technical content.
Opinion essay task
Target density: 0.54 to 0.58. The opinion-essay task rewards higher-density argumentative writing with clear claim-evidence-warrant structure. Densities above 0.60 begin to lose argumentative clarity; densities below 0.52 read as underdeveloped. The mechanism mix should emphasize nominalization for converting claim-clauses to noun-phrase subjects and participial reduction for attaching evidence to claims efficiently.
Summary task
Target density: 0.56 to 0.62. The summary task rewards the highest density of the four task types because the candidate must compress source material into a limited word count. Densities above 0.65 begin to lose readability; densities below 0.54 indicate underuse of the compression mechanisms the task is designed to test. All three mechanisms should be deployed actively, with compound nominal construction particularly valuable for capturing source-text terminology efficiently.
Integrated reading-writing task
Target density: 0.55 to 0.60. The integrated task rewards academic-register writing that synthesizes reading-source material with the candidate's own analysis. Densities should rise slightly above opinion-essay levels because the candidate is paraphrasing source material in addition to making original arguments. The mechanism mix should emphasize nominalization for converting source-text claims to noun-phrase form and compound nominal construction for capturing source-text key terms.
Common density mismatches and their score penalties
A candidate writing at 0.42 density on an opinion essay task is signaling conversational register where the rubric expects professional register. The score penalty typically falls in the two-to-three point range at band 26 and above. A candidate writing at 0.62 density on an email-response task is signaling academic register where the rubric expects professional register. The score penalty typically falls in the one-to-two point range at band 25 and above. A candidate writing at 0.45 density on a summary task is signaling underuse of the compression mechanisms the task is designed to test. The score penalty typically falls in the three-to-four point range at band 24 and above.
Measurement protocol for self-assessment
The candidate should adopt a weekly density measurement routine that runs in parallel to all other drill work. The routine is to take three full task responses per week (one of each long-form task type), count content words and total words in each response, compute the density, and log the result in a personal density-trajectory spreadsheet. Over twelve weeks, the trajectory should show a steady climb from baseline density toward the genre-target densities, with the climb plateauing as the candidate's instinct for genre-matched packaging stabilizes.
What success looks like at band 28
A band-28 candidate produces an email-response task at 0.52 density, an opinion-essay task at 0.56 density, a summary task at 0.58 density, and an integrated reading-writing task at 0.57 density. The candidate has internalized the genre-density mapping and produces densities within target ranges without conscious counting. The candidate also produces densities that vary across paragraphs within a single response — typically 0.48 for opening paragraphs (which front-load orientation) and 0.58 for body paragraphs (which carry the argumentative load). The intra-response variation is itself a band-28 signal because it demonstrates conscious packaging control rather than mechanical density-maximization.
For the broader writing-skill framework that lexical density fits into, see the writing task types and scoring criteria guide and the writing claim-evidence-warrant paragraph construction guide for the argumentative-paragraph engineering that packaging control sits on top of.