What “band 7 vocabulary” really means

The advice is everywhere: to reach band 7 in Writing, learn “advanced” vocabulary. So learners collect rare words and swap them in with a thesaurus. It is one of the most reliable ways to not reach band 7. The descriptors and the research both point somewhere else—toward depth, collocation, and precision.

What the descriptor actually rewards

The public band descriptors describe band 7 Lexical Resource in terms of range used with control: the writer handles some less common and idiomatic vocabulary with an awareness of style and collocation, while occasional errors are tolerated. Band 6 already credits an adequate range and an attempt at less common words—but with noticeable inaccuracy in choice and word formation. descriptor · paraphrase · ielts-descriptors The step up from 6 to 7 is not more words; it is the control with which they are used. The Writing Lexical Resource guide lays out the band-by-band moves.

What Milton and Boers actually measured

It is worth being precise here, because this research is often overstated. Milton, Wade and Hopkins (2010) measured learners’ recognition vocabulary size—how many words they could identify on a yes/no test—and correlated it with IELTS. Written vocabulary size correlated most strongly with the Writing sub-score (about .76, roughly 58% of its variance) and with Reading; aural vocabulary size tracked Speaking and Listening instead. research finding · milton-2010 That is a real relationship—but it is recognition, it is correlation, and it says a bigger vocabulary tends to come with a higher band, not that inserting rarer words raises a score.

Boers and colleagues (2006) measured something different. In a small experiment with 32 students, learners who used more formulaic sequences—collocations and set phrases—were rated as more proficient speakers by blind judges. research finding · boers-2006 The lever there was word combinations used naturally, not single rare words.

And Banerjee, Franceschina and Smith (2007), analysing real scripts across bands, found that lexical diversity and sophistication do rise with band—when the words are used appropriately. research finding · banerjee-2007 Sophistication that is misused does not read as sophistication to an examiner.

Depth is four things, not one

Knowing a word well enough to earn band-7 credit means more than recognising it. It means controlling its research finding · nation-word-knowledge:

• form—spelling and the correct part of speech;
• meaning—the precise sense, not an approximate synonym;
• collocation—the words it naturally combines with (make a decision, not do a decision);
• register—whether it fits a formal essay or belongs in casual speech.

A thesaurus gives you the word and none of the other three. That is why swapped-in rare words so often break collocation or register—and an examiner grading precisely those things notices.

Why rare words backfire

Take a real trade-off. Important is a plain word; paramount is an impressive one. But paramount collocates narrowly and carries a heavier register; used loosely it reads as inserted, and a band-7 rater is grading style and collocation. The precise common word beats the imprecise rare one. Impressive is not the same as indispensable: the words that carry an argument are usually ordinary—because, argue that, the topic’s core noun—used exactly.

What to do instead

1. Learn words inside their collocations, from real input. Meet a word in reading and note the phrase it lived in, not the word alone.
2. Prioritise precise use of the vocabulary you already control. Above band 6, accuracy and fit move the score more than novelty does.
3. Check register and collocation before reaching for a rarer word. If you are unsure it fits, the safer precise choice scores higher.
4. Grow recognition vocabulary by reading widely. That is what Milton’s correlation reflects, and it compounds slowly—not through last-minute lists.

What this does and doesn’t say

This is not “vocabulary doesn’t matter.” Milton’s correlations are real: a bigger vocabulary genuinely tends to come with a higher band. The point is how it helps—through recognition that grows from reading, and through precise, well-collocated use—not through rare words inserted for effect. The evidence is also mostly correlational, and Boers is a small experiment about perceived speaking proficiency; none of it promises a specific band from a specific word list.

So when you next meet a “band 7 word,” the useful question is not “is it rare?” but “can I use it precisely, in the right collocation and register?” If not, the ordinary word used exactly will score higher. The Writing Lexical Resource guide shows what each band actually rewards.