What actually builds a vocabulary.

What word lists can and cannot do

A systematic review of 36 vocabulary-intervention studies found that directly teaching word meanings reliably improves comprehension of passages containing those specific taught words, but found only weak evidence that vocabulary instruction improves generalised, untrained reading comprehension VOC-02.

That is worth saying plainly on a site that publishes a 617-word scholarship vocabulary list. The list is built for a particular job: the lexicon a scholarship or selective-entry paper actually draws on. Learned before that exam, it does the thing the evidence supports. What it will not do is raise comprehension in general, and a student who finishes it has not finished building a vocabulary.

Why it still matters in high school

Most vocabulary research is about young children, which makes it easy to assume the work is done by the time a student reaches Year 7. Australia’s own education research body says otherwise. In guidance written for secondary teachers it states that vocabulary knowledge alone accounts for 50–60% of the variance in reading comprehension, and names background knowledge and vocabulary as the most critical components of comprehension VOC-14.

That figure is AERO restating an earlier synthesis rather than measuring it, so treat it as the position Australia’s research body publishes for teachers rather than as a number this site has independently checked. The direction is the part to act on. A Year 10 student who understands every word of a text and still cannot write about it has a different problem from one who is quietly guessing at a quarter of the vocabulary — and the second is far more common, because the words that carry meaning in senior texts are exactly the ones nobody stops to define.

Which words are worth the effort

The most useful sorting tool is Beck, McKeown and Kucan’s three tiers VOC-06. Almost all the value sits in the middle row.

Tier 2 words are the ones that recur across subjects and are rarely picked up from conversation. They are also the words that make academic writing sound like academic writing, which is why a student can know a topic well and still write about it thinly.

Teaching word structure as well as words

A meta-analysis of morphological instruction across 22 studies from preschool to Grade 8 found that teaching students to analyse prefixes, suffixes and roots benefits literacy outcomes, with particular benefit for less able readers VOC-07. A student who can take a word apart meets fewer genuinely new words, because an unfamiliar word built from familiar parts is already half known.

That has a ceiling worth knowing, from a larger and more recent meta-analysis — 28 studies, 177 effect sizes, pre-registered. It found average effects distinguishable from zero for reading and spelling outcomes, but also that [e]ffect sizes were larger for trained words than untrained words, that there was evidence of transfer to untrained words for spelling outcomes, but not for reading outcomes, and that [t]here was also no clear evidence of effects on reading comprehension outcomes VOC-13. So word-structure teaching earns its place on spelling and on reading the words themselves. Treating it as a route to comprehension is claiming something the evidence does not show.

Where most word knowledge comes from

Vocabulary and comprehension develop reciprocally — better vocabulary produces better comprehension, which improves faster over time — and written text is a richer source of new words than everyday speech, because it carries less common vocabulary VOC-10. That is the mechanism behind the reading-volume figures: wide reading is where most vocabulary actually comes from, and direct word work supplements it.

The 30 million word gap, three ways

Any search on this topic meets the “30 million word gap”. The register holds the full sequence rather than the headline, and the sequence is the useful part.

1. The finding.Hart and Risley’s 1995 study of 42 Kansas City families estimated that by age four, children from professional families had heard about 30 million more words addressed to them than children from families on welfare VOC-03.
2. The challenge. A 2019 replication across five communities found little to no significant socioeconomic difference in words directly addressed to children, once fuller observation methods were used VOC-04.
3. The rebuttal. A peer-reviewed response argues the replication measured overheard rather than child-directed speech and lacked a high-income comparison group, and maintains that the quality of child-directed language — not raw word count — is what predicts outcomes VOC-05.

The practical residue, once the three are held together: talking with a child in a way that expects a reply matters more than the number of words spoken near them. That is a smaller and more useful claim than the headline.

Where these claims come from

Every figure above is drawn from the vocabulary domain of the Evidence Register, where each claim appears with its verbatim quote and full citation — including the contested ones, shown with that record intact.

Working on it with someone

Argo Academics is a one-to-one English tutoring practice in Melbourne, where Callum Hay teaches every lesson personally. Tier 2 vocabulary is taught where it belongs — inside the reading and writing a student is already doing, rather than as a list to be memorised beside it.

Not sure whether vocabulary is the thing holding your child back? The two-minute diagnostic asks about what you have noticed in their writing, which is a better starting point than a word count.