How to Learn Academic Vocabulary That Sticks

Why Do Word Lists Fail to Improve Your Writing?

Recognising a word on a flashcard and producing it correctly in a sentence are two different skills, and a plain word list only trains the first one. You can read "ubiquitous" a dozen times, know it means "everywhere," and still write "the problem is very common in all the schools" in your essay because the word never got attached to how it actually behaves in a sentence. The techniques in the site's vocabulary memorization guide, keyword mnemonics, spaced review, and self-testing, work well for locking in a single meaning; they do not by themselves teach a word's forms, its partner words, or its register, which is where this page picks up. Three failure modes explain most of the gap between a studied word list and usable academic writing.

If unfamiliar academic vocabulary, rather than raw reading speed, is what makes your course readings feel slow, the site's reading speed test is a quick way to check which one is actually holding you back before you invest time in either.

• 01Passive recognition, not active recall. Multiple-choice flashcard apps ask you to pick the right definition. Writing an essay asks you to generate the word from nothing. Recognition and generation use different memory pathways, and only generation practice builds the second one.
• 02One form learned, four needed. You memorise "analyse" as a verb but never practise "analysis," "analytical" or "analytically." In your own writing you default to whichever form you happened to study, even when the sentence needs a different one.
• 03No sense of which words go together. A word list teaches meaning in isolation. It does not teach that you "conduct" research but "carry out" an experiment, or that a hypothesis is "tested," not "proved right." Getting the meaning correct while getting the partner word wrong is what makes a sentence sound translated rather than written.

What Are Academic English Word Families?

A word family is one root plus its regular derivations across parts of speech: the noun, the verb, the adjective and the adverb built from the same stem. This is exactly how the field's best-known resource, the Academic Word List, is organised. Averil Coxhead built the list for her MA thesis at Victoria University of Wellington by analysing which words recurred across a 3.5-million-word corpus of 414 academic texts spanning arts, commerce, law and science, then published it as Coxhead (2000), "A New Academic Word List," in TESOL Quarterly. The list groups its 570 entries into 10 sublists of word families rather than 570 flat, disconnected words, precisely because a learner who knows "analyse" is most of the way to "analysis," "analytical" and "analytically" once the pattern is drilled explicitly (Victoria University of Wellington, accessed September 2026).

The practical payoff: study one headword's whole family and you can usually fill four slots in an essay, not one. Below is the family for "analyse," the exact example the university's own guidance uses to illustrate the concept.

Word family mode in the drill above works the same way: it shows you one headword and a sentence frame tagged with the part of speech the blank needs, and you have to produce the matching form from memory, not pick it from a list.

Why Is Academic Collocation the Real Difficulty?

Once you know a word's forms, the next wall is collocation: which verbs and adjectives conventionally pair with a given academic noun. English speakers "conduct" an analysis, "draw" a conclusion, "test" a hypothesis and "reach" a consensus. Swap the pairings and the grammar stays perfectly correct while the sentence stops sounding like it was written by someone fluent in the register. "We did an analysis of the data" is grammatical; "we conducted an analysis of the data" is what actually appears in published academic writing. Neither a dictionary entry nor a bare word list tells you this, because collocation is a property of usage, not of definition. It is also a different category of problem from the errors on the site's common grammar mistakes in student writing list: those are rule violations a proofreading pass can catch, while a wrong collocation is grammatically clean and only sounds foreign to a reader who already knows the register.

Collocation mode in the drill presents an academic noun with four candidate partner words, one authentic and three drawn from other headwords' collocations, so a wrong choice always gets contrasted against the noun it actually belongs with.

How Do You Swap Informal Verbs for Formal Academic Ones?

Register is the last piece: choosing the academic equivalent of an everyday verb without tipping the sentence into sounding inflated. The goal is not to replace every plain word with a longer one. "The results show that emissions fell" does not need to become "the results demonstrate that emissions have exhibited a decline." Register shifts earn their place when the everyday verb is genuinely vague about the writer's stance: "show" can mean prove, illustrate, indicate, or reveal, and picking the precise academic verb tells the reader which one you mean. Overcorrecting, by formalising words that were already precise, is its own error and reads as padding.

Notice what did not change: sentence length stayed similar, and no word was swapped just to sound impressive. Five verbs moved from vague-but-common to specific-but-formal, and the paragraph reads as a report rather than a conversation. Register mode in the drill trains exactly this swap, one verb at a time, in a sentence specific enough that only the precise academic verb fits.

How Do You Build a Study Set From Your Own Reading?

The most durable vocabulary comes from words you actually met in your own course reading, not from a list a website chose for you. Use this five-step routine on one article or textbook chapter a week.

1. Mark, don't stop. Read the text once for meaning. Underline (or highlight in a PDF) any word that is clearly academic in register but not part of your everyday vocabulary. Keep reading; do not look anything up yet.
2. Cut the list to five. Go back and pick the five most useful marked words, the ones you can imagine needing in your own essay on this topic. A list of thirty overwhelms; a list of five gets learned.
3. Find the whole family. For each of the five, look up the noun, verb, adjective and adverb forms and note which ones do not exist. A dictionary that shows word forms (most learner dictionaries do) makes this a two-minute step per word.
4. Copy the real collocation. Find the exact sentence from your source text that used the word, and note which verb or adjective it was paired with there. That authentic pairing is worth more than a generic example sentence, because it is one you have already seen work.
5. Write one original sentence per word. Use the word about your own course material, not the source text's topic. This is the step that turns a noted word into a usable one, and it is also where the vocabulary list template gives you a column for it so the sentence does not get lost.

Run this routine weekly and you accumulate roughly 20 new headwords a month drawn entirely from material you are already required to read, each one already anchored to a real sentence and a real collocation rather than a generic dictionary example. Once a week's set is written up, the flashcard maker turns the same five rows into a reviewable deck in a couple of minutes.

How Often Should You Review a New Academic Word?

A word family is not "learned" after one good study session; it needs to be retrieved again while you are still likely to forget it, which is what makes each successful retrieval stronger than the last. The Academic Word List's own guidance for learners recommends reviewing a new word a few minutes after first meeting it, then about an hour later, then the next day, then a week later, and again after a couple of weeks (Victoria University of Wellington, accessed September 2026). That schedule is the same expanding-interval logic behind the site's spaced repetition method: each successful recall doubles or triples the gap before the next one, so a word that keeps getting produced correctly needs less and less deliberate review over time.

In practice this means a headword from this week's reading should reappear in your drill rounds this week, once more next week, and then only occasionally after that, exactly the rotation the drill's shuffled item pool is built to support when you keep coming back to it across sessions.

What Does Each Drill Mode Actually Train?

What Errors Does This Drill Catch, and Why?

When the drill marks an answer wrong, it always states the target form and the specific rule above, rather than just a red mark, because the rule is what transfers to your next essay.

Should You Memorize the Whole Academic Word List?

No. Working through all 570 word families of the Academic Word List end to end, in sublist order, treats it as a list to finish rather than a resource to consult. The list itself is not evenly useful: Sublist 1's 60 families cover about 3.6% of the running words in the academic corpus it was built from, while Sublist 9's 60 families cover only about 0.2%, meaning a learner who reads Sublist 1 will meet those words roughly every 4 pages, and a learner deep in Sublist 9 will meet those words roughly every 67 pages (Victoria University of Wellington, accessed September 2026). Memorising Sublist 9 before you have consolidated Sublist 1 spends the same effort for a fraction of the payoff.

What Does a Weekly Academic Vocabulary Routine Look Like?

Fifteen minutes a day is enough if the fifteen minutes are spent retrieving, not re-reading. One workable split across a seven-day week:

That is four short drill sessions across the three modes and one longer review session, which matches the extraction routine's roughly 20 new headwords a month at a pace that leaves room to actually retrieve last week's words instead of only meeting new ones. The payoff shows up under conditions the drill cannot simulate: a timed essay exam, where the site's essay exam tips for writing under time pressure only work if the academic word you need is already available, not something you have to hunt for while the clock runs.

Sources

• The Academic Word List, Victoria University of Wellington. Overview of the list and its structure. Accessed September 2026.
• Academic Word List information, Victoria University of Wellington: origin (Averil Coxhead's MA thesis), the 570-word-family count, and the citation Coxhead, Averil (2000), "A New Academic Word List," TESOL Quarterly, 34(2): 213–238. Accessed September 2026.
• The sublists, Victoria University of Wellington: the 10-sublist structure, per-sublist family counts and corpus-coverage percentages cited above. Accessed September 2026.
• The Academic Corpus, Victoria University of Wellington: corpus size (approximately 3.5 million running words, 414 texts across Arts, Commerce, Law and Science). Accessed September 2026.
• How to use this list, Victoria University of Wellington: the university's own spaced-review guidance for learning new words, cited in the spacing section above. Accessed September 2026.
• Table 1 (CEFR 3.3): Common Reference levels, global scale, Council of Europe. Official B2 and C1 level descriptors used to set this drill's difficulty ceiling. Accessed September 2026.

FAQ