Extensive reading means reading a lot of easy, enjoyable text for the meaning rather than to study it. Paul Nation estimates that a learner who knows about 1,000 words should read at least 10,000 words a week. Studies of vocabulary coverage suggest choosing texts in which around 98% of the words are already familiar.
How easy should the text be?
Researchers measure this as coverage: the share of the running words in a text that the reader already knows. These are the figures you will see quoted most often, and what each study actually found.
| Study | What it found |
|---|---|
| Laufer (1989) | About 95% coverage best separated learners who reached the pass mark on a comprehension test (55%) from those who didn't. |
| Hu & Nation (2000) | A 633-word story was rewritten with 100%, 95%, 90% and 80% known words. No reader of the 80% version understood it adequately, and only a minority at 90% and 95% did. About 98% looked like a satisfactory level for unassisted extensive reading. |
| Schmitt, Jiang & Grabe (2011) | Across 661 learners in 8 countries reading academic texts, comprehension rose steadily with coverage and showed no single threshold. For these advanced readers, 95% coverage was enough for about 60% comprehension, while 70% comprehension needed 98–99%. |
| Kremmel and colleagues (2023) | A replication of Hu & Nation in Sri Lanka concluded that one uniform threshold may be questionable, and that question format and text type may change how many words a reader needs. |
So treat 98% as a guide rather than a law. In practice it means about one unfamiliar word in every 50. On a Contos A1 chapter of around 180 words, that's three or four words that make you pause, and a tap gives you their meaning without leaving the page.
How many words a week should you read?
Nation's case for volume rests on repetition. A half-learned word fades unless you meet it again soon, and in ordinary text the less common a word is, the longer the gap between meetings.
| Words you already know | You meet each new word about once every |
|---|---|
| 1,000 | 10,000 running words |
| 2,000 | 20,000 running words |
| 4,000 | 43,000 running words |
| 6,000 | 125,000 running words |
His worked example: if a learner with about 1,000 words remembers a new word for roughly a week, they need to read at least 10,000 words a week to see it again in time. That is about 40 pages of 250 words, or a graded reader every week or two. Up to the 2,000-word level, he suggests about a book a week. He adds that simplified texts repeat common words more often than the unsimplified ones behind these figures.
In Contos terms, a reader at A2 (between 500 and 1,500 known words) would cover 10,000 words in about 40 chapters, at the average Spanish A2 chapter length of 255 words. Few people manage that every week, so treat it as a direction rather than a quota: the more easy text you read, the sooner new words come round again.
How is extensive reading different from intensive reading?
Bamford and Day trace the term to Palmer, who used it to contrast with intensive reading and valued both. The two do different jobs.
| Extensive reading | Intensive reading | |
|---|---|---|
| Purpose | Pleasure, information and the gist | Studying the language of one text closely |
| Difficulty | Well within what you already know | Shorter, harder texts |
| Dictionary | Rarely needed | Dictionaries, notes and translations to hand |
| Amount | As much as possible | A passage at a time |
| Choice | Your own, and you can drop a dull book | Usually set by a course |
Waring argues that learners who only ever decode hard texts never build speed, and that fluent reading in turn makes intensive work easier. Most learners benefit from some of each.
What does extensive reading improve?
- Fluency. Waring lists faster reading, quicker word recognition and a shift from decoding single words to taking in whole ideas.
- Broad language skills. In Elley's "book flood" studies, children aged 6 to 12 given plenty of books for 12 to 36 months improved in reading comprehension, writing, word recognition, grammar and oral language. The biggest writing gains appeared after about two years.
- Reading proficiency in adults, too. Jeon and Day's 2016 meta-analysis found a larger effect for adults than for children and adolescents.
The limits are just as clear. Nation stresses that the benefits come over the long term, not in a few weeks, and that words picked up incidentally need reinforcing, so he pairs reading with direct vocabulary work. In Contos, that's the job of the flashcards built from the words you tap.
Read three chapters at your level this week and count how often you needed to tap.
Read free in your browserFrequently asked questions
What counts as extensive reading?
Reading plenty of material that is easier than your current level, fast enough that you don't need a dictionary, and chosen because you want to read it. The Extensive Reading Foundation puts the volume of reading ahead of understanding every detail.
Is it a waste of time to read books that feel too easy?
No. Waring notes that very easy material builds reading speed, which is one of the aims. Krashen reports that readers who start with light books tend to broaden their tastes over time rather than staying with easy reading.
Should I look up every word I don't know?
No. Constant look-ups slow you down and break the story. Look up a word when it blocks the meaning or keeps coming back. In Contos a tap shows the meaning without leaving the page.
How long before extensive reading shows results?
Expect months rather than weeks. Elley's book flood studies ran for one to three years, and in Krashen's survey of in-school reading, the programmes lasting longer than a school year were the ones where readers nearly always came out ahead.
Does 98% coverage mean knowing 98% of a language?
No. It's the share of the words in one particular text that you know. The same learner can be near 98% in an A1 story and far below it in a newspaper, which is why choosing the right level matters.