Does Extensive Reading Work? What the Research Shows, and Why Some Say It's Useless
Does extensive reading work? Many studies report that it does. The skills that tend to improve most are reading speed and reading comprehension; a single book leaves only a few new words behind, and reading alone is unlikely to build speaking skills. This article sums up what research reviews have found, checks the common reasons people call extensive reading useless against the studies, and shows a simple way to measure your own progress in numbers.

Key takeaways
- In meta-analyses of extensive reading studies, groups who read extensively outperformed groups who did not, with a small-to-medium effect
- The skills that improve most readily are reading speed and comprehension. One book leaves only a few new words, so you need to read a lot
- Building speaking and writing skills is thought to require time spent on more than reading
- Feeling that extensive reading is useless usually comes from a mismatch in one of five areas: vocabulary expectations, speaking expectations, reading volume, book difficulty, or post-reading activities
- You can check the effect by measuring words read per minute together with comprehension, once a month
Does extensive reading work? The short answer from the research
For learners of English, many studies report that extensive reading works. The skills that tend to improve most are reading speed and the ability to understand what you read. On the other hand, a single book leaves only a few new words behind, and speaking and writing skills are unlikely to grow from reading alone. Most people who feel extensive reading is useless are not failing because of extensive reading itself; rather, one of five things differs from the conditions under which studies found an effect: vocabulary expectations, speaking expectations, reading volume, book difficulty, or post-reading activities.
In this article, we look at two papers that pull together extensive reading research (meta-analyses, which collect the results of many studies and average them) to sort out what extensive reading helps with and what it helps less with. Then we check the common reasons people call it useless against the research, one by one, and finally show how to measure in numbers whether your own extensive reading is paying off. For how to get started and how to choose books, see How to start extensive reading in English for free (in Japanese).
What research reviews have found
There are dozens of studies worldwide on the effects of extensive reading. Because individual results vary, here we look at two papers that combined and averaged many of them.
| Paper | Studies included | Findings |
|---|---|---|
| Nakanishi (2015, TESOL Quarterly) | 34 studies, 3,942 participants | Groups who read extensively outperformed groups who did not. The effect size was "medium." Comparing before and after reading showed even larger gains. Effects were larger for older learners and for longer programs |
| Jeon and Day (2016, Reading in a Foreign Language) | 49 studies (1980–2014), 5,919 participants | The effect was "small to medium." The average person in the extensive reading group scored above 70% of the group that did not read extensively. Adults showed larger effects than secondary school students, and programs that combined reading with other activities showed larger effects |
The Jeon and Day paper re-ran the analysis with stricter criteria, on the grounds that Nakanishi's analysis had included studies that did not fit the definition of extensive reading, such as studies in which learners read assigned long texts instead of choosing their own books. Even so, the overall direction was the same: extensive reading groups outperformed groups taught with intensive reading (translating sentence by sentence) or conventional reading classes.
To put "above 70%" in everyday terms: if you lined up 100 people who did not read extensively and placed the middle person of the extensive reading group among them, that person would stand at about 28th from the top. It is not a dramatic transformation, but it is a difference that is hard to put down to chance, and it shows up again and again.
Researchers have also looked at which skills improved. In the Jeon and Day analysis, the effect was largest for reading speed, followed by reading comprehension and then vocabulary.
The paper itself notes, however, that the differences among these three are not clear-cut and may reflect the small number of studies on reading speed. The reading most faithful to the original is: "It tends to help reading speed, and it helps vocabulary too, though a little more modestly than the others."
Underlying the idea of extensive reading is the "reading hypothesis" proposed by the linguist Krashen. As the title of his 1989 paper puts it, the idea is that we acquire vocabulary and spelling by reading. It remains a hypothesis, but the two reviews above show, as an average across many studies, that groups who read extensively improved more than groups who did not.
Why some say extensive reading is useless, checked against the research
Alongside the reports of positive effects, you often hear that extensive reading is useless. Below we take five commonly cited reasons and pair each with what research has found and what to do about it. Reasons 1 and 2 are mismatches in what you expect extensive reading to build; reasons 3 to 5 are mismatches in how it is done.
1. I can't remember new words
This matches the research. Waring and Takaki (2003) had 15 Japanese learners of English read one easy book written with 400 basic words, and tested how many of 25 words planted in the book they retained. Three months later, the learners remembered on average about 1 of the 25 words, and words met fewer than 8 times in the book were not retained at all after three months.
But the paper's conclusion was not "extensive reading is a waste." It was "one book leaves almost no new words, so you need to read a lot." In addition, Pigada and Schmitt (2006), who followed one learner of French for a month, found that when they measured not only meaning but also spelling and usage, some knowledge had increased for 65% of the words studied. The paper points out that measuring only meaning, and not counting partial knowledge, tends to underestimate gains.

What to do: if you want to build vocabulary, the shortcut is to keep reading the same series or the same genre, so you meet the same words again and again. If you have specific words you want to learn, adding separate practice with a word list also fits the thinking behind the research (explained in reason 2).
2. I'm not learning to speak
This, too, is a mismatch in what you expect from extensive reading. Nation (2007), known for his vocabulary research, divides language learning into four strands: receiving meaning (reading and listening), expressing meaning (speaking and writing), deliberately studying the language itself (grammar, pronunciation, vocabulary), and becoming fluent with what you already know. He recommends spending roughly equal time on each. Extensive reading belongs to the first strand.
Some studies have found that extensive reading improved writing, but it is natural to assume that building speaking skills takes time actually spent speaking. Adding small bits of output, such as summarizing the story you read in one minute out loud or writing three lines about what you thought, lets you start using the English you have read.
3. Nothing has changed after months
Jeon and Day say that the effects of extensive reading are slow to show and take time. When they compared programs by length, there was no clear difference between programs of under six months, about six to ten months, and more than a year (only four studies lasted over a year). Nakanishi's analysis found larger effects for longer programs, but Jeon and Day, with stricter criteria, found no difference. How much you read during that time is a better clue than how long you have been doing it.
For a benchmark on volume, there is a study by Beglar and Hunt (2014). The 14 Japanese university students whose reading speed improved most over one year had read about 209,000 words on average (counted in "standard words," with 6 characters counted as one word). From this result, the two researchers proposed 200,000 words a year as an evidence-based minimum. Spread over a year, that is about 550 words a day. The 100,000-word and 1,000,000-word goals in the getting-started article are targets for keeping you going, whereas 200,000 words a year comes from the average amount actually read by students whose reading speed improved. If you feel no change after a few months, first check whether your reading volume reaches this benchmark.
4. The books were too hard
In the same study by Beglar and Hunt, the students who improved had mainly read books rewritten in easier language for learners (graded readers), and these were found to be more effective for building reading speed than unsimplified books. The study's conclusion is that, for learners below the intermediate level, reading books they could read comfortably, even if below their level, was what worked.
Reading slowly while looking words up in a dictionary is useful as intensive reading, but it is not the condition under which extensive reading studies found an effect. For guidance on how easy a book should be, see How to start extensive reading in English for free (in Japanese).
5. I only read, and did nothing afterward
In the Jeon and Day analysis, programs that built extensive reading into the whole course had the largest effect, and stand-alone extensive reading classes had the smallest (only one study looked at stand-alone classes, so this cannot be stated firmly). The paper says effects were higher when reading was combined with activities such as vocabulary exercises, discussions, and presentations.
Even when you read on your own, adding a little "after reading," such as talking with someone about the book, or noting an expression you liked and trying to use it the next day, helps the English you read stick.
Here are the five reasons in order, so you can check which one applies to you.
Measure whether your extensive reading is working
Reading speed, where studies found larger effects, is hard to see unless you measure it yourself. So once a month, measure two things together: how many words you can read per minute, and how well you understood the content. For how to keep a running total of words read, see the getting-started article (in Japanese).
Reading speed means nothing if your eyes race over text you don't understand. Research, too, has measured speed while checking that comprehension did not drop below 70% (Shimono, 2018). Always record speed and comprehension as a pair.
Here are the four steps.
- Choose one easy text whose word count you know (such as one chapter of a graded reader)
- Read it silently at your normal pace and time how long it takes
- Work out your words per minute: word count ÷ seconds taken × 60
- Without looking at the text, answer questions that check your understanding
Use a text you have not read before, don't reread too much, and read the way you usually do. A stopwatch on your phone is enough for timing. For example, if you read 400 words in 3 minutes 20 seconds (200 seconds), that is 400 ÷ 200 × 60 = 120 words per minute.
Graded readers sometimes come with questions; if yours doesn't, you can ask an AI to write some.
Please write questions to check whether someone who read the English text below understood its content. - 5 questions in English, each with 4 answer choices - Only ask about things clearly stated in the text - Do not write questions that can be answered by common sense without reading the text - Put all the answers together after the questions [Text] (Paste the text you read here)
If you look at the questions first, you end up reading to hunt for answers, so open them only after you finish reading. To match the 70% used in research, aim here for at least 4 of 5 questions (80%) correct.
You don't need to measure every day. Once a month, measure with a different text of similar difficulty, and write down speed and comprehension side by side.
| Month | Word count | Time taken | Words per minute | Questions | What it tells you |
|---|---|---|---|---|---|
| Month 1 | 400 words | 200 sec | 120 words | 4 of 5 | Use this as your starting point |
| Month 2 | 380 words | 175 sec | About 130 words | 5 of 5 | Both speed and comprehension went up. You're improving |
| Month 3 | 420 words | 180 sec | 140 words | 3 of 5 | Speed went up but comprehension dropped. Restore comprehension before speed |
Date measured: Word count: words Time taken: min sec (= sec) Words per minute: words (word count ÷ seconds × 60) Comprehension questions: of 5 correct
If, as in month 3 of the example, your speed rises but comprehension falls, you are focusing too much on reading fast. You are truly improving when you keep both speed and comprehension.
To read a lot, choose stories you want to keep reading
What the research tells us is: "easy English, and lots of it." The thing that helps most with that may be a story that makes you want to read the next page. Original Novel (original-novel.com), a collection of stories made with AI, also has works you can read in English. They are divided into chapters, which makes it easy to read a little each day.
Stories that make you want to keep reading
See more: Stories that make you want to keep reading ›If you are going to read a lot, start with stories you can't put down. On the story search page, you can filter for works you can read in English by text language.
The Only Voice I Can't HearFor teens · Showbiz & idolsCompletedMay We Begin With a Lie?For adults · RomanceThe Jingjok Behind the BlackboardFor kids · SchoolSources (date checked)
- The effectiveness of ER on reading proficiency: A meta-analysisReading in a Foreign Language (Jeon & Day, 2016) / Checked Nov 18, 2026
- At what rate do learners learn and retain new vocabulary from reading a graded reader?Reading in a Foreign Language (Waring & Takaki, 2003) / Checked Nov 18, 2026
- Vocabulary acquisition from extensive reading: A case studyReading in a Foreign Language (Pigada & Schmitt, 2006) / Checked Nov 18, 2026
- Pleasure reading and reading rate gainsReading in a Foreign Language (Beglar & Hunt, 2014) / Checked Nov 18, 2026
- L2 reading fluency progression using timed reading and repeated oral readingReading in a Foreign Language (Shimono, 2018) / Checked Nov 18, 2026
FAQ
Does extensive reading work for children and teenagers too?
Positive results have been reported, but in the meta-analysis by Jeon and Day (2016), the effect was largest for adults, followed by children and then secondary school students. There are few studies of children and teenagers, however, and the authors themselves say more research is needed. As a possible reason for the smaller effect among secondary school students, the paper suggests they may find it harder to take an interest in reading that does not directly affect their grades.
Is extensive reading better on paper or online / as e-books?
In the analysis by Jeon and Day (2016), stories read online showed a larger effect than paper books. However, there were only 10 online studies, and most of them involved adults, who show larger effects anyway, so the paper warns that this result should be interpreted with caution. The question is not settled yet, so choose whichever format you find easier to keep up.
Which works better for English skills: extensive reading or intensive reading (translating sentence by sentence)?
The meta-analyses report that groups who read extensively outperformed groups taught with intensive reading or conventional reading classes. Still, time spent consciously studying grammar and vocabulary is also considered essential to language learning, so you do not have to choose one or the other. A realistic approach is to read difficult texts intensively and easy texts extensively.



