Skip to main content
SpeedReadTest

How Our Reading Speed Test Works

Exactly what we measure, how we score it, where the benchmarks come from, and what a short test can't tell you.

Updated by the SpeedReadTest team

SpeedReadTest times you reading a passage, then asks 4 questions about it. Your Effective Reading Speed (ERS) is your words per minute multiplied by your comprehension, so skimming doesn't earn a high score.

Test your reading speed →

The Passages

The test has 42 original passages in 5 languages, at three levels of difficulty. Each passage has a bank of 8 questions, and each test draws 4 you haven't seen yet, so retakes use different questions.

Number and length of test passages by language
LanguagePassagesLength (words)
English30179–283
Español3260–310
Deutsch3250–305
Français3265–310
Português3260–315

How We Score It

  • Words per minute: the passage's word count divided by your reading time. The timer pauses if you leave the tab.
  • Comprehension: the share of quiz questions you answer correctly.
  • Effective Reading Speed: WPM × comprehension.
  • Typical speed: the median WPM of your last 3 passing tests (at least 75% comprehension) on different passages, because one short test is noisy.
  • Not scored: speeds above 1,000 WPM, which are too fast to be reading.

Answer options are shuffled, and we keep the wrong answers about as long as the right one, so length doesn't give the answer away.

Where the Benchmarks Come From

We compare results with a meta-analysis of 190 studies and 18,573 participants, which found that adults read English non-fiction silently at about 238 WPM, with most between 175 and 300 WPM (Brysbaert, 2019). The averages we use for other languages come from the same paper.

Longer words take longer to read, so English passages use Brysbaert's formula: expected speed = 238 × 4.6 ÷ average word length. Our passages use slightly longer words than typical non-fiction, so this gives a fairer comparison. You can see all of the figures on our average reading speed page.

What the Test Can't Tell You

  • It is short. Short tests are noisy. Longer tests are more reliable: in the research, a standard test had a reliability of about 0.7, while two long texts reached 0.95 (Brysbaert, 2019).
  • Four questions is a rough measure. One question is worth 25 points of comprehension.
  • Topic matters. You will read faster on a subject you know well.
  • It isn't a diagnosis. The test can't identify dyslexia or any other reading difficulty. Please talk to a specialist if you are concerned.

Your Data

Your results are stored only in your browser. Text you paste into the custom test never leaves your device. A score is sent to our server only if you choose to add it to the leaderboard. Leaderboard entries are checked for timing and comprehension, and "House" entries are examples, labelled as such. See the privacy policy for details.

Frequently Asked Questions

How does the SpeedReadTest reading speed test work?

You read a short passage while a timer runs, press finish, then answer 4 questions about it. Your score combines your words per minute with the share of questions you got right.

How reliable is one test result?

A single short test gives a useful estimate, but it is noisy. Your result will change with the passage, your mood and how familiar the topic is. Taking several tests gives a better picture, which is why we show your typical speed as the median of your last three tests.

How is my score worked out?

Your Effective Reading Speed (ERS) is your words per minute multiplied by your comprehension. If you read at 300 WPM and get 3 of 4 questions right, your ERS is 300 × 0.75 = 225. It rewards reading with understanding rather than skimming.

Sources

  1. Brysbaert, M. (2019). How many words do we read per minute? A review and meta-analysis of reading rate. Journal of Memory and Language, 109, 104047. Link
  2. Rayner, K., Schotter, E. R., Masson, M. E. J., Potter, M. C., & Treiman, R. (2016). So much to read, so little time: How do we read, and can speed reading help? Psychological Science in the Public Interest, 17(1), 4–34. Link