Every time you rewordify something, we display a READ score (the Rewordify.com Estimated Average Difficulty score) of the original and rewordified versions of the text. This helps writers and educators who need a valid complexity measure of a text passage.
How the READ score is calculated
- Our site obtains the text block's average word frequency from within Brigham Young University's 450-million-word Corpus of Contemporary American English. That corpus comprises 190,000 texts from fiction, magazines, newspapers, spoken interviews, and academic journals.
- It looks at each word in the text block you entered (or the document you're reading), looks up how frequently that word exists in the corpus, and assigns the logarithm of that frequency. It does the same for each word, and then calculates the average log frequency.
This is a highly valid way to determine text difficulty: the READ score determines, on average, how likely it is that a reader will have been exposed to the words in the passage based on a huge collection of real-world writing. (The words needed and sister are scored as easy because they're very common. The words consume and vicious are scored as harder because they're not as common. And, the words abet and retorts are scored as difficult because they're rarely seen.) Notice that these words are all two-syllable words; measures like the Flesch-Kincaid, that only look at syllable length, would not be sensitive to the significant difference in difficulty between these words.
- The site makes an adjustment based on sentence length. The READ level is adjusted down for shorter average sentence length, and adjusted up for longer average sentence length.
- Finally, the site makes this adjustment: if a hard word is in a passage, it only counts it twice. This prevents some articles from getting an artificially high READ score. For example, a simply-written article about presbyopia might get too high a READ score because the word presbyopia is mentioned ten or fifteen times. This adjustment prevents that from happening.
Using the READ score
If you're familiar with the Lexile measure, you're familiar with our READ score, because the scales are the same: from 100 (extremely easy) to 2000 (impossibly difficult). It's very rare to encounter real-world writing that scores at either end of that range.
In a random sample of 25 varied writing passages, our READ score correlated with the Lexile measure with a Pearson correlation coefficient of .87, which is an extraordinarily strong correlation. Why are the scores so strongly correlated? Because we use the same procedure as the Lexile measure (described above). The difference between the READ score and the Lexile measure is likely due to their using a different corpus than us.
Don't put too much emphasis on any computerized text difficulty measure. Here are some guidelines when using any text difficulty measure, including our READ score:
Statements that reflect sound educational practice:
- My students in the same grade have a wide range of reading skill
- In general, many texts with higher difficulty scores are more difficult than texts with lower difficulty scores, but there are exceptions
- I will not systematically restrict my students from reading texts that appear to be too easy or too difficult based on a computerized reading difficulty score
- My students’ interest and my intuition carry greater weight in reading decisions than a computerized reading difficulty score
Statements that reflect unsound educational practice:
- I will give my seventh graders only seventh-grade-level books.
- I teach ninth grade. I need Rewordify.com to simplify Frankenstein to a ninth grade level. (Read this.)
- My student wants to read passages that have a READ score of 1300. I won't let her; they're too difficult.
- Passage A has a READ score of 800. Passage B has a READ score of 1000. Therefore, Passage B is more difficult.
- I want to match a READ score to a grade level. (Read this page on the Lexile site. Both they, and we, discourage educators from limiting readers to what they're "supposed" to be reading based on their grade.)
Your feedback is important to us. Please tell us what you think about our READ score. We really do want to hear from you.