Readability metrics and reading skills

Several readability checkers using a variety of metrics are available to help you assess the readability of your text. Some are more sophisticated than others. Many checkers offer a grade score to suggest what age or level of education is needed to read the text. Others give you deeper analysis of the language, identifying the interplay of words and sentence structures that make the text difficult for readers to process.

Flesch–Kincaid reading difficulty grade scores

The most widely used readability checkers are based on the Flesch–Kincaid system. This system uses a formula that combines the average word length with the average sentence length to show reading difficulty. It produces readability scores that are aligned with American school grades 3–12. The scale was later expanded to include grades 13–16 for college students and a 17th grade to identify specialised technical texts.

The Flesch–Kincaid system is inaccurate at upper grade levels because of the great variety among readers with higher levels of education, reflecting their individual interests and aptitudes. It also does not take account of different language or cultural backgrounds.

Flesch reading ease scores

Readability scores are sometimes shown as scores on a scale of 0–100, with 0 hardest to read and 100 easiest. This was Flesch’s original approach to readability, based on word and sentence lengths. The scale of reading ease is inverse to the Flesch–Kincaid grade scores, so that lower scores represent the difficult end of the scale. Texts scoring:

• under 50 are ‘difficult’ or ‘very difficult’

• between 60 and 70 are ‘standard’

• above 80 are ‘easy’ or ‘very easy’.

The reading ease index offers a wide scale for discriminating between similar types of reading material. For example, research(Opens in a new tab/window) on the readability of 15 topics in 2.5 million articles from major online British and American newspapers found that sports reporting was the easiest to read, and politics the most difficult.

SMOG

The SMOG – simple measure of gobbledygook – also assesses text using school grade levels, but uses 2 measures of word difficulty. The metric includes both the count of words of 3 or more syllables and measures of word frequency. Compared with Flesch–Kincaid, this combination more accurately grades material at higher levels of reading difficulty and assesses the level of technical language in the text. This makes SMOG more suitable for assessing texts used in professional and technical education.

Coh-Metrix

The Coh-Metrix™ readability checker is among the few that analyse the variations in grammar that make a text easier or harder for readers. Coh-Metrix reviews the structure of successive sentences to show how readable they would be for both first- and second-language users of English. It also analyses other factors that support readers’ comprehension of text, including the relative frequency of words (how familiar they are likely to be to adult readers), and the connectivity between words across the text. 

School grades

Readability benchmarks are typically expressed in terms of American school grades. These gradings seem to line up with Australian school grades 3–12.

However, these benchmarks should be treated with caution, because referencing school grades oversimplifies the range of reading skills at any level. Students’ reading skills cover a very wide range at any level, as shown in the following figure.

The Australian NAPLAN scheme sets out bands for the reference years in primary and junior high school, showing the range of student reading skills in each year. The 2 middle bands of readers in years 5 and 7 are roughly equal to American scores for school grades 5 and 7. But the American grade 9 lines up roughly with the upper bands of the Australian year 9, suggesting that it is slightly higher than the Australian school grade.