WebAIM - Web Accessibility In Mind

E-mail List Archives

Re: Help With Plain Language And Learning Disabilities

for

From: Tyllick,Cliff S (HHSC/DADS)
Date: Nov 1, 2016 1:17PM

Hi, Jim!

I think we've already covered what you're asking for, but here's a quick overview of why it's unwise to rely strongly on readability scores, especially when it comes to evaluating accessibility.

Grade-level calculations consider only two factors:
- Words per sentence
- Syllables per word

The calculations don't recognize techniques that improve readability without adding more periods (full stops). For example, it's easier to find one item in a bulleted list like this one:
- grapes,
- apples,
- cherries,
- mangoes, or
- pineapples.

Than it is to find the same item in a list that is not broken out into bulleted items: grapes, apples, cherries, mangoes, and pineapples.

But both formats get the same readability score.

The calculations also don't factor in the adverse impact of having too many short sentences in a row. If you present a concept in two or more short sentences, the reader has to figure out how those statements work together. Putting the same information in one longer sentence might save them that effort.

Here's a really dumb example. If we present our list of fruits in a series of short sentences, the Flesch-Kincaid reading level is grade 0.2:

Choose a fruit. You may choose grapes. You may choose apples. You may choose cherries. You may choose mangoes. You may choose pineapples.

But the reading level is grade 20.2 if we get rid of the unnecessary words and rewrite it as a bulleted list:

Choose a fruit:
- grapes,
- apples,
- cherries,
- mangoes, or
- pineapples.

Which version is easier to use? Which is more accessible to all readers?

Finally, the calculations don't tell us when any sentence is too long, no matter the reading level. Consider these:
- "Go no more than sixty-five miles an hour." (Grade level 2.2)
- "Speed limit 65" (That is not a sentence, so you can't do the calculation. In fact, our list of fruits would be better without the "or" and the end punctuation, but then we technically don't have a sentence, so the computer won't do the calculation.)

Which is the better way to get that information to people driving down the highway?

So what is the best approach? First, I strongly encourage you to have your writers read "The Audience You Didn't Know You Had" <http://contentsmagazine.com/articles/the-audience-you-didnt-know-you-had/index.html>;. The basic principles that Angela Colter presents will help anyone improve their writing. If your writers have less than an hour available to learn more, it's a fantastic resource.

Second, in her book "Letting Go of the Words: Writing Web Content that Works," Ginny Redish uses clear examples to teach the fundamentals. For example, consider this advice about the best number of sentences in a paragraph on a Web page:
- Three sentences is good.
- Two is better.
- One is best.
- Unless a sentence fragment will do the job

If your authors have time to read more than a short article, I suggest that they get a copy of Ginny's book. They should skim it, read one or two chapters, and then keep it available as a quick reference whenever they have to polish up a piece of writing.

As with so many aspects of accessibility, assessing success requires more than a measurement. Indeed, when it comes to assessing readability, too often the measurement can lead us astray.

Cheers!

Cliff

Cliff Tyllick
EIR Accessibility Coordinator
Texas Health & Human Services Commission
512-438-2494
<EMAIL REMOVED>