Earlier this week, we wrote about our expanding understanding of the ways Lexile measures misjudge books that don’t fit narrow parameters, and we promised to look more closely at a couple of texts that are “Lexile confounders.”
This delightful and whimsical story is quite complex. When we read it, we fell in love with the clever characters and the opportunities for inference. However, when we checked the Lexile level, we discovered that it was 250L!
This made no sense to us, we knew it was too difficult to be a beginning reading text! Some of the sentences are quite complex; the one-word exchanges between the characters (set as dialogue) are also complex, proving that less is sometimes more.
We typed the text into Microsoft Word to check their readability statistics, only to find similar results. Here’s a sample page from the text:
“Wait. Finding gold takes planning,” said Crocodile.
“It does?” said Anna.
“No one else can know what we are doing, or they will find it first,” said Crocodile. “Look at my face. Can you tell what I am thinking?”
“That’s because I am making a secret face,” said Crocodile. “Try it.”
“Can you tell what I am thinking?” Anna asked.
“You are thinking about gold.”
“Probably still gold,” said Crocodile.
“Oh, but you know me very well,” said Anna. “It’ll do. Let’s go.”
What we figured out is that, formulas for calculating text difficulty rely heavily on the length of the sentences. This text is mostly made up of very long sentences and very short sentences, both of which introduce complexity. The average of those sentences–which is four words per sentence, in this case–informs the algorithm. So, for How to Find Gold, the average sentence is four words, explaining its low lexile score. What that score doesn’t account for, however, are qualitative measures that make it much more difficult than a beginning reading text.