In my last post I mentioned how while listening to Irene Fountas and Gay Su Pinnell speak yesterday during the focus group I was at I could have literally sat and quoted them all day. Well, there is one other thing that they addressed yesterday that really stuck with me that I feel is "worthy" of a blog post. That is the idea of lexile score correlating to benchmark score. From the mouth of Irene Fountas, "There is no correlation between lexile and benchmark." She went on to explain that lexile is a mathematical formula that takes sentence length and word frequency into account. That's it. That's all it does. The Old Man and the Sea, for instance, is lexiled at a 4th grade reading level because of Hemingway's writing style. For any of you out there who have read it, you know that this is just plain silly. It's so widely used because it's easy, but the real question is: Is it an effective measure to use when having students select books? The answer is clearly no.
Benchmark scores (Levels A-Z) are determined by human beings (teachers, literacy experts) who use the ten factors of text to determine the level. These factors are: genre, text structure, content, themes and ideas, language and literacy features, sentence complexity, vocabulary, words, illustrations, book and print features.
So my advice would be to forget lexile and teach your students how to select a just-right book for them based on what they know about themselves as readers. As far as guided reading goes, I would definitely recommend using the benchmark leveling system to match books to small groups of readers who you are working with at their highest instructional reading level.