Gifted Issues Discussion homepage Forums Recommended Resources "Reading levels" -- how meaningful are they?

DD6 took a big leap forward in reading, and out of curiosity I looked up the level of some of the books. According to this site, they range from late 2nd grade through 5th grade. How meaningful is this?

Can a few difficult words bump a book up to the higher grade levels? (We're using advanced picture books rather than Easy Readers, hence not a carefully controlled vocabulary climate.) Also, the "interest level" ratings on this site are really whack, which makes me doubt the whole thing.

I heard somewhere that the main thing used to level books at the lower levels is the spacing of the page. That is, how they look.

I have never found reading levels to be very meaningful for our kids (who were early and self-taught readers). If I had a child with a reading disability, I bet I'd be watching them like a hawk, though, for the off chance they were revealing...

I'm a bit dubious about those rating tools and tend to see them as solutions in search of problems. For example, Accelerated Reader (AR) gave Dune a reading level of grade 5.7. The Lexile level for this book is also fifth grade. The interest level AR gives this book (grades 9-12) is better, but anyone who's ever read Dune would know that it's way above a fifth grade level. Interestingly, the Lexile rating talks about "power words" in each book that "each student should know." One of the power words in Dune was fluffy. Adult and bald were two others. I can't help but wonder if Lexile was analyzing the same book I was reading:


Fifth grade? Seriously, I would hope that fluffy and bald haven't become fifth-grade power-words, and as for compressed suspensor field? I don't even know what that means.

Tools like AR and Lexile give scores that reflect "text difficulty:"


Frank Herbert used advanced vocabulary in Dune and he made up umpteen words. However, he re-used them, which may explain why the book got its fifth-grade rating.


The AR and Lexile scores can be useful if you're trying to get a basic idea about a book's grade level score, but honestly, I'm not sure what kind of value that number really offers. The scores for Henry and Mudge and the original Nancy Drew books seem about right, but it isn't like the audiences for these books were big secrets until Lexile came along.

I tried to use AR and Lexile to find books for my kids when they were younger, but I gave up because I didn't find them to be terribly helpful. However, I'm cynical about this kind of thing and see it as an industrial approach (calculating "grade level" using a standardized algorithm) to a challenge that's very much about an individual ("What kind of books will my kid like to read?"). YMMV.

I think it's a good question- my inexpert thoughts- reading levels may have meaning if kids are struggling with reading.

I have found that people (kids and people in general) who enjoy reading and don't struggle with it, tend to read at many levels with variety. What advanced readers don't seem to enjoy is formulaic plot- unless, like me, they are a sucker for a happy ending, and occasionally read such a book for escape.

My DS7 is supposedly way above grade level in reading- but I know it's only certain kinds of reading. The kind that involves adventure/mystery/fantasy and is primarily plot driven. He is probably only a year or two above level when it comes to realistic fiction- especially understanding a character. I think that character driven books tend to be more difficult for most people because it requires advanced understanding- and for kids, more experience than they may yet possess.

Just thoughts:)

I played around with Lexile levels several years ago. What I found was that sentence length is huge, and what is considered a sentence is what is surrounded by periods. So, for example, I typed up a chapter of a Magic Tree House book. MTH is notorious for having sentence fragments everywhere. I used the Lexile analyzer to find the Lexile level of the chapter as written and then I went through and normalized the text, so that the fragments were incorporated into complete sentences. The only words I added were things like "and." Anyway, the grade level of the original text with all the fragments was something like 2.0 and the grade level of the properly written text was something like 4.0--which was what I was expecting.

The Reading Counts levels are pretty good. They show Lexile level, which is sort of a brute force method for calculating reading level, as well as interest level and their own grade level, which is based on Lexile level and other intangibles.

I found all of this helpful when I was trying to track my HGish dyslexic son's reading progress. I found it less helpful with my younger son, who is HG+ and doesn't have dyslexia.

I think some of the reading level systems like Fountas and Pinnel or Guided Reading take into account the complexity of the text in terms of theme, concepts, vocabulary etc. whereas the lexile system seems more interested in measuring decoding or how difficult a text would be to read fluently. So there may be a certain book that is very simple to read but hard to understand, and those would be given a higher reading level than what it would appear on the surface, or by the lexile value.

Such an interesting thread -- so many good points. Val, YES to everything you said! Kai, I love your MTH experiment. Very revealing. And Blackcat -- "simple to read but difficult to understand." How about the emotional depth in a book like The Miraculous Journey of Edward Tulane"? The language isn't difficult but the themes are mature in the very best sense.

Kai, I'm gonna pm you when tomorrow. (Too late here for me to be coherent.) I would really love to pick your brain.

The experimentalist in me would love to fool around with this. What happens if you take, say, a relatively short "5th grade" book and remove the single most difficult word? The two most difficult words? Etc. I bet these ratings don't exhibit "graceful degradation."

Karma, by Ostlere, has a Lexile of 400. It's a wonderfully complex book, full of imagery and allusions. It's also written in diary format with free verse and sentence fragments. The small number of words on each page evidently give it the low Lexile number.

In short, no reading level scheme is perfect. The ones that appear best for my use, are those that use a variety of measures, including vocabulary, complexity of sentence structure, and aspects of comprehension. When my kids were younger, Fountas and Pinnel and DRA levels were most useful. Once they each hit ~4th or 5th grade level, it doesn't seem to matter as much.

When it comes to assessing a reading level for school, reading levels sometimes appear random, as they can be so sensitive to subjective judgement from the teacher and from one text to the next. We've found that progress assessed by one teacher through a year is reasonable feedback. Correlating scores across years is problematic, particularly when disabilities are considered (e.g., DS has a speech disability that affects his oral reading accuracy and rate.)

That hasn't been true for us, because 2E DS had strange patterns of strengths and weaknesses that made the test non-functional for him.

Tests that rely on retelling are terrible for him, because he has sequencing issues. We've had a history of enormous underestimations of reading level by schoolteachers (even those of good will) because the tests chosen by the district actually measured retelling skill instead of reading skill.

We work on retelling, and work to get the school to see what he actually understands in other ways, and by using different assessments.

Geofizz, I use the scholastic site as well to keep track of dd4.9's reading progress. I record the guided reading level, dra, Lexile measure and grade level equivalent data. I have found that many times the 4 don't coincide at all. However, they give me a general idea about where dd is. She started reading only this spring and is reading mostly at end of 2nd grade level no matter which scale I use. However, I recently had her assessed by a reading institute as I wanted to make sure she was okay for the teacher-recommended grade skip. The reading instructor assessed dd at grade level 1.8 to 2.8. She basically used a couple of paragraphs pulled from different books and asked dd to read. She also had dd read aloud 2 books at home and I had to fill out a survey questionnaire on how easily dd could decode unfamiliar words, reading with expression and if dd enjoyed reading the books. Even though it was a brief assessment, what I got was that there is a difference between how a first grader is supposed to attack new words vs. a second grader. The child should be guessing less and using various strategies to really read the word.

Eh.

What Val said.

I gave up using this sort of tool when Lexile recommendations for my 7yo included Tess of the D'Urbervilles and Madame Bovary.

Sure, Oliver Twist might have been fine for my particular 7yo (and was, in fact, fine for HER), but I had to draw the line at Flaubert.

It was then that I realized that such tools are probably entirely UN-helpful for parents or teachers of children who are fluent or advanced readers.

Which begs the question; why are such children forced to participate in evaluations intended to provide this kind of feedback, hmm? I digress.

We never paid any attention to levels here - partly because they never seemed relevant, partly because they were not emphasized at our children's school.

I just had to jump in though and say… um…. Dune for 5th grade? Yikes!


I wonder sometimes if there is a bit of a geographic/school-district dependence relation that impacts reading levels - Magic Tree House books are definitely 2nd grade level here and many kids are reading them by the end of first (not talking gifted classes either). By third grade very few students in the classes I've helped out in were interested in them anymore -

Best wishes,

polarbear

I'll agree with that idea, except for their Research Guide series. My eldest loved those books when he was 9-10 or so.

While the levels for any individual book may not (and most likely will not) be accurate on at least one or more of the scales used, I have found them (all scales taken together over 150+ books read) to be meaningful to track progress in the case of my dd. I find that the grade level equivalent and the guided reading level are correlated with the book's length, the sentence length, and the number of non-repetitive words. The lexile measure is more closely correlated with number of difficult-to-decode words in a text.

DD is a late reader compared to children in this forum. So maybe the scales are only valid for average to slightly above average readers. My guess is that once dd gains reading stamina, she may be able to read much higher levels as her decoding is pretty good and her comprehension is way up there. Anyway, at that point, I may stop using the various reading scales as well. For now, I do find them very useful.

I've never seen the Research Guides, I was thinking of the basic books

pbear

DD is a late reader compared to children in this forum. So maybe the scales are only valid for average to slightly above average readers.



Yes, this is what I think, too, LovemyDD.

Oh, they're great.

Mummies and Pyramids!

Pirates!

Space!

The list goes on....

I have mentioned this before....it isn't the level that is the most important thing. My ds reads books from 3rd grade level to 6th grade level for battle of the books...no problem. He doesn't like them all but most of the twelve he will enjoy. The thing that makes the easier ones tolerable is that he can go at his own pace. He read Waiting for the Magic last year (by the Sarah plain and tall author I think) and it was a sweet, sweet little book. He read it in less than two hours. Same book if it had been an assignment in class and he would have had to slowly read a chapter a day and painfully discuss vocabulary and comprehension questions...would have been torture and he would have hated the book.

What is assessed reading instructional level? High school . I don't think he needs to be taught high school books at 9 but I don't think you have to watch pacing on books that are easy and give him the highest level novels you have available (that he hasn't already read). He takes work.

I've noticed two things on reading levels:
1) in the elementary reading levels, the 'level' of the book is almost always at least one grade harder than the 'interest' level - and often up to three grades higher. This doesn't seem to be dependent on the scale used.
2) teachers and reading scales seem to assume that remembering the sentence you've read all the way until you have decoded the end of it, so you can make sense of it, is a major determinant of reading level. This is why using lots of sentence fragments (in MTH or elsewhere) lowers the reading level dramatically: the meaning is somewhat pre-digested. I have to believe this is true for many kids or it wouldn't be such a universal belief.

My younger one can decode an entire paragraph syllable by syllable and then just remember it all to make sense of it. For her, the ideal rating scale would include a LOT of vocabulary, some phonics/decodability measures, and clause complexity, but not sentence length.