Book leveling and the Flesch-Kincaid grade-level equivalent

Posted: March 11, 2011 in link, post
Tags: , , , , ,

The Process of Readability Formulas

The Process of Readability Formulas (Photo credit: Brian Scott Designs)

I am in the process of leveling some of the “summer readers” for the summer reading program at the school where I am working as a volunteer literacy tutor. The primary resource I use to level books is the “AR” score, a grade-level equivalent measure from Renaissance Learning. If you are familiar with “AR Quizzes” that is the tool I am using. AR is great because it has most of the books that we have purchased for our program, but once in a while, I find myself stumped, so my next resource (because most of the books we buy come from Scholastic) is the “Book Wizard” from The reason I tend toward AR first instead of Scholastic is because Renaissance is a “what works” in education program and the scores are quite reliable. Look up the same book between Scholastic and AR and you are likely to find a discrepancy; though usually not a great discrepancy, consistency is best!

When I really get into trouble is when I cannot find a score from either AR or Scholastic…. Sometimes a Scholastic book will have a level written on the back. If AR has no level, and I can’t find one from Scholastic’s “wizard,” I will just work with what is printed on the back; however, sometimes (usually in the case of an older or brand new book) I can’t find any level, but I have located some new–to me–online resources that I have found helpful. Rather than suggest any specific site for this purpose, let me instead suggest the following search string: “grade level equivalent analyzer.” This search string will yield results whereby one can enter a bit of text…I would suggest two or more representative paragraphs…and the analyzer will run the calculations to determine an approximate readability level. The caveat here is that not all analyzers are created equally; hence, why I did not wish to recommend a specific source at this time. Even those that take the same factors into consideration may have different formulas that will yield disparate results.

Again, this is new to me, and perhaps some reader will have an idea of a specific and reliable site. If there is one you have used, please mention it as I would love to try several sites.

UPDATE: 03.18.11–In addition to finding some of the online analyzers, here are some other sites I have found that will teach you about leveling books and how to level on your own. It is more time-consuming, especially at first, but you will have the satisfaction of knowing that you crunched the numbers and can feel confident about the grade-level equivalent assigned. For those with tech-savvy, you could probably create your own analyzer using formulas in a spreadsheet program. Here are some great links:

First, here is a great white paper describing ATOS and other leveling systems. It is only 8 pages. Report from the School Renaissance Institute. is another helpful site that provides the Flesch-Kincaid Grade-Level Readability formula.

Kathy Shrock’s Guide for Educators provides detailed information regarding Fry’s Readability Graph, including directions and a sample of a Fry graph.

Grace Fleming, an guide, provides the Flesch Readability formula as well as additional links here. provides a wealth of info regarding readability formulas and even includes a unique search-engine that allows you to determine the readability of a website by entering the URL!

A site called StreetDirectory provides a “down and dirty” overview of the topic along with the Flesch formula.

A site I can’t quite identify by name, provides a detailed overview of Flesch-Kincaid and other Readability formulas here. The site lists some books on the subject if you just can’t get enough!

All About Readability by Cheryl Stephens provides an historical overview of readability formulas and discusses the uses and pitfalls of readability tests. It is written from a writer’s perspective but is useful for understanding the history and uses of readability formulas.

If you are interested in “Plain Language” training for writing, visit for an overview and resources, including the free “Plain Train” course.

I hope you will find the above resources useful. I am sure there are many, many more; however, these seemed to be a few of the best I could locate. Please leave a comment regarding your experience with any of these sites if you use them.

UPDATE, December 8, 2011–

As I was leveling some new books for the school “leveled library” (a resource for teachers to use in their classrooms) I discovered that ATOS, the readability standard developed by Renaissance Learning, now allows users to enter their own text! This is fantastic because prior to this point if one used the ATOS system, the book had to be available either through the AR Bookfinder or the Renaissance Learning Quiz Store. If the book you wanted to level was not available in either of those two places, then another system would need to be used. As noted, there are various reliable methods around; however, being able to enter text directly and analyzing it using the ATOS system will, in my case anyway, lead to greater uniformity.

Briefly, to conclude this update here are the ways in which the ATOS Readability measure can be used for books not in the database:

1. You may submit the complete text of a book for instant analysis. This is probably best for very short books for early readers.

2. You may submit text with an estimated word count. The directions for estimating the word count are available on the site. Just select three different portions of text from the book (each about 150 words or so of representative text) and then upload your text along with your estimated word count for the book.

3. ATOS for Text “…works best for short stories, magazine and newspaper articles, test items, and other classroom materials.” Accessed 12-8-2011.

I’m excited about this great new feature from Renaissance, and I hope to make use of it quite frequently.

  1. From Big Sister: when I see you involved in something like this–a “practical” activity that presupposes a high level of education, abstract thinking ability, familiarity with professional resources, etc.–I have to wonder: do you still think you would have been better off going to a Vo-Tech school rather than college? You can guess what I think!

  2. Ed Griffin says:

    Good blog. Interesting. Thanks
    Ed Griffin

  3. AR has a web page that lets a user insert about 20K of text for evaluation. I tried Mary Shelley’s Frankenstein but took multiple samples from different areas in the work. Only 2 of 6 sample gave the same result 9.8, 10.1, 10.1, 10.6, 11.0 and 11.5 giving an average of 10.5.

    But Frankenstein was written in 1818. The word ‘scientist’ had not been coined. I wrote my own program to evaluate science fiction on the basis of science words used. Frankenstein got 0.169.

    H. G. Wells’ War of the Worlds got 0.411. The relatively modern book A Fall of Moondust by Arthur C. Clarke got 0.994. A score of 1.00 would mean one science word for every 1000 characters. The ATOS scores for Moondust ranged from 8.0 to 9.4.

    Frankenstein scored higher on ATOS because of longer sentences. Possibly a different style of writing 200 years ago. Would a high science word density actually be more useful for students?

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s