Earlier this week, I came across a headline to raise the eyebrows of all of us lovers of reading: “American High School Students Are Reading Books At 5th-Grade-Appropriate Levels: Report.” Say what? As I skimmed the article, my bias meter went off–the report in question was authored by Renaissance Learning, Inc., creator of the Accelerated Reader (AR) reading assessment software. Wouldn’t Renaissance Learning have a vested interest in continued use of the AR software to improve reading abilities assessed as too low by … Renaissance Learning? Plus, I often resented AR for causing my most enthusiastic booktalks to be met with the response, “It sounds good, but I’m only allowed to read books at my AR level.” Arrow to a librarian’s heart, those words!
Still, I realized that–I can admit it–I have my own biases. I gravitate toward the cautionary views of Jim Trelease and Susan Straight; I like students to be able to take a book that sounds interesting to them without feeling like it’s worthless without a measurement or reward. But I’ve never tried to get to know or use AR, so I took it upon myself to do so, as well as to read their whole gosh-dang report. Phew! And with the caveat that I am not a teacher, reading specialist, or mathematician, here are some of my thoughts in the spirit of discussion:
- This is actually a seriously interesting report. Its data sample is huge: reading records from over 7.6 million students who read 241 million books during the 2010-2011 school year. Much of it is charts and tables, so it’s not as long as it looks, and the text includes essays answering the question, “What should kids be reading?” Contributors include authors (Dan Gutman and Ellen Hopkins, among others), teachers, reading experts (Dave Coleman, a Common Core State Standards Contributing author), and a librarian. These contributors are allowed to disagree and present all sides, thus helping to allay my fear of bias.
- ATOS means what again? I finally got the scoop on something that’s been shrouded in mystery to me: how the levels are assigned to AR books. The ATOS readability formula accounts for “the most important predictors of text complexityâ€”average sentence length, average word length, word difficulty level, and total number of words in a book or passage.”
- Now, about that headline: As the Huffington Post article gleans from the report, “A compilation of the top 40 books teens in grades 9-12 are reading in school shows that the average reading level of that list is 5.3–barely above the fifth grade.” This average includes books likely assigned (To Kill a Mockingbird) as well as chosen freely (The Hunger Games), yet the readability formula shows that these texts are hardly more complex than a 5th grader should be able to read. Obviously, text complexity does not equal idea/content complexity, and to account for this, all of the books on the list are from the “Upper Grade” interest level category.
- Worth mentioning: In my opinion, two additional pieces of data need to be pointed out. First, grades 1-8 each have their own breakdown of top titles, showing them closer to grade-level reading ability, whereas grades 9-12 are averaged as a whole, providing a broad look at the high school averages that is inconsistent with the lower and middle grade method of averaging. Second, using the grade-level breakdown, it is easy to see that sixth grade is the year when students’ reading level begins to plateau. Having read on grade level each year up through 5th grade, the average ATOS book level of the top 40 books read by sixth graders was 5.2 overall–and remains 5.2 in seventh and eighth grades, ending with the 5.3 average for grades 9-12. High school teachers can be faced with the impossible task of raising a student as much as 7 reading levels–and getting all the criticism for not turning out college-ready, career-ready students–when this data shows that perhaps middle school is where the problem begins.
- Putting AR to the test. One of the statements that jumped out at me was from Professor of Education Reform Sandra Stotsky: “…only the top 5% of our students will be able to read this country’s seminal founding documents in grade 11 or 12.” Surely, I thought, the founding documents aren’t that hard to read. I found the texts and plugged them into the Immediate ATOS Analyzer, where I was immediately proven wrong: the Constitution is a level 12.9; the Declaration of Independence, 13.8. The only book in the AR Bookfinder at level 13.8 is The History of Tom Jones, A Foundling, by Henry Fielding. 896 pages, written in 1749. Complexity: 1, teen appeal: 0. Curious if I could even this score, I grabbed the 2011 Top Ten Best Fiction for Young Adults titles and averaged their ATOS levels. The result was an even-lower-than-predicted 4.6.
I’d love to hear comments from anyone else with opinions on this report. Any teachers, parents, or teens using AR? Anyone go through grade school using a quiz or incentive program? Anyone just starting college wishing you’d read more difficult books to prepare you? Anyone living proof that you can read what you want and succeed in your chosen career? As a reader and a librarian, I am saddened by data from multiple sources, in addition to Renaissance Learning, that points to reading scores falling on a national level. I retain much of my skepticism toward oversimplifying reading as a mechanical formula, especially by a for-profit company. But having learned more about AR, I can appreciate it as a tool that helps parents, teachers, and students begin the daunting task of facing an entire library of options and knowing that choices should become progressively more complex to continue to grow their reading ability. As another ATOS report noted, “Matching students and books is always a two-step process in which readability measurement is a starting point.” I hope to meet many readers on the stepping stones beyond!
— Becky O’Neil, currently reading Lola and the Boy Next Door by Stephanie Perkins