To really learn a new skill, you need to practice. That theory drives much of Katie Farrell's reading instruction.
In her 1st grade class at Bauer Elementary School in Hudsonville, Mich., Farrell teaches students phonics-how letters on the page represent the spoken sounds children hear.
But for some kids, the learning only really clicks once they practice these patterns in decodable books. These short texts are written with a high proportion of words that are phonetically regular-meaning they follow common sound-spelling rules-and mostly include words with phonics patterns that children have already learned.
"When you can make that match ... that's where the power lies," she said.
Research has long shown that teaching early elementary students phonics is the most reliable way to make sure that they learn how to read words. And much of the current debate around reading instruction has focused on phonics teaching, as many schools don't currently follow research-based best practice in this area.
But text plays a big role in the reading classroom, too. Decodable text, specifically, is a "crucial learning tool," said Wiley Blevins, who has written several books on phonics and currently works as a consultant training teachers.
Even so, teachers are divided when it comes to decodable books.
In Education Week's recent national survey of early reading teachers, only 23 percent said that beginning readers should be using these texts most often. The majority, 61 percent, said that students should be reading books with high-frequency words, predictable sentence structures, and pictures that emphasize meaning. Often called leveled books, these texts are rated on a difficulty scale. Teachers aim to match students with books at their level.
There's also a common criticism that decodable books, because of their inherent language constraints, are boring and stilted. Why subject students to these contrived stories, the argument goes, when they could be reading something more engaging?
But many experts agree that kids need that targeted practice. "When you are teaching phonics, the way to get that learning to stick is to apply it in connected text," said Blevins.
"It builds the right strategies," said Farrell. "They're not reading books that they're not ready for, and using the pictures to guess."
Still, decodables aren't the only books that young students should read. Most experts suggest a varied text diet. And, decodables are ultimately a stepping stone.
Eventually, Farrell says, "I want them in that authentic text using the strategies that they practiced when they're using the decodable books."
Researchers agree that decodable text is meant to be used during a short window, when students are first learning to sound out words.
Studies have shown some benefits for early readers. When kids read decodable books, they're more likely to try to decode-to sound out the words. Some studies have found that they're also more likely to read words accurately.
But other research suggests that it may not matter what kind of text students read, as long as they're getting strong phonics instruction. In one 2004 study, two groups of struggling readers in 1st grade received one-on-one phonics tutoring. One group read books that were mostly decodable; the other read books that were mostly not decodable.
There wasn't any significant difference in the word reading or comprehension of the two groups at the end of the study.
Still, there's more research on decodable text than on other types of early reading materials, like leveled readers, said Heidi Anne E. Mesmer, a professor of reading at Virginia Tech.
She suggests that decodable books be used like "a set of training wheels on a bicycle."
"If you think about the amount of time that children learning to ride a bike use training wheels, it's not long," she wrote in an email to Education Week. "Also, not all children need training wheels."
These "training wheels" help students practice their phonics skills in a controlled environment. But just as importantly, they teach students to try to sound out words, Blevins said.
He pointed to a 1985 study by researchers Connie Juel and Diane Roper-Schneider, which found that the texts students were exposed to early on could affect how they tackled words.
In the study, students who read decodable text tried to sound out words more often than students who read text that prompted students to use other cues.
When students are mainly reading leveled text with predictable sentence structures, "they're undervaluing and underusing their phonics skills," Blevins said. "This creates a really bad habit. Every book they pick up, their first strategy is, try to look at patterns, look at pictures, memorize." Decodable books encourage the right strategy of sounding out the words, he said.
In Claudia Margaroli's 1st grade class, decodable books help remind students that they should be focused on sounding out the words.
"This year, I've been trying to be more specific with teaching sounds in a sequential order," said Margaroli, who teaches at Charlotte East Language Academy in Charlotte, N.C. She teaches sound-letter correspondences explicitly in her phonics lessons, and then students practice in decodable books.
"They know-and I make them say it and verbalize it-that these are sounds they've been working on, these are words they can read," Margaroli said.
Decodable books should follow the progression of a phonics program, focusing on new sound-spelling patterns and "folding in review and repetition," said Blevins.
But some teachers balk at the idea of using these books, even for practice of key skills, said Blevins, who does training with schools. Why? He remembers one group of teachers who were especially blunt about decodables: "They're boring and stupid," they told him.
Margaroli says it's true that some decodable books "just don't have a storyline." She looks for decodables "that you can actually use for comprehension," she says, "rather than a weird story about a cat and a mat, where at the end nothing happens except that cat is on the same mat."
How did we get "weird" stories about cats and mats, with thin plots and stilted language? Researchers trace the trend back to the late 1990s and early 2000s, when Texas and California both required decodable texts in their reading program adoptions. The states set decodability thresholds for texts: In Texas, 80 percent of the text had to be sound letter correspondences that students had already learned; in California, the number was 75 percent.
In response, publishers got competitive, each trying to make the book that was the highest percent decodable, Blevins said. Irregular words, like "the," often disappeared, even though they're highly common in the English language.
But there isn't evidence to suggest that a 90 percent decodable book is more effective than one that's 75 percent decodable, or 60 percent, said Timothy Shanahan, professor emeritus at the University of Illinois Chicago. There's no "magic level," he said.
In the rush to fill texts with only decodable words, the number of unique words per hundred in these books also increased during this time, said Elfrieda H. Hiebert, a reading researcher and the president and CEO of TextProject. So instead of seeing the same word multiple times throughout a story, students would see different words that all had the same spelling patterns.
To clear the high decodability bar, publishers started using sentences that English speakers wouldn't say or write under normal circumstances, said Blevins-like, "Let Lin dab a lip."
"The problem is, these stories made no sense," he said. "These books aren't Shakespeare, but they should be good stories that children enjoy reading."
There's also value in repeating some of the same words throughout the story, said Hiebert. Decoding the same word several times helps kids link the sound to the spelling in their minds, Hiebert said, and can lead to more fluent reading. "There has to be a really strong component of consistent data that kids are getting," she said.
Hiebert looks for a few criteria when she's evaluating decodable books.
She wants to know if they're exposing students to "highly consistent and prolific patterns" in the text, getting practice with letter-sound correspondences that they can apply to other texts.
She also wants to know if the texts make sense as stories, and are building student knowledge. What are they teaching students about the world? A lot of decodables still fall short in this category, she said.
But when a decodable book has a story, it doesn't have to be relegated just to sounding out practice, disconnected from the rest of the lesson, said Blevins. He suggests that teachers have rich conversations about the stories with students, asking comprehension questions to demonstrate that reading is about meaning. Students can also write about the books.
In Margaroli's class, students do just that, writing responses to questions about the text. Still, reading and writing about decodable text is only one part of Margaroli's literacy block.
Her students also listen to read-alouds, have conversations, and read books from their class library.
There are no research-based rules on how much time beginning readers should spend with decodable text, said Shanahan. It would be "very reasonable," though, to spend some portion of phonics instruction on practice, he said. This includes decoding individual words, spelling words, and reading decodable books.
Shanahan, Blevins, and Mesmer all said that decodable books aren't the only kind of text that students should have access to in these early elementary years. And though Margaroli's students practice in decodables, they have other time in the day to read books of their choice from the class library.
This kind of diverse reading diet is important for students because it exposes them to a broader representation of the English language, said Shanahan. Decodable books are usually constrained to phonetically regular words. Letting kids read books without those constraints can give students some experience encountering words that don't follow normal patterns, and help them "figure out the statistical properties of the language," he said.
How can teachers know when students are ready to take the training wheels off, and stop practicing on decodables altogether?
Farrell, the 1st grade teacher in Michigan, watches how students are segmenting and blending words as they read.
Once they can consistently apply the skills they've learned in their phonics lessons, "that's my first clue that I think we're ready to move on," Farrell said. It shows her that, with her guidance, students could apply the same strategies when they read more authentic text, she said.
By the spring of 1st grade, "almost no one in the class is using decodable books," said Farrell. "I love them, and then we get to a point where we just don't need them anymore."