How much should you challenge your students in reading? Should you give them texts they can read relatively easily or should you give them texts that are beyond their reach?
The Common Core State Standards (COSS) says step up text complexity to increase student learning. But, a recent review of the research in The Reading Teacher by a group of researchers from the University of Tennessee in Knoxville disputes this approach. Their review of the available research on text complexity shows that texts students can read with at least 95% accuracy produce greater learning gains than harder texts.
“We contend that in order for students to become proficient readers who are engaged in text while self-regulating and building vocabulary knowledge, the text must appropriately match the student’s reading level,” the researchers write. “We fear that the push from the CCSS to promote the use of more complex texts will result in decreased reading engagement and less time spent reading, with a potential decline in reading achievement the ultimate outcome.”
The researchers advise teachers to be cautious about increasing text complexity until there is clearer guidance from the research. They recommend sticking with grade-level texts in the meantime.
“We recommend that elementary-grade teachers continue to adhere to the traditional oral reading accuracy criteria of instructional texts that can be read at 95% accuracy or higher until the outcomes of research on both issues is available.”
In a one year-long study of 1st graders and another study of 2nd and 5th-grade classrooms, researchers concluded that the average number of oral reading errors was negatively correlated with reading growth. The latter study also reported a negative correlation between high error rates and engagement.
But, researchers say keeping texts readable was more important for the 2nd-graders than the 5th-graders so more accomplished readers may be able to better handle complex text.
In another study of struggling grade 3-5 students who read at below the 2nd-grade level (fewer than 80 words per minute), students were randomly assigned to 18 weeks of daily tutoring with grade-level texts or reading-level matched texts (There was also a control group). Text complexity capabilities of students ranged from 3.5 grade level to 7.0 grade level—roughly 1 ½-3 ½ years above the participants’ reading level.
Reading-level texts were selected based on students’ oral reading according to criteria developed by Emmet Betts in the 1940s. The researchers note that Betts’ criteria of text complexity were based on oral reading accuracy. “Precisely how he set the levels is a bit of a mystery and cannot be said to have involved extensive research, “ the authors write.
Tutoring sessions for both the grade-level and the reading-level groups were organized so that students spent 5 minutes on phonological or word-work activities, 20 minutes focused on text reading and 5 minutes on comprehension support.
Students in both groups did better than students in the control group. However, reading-level tutoring students did significantly better than grade-level tutoring students on word identification, word attack and fluency. Researchers note that keeping to grade-level texts seem to be even more important for struggling readers.
In yet another study of self-regulating behaviors during reading, such as self-corrections and use of decoding strategies, the authors report that readers display these behaviors when they can read orally with at least 95% accuracy.
The only studies where students making gains in achievement with frustration-level texts, according to the researchers, were studies in which students were well-supported. In one study, the teacher read the text aloud first and did a story map introduction, students reread the text on their own multiple times and participated in partner reading. In another study, students engaged in dyad reading for a total of 24 hours over 95 days. The weaker reader of the two followed the lead of the strong reader reading aloud.
The researchers report that they found no studies comparing the outcomes of students who worked with texts they could read with at least 95% accuracy with the outcomes of students working with harder texts.
Before working with more complex texts, teachers need more evidence of what instructional scaffolding is needed to enhance reading development, they write.
“What research says about text complexity and learning to read,” Richard Allington et al., The Reading Teacher, 2015, Volume 68, Number 7, pp. 491-501.