An innovative study of 17 schools across the country suggests that putting literacy coaches in schools can help boost students’ reading skills by as much as 32 percent over three years.
The study, which was presented here on May 1 during the annual meeting of the American Educational Research Association, is as notable for its methods as for its results. It’s among the first of what many scholars hope will be a new generation of studies that offer solid clues not only to what works but also when, under what conditions, and to some extent, why.
The study finds that reading gains are greatest in schools where teachers receive a larger amount of coaching. It also finds that the amount of coaching that teachers receive varies widely and is influenced by an array of factors, including relationships among staff members and how teachers envision their roles.
“This shows that this initiative can build networks and build social capacity in schools, and you can actually measure these things,” said Anthony S. Bryk, who led the four-year study with current and former Stanford University graduate students. He is currently the president of the Carnegie Foundation for the Advancement of Teaching, which is located on the university’s campus in California.
The study, which was paid for by the federal Institute of Education Sciences, focused on the Literacy Collaborative, a program jointly developed by researchers at Ohio State University in Columbus and Lesley University in Cambridge, Mass., with assistance from researchers from the University of Chicago.
Used in more than 700 schools nationwide, the program trains teachers to become literacy coaches, who then work one-on-one with their colleagues on a half-time basis to spread a set of teaching routines drawn from principles of cognitive science.
Teachers in Literacy Collaborative classrooms might, for example, help walk students through decoding processes as they read aloud or lead children in groups as they read progressively more-difficult texts.
The researchers tracked the implementation of the program in K-2 classroom in 17 schools. The total number of 8,520 students included in the study represented a mix of social and economic characteristics. For example, even though 45 percent of the students in the total sample came from low-income families, the percentages of students in each school who qualified for federally subsidized school meals—a commonly used indicator of family poverty—ranged from a low of 19 percent to a high of 86 percent.
To calculate the program’s learning impact, the researchers used value-added techniques to compare students’ progress on various reading-related tests and tasks with how much students would have been expected to gain on those measures with more-typical instruction.
They found that students’ reading skills grew 16 percent beyond predicted levels the first year, 28 percent more than expected by the second year, and 32 percent more than predicted by the third year.
But, as with many school improvement measures, the results varied widely from school to school, and even more from teacher to teacher within the same schools, said Gina Biancarosa, an assistant professor of education at the University of Oregon, in Eugene, who co-authored the study with Mr. Bryk, along with Allison C. Atteberry and Heather J. Hough, both doctoral students at Stanford University.
One explanation for that variation, the researchers learned, was coaching. Teachers and schools that experienced more coaching sessions tended to spur bigger learning gains in their students. Some teachers received no coaching over the course of the study, while others had as many as 43 sessions.
The teachers who got the most coaching were new teachers, teachers committed to the school and the reform model, and those who were found, through baseline surveys, to be more likely to initiate work-related interactions with other teachers. “So in some ways, coaching is a voluntary activity,” Ms. Atteberry said.
Faster Teacher Learning?
The schools where the most coaching took place were smaller, possibly because coaches were stretched more thinly in larger schools. They were also places where teachers felt they had a voice in what went on in their building and where professional networks among teachers were already strong. (Those network connections also grew over the course of the study, one of the papers found.)
And, likewise, teachers who had had more coaching were using the targeted teaching routines more often in their classes by the end of the study. The rate at which teachers picked up the new practices was highest for new teachers and those who came to their schools in the later years of the study, which spanned from the 2004-05 school year to 2007-08.
Across the board, teachers went from using Literacy Collaborative practices an average of 2.88 times a week to 3.36 times a week over three years, according to the study.
“I think there’s good reason to believe that having this instructional system in place is accelerating the process of teacher learning,” said Mr. Bryk.
Jennifer Sloan McComb, a policy researcher in the Washington office of the RAND Corp., a think tank with headquarters in Santa Monica, Calif., said the new study adds to a sparse body of research on literacy coaching, a practice that was one of the foundations of the federal Reading First program and is widely used in other school improvement programs.
The only two randomized controlled studies of the technique so far found that it yielded only small or no gains in students’ learning. Ms. McComb said the new study is important because it tracks progress over a long period, drills deep, links students to their teachers, and focuses on a high-quality coaching program.
“You do have quite a bit of coaching going on in this study,” she noted, in commenting on the new research at the AERA conference. Yet, she added, on average only about half the amount of coaching that the model called for actually took place in the schools included in the study.
Despite the findings of positive results from the Literacy Collaborative, half the schools in the study have since withdrawn from the program. Mr. Bryk said the program’s expense was a factor in those decisions.
A report on the study is due out later this year in The Elementary School Journal.
A version of this article appeared in the May 12, 2010 edition of Education Week as Coaching of Teachers Linked to Stronger Gains in Reading