What Works?

What Works?

Administrators are learning that the art of teaching reading isn't fundamental.

Sherlock Holmes would develop a migraine deciphering this one:

This spring, the Denver press jumped all over the fact that three years after the Denver Public Schools sank $13.5 million in the first year alone into a literacy program, students in the poorest schools still weren't reading or writing any better than before. Meanwhile, the wealthier kids did respond, widening the achievement gap between rich and poor.

"There's a long way to go--a lot of people have pointed that out. What would be more helpful is if someone pointed out a model that works," outgoing Superintendent Jerry Wartgow told a Denver Post reporter.

Then there is the Santa Rosa County School District in Florida, which implemented Scholastic's Read 180 program during the 2001-02 school year for its middle school and high school students reading significantly below grade level. Schools had successfully moved between 42 percent and 80 percent of these students out of the lowest reading category, according to the Florida Comprehensive Assessment Test scores at the end of that year. By 2004, every school in that district using this intervention tool and a 90-minute instructional model made adequate yearly progress in reading with every population.

Reading scores across Anne Arundel County Public Schools averaged 64 percent in 2003: a statistic Superintendent Eric Smith never wanted to see again. Unfortunately, Anne Arundel lacked a district-wide defi ned reading strategy.

The clue to solving cases like these, in Ted Rebarber's opinion, lies with applying best reading practices. "Reading is an area where we have a wealth of research and evidence that certain practices work with a much greater proportion than the practices commonly used today. For the money and time we are spending, if we use practices that consistently work, we would get better results," says the CEO of the Education Leaders Council in Washington, D.C. And in these days of No Child Left Behind requirements, educators' ears perk up at the mention of positive results for the same amount of money.

However, Rebarber admits, he's not a reading researcher. That breed tells a far more complicated story. But happily, Rebarber is dead on in one aspect: Most school districts can impact reading scores positively with the dollars at their disposal, says Timothy Shanahan, professor of urban education at the University of Illinois at Chicago and president-elect of the International Reading Association. As the former director of reading for the Chicago Public Schools, he engineered a literacy program that saw the biggest test score gains in that city's history. Approximately 450 of 600 schools improved during the 2001-2002 school year, with the bottom 100 schools gaining as fast as everyone else--without any new dollars in the coffers.

"If most districts reallocated a half to three percent of their budget, they could have a pretty dramatic impact," he says of the school systems he consults today.

"But the idea isn't, 'Boy, if we just fixed the materials, everything will be fine,' " Shanahan adds. "The real key is a combination of whether you teach the right curriculum, how long you teach and how well you teach it."

The Foundation

According to Robert Wortman, adjunct associate professor of language, reading and culture at the University of Arizona, 75 percent of kids learn to read no matter the program--the hoopla surrounding best reading practices centers around that 25 percent that don't grasp the skill as easily.

Eric Smith, outgoing superintendent of Anne Arundel County Public Schools in Annapolis, Md., is living proof of those statistics. Reading scores across the district averaged 64 percent when the kids left for summer break in 2003, a statistic he never wanted to see again. Unfortunately, Anne Arundel lacked a defined reading strategy district wide; each school functioned as an island to apply balanced literacy and whole language tactics. Smith instead established the Open Court core reading program, and saw reading scores in his district jump to 84 percent by spring 2005. Digging deeper, black student test scores rose from 45 percent to 70 percent from 2003 to 2005.

Yet today, Smith worries about the kids who didn't respond to the Open Court program, so he's on the hunt for the answer to that puzzle. "It's a smaller number than before, but we still have to find solutions to this," he says.

So even the case studies beg the question: just what is a best reading practice? "Basically a code word for whatever you want to do, or whatever you've done in the past," says G. Reid Lyon, research psychologist and the former chief of the Child Development and Behavior Branch at the National Institute of Health. He prefers the term "evidence-based practices."

Instead of criticism, it "would be more helpful if someone pointed out a model that works." -Jerry Wartgow, superintendent, Denver Public Schools

But whatever the nomenclature, most researchers start with the National Reading Panel's elements, a report conducted for Congress in April 2000 by a panel of experts that included Shanahan. Ironically, the panel's report isn't his launching pad when consulting with administrators. Shanahan first digs into the sheer amount of instruction teachers devote to reading (Chicago Public Schools require two to three hours a day) before examining and applying the National Reading Panel focus areas (see below).

In his role as an educational consultant, Wortman refuses to confine the best practices to these five pillars. He throws instructional elements like demonstration, guided practice and assessment into the mix as well. "Best practices keep being invented by good teachers," he says.

In fact, the industry is fairly stable from a research standpoint on what's necessary for students to learn how to read, says Doug Carnine, director of the Western Region Reading First Technical Assistance Center at the University of Oregon. It's the variables-children from home environments that don't use a substantial vocabulary or students with learning disabilities in phonics, for example--that become today's monkey wrenches. With these in the mix, a single reading program probably won't cut it even at the individual classroom level, he says.

So the crux of the reading program dilemma is that some programs could work for some schools and not others. And, as Santa Rosa County School District discovered, it can even depend on how well you follow directions. One of its schools shunned the 90-minute instruction model of READ 180, and those students failed to meet AYP, reports Scholastic spokeswoman Rachel Cruz. "Our studies have shown that off-model results are not as strong," she explains.

Lyon has been searching for the answers to this basic question since 1975: For which kids are which reading approaches most beneficial within which settings at which phase of reading development?

Carnine's research indicates that the more difficulties kids have, the more comprehensive and intense the program must be, and the more minutes per day teachers should devote to reading. "And the better prepared the teacher should be, because no one program in any clinical trial will be equally beneficial for all kids," he points out. "The teacher has to clearly understand where the program is breaking down with a particular kid."

Teacher Training

Now Carnine is singing Wortman's song. "My idea on using the money wisely in a school district is to apply it to professional development. You have to help teachers understand the reading process, reading strategies they could use, and how to make decisions about what's best for their classroom of kids," says Wortman. "Otherwise, any program you implement won't be effective."

For example, say two second-graders from the same background can't read well. The teacher uses a strong evidence-based program and Joe catches on, but Johnny continues to struggle. Now the teacher must have the knowledge to drill down to find the break-down in phonics, comprehension or vocabulary.

Unfortunately, says Rebarber, schools of education traditionally focus on philosophy and interesting theories about instruction as opposed to the more scientific, research-based skills needed in a real-world classroom.

It falls to the district to bridge the gap, which is the conclusion Smith drew from his meetings with Haan Foundation for Children educational researchers near Pittsburgh. This school year, he is launching a comprehensive tier strategy for Anne Arundel, which is basically a series of programs and directions teachers can turn to for students still not reading up to par. Smith refers to it as "triage."

"I try to picture what it's like if I'm a brand new principal: 30 years [old], a former elementary school teacher, maybe trained in reading but probably not. I'm assigned to a low-performing, low-income school where children have a great deal of reading difficulty. What am I going to do?" he says. "Most districts don't provide a real explicit road map of how you approach this."

When he served as director of reading for Tucson Unified Schools in Arizona from 1999 to 2003, Wortman replaced citizen volunteers with trained reading tutors, then built up a cadre of classroom teachers well-trained in reading and writing strategies to serve as benchmark models and internal consultants throughout the district. He brought in reading experts to give seminars and bought a copy of the speakers' books and materials for each teacher who attended to encourage incorporating the ideas in the classroom.

Shanahan encourages principals to take a page from Wortman's plan. "A good administrator has to go into a classroom and see how things are going, interact with teachers, make sure the teacher is making judgments and decisions the whole time," he says. "The idea isn't to catch them to write them up. It's to make sure nothing is overlooked, to keep everyone's head in the game." Superintendents, too, need to adopt this role with the principals. "It comes down to making sure everything you do actually supports the outcomes you're trying to make happen," Shanahan adds.

"The whole reading direction is very much the cooperative learning from 20 years ago, but we've taken it to a new level," says Joan Cunningham, who handles customer and industry relations at Kurzweil Educational Systems based in Bedford, Mass., and is a 15-year classroom and school administrator veteran. "I find it really refreshing. I wish I had a classroom to go back and apply it to."

Tire-Kicking Guidelines

So if context and a teacher's teaching strengths play such a crucial role, does this mean there's no such thing as a bad reading program? Well, sometimes programs purport to teach best reading practices but don't offer enough thoroughness, Shanahan hedges. Perhaps they have too few lessons or they don't extend clear explanations.

Cunningham joins a large number of educators who deride yesterday's whole language approach. "I had students who were basically clueless in terms of encountering an unknown word. They'd look at the initial letter and beyond that, have no idea what the word said," she describes. But even within that failed experiment, she can pick out good tenants.

"You can have a poor reading program, there's no doubt about it. But the more common problem is that the program isn't being applied properly," says Shanahan. In other words, he applauds a good vocabulary program, unless a district buys it with the idea it's a core reading program.

Publishers trot out a stream of programs and updates annually for administrators' consideration--and they do it because we live in a capitalistic society, not from an altruistic motivation. "Certainly when the vendors started hearing that we were thinking about making a purchase of their product, they all got really excited," Smith relates. "Then they found out there are three tiers to this plan and they didn't know if it was a good thing or bad thing to be on tier two or tier three. But it's not about the vendors. It's about the success of our children. At the end of the day we need to be able to give better, stronger evidence to Mom and Dad that when they trust their kids to us, we're going to find the right answers for them."

Thanks to copyright issues, no two groups do things exactly the same way, so some teach short O before long O while others teach short A before short O. The competition, in Wortman's words, "gets to be a little much. That's why in the long run, the more teachers know about the reading process, child development and teaching strategies, they're much better prepared to teach any program."

"Every time someone comes up with a new program, yes, they have to field test it and yes, it's based on research just like all the other programs. But who you research it with and where makes all the difference," Wortman continues. That's partially why reading centers like the Florida Center for Reading Research at Florida State University in Tallahassee review research claims and post their findings. (About Santa Rosa's Read 180, the Florida center notes that in the Los Angeles Unified School District, low reader scores improved only slightly. It lists classroom management as one of the program's weak spots.) Likewise, school districts turn to the What Works Clearinghouse, a federally funded initiative that lists the findings from programs it reviews at its Web site.

"But when push comes to shove, whether you have a Houghton Mifflin program or a Harcourt Brace, life will go on," Wortman says. After all, he notes, a best reading practice is not the same as a reading program. "Most administrators, when they're reviewing a program, go by what they know. I want a program built on research, but at the same time, look like it's using some common sense and looks user-friendly for a teacher."

Julie Sturgeon is a contributing editor.

Related Information


Advertisement