Sunday, September 25, 2016

Teaching Struggling Readers: Focus on Meaning

Why do we read? Of course, you will say we read for pleasure and we read for information and perhaps even we read to become better writers and thinkers. In order to enjoy and learn from reading we must make sense of the words on the printed page. This is true of all readers. The driver behind all reading activities is meaning.

Struggling readers have difficulty accessing meaning. This difficulty can lead to a cascading accumulation of reading difficulties. Kids who have difficulty making sense of what they read, read less. When they read with the teacher, they are more likely to be interrupted than skilled readers and so they read less. When children read less, they get less chance to develop the decoding abilities, vocabulary knowledge, syntactic understanding and content richness that good readers need.

Reading researcher, Keith Stanovitch has called this, "Matthew Effects in Reading", the idea that in reading the rich get richer and the poor get poorer, mainly due to differentiated exposure to text. Motivational issues then compound the problem. When you are not good at something, you tend to avoid it. Children who are not good at reading read less.

Stanovitch and others have identified one primary cause of reading difficulty - phonological processing. Children with  phonological processing problems have difficulty decoding words - breaking words down to their component parts or "sounding them out." Often these children are labeled as "dyslexic" a term fraught with mythology and mistaken connotations. I do not think the term is very helpful. Better I believe to think of struggling readers on a continuum of struggle -from those who have a great deal of difficulty taking on reading, to those with more minor concerns.

Whatever the label, however, the identification of phonological processing problems as a major cause of reading disability has led to a cottage industry of programs designed to solve the problem. These programs are, for the most part, based on what is commonly known as the Orton Gillingham Approach. We have Wilson Reading, Spell Read, Stevenson, Corrective Reading and on and on. These programs have two things in common. First, they are heavily focused on phonics instruction. Second, they do not work very well (See below)*

In 2008, the 9 million dollar Power4kids study was lauded as having the "highest possible standards for a research study." The purpose of the study was to find out which of four programs (Spell Read, Wilson Reading, Corrective Reading, and Failure Free Reading) designed for struggling readers provided the best results. Struggling readers in 3rd and 5th grade were studied. The researchers found, to their surprise, that students in the treatment group who received instruction from one of these programs fared worse than students who received no special programs. While the lead researcher, Dr. Joseph Torgeson of Florida University, said that this was an "extremely well-designed study", he backpedaled on the results citing a litany of reasons why the study did not yield the results he had hoped. One reason Torgeson cited for the failure of the study caught my eye - the interventions did not include enough comprehension instruction.

What Torgeson seemed on the verge of discovering, though he did not state it directly, is that any program that focuses on any one component of literacy is doomed to fail. In reading instruction, we tend to want to identify what is wrong and fix it. Since we have identified decoding weakness as a difficulty for many struggling readers, we tend to want to do lots of focused decoding instruction, ala Orton-Gillingham or Wilson. We apply the metaphor of the machine. After all if your car is not running and you identify a faulty fuel pump as the reason, you replace the fuel pump.

But children are not machines and language is not a fuel pump. The key thing to understand in designing a support program for readers is that reading is communication. If we begin our search for the best way to help a struggling reader with the idea that language is meaningful and reading is about making sense of written language, then we have a better chance to help struggling readers.

What does this mean for instruction? One thing it means is we need to provide interventions early, before children experience too much failure and adopt too many "confusions" about how reading works. Secondly, it means that rather than doubling down on phonics instruction, we need to double down on meaning making.  If a student struggles to make meaning from text, we must scaffold the meaning sufficiently to assist the student in decoding the words.

Most instruction for struggling readers, in other words, has it backward. The Orton Gillingham approach and all of its imitators like Wilson Reading and Stevenson, note that struggling readers have difficulty with phonics so they pitch right in to provide large doses of phonics instruction. The study of that great communication skill, reading and writing, is divorced from communication and centers on the tiny bits and parts of words. This fails because the driving force behind reading, making meaning, gets lost in a forest of sounds, symbols, and step-by-step study of one reading skill after another. "Let's focus on the schwa sound today, children and we'll look at 'ed" endings tomorrow."

Now let's imagine an instructional design where we turn this on its head. Reading is chosen for a group of struggling readers that is fairly easy for them to read. The text is deliberately chosen so that the children will likely stumble on only a few words. Literacy researcher Richard Allington says that most struggling readers spend most of their reading time in school reading text that is too difficult for them. So let's make the reading fairly easy.

Next let us suppose that we introduce the story to the children in such a way that it helps them activate their background knowledge for what they will read, builds expectations about what they will read and allows them to predict what they will be reading about before they begin reading. Let's suppose that we get the children talking about the story, noticing things from the pictures that accompany the story and making personal connections to the story before they read.

A good book introduction helps children get ready to read and anticipate what they will read. They have built up expectations and those expectations can help them power through unfamiliar words.

Next let's suppose that while all these struggling readers are reading this story that is reasonably easy for them and that they have been well prepared to read, that we listen in while they are reading (all children reading at the same time, not round robin) and we notice they stumble when trying to identify a word. At the point of difficulty we can then prompt them to use all the cues available to them to identify the word: What would make sense? What would sound right? What would look right? And when they try to make it look right, but limited ability to apply phonics rules interferes? We try other prompts. What do you see that can help you? What is the first letter? Does this look like a word you know?

After the reading and after we have observed what children have struggled with, we can follow up with a decoding lesson based on the struggles that the students had with individual words, but we must also follow up with a discussion that builds the students understanding of what they had read.

Of course, what I am laying out here is the design for guided reading. Guided reading puts meaning at the center of the reading instruction and therefore puts struggling readers in a better position to use all the cues available in reading to decode the text. If we front load meaning, we give struggling readers a better chance to read meaningfully and if children are reading meaningfully they have a better chance to decode successfully, read more and improve their decoding abilities.

Guided reading is based on one reading intervention strategy that has been proven to be effective for struggling readers: Reading Recovery. Like all programs for struggling readers, decoding instruction is an important component of Reading Recovery instruction, but Reading Recovery puts meaning at the front of the instruction.

Literacy expert, researcher and writer, Richard Allington says, "We’ve known for two decades that when classroom reading lessons for struggling readers are meaning focused, struggling readers improve more than when lessons are skills focused."

Enough said. Want to help our most struggling readers? Focus on meaning.

*I know that the statement above is going to anger a number of my hard working colleagues who labor mightily everyday with struggling readers employing one of these Orton-Gillingham style programs. I know they believe in what they are doing and are doing it to the best of their ability in the service of children. But faith is not research and the research evidence is just not there. You can look hereherehere, here, here and here for summaries of that research. The research into these programs indicates that these programs might have some positive effect on "alphabetics", that is phonics, and very limited or no impact on comprehension. This is to be expected. If teachers focus their attention on one thing, kids are likely to get better at that one thing. According to the University of Michigan website on dyslexia, children with dyslexia can learn phonics, their difficulty is in applying phonics in real reading situations. They say "increased instruction in phonics will not help dyslexics" Phonics is not reading any more than spelling is writing.

Monday, September 19, 2016

Stop the Nonsense (Words)

Let's get the main idea of this blog post out of the way early. Nonsense words should not be used for decoding instruction. Period. No how. Not ever.

A while back, I wrote a post that argued that the commonly used early literacy assessment called DIBELs, led to poor literacy instruction because it focused on the bits and pieces of literacy learning rather than looking at the big picture. Paraphrasing literacy luminary, P. David Pearson, I wrote, "the widespread employment of DIBELS has had dire consequences on the actual teaching of reading."

Like many early literacy assessments, DIBELs uses nonsense words to assess student ability to decode. This is a well-validated practice and can provide useful information for diagnostic purposes. But diagnosis is not treatment and nonsense words should never be used for instruction. When a doctor suspects a broken bone, that doctor will often order an X-Ray. If the X-Ray shows a broken bone, the doctor treats the bone with a cast, a wrap or surgery, not with another X-Ray. So it is with nonsense words - they point to a problem, but are not to be used to treat the problem.

Why not?

Because learning to read is an act of communication and communication only happens with real words. In order to develop skilled decoding abilities, children need to be exposed to lots of real words. Real words have a certain set of finite spelling patterns. Yes, in English this is more complex than in most languages and this is a source of much difficulty, but still the patterns are there. The human brain is a pattern identifying machine. And young minds are particularly adept at intuiting patterns. The detecting of patterns in writing is mediated by the child's oral language. A young developing reader learns that the word "man" begins with the sound "mmm" and then learns that that sound can be represented by the letter "m." This can only happen through exposure to real words that are in the child's oral vocabulary.

As we expose children to real words, they get more information to store in the pattern detecting parts of their brain. We can expose the children to words in isolation, in real reading contexts, in word families, or as onsets and rimes (sp+ot), but no matter how we are presenting words to children, we must be presenting real words, so that children can discern the patterns. Of, course we can also teach those patterns explicitly through word families and spelling instruction.

Literacy researcher, Marilyn Jager Adams says that, no matter how we are exposing children to real words,  we can optimize student understanding by making sure that the children see the word, say the word, understand the word and know its meaning. All of these contribute to a child learning a word and building the ability to decode the word and other words with similar patterns.

Of course, not all English words follow regular patterns, so sight word instruction is also key, especially for function words necessary for early reading like the, of, was, do. These words should be the focus of early instruction and learned by sight.

Nonsense words do not give children the opportunity to intuit patterns. They violate patterns and make learning to decode more difficult. This characteristic makes them useful for diagnostic tests, but disqualifies them for instruction.

Literacy researcher, Tim Shanahan, believes that the spread of the use of nonsense words can be attributed to administrators mistakenly using diagnostic tests to evaluate teacher performance. If teachers are going to be assessed on these tests, then teachers can hardly be faulted for teaching kids how to read nonsense words. Simply put, using diagnostic tests in any way to evaluate teachers is, well, nonsense. On using nonsense words in instruction, Shanahan says simply, "Don't do it."

As Adams puts it, "The brain does not grow block by block from bottom up. It grows through its own efforts to communicate and find coherence within itself." Nonsense words interfere with our natural desire to communicate and lack any coherence with a child's spoken language.

Stop the nonsense!