Rocket Science Revisited: Where Should Reading Research Take Us Next?

Narrator:

Welcome to EDVIEW360.

Dr. Louisa Moats:

So, we need more of those leadership development opportunities that are run by people with real expertise. And then, I think when people understand what needs to be done, if they have the temperament to stick their neck out and assert themselves, they will succeed. Of course, they have to have a lot of personal characteristics of encouraging people and bringing them along, getting buy-in, and all those things that have always been the characteristics of effective leaders.

Dr. Tim Odegard: 

I know that right now what we need is fearless leadership who’s willing to lead through transitions and paradigms. For too long, it’s been a specific reading disability. It isn’t. It’s one in written language. That means reading and writing. I’ve been using some basic examples. I don’t know a child who can spell Constitution, but can’t read it. And I don’t know a child who can write an essay about a grade-level understanding of Constitution who can’t read and understand a grade-level essay on it. Those two things are parallel. They’re based on language. And for too long, we’ve siloed those and we haven’t put those together.

Narrator: 

You just heard from renowned literacy experts Dr. Louisa Moats and Dr. Tim Odegard. Dr. Moats and Dr. Odegard are our podcast guests today on EDVIEW360.

Cassondra Mantovani:

Welcome to Voyager Sopris Learning’s EDVIEW360, the podcast where we bring together leading voices in literacy education to explore what works, what matters, and what’s next. I’m your host Cassondra Mantovani. On our show today, we are honored to have two of the most respected figures in the field, Dr. Louisa Moats and Dr. Tim Odegard. This episode, Rocket Science Revisited, Where Should Reading Research Take Us Next, takes a candid look at the urgent need for systemic change in how reading is taught and supported across the U.S. We’ll revisit Dr. Moats’ seminal work, Teaching Reading Is Rocket Science, and hear Dr. Odegard’s perspective on dyslexia, teacher preparation, and the moral imperative to serve all learners. Together, they’ll challenge assumptions, clarify misconceptions, and call for courageous leadership at every level of education. Let’s dive in. So, first, to start things off, I’d love to hear from each of you on a personal level. Dr. Odegard, can you share the personal turning points that made literacy and especially serving learners with dyslexia such a deeply personal mission for you?

TO: 

Yeah, thanks for having me. The turning point really was early in life. Learning to read and write, and then reading and writing were always extremely challenging for me. Given how prolific I’ve been with writing now and how much I have to read, most people would probably find that a little incredulous for me to say. But I was identified very early as having real struggles with the basic mechanic of reading the English language or language, which is reading, and then the basic mechanic or one of the basic mechanics of writing the English language, which is spelling. And we know that those are defining characteristics across all writing systems of what we call dyslexia. Now, for me, it really set a spark that I wanted to prove to myself my worth, because part of my history is that I was perceived to be very less than when it comes to intellectual capability of learning more, which again, I’ve become a brain scientist. So, it’s a little ironic that the child who in early elementary was thought as being the class dunce grew up to be a well-published and well-funded neuroscientist and helped to develop actual brain imaging protocols used on brain scanners. 
So, it’s a little ironic in hindsight, which makes my story very compelling because you have to realize that test given so long ago was not a characterizing … it wasn't an accurate characterization of my true potential. Now, I didn’t start off in literacy. What I did was I started off as being a developmental cognitive psychologist and I was trained in that area. And I looked at learning and memory, potentially how you’d apply that to forensic settings as well as basic learning mechanisms across the lifespan, early childhood and older childhood. And it was when I started to draft my post-doctoral fellowship grant proposal for NICHD that was funded, I shifted and I put research into it and cross-training and dyslexia and literacy. So, I shifted coming out of my doctoral work into my post-doctoral funded research from NIH to look at that. And that was the turning point for me.

CM: 

Wow. Well, I just want to say congratulations to you for not allowing those early sort of stigmas to shape your passion for your work. And to go a little off script, if you don’t mind, I’d like to just ask a follow-up question to you on what you shared because it is so personal. If you were going back and you could talk to your younger self, just entering your dyslexia research, what question would you tell younger you to chase more boldly or to stop avoiding?

TO:

Oh, the younger me. That’s an interesting question because the younger me was very much in the basic mechanics and the neurobiological mechanisms and also the basic underpinnings of how we group letter chunks. And I was using very basic perceptual processing tasks developed by Ann Treisman, a world-renowned researcher on attention. She was positioned towards the end of her life at Princeton. She was the wife and partner of Danny Kahneman. So, Danny won the Nobel Prize. Anne won my heart with her research and methodology. I was really interested in the orthographic patterning and what Anne’s question was: What is a feature? And my hypothesis, which I think has been borne out more and more, all those years ago, back when I was writing up my dissertation and looking at this in a memory perspective, I was thinking this … it’s going to be scalable. So, that early in development, I bet the orthographic chunks and features might be the smallest units of writing, such as graphemes or even letters. But then as we start to develop and get more advanced, I bet we start to get to larger chunks. So, that if I had divided this line of research that was going to use really quick perceptual processing in both proficient adults as well as children with and without learning differences to see the scalable differences to where now we’re at final stable syllables such as T-I-O-N or S-I-O-N. We’re at these different chunks to do it. And I would be able to use 100 millisecond precision to look into those. I think that was well ahead of the curve. But then I would tell myself that only matters if it makes it into classrooms, which I think is where my passion really led me. So, it would be to also be brave and realize that a crooked path might lead to something very important. And so, don’t be afraid to leave academia, come back to academia. Don’t be afraid to do more translational research. And don’t be discouraged if people tell you that the questions you’re asking aren’t necessarily meritorious from a scientific standpoint. Because sometimes the most obvious questions to you are the ones that seem like they’re the least important for anybody else. And so, that’s been really a defining feature of what I've tried to have the courage to do.

CM:

Absolutely. Well, thank you so much for sharing. And I know it’s very personal. And so, I appreciate you sharing so candidly with us. And Dr. Moats, I want to turn the question a little bit to you. You have been such a pillar of this industry. And so, you have seen the ups and downs and twists and turns and the crooked road, as Dr. Odegard mentioned. So, what really keeps you so deeply involved in this work?

LM:

OK. Well, first of all, it’s a privilege to be having a dialogue with Tim and with you. Thank you for encouraging us to do this. So, I am not dyslexic. I learned to read easily. And in my first job, I was hired as the secretary in the neuropsychology laboratory at the New England Medical Center out of Wellesley College. I was a little bit underemployed, but my boss thought I had more potential. So he said, “Well, you can be a technician in neuropsychology if you also type all the reports.” So, I agreed to do that. And my boss put a white coat on me and I started giving an extensive battery of neuropsychological tests to all these people who came to the clinic. And this was before neuroimaging existed. So, when we had patients with neurological disease or injury or suspected pathology of any kind, we would use these tests to try to make inferences about where that problem lay in the brain and which brain systems might be involved. So, that if there was an actual treatment possible for the person, that the neurosurgeon or the neurologist would get some guidance. It was a very primitive science at the time. But what interested me the most was the kids who were sent to the clinic because they couldn’t read or write. And the school districts in good faith would send them to us hoping we would provide them with some insight into what was the matter. And I remember my boss, who was the licensed neuropsychologist of note, would put a report about three lines long saying we couldn’t find anything the matter with this child’s brain. So, it was not our problem, kind of. So, fast forward when I had an aborted career in music songwriting, went back to the neuropsychology clinic, and became the educational specialist. I was told I needed a doctorate. So, I enrolled at Harvard in the graduate program in reading and language. And that’s when my big ahas occurred because I began to study language. And all of a sudden, I gained all these insights that I’ve been writing about ever since about the fact that reading is dependent on language. Reading is language. So, then fast forward a long time through I had a number of different jobs and research and clinical work and policy, this and that and the other thing, and I became really passionate about the fact that there were a gazillion teachers, like the teacher I was, who had not had the requisite training and background that would allow them to do meaningful, effective work with their students. And that was when in 2000 I published my Speech to Print book. I began writing a lot more. Well, actually, I'd been writing for a decade about all this, but I didn't have a wide audience. But when I really stuck my neck out and said, “Look, we’ve got to develop courses for teachers, that’s when LETRS^® became a reality. And that’s been my primary focus, still trying to campaign for teacher education, educator education, because we can write policies and pass laws and mandate this and that and have standards and whatever … If the teacher doesn’t understand what they’re doing, who they should do it for, how they should do it, and whether or not it’s being effective according to their intentions, uh, we’re not going to make the gains with the promise of the science that we have.

CM: 

Absolutely. And we’re going to actually talk a little bit about some of the laws, but before we do that, I wanted to go back to to talk about Speech to Print because it has shaped how thousands of educators understand language. But when you first published it, was there an idea you worried might be too disruptive, but then turned out to be an essential component of later work?

LM:

Oh, I was worried that no one would pay any attention to it. And that I would never have any influence beyond a very small group of educators that I encountered through the International Dyslexia Association^®, mainly where I was a board member and an officer for a long time. And I would meet absolutely stellar individuals who knew a whole lot and who were very effective. But I was aware that they were like 1 percent of the teacher population. So, I thought I was writing for a very small audience. In LETRS, what LETRS became, I had absolutely no clue that we would find the audience that we have found with that.

CM:

And boy, have you. I mean, it’s been adopted at a scale that few professional learning models have ever reached. So, I’m so glad that you didn’t let the concern of a small audience stop you from pursuing that work because you have made such a difference. So, I want to circle back to some of the reading laws that have come across the country. So, we’ve seen a wave of dyslexia and science of reading laws come out across the country, but their impact varies widely. So, I'm going to ask the both of you to share your thoughts on what these laws reveal about the progress we’re making nationally and what you both feel that the gaps are that remain in research and implementation. So, Dr. Odegard, I’m going to start with you, if you don't mind answering that one for me, please.

TO:

Oh, it’s an interesting way of phrasing the question. So, what do these laws say about the progress that we’ve made? Well, I think it kind of echoes back to Louisa's point, which is there is a small, devoted group of people who know quite a bit and are very passionate. And, I think what we’re seeing is that there’s still a very small group that has grown across the states, across the context of north of the United States, and also up to Canada and north, so all of North America, who are very devoted to finding levers that they think they can pull to make that change happen in the classroom. So, what we’re, I think what we’re seeing, is that there’s a very select group of people who are highly motivated, and are coming to these conferences like The International Dyslexia Association, The Reading League, and other more grassroots types of groups like Decoding Dyslexia, who are energized and out there, and they’re pulling what they think is a lever that makes sense. And they’ll go as far as to say that there could be no change without a legislative context in place. What the research is showing is saying that the impact these are having has been very negligible. Unfortunately, when you actually do research like I've done to actually ask and use objective data to say what has been the impact. A review done by Elizabeth Zagata and Michael Coyne says that sadly, when it comes to this type of research, impact studies like I've done are very scarce. And so what we’ve learned and what we need more from research is exactly what Michael and Elizabeth called for. More research that can concretely tell us and provide guidance on what we’re learning and what we’re seeing. And what we’re seeing is limited application and translation into classrooms. So, that means that we’ll have limited ability to make the actual objective change that we want, which is the lives of children and literacy for children in the nation.

CM: 

Dr. Moats, same question to you. 

LM:

Well, yeah, if I could say that it is Tim Odegard’s research that has reinforced for me some realities that I’ve observed informally, but that have great importance for the way we proceed. First of all, kids who are less well off economically are far less likely to be identified as having reading problems that need to be treated with extra support or intervention, let alone a diagnosis of dyslexia. And the more a school is afflicted with schoolwide low performance, the less likely it is that the problems of individuals are going to be addressed with the what we call research-based instruction. And furthermore, as we know from multiple studies, that the kids in those populations of less economic advantage tend to be kids of color. And so, we have these laws on the books now, which seem like a step forward, and they are a step forward in building awareness. But in many states, those laws came about because of the advocacy of I hate to say it, but it’s true, white privileged parents who expected more of their schools and who had the means, the time, and the strategic approach to affect statewide legislation with their advocacy. But what has happened is we still see just huge disparities in the extent to which schools are embracing the real science and not just giving lip service to it. So, there are those realities there. And then, if I could talk a little more about just in terms of instruction, what I think the messages of research are, just to be brief about this. 
We could go on for a long time, but just to be brief about this, there are two major aspects of reading instruction. I'm not even talking about spelling and writing yet. But you have to teach the kids how to read the words, that’s word recognition, and they have to comprehend the language in the text, academic language. That means their vocabulary, their ability to process the syntax, their background knowledge, their knowledge of what to expect when they read a certain kind of a text, and all those things. That’s in the comprehension bucket. Our science is telling us about good practices that are very far from being widely embraced. And instead, what we have is, I think, on the word recognition side, an earnest but naive attempt to teach kids word reading using a kind of piecemeal approach, like, “OK, I’m going to do some phonics, and I’m going to do some morphology, and I’m going to do some phoneme awareness with a separate manual.” And those things don’t add up. The research tells us that this lesson needs to be integrated with those components in a meaningful sequence, feeding off of one another. 
Furthermore, that word recognition work needs to be integrated with the actual reading that kids do. So, there’s transfer. And very often what we’re seeing is real compartmentalization of these components that are now enshrined in our laws. Again, an earnest attempt in the right direction because it’s better than what was going on before. But in order to translate effectively, we have to get across the idea of integrated, language-focused instruction. And then, on the comprehension side, let’s see, a short way of saying this is: We're not in our classrooms doing nearly enough with in-depth, reflective reading of challenging material that requires the student to become comfortable with book language and academic language, which is not the same as conversational language. And we’re doing a lot of superficial instruction in the name of comprehension, which is not going to boost our national profile at this point.

CM:

Well, it’s interesting that you bring that up because it was something I wanted us to dive into as a group, which is that the development of oral language and comprehension are frequently overlooked in screening and in curriculum adoption. So, you’ve already touched a little bit about what some of the consequences of this neglect are, but how can some of these systems correct that? I’d like for you both to share your thoughts on this, if you would.

LM:

I'm going to pass that one to Tim because you, in that issue of Annals of Dyslexia, have some profiles of states that seem to be ahead of the game. So, what are they doing to bring practice into line with research?

TO:

Yeah, the two profiles were done by two different groups. One working in Kansas, the other one working in the context of Ohio. Actually, the lead authors were not necessarily in Ohio, but were working with the state of Ohio. What they’re really good examples of is putting systemic level things at place that would provide structural resources that would be able to help make alignment. So, one of the things that really popped out at me as I was reading my colleagues’ work as they were reporting on these was I do think that a coordinated effort to make sure that the intent and the good faith intent by these advocates is followed through. Because what Louisa was describing of this poorly constructed, naive approach to this, creates the ability for research scientists to say that the advocates don’t know the science because that’s not what the science would support. And so, that creates one group that should be in support of this type of initiative because it should be them seeing the translation of their science, but we’re not seeing it. And then the second group would be the people who were naysayers all along saying, and when we don’t get the results, “We told you it would be this all over again.” And so, the consequences of us not having systemic and structural pieces that can integrate and coordinate with higher education, to look at syllabus, go in and look at content, create alignment to the intentions, to have a pipeline of pre-service educators coming out with better knowledge of the linguistic components and the language components that they’re wanting kids to be able to understand orally and be able to understand in the written form and be able to express themselves in both oral and written form. You want them to be able to do those things and some practical awareness about what good practice looks like. So, they’re actually instantiating that in the practical components, working to make sure that there's now in-service support so that the teachers who are in the field who may have been let down in the context of their higher education, and or, as shifts in policy are resulting in large-scale adoption of new approaches in their context to let them have a fighting chance to have the resources they need to understand that. And then, also thinking about the interface of how the leadership will go in, the challenges that I will say is that those are examples of kind of intermediate-level support. The legislation pushing down to the Department of Education at the state level, creating intermediaries at the local context. It all happens at the local context. So, if your schedule doesn’t allow for the time for something to happen, it can’t happen. If your screening isn’t aligned in a way, one of the other groups of researchers, actually, two different groups of researchers, screened educators, understanding and use of the screeners that have been launched in response to the dyslexia and the literacy legislation. They’re ill-informed. They don't feel prepared. They don't know how to use the data. They find that the technology is often missing and in place for the way that these are being administered, and they find that there’s not the time they need to administer it. And so, then they devalue it or they don’t even know how to value the data that they’re being handed in this mandate. And so, I think that Kansas and Ohio highlight how the intermediates can come in. But again, the one level down that I know from working with folks in those states historically into today is that the follow-through and the accountability word isn’t the right one. The support and the realization of that into the local context will be the driver of change.

CM: 

Dr. Moats, did you have some thoughts as well that you want to share?

LM:

Yeah. As I think about what I would want to see 10 years down the road, I’ve been around a long time and have seen a lot of things come and go. And now, everybody’s spending a lot of energy talking about AI and this and that, the latest thing. What I want us to do is remember some very fundamental things and not have these very fundamental things get lost in the maelstrom of fads, fights, whatever goes on in education. No. 1 there is no substitute for the teacher when it comes to language learning in the classroom. Kids do not learn language from screens. They don’t learn language from each other. They don’t learn it from TV. They learn from interaction with better models and who can also encourage their language development. There’s no substitute for reading aloud to kids. That’s not gonna go away. And if any parent is listening to this, an iPad is not a substitute for a lap and a book. Really important, and let's not sacrifice that. And then, in the classroom, we find now some growing advocacy for this. Kids need to be reading real books, the whole book, not a little excerpt. And they need a teacher who has been shown how and supported in deep reading of the book. So, the book has to be worth deep reading. Yes, it can't be superficial. It has to have layers of meaning to uncover and talk about. And we need to get back to having a classroom be a place where people talk about meaningful things and the book and the shared experience of reading the book is the focus. And the teacher is capable of managing that or guiding that experience. That’s the heart of reading instruction. So, as we look ahead, I just, I want to take some of the fluffy stuff out. I would. And then, people say: “Well, where do we get the time to do this?” Well, you make the time for the most important things and don’t waste time with the ancillary stuff. Like, oh gee, if you’re teaching kids how to spell, they don’t need to write every letter in a different color. That is not efficient. It’s popular. I could go on. I'm not.

CM:

I’m so glad you touched on the topic of AI because we see it everywhere we turn these days. And I know it does provide bandwidth in certain capacities and can become a really great tool. But as we talk about the topic of dyslexia and we talk about students who are in need of intervention, firmly believing that there is a place for print and an actual textbook in hand and how the research shows that’s so much more impactful, especially for that student population. So, Dr. Odegard, would you mind weighing in a little bit on your take on sort of the AI’s place within the classroom and how to balance that AI and print within the challenges that the classroom faces today?

TO:

Well, I appreciate that. I’ve been working with my colleague Megan Gierka and a set of advisers. We’ve actually written a brief for the International Dyslexia Association thinking through this, and the premise being what Louisa started with technology can’t be a replacement for human interaction because it is language. And so, that interaction is going to be key to it. Really, what it forced us to do is to start from the premise of: Well, how does good learning happen? And what would good instruction look like? And where would the interface be to where, in a time-strapped position, where would it be? And most importantly, what I started to get a better appraisal with as I was looking at it is we have so little time for kids to do the thing that you want them to do and the ultimate end of everything. So, if it comes with work with first- and second-graders, really working with reading words and reading them in context and putting it into practice and having those. So, it really kind of boiled down to thinking through would this be best to use in the way that Carol Connor and her research looked at … with using machine learning and algorithms to use formative data that might be captured to let the teachers know where children's strength and weaknesses are. And what Carol’s research really showed was when she put teachers up against her algorithm, her algorithm was better at doing what computers do good. Look at good big data sets and find patterns. They’re really good at it when they’re trained well. And so, Carol was showing early, I mean, a long time ago, with her A2i platform, some of the promise of using technology to augment and give data and to think about what the time a child might benefit the most in a program might be. Also, would there be a way of using AI to generate ways for teachers to think about their instructional approaches and to create probes, prompts, and types of items that would fit in to make it really good, then review those with deep knowledge to look to see if those were generated in a way that would be beneficial? So, creating the content that would be beneficial to differentiate and intensify instruction for kids. So then, to be a teacher’s assistant in that way to build out time. There is also going to be the need for our diverse learners who take so much more time and practice to answer a really simple question. Are we at the time now, since so many schools do put so much money into computers that we’re going to think about what those are doing? Very rarely are they actually having them read words. Very rarely are they having them spell words. They’re normally proxies for those things with mouse clicks on a screen. That learning from some industry data that I’ve been proxy to behind the scenes isn’t as good as a spelling task. So again, that comes back to my basic principle. If you want kids to get better at doing the thing, have them do the thing.

Voyager Sopris Learning: 

Voyager Sopris Learning^® is the reading, writing, and math intervention specialist. With four decades of results, we provide evidence-based interventions and assessments that help educators ensure academic success for all students. Learn more at >voyagersopris.com.

TO:

And so to the extent to where technology might be in the position where we could build off of existing models, such as Art Graesser’s AutoTutor platform, for example. Or, older work from Wise up in Colorado, to where voice recognition systems could allow for some autonomy for children to go off and have scaffolded practice in a way that would benefit them in real time. I know that Mark Seidenberg’s name is thrown around a lot, and he’s known for a few things as far as the research side of it. And he creates these computational models. From a computational standpoint, they’re actually pretty simplistic because they’re based on these layers of representations. So, they’re not the most complicated of neural models that we have in this space from a science standpoint. But the thing that he never identifies for you is those models learn from feedback, just as children learn from feedback. So, when a child is able to be in a situation where they’re able to stumble along for themselves and self-teach them as autodidacts because they’re proficient enough to be able to do that, they’ll actually pick up a book and with a little bit of instruction, they can kind of take off. I know that Mark likes to say things like this. But for a lot of us and a non-trivial amount of us, it’s going to take a lot more because spell check doesn’t work for me for a reason. I spell words that are completely different words that I can’t recognize and read. So, to the extent to where I can’t self-correct myself, as David Share would talk about in his self-teaching hypothesis, it means that certain learners won’t benefit as much and they can’t give themselves that corrective feedback and fix up what they need in the moment when it comes to word recognition practice. So, to the extent that we might be able to leverage in those settings, that would be really good. The AI is also probably going to find real traction in the screening and the identification, as researchers like Laura Steacy and Don Compton continue something that I know that Louisa and I have thought a lot about, which is which items and orthographic patterns and what types of situations would give you the most diagnostic and how much you customize screeners to pull those apart. So, as we move into a new generation of adaptive testing and computer adaptive testing that’s going to be fed linguistic data and will use large data sets, being the global we, because researchers are being funded right now to do a developmental lexicon project based off of Dave Balota’s work at Washington U. And I think that’s really promising as we think about AI and its benefits. It can see the patterns. And when we know how to structure it, there’s a lot of risks. So, we’re also really hesitant to say it. And a computer is not a teacher. But where it can fit into good practice with good models of instruction, that I think that’s key is we need to think about how we take really good instruction and we create the opportunity for AI to facilitate educators to do that and then give students the facility to do better practice to create that information that they need in their heads … language.

CM:

Yeah. Love that. Dr. Moats, any other thoughts you want to share regarding your take on AI?

LM:

I am such a non-techie person. I’ve never used it for anything that I know about consciously. I mean, I know it is using me in multiple ways all the time. And I know I will be dragged into the modern age where AI is perhaps a useful tool. But, I really, it’s interesting that all that’s going on, the fervor and the discussion of AI. And Tim, I can't argue with Tim about the potential for AI to be used to help teachers be more skilled and perhaps give them more immediate feedback about the dialogue that’s going on between them and the students. For example, if we want that dialogue to be getting at more substance and less superficiality, perhaps there can be a little bug in the teacher’s ear driven by a computer that’s listening to this interaction, and say: “Why don't you ask how come the students know that the character is this way or that way?” And because trying to teach teachers how to do that without a whole lot of training and feedback has proven to be difficult. So, if there’s a way to do it kind of in the moment, that might be promising. And then the focus could still be on the worthwhile text where shared reading is a pleasure and sets kids up to be lifelong readers. So, that’s the other thing we’re seeing as a society is that while we grapple with how to teach kids to read, to learn to read, more and more we are all concerned about the fact that people don’t read as they get older. And even in college, we’ve had studies from institutes of higher education saying that college students come in not ever having read a book or not being able to handle a reading list. That was a white paper from the whatever it is on higher education just a couple of years ago, that professors have noticed a huge difference in their students from even 10 years ago because of the amount of screen time, No. 1, and also the way reading is taught in our schools.

CM:

Absolutely. Wow. Wow. Well, I”m going to shift gears a little bit and transition to talk a little bit about data. So, data transparency is often talked about, but rarely implemented well. So, how do you feel that transparent data at both the classroom and state levels can be leveraged to hold systems accountable for the literacy outcomes that we hope for?

LM:

Well, I'll say something fairly general. And I know Tim, this is really Tim is much more of a researcher and statistician than I am at this point. But while he’s thinking, one large goal that I think we need in education is to align what we teach with what we test. Because right now we’re testing kids a lot on stuff that they haven’t necessarily been taught. And there’s this misalignment. And I don’t think this happens in European countries, for example. I think in European countries there’s a curriculum that outlines content that teachers should be teaching instead of these general goals that we have in our standards about kids finding the main idea. Well, finding the main idea is very different if you’re reading about the American Revolution or if you’re reading about migrating habits of birds. And what we do is we put these tests together that have short passages about any topic under the sun, and the kids may or may not know. And then we know that their background knowledge affects their performance on the test. But when you’re done, you get a generic picture of a kid’s ability to perform on that test relative to other kids who took the test. But we don’t have a good alignment between, let’s call it a triangle, what teachers are taught to do with the content that they’re supposed to be teaching and the tests that we give to kids to evaluate whether or not they’ve learned these important things in social studies, science, literature, and then math, the arts, you name it. So, that’s my thinking. That I don’t, it’s like this big readjustment needs to happen in the whole system to have that alignment occur. 

CM:

Absolutely. Dr. Odegard, your thoughts?

TO:

Well, I think that’s a good set up. And I didn't quite know where I was going to go. Because the spirit of the question really is that there’s a lot of movement in legislative houses for accountability for education. Different states are passing these and implementing these dashboards. I know that I’m here in Tennessee, and our group, SCORE, was just holding a webinar on using the new Tennessee dashboard so that parents and others can see into the classrooms and into the schools to see kind of performance data. I know that Florida has been and is aligned and got a screener that was going to hold accountability even earlier in the process, starting in kindergarten and first with their state-adopted screener. So, I understand the spirit of it, but I think what Louisa is highlighting is really critical to think about what we measure and then what the ultimate outcome is. And it highlights what we know about research and also where some of our blind spots are. And also, when we think we get too concrete where we miss the opportunity. And I know that oral language came up earlier. And so, one of the challenges has been that when I work with high-level folks in large districts, some in Florida, some in other places, or states, some in the south, some in the east, some in the west, they’re identifying that the diagnostics of their screeners aren’t capturing the kids who are going to go on and not do as well as they need to on those high-stakes tests what Louisa is talking about. And so, one of the working hypotheses are is based off of our >LARRC study that the U.S. federal government funded at the very high dollar amount, which was longitudinally following language and comprehension development. And one of the interesting things is that in the early grades, your word abilities are going to be the best predictor of your reading comprehension. And so, they account for everything. Of course, we’re going to statistically or we’re going to follow where the data tell us to go. So, the best diagnostics from a screening standpoint and from a measurement standpoint is to be measuring: Can they read the words? Now, why is that? Well, it’s because in the early grades, late kindergarten, first grade, they’re really simple text. The word structure, the text structure is so simple. The background knowledge doesn’t really need it. You’re going to be able to read it, understand it, and you probably have a little bit of background knowledge because it’s going to be kind of a passage about Tim and his dog walking to the woods, going on an adventure, things that kids are likely going to know about. There might be some cultural differences that we need to take into account, but often those are kind of washed out of that as far as like making sure that they’re OK. What we’re missing is that, and what they say is that: “Well, they’re doing just fine on those screeners.” A lot of kids, and they get to that third-grade test. Now, we have retention laws and they don’t do well a lot. What happened? Well, what we find from, let’s say, >Tiffany Hogan’s research, Suzanne Adlof’s research with Tiffany and others is that all of a sudden, measures that were in kindergarten weren’t predictive of reading comprehension early, are now more predictive than the word reading ability later on, let’s say in late second, third grade. And why is this now? Well, the text structure has gotten so much more dense. The words in it are much bigger and more complex. And simply reading the words and maybe knowing what a story is about isn’t going to be sufficient anymore. Those earlier abilities in oral language and those comprehension strands of oral language now mean a lot. And so, the predictive value is developmentally shifted. And so, it raises this interesting question. We shouldn’t necessarily be reading measuring oral language to find your classic dyslexic better in those grades, but we probably should do a better job to make sure that we are teaching to the full spectrum of language development and giving them the skills to set them up for success later on. Now, Louisa made a good point. It might be that we’re not teaching the text structure in a disciplinary way early enough to really make them understand how they should attack different texts and the approach they should use with different ones. That’s what she’s highlighting. Now, that might matter and that might account for some stuff. But I will say that my hunt tells me as I talk to certain states, why they’re motivated to maybe bring in a language screen earlier is because they’re hopeful that by identifying kids whose language needs to be elevated so that their comprehension skills are being elevated and not relying on written language as a gatekeeper to tell you that they will be able to better find kids who will not do all that high-stakes test, might be able to develop simple decoding abilities, but later on start to struggle. Historically, Hollis Scarborough and Don Compton and even Hugh Catts have called this as a late emerging type. Me and Tiffany Hogan joke because it’s like, well, if you would have looked for them a little earlier, you might have found them. So, I’m hopeful that I know that I’m motivating and trying to get into some different types of research questions to say, I’m not arguing that the research wasn’t right. I’m arguing, were we not measuring the right things early to see the benefit of instructing certain things and bringing it online so it’ll continue to develop and we can elevate it in certain classes of kids. It’s a good hypothesis, which needs to be tested.

LM:

That’s really important. I just want to second what Tim said that the measurement of language in preschool kids or before kindergarten is a powerful predictor of who’s going to be able to comprehend what they read after third grade. That’s that, and that is a finding that’s been replicated a number of times. But because it’s even more clear that’s the case with better measurements, it also points us in the direction, once again, and this is not new news. It’s just that we’ve got to get people to pay attention to the importance of finding kids early and giving them instruction. I could use that word, in an early childhood to develop, design to develop their language abilities. And that we see from studies we were hearing about from the language conference at Big Sky this summer, those interventions make a significant difference, especially with kids who come from more impoverished backgrounds, who are not exposed to the language models or not read to as much and all that. The way we level the playing field is not to wait until third grade to retain them because they haven’t met the standard. The way to level a playing field is to get those kids as early as we can in their development.