FAQ: Science of Reading – Partnerships for Literacy and Learning

What is the Science of Reading?
What is the Active View of Reading?
What are the Foundational Skills?
What is the debate around early reading?
1. What is the argument about how phonics should be taught?
2. What does the research about teaching phonics actually say?
What texts should be used in early reading instruction?
What are high frequency or sight words?
What is evidence-based literacy instruction?
What is research-based literacy instruction?
How do I know if something is evidence or research based?

What is the Science of Reading?

“Science of Reading (SOR)” refers to a body of research about the reading process and learning to read. As used currently, the term refers to studies and commentary that have been driven primarily by the theoretical model known as the Simple View of Reading (SVR). In its earliest iteration, the Simple View of Reading (Gough & Tunmer, 1986) stated that there were two components that predicted skilled reading: decoding and listening comprehension. It also theorized that decoding preceded comprehension and leads to comprehension. “The simple view presumes that, once the printed matter is decoded, the reader applies to the text exactly the same mechanisms which he or she would bring to bear on its spoken equivalent” (Gough & Tunmer, 1986, p. 9). While this view has been updated in recent decades (the so-called Rope View, Scarborough, 2001), it still posits that decoding and comprehension are separate strands until after phonics knowledge/skill is firmly established.

The research that is generally used by advocates includes descriptions of the reading process derived from traditional research studies as well as neuroimaging. These studies have contributed to our understanding of the reading process by confirming the important role of decoding (and its underlying abilities) to skilled reading. However, advocates of the SVR generally dismiss research that demonstrates that the reading process–and learning to read– are not actually that simple (see Duke & Cartwright, 2021; Snow, 2018). This, in turn, has made the “Science of Reading” a flashpoint for conflicting views of reading instruction.

Importantly, this version of the science of reading is being used as a proxy for specific practices – even when there is little or no research to support those specific practices. Thus, for example, instruction in decoding (phonological awareness and phonics in particular) is viewed as a necessary precursor to any instruction focused on strategic behavior or comprehension (except listening comprehension); an assertion that has no research base. See Foundations of Literacy.

Critics of these ideas argue that there is more science to be considered. They, too, believe in using research and evidence to understand reading and reading instruction. However, they advocate considering all of the applicable research. This position has recently been articulated in a comprehensive review that describes how advancements in research must be incorporated into our understanding and our practice (see Duke and Cartwright, 2021). After examining a far-ranging body of work, they describe the Active View of Reading as a counterweight to the SVR.

What is the Active View of Reading?

The Active View of Reading describes a model of reading that takes into account the findings of contemporary research in reading and learning to read.[1] It expands the “science of reading” to include new insights from research over the past 35 years since the simple view of reading (SVR) was published. We now know a great deal more about cognitive science, neuroscience, linguistics, and learning. Duke and Cartwright describe how the convergence of an extensive body of research highlight three (3) important conclusions:

Reading difficulties have many possible causes – within and beyond word recognition and language comprehension. Studies of reading performance among students with difficulties reveal diverse patterns (subtypes), with only some accounted for by poor word recognition. Research conducted from an interactive view of reading difficulties has long shown how a mismatch among reader, text, context and purpose can result in poor reading performance (see e.g. Lipson & Wixson, 1986; Wixson & Lipson, 1991; Valencia & Buly, 2004).
Research studies indicate that word recognition and language comprehension interact right from the beginning. In fact, variation in the areas of overlap predicts reading performance more strongly than does variation in either word recognition or language comprehension alone (see Lonigan, Burgess, & Schatschneider, 2018). As a result, reading and learning to read require bridging skills/abilities. Among these bridging processes are vocabulary, fluency and morphological awareness.
Self-regulation is a critical component of skilled reading. “…readers must learn to regulate themselves, actively coordinate the various processes and text elements necessary for successful reading, deploy strategies to ensure reading processes go smoothly, maintain motivation, and actively engage with text” (Duke & cartwright, 2021, p. 6).

Because the Active View of Reading takes into account a broad range of reading and reading-related abilities, it deserves the serious attention of educators.

What are the Foundational Skills?

The term foundational skills (or, foundations of reading) came into broad use in the 1980s as researchers began to seriously explore the reading process and learning to read. One of the earliest uses, in fact, is in the iconic Becoming a Nation of Readers (1985). The importance of these skills has been widely acknowledged for decades and have typically included significant attention to language development (see e.g., Lipson & Wixson,1991). More recently the term has been used to designate print concepts, phonological awareness, phonics and word recognition and fluency.

It is unequivocally clear that children need to come to realize that words have specific boundaries, and that there is a speech-print match word-for-word. Further, they must learn that speech is segmented into sounds (phonemes) and that these phonemes are represented in print by letters or letter combinations. Today, there is very little argument among educators or researchers that these skills and abilities are critically important to skilled reading – nor that they should be part of every K-3 literacy program. The confusion and debate surround not whether, but how, these should be addressed. See Debate: Instruction in Foundational Skills.

What is important to remember is that foundational skills are not limited to phonics/decoding. A comprehensive approach to teaching the foundational skills would also provide instruction and support for working with print concepts (including concept of word, print direction and book handling, etc.) as well as significant attention to the development of meaning vocabulary, oral language comprehension and, a bit later, fluency.

Can you explain the debate about early reading (and/or phonics instruction)?

There is not one, but a series of issues, that are typically wrapped up in the debate. The first challenge is to position the discussion in a much less contentious space than is typically provided. We start with the areas of broad agreement:

Should phonics (and phonological awareness) be taught?

Yes, of course, phonological awareness and phonics should be included in every early reading program. As noted elsewhere (see What are Foundational Skills), the debate is not about whether foundational skills are important and should be part of every early reading program, but how that should occur. Despite comments in the popular press, most educators and researchers have a longstanding view of the importance of phonics (see e.g., the 1997 position paper by the International Reading Association (now LRA).[2] (see What is the Backstory Behind all of this?). What that instruction should look like is more problematic.

What is the argument about how phonics should be taught?

This is a complex area that cannot be fully addressed in a set of FAQs. Please see the reference list for some resources for further reading. In brief, there are two major differences among the “purists” in this debate – but, remember, that most educators (and researchers) are not at either of these extremes.

The SOR group (see this FAQ) asserts that phonics and comprehension are separate and distinct aspects of reading and learning to read. Therefore, they propose that phonics (and its precursor skills of phonological awareness) must be taught first, and apart from, comprehension (meaning). This proposition impacts all other decisions. For example, it accounts for the assertion that using context to support word identification is “guessing” because the decoding must precede before meaning and anything else is just a guess. It also leads some advocates to suggest that children should not use any books in the earliest stages, but rather attend to only isolated decoding. When they are used, most suggest that the books must be highly decodable (see Early Reading Texts). Note: most advocates of SOR do, also, suggest spending time reading to children and developing listening comprehension. Finally, this group typically insists that only systematic, explicit and structured phonics instruction are effective.
The “limited phonics” group (the “whole language” of the 1990s) argues that phonics should never be taught in isolation from text reading. Further, they argue that learning to read is similar to learning to talk and that it should not be broken into small pieces for early learning. Instead of sequences of learning, Goodman (xxx) argued relevance, purpose, and meaning were central to learning to read.[3] Skills should be taught in the context of authentic reading and writing. Phonics should not be learned or practiced in isolation.

These views anchor the ends of the continuum of theories and beliefs about reading. Each has always been controversial, and each has contributed insights into effective practices (e.g. the importance of explicit phonics instruction and the need for purposeful, authentic texts, etc).

Along the way, many educators adopted a Balanced Literacy approach in an attempt to incorporate the important aspects of each of these more extreme views. We address that question later. For now, it is important to note that balanced literacy is not the same as “whole language” but a considered effort to accommodate sometimes conflicting evidence.

The so-called “Reading Wars” are definitely not new – but, they have become very heated in recent years (See What is the Backstory). They started in the 1950s, pitting the “look-say” approach against “phonics”. In other words, this was a contrast between phonics and “no phonics.” At that time, there was virtually no reliable research focused on children’s early reading to help resolve this issue. Today we do have research. Lots of it.

So, what does the research about teaching phonics actually say?

Many of the contentious arguments surround the type of phonics (e.g., synthetic, analytic, or embedded); the specific sequence of teaching and learning; and the relationship between word reading and text practice (see Early Reading Texts). There has been an exceptional surge in high-quality research related to this question in the last decade. The research has included meta-analysis as well as individual empirical studies. Meta-analysis is statistical analysis that combines the results of multiple scientific studies. Its aim is to derive conclusions and insights into an area that has a significant body of research. The well-known phonics section of the National Reading Panel Report is an example of this type of research. We have referenced some of the most important research of both types and provide a set of conclusions below.

Students do need to acquire both phonological awareness and phonics skills – English is an alphabetic language
There is no evidence that one specific approach is superior to others – for either normally-developing Ss or those who are struggling or at risk
Intentional instruction is required but there is no particular advantage to synthetic, analytic or embedded phonics – as long as teachers have a clear understanding of the trajectory of students’ learning and provide intentional and responsive instruction
The inclusion of multi-sensory activities does not appear to improve students’ reading abilities
There is increasing attention to morphological elements (as well as phonological ones)
The advantages for most phonics instruction are, at best, moderate and often extinguish themselves after grade 3
We point you to the following good advice: “The appropriate question to ask of a twenty-first century science of teaching is not the superiority of phonics versus alternative reading methods, including whole language and balanced literacy, but how best to combine different components of evidence-based reading instruction into an integrated and customized approach that addresses the learning needs of each child.” (Fletcher, Savage, & Vaughn, 2020).

Early Reading Texts: What kinds of texts should be used for early reading instruction?

In the earliest stages of learning anything, we try to simplify things so that learners can get support for trying new things and can experience some success (e.g., a smaller field for soccer, T-ball for baseball, smaller scissors for learning to cut, etc.). Learning to read is no different. The precursor skills needed to be successful are typically taught intentionally and explicitly and students get modified texts to practice their emerging skills. In order for students to be able to read these texts – something needs to be controlled. These early, controlled, texts are designed to support young readers as they acquire and practice emerging skills.

Not surprisingly, there are differences of opinion about the characteristics and purposes of these texts and, also, the role they should play in the overall program for early readers. The controversy closely tracks the various theories of reading and is often played out at the extremes of these positions. For example:

One group asserts that beginning readers should read only decodable texts. Decodable texts control the types of words used — limiting them to phonically-regular words. This may be limited to only one phonic element (-at for example). Over time, they include more phonic elements and these are often only the words/elements that have been explicitly taught. The idea is to encourage students to rely solely (or almost entirely) on letter-sound relationships to read words.

Another group argues that beginning readers should use predictable texts. These are texts that may include phonically-regular words but they also include words that may be partially decoded using (in addition) context (especially in the earliest levels, pictures). Here, the idea is to encourage students from the beginning to use both word recognition and comprehension strategies and to be strategic in using their knowledge and skill.

Neither the entirely decodable texts nor the completely predictable texts are optimal for teaching early reading. The most decodable texts deprive students of the chance to think about their reading and develop strategies for reading new words/texts. The most predictable texts deprive students of the chance to learn about the regularities of the sound-symbol system.

One of the best resources for teachers of early reading is the Hiebert’s Text Project http://textproject.org/teachers/students/beginningreads/. It includes thoughtful discussions and research-based analysis of early reading materials. As well, there are many downloadable books, videos, and lessons. In particular, she demonstrates how early reading texts should contain both decodable and predictable elements. Texts that control for both aspects, and make judicious use of high-frequency sight words, tend to be both more interesting and more supportive of a wider range of reading abilities.

In any case, it is important for students to have opportunities to interact with meaningful texts and have a comprehensive repertoire of reading skills. We suggest that you think about having different texts for different purposes:

Decodable texts to provide additional practice in applying newly-taught phonic elements;
Predictable texts to ensure that students are engaged with interesting reading, that they are using appropriate strategies; and that they are regulating their own success;
Age-appropriate texts for teacher read-alouds to ensure that students are interacting with complex texts, content, and ideas.

In the end, the most important thing is to consider how texts are used within the total reading program – and to consider what outcomes are being developed with these practices. In both cases, students will encounter high-frequency or sight words. These, too, may be the source of debate and we have provided an FAQ about rapidly emerging concepts.

Early Reading Texts: What kinds of texts should be used for early reading instruction?

Decodable texts to provide additional practice in applying newly-taught phonic elements;
Predictable texts to ensure that students are engaged with interesting reading, that they are using appropriate strategies; and that they are regulating their own success;
Age-appropriate texts for teacher read-alouds to ensure that students are interacting with complex texts, content, and ideas.

What are high-frequency or sight words?

These terms are often used interchangeably. However, there are important distinctions to be made, especially in early reading. As its name implies, sight words are any words that an individual can read “at sight” – without needing to decode the word. For adults, this is a very large proportion of everyday reading. It should be noted, also, that skilled adults engage in such rapid and automatic decoding that they may think the word was recognized at sight, when it was actually scanned for each letter and sound.

High-frequency words are words that appear frequently in written English (e.g., the, and, that, in, to, you, she, my, is, are, do, does). Word frequency lists (like Fry and Zeno) are based on statistical analyses of frequency. However, the frequency of words in early reading texts poses a special case.

Early reading instruction traditionally was distributed, with some time spent teaching phonics patterns and relationships and time spent on teaching “sight words,” which were typically considered “irregular.” Recent analyses suggest that many of the so-called irregular sight words are actually decodable. Words such as “like,” “play,” and “that” follow regular phonics-spelling patterns. The problem is this: children frequently encounter and/or need to read some of these words before they have acquired the requisite sound-symbol knowledge. Silent -e and the digraph -ay are 2 good examples. Typically, these are introduced as “sight words” to get a corpus of words that can be used to teach the pattern they represent. Many experts suggest teaching 10-15 words so that children can do some initial reading and also so that they can attend to both decoding and the acquisition of phonic patterns.

In other words, as with many of the contentious issues in early reading, there are arguments on all sides. What is clear is the kindergarten and grades 1-2 and special cases of reading that require careful and thoughtful attention to development.

What is Evidence-Based literacy instruction?

The terms “scientifically based,” “evidence-based” and “research-based” have often been used interchangeably. However, in recent years, distinctions between and among these terms have taken on a more nuanced and also a more significant meaning. That is because the federal government decided to prioritize its spending to focus on the use (and development of) approaches that have been demonstrated to work (have evidence of efficacy) and, in the process created and endorsed several, specific definitions (see Pena and Behrens, 2019)[4].

According to government definitions[5] evidence- based practices or evidence-based programs are approaches … that are considered effective based on documented scientific evidence. This includes findings established through controlled clinical studies, but other methods of establishing evidence are valid as well. Evidence-based practices and programs may be described as “supported” or “well-supported”, depending on the strength of the research design and the weight of the evidence[6].

What is Research-Based literacy instruction?

Research-Based Practices are practices that were developed based on the best research available in the field. As Shaywitz (2014) has noted, research-based means there are theories behind it, but that they aren’t always proven true. She tells how evidence-based means there is efficacy to back it up.” https://www.youtube.com/watch?v=nbQ9wAtTxlU. This means that users should expect that the content and practices have been shown to be important to reading development. But, unlike evidence-based practices or programs, research-based practices have not themselves been researched in controlled settings or compared to other approaches designed to address that same important content or practices.

How do I decide if something is evidence or research based?

As a practical matter, districts and schools use a combination of approaches for the simple reason that not everything that is important nor everything we need to teach has an appropriate evidence base. For that reason, educators need to examine the claims made by approaches and programs. Commercial programs, especially, are almost always “research-based.” Some of these have quite extensive scholarship and some data to support the efficacy of the approach. However, most do not. Instead, their claims rest on identifying aspects of literacy that DO improve reading and then assert that they address that area in their program or approach.

Phonemic awareness is one area where this is especially likely to occur. Research is unequivocal that, when children who do not have good phonological awareness skills are provided with instruction in these areas, they are much more likely to learn to read more readily. However, there is almost no evidence that a particular program is superior to others in achieving this. When the approach says it is research based it is simply saying that research says PA is important and we do it! Importantly, there will also be little or no evidence to show that all of the content in their approach is equally important or “research-based.”

Choosing an approach that is evidence based does not avoid the challenge of looking closely at the claims. For that reason, in the area of education, the federal education law details several ways to determine efficacy. It describes the rigor of the research designs themselves by defining 4 tiers of evidence (see xxxx). These tiers are distinguished by the sorts of studies providing evidence that a project component has a positive (that is, favorable) effect on a student outcome or other relevant outcome.

It also provides guidance and principles for conducting the research (see Pena & Behrens, 2019). The Iowa Reading Research Center summarizes these in the following way: “First, it requires that the program was studied by researchers who were not involved in creating the program. In addition, the researchers cannot stand to profit from the outcomes.

The research itself should have the following characteristics:

The program was compared to another type of program or a different kind of instruction.
Improvements in students’ reading abilities were measured with valid and reliable instruments.
There was a thorough description of how the program was implemented so that others could follow those same procedures and include the same elements.

The effect sizes were reported, and those revealed an improvement that was significantly greater than any improvement in the comparison condition.” [7]

[1] Duke & Cartwright (2021).

[2] IRA (not LRA). (1997). The role of phonics in reading instruction. Newark, DE: International Reading Association.

[3] Goodman, Ken. 2005. What’s Whole in Whole Language. Berkeley CA: RDR Books, pp.3-7. || Amazon || Worldcat, https://newlearningonline.com/literacies/chapter-5/goodman-on-whole-language

[4] Pena & Behrens (2019). Evidence-Based Approaches for Improving Federal Programs and Informing Funding Decisions. https://www.ida.org/-/media/feature/publications/e/ev/evidence-based-approaches-for-improving-federal-programs-and-informing-funding-decisions/d10703final.ashx)

[5] https://ies.ed.gov/ncee/wwc/Docs/Multimedia/wwc_definitions_transcript.pdf

[6] See Childwelfare.gov: https://www.childwelfare.gov/topics/management/practice-improvement/evidence/ebp/definitions/#:~:text=Evidence%2Dbased%20programs%20use%20a,strength%20of%20the%20research%20design.

[7] . SEE: https://iowareadingresearch.org/evidence-based-vs-research-based-interventions-update.