Whole Language Research Base

Jill Kerper Mora

Whole Language Research Base

The purpose of this webpage is not to propose a defense of Whole Language. My purpose is to affirm the position of the theoretical framework and empirical research base for Whole Language instruction as credible and legitimate Science of Reading research. It is my impression that the SoR Movement is needlessly fighting the ghost of Whole Language in an attempt to establish its own credibility, but in the process, is detracting from it. This is because the SoR Movement has focused its ire on the notion of “three-cueing” and has made this a target of policies and pronouncements designed to discredit instructional and assessment practices that utilize data from the large body of study of oral reading and language processing based on miscue analysis. Proponents of SoR create straw man arguments against whole language in order to set them on fire to burn them at the stake, thereby producing much more heat than light. The criticisms of cueing systems based on a dichotomy between phonological decoding and the use of “context cues” is the product of a misinterpretation of what whole language (psycholinguistic) researchers meant by “context.” Consider that the term/name Whole Language contains the words “whole” and “language.” I have always interpreted this to mean that WL is based on research that rejects the notion that phonology is the most important, and possibly the only important, source of cueing about words’ meanings (semantics) and therefore, the only part of language that is really needed for decoding words. The term “whole language” emphasizes that all of the subsystems of language that make language whole in order for it to convey meaning must be considered in teaching children to read and write.

Let us consider the nature of the ongoing war against Whole Language. We examine the Indiana Department of Education (2023: 29) titled Indiana’s priorities for early literacy. This document states that the “three cueing system” is a “whole language theory of reading.” To quote the document verbatim: “These approaches are often based in the whole language theory of reading and three-cueing system, which emphasize meaning-based instruction and the belief that readers use cues (e.g., semantic, syntactic, and grapho-phonics) to pronounce words…” The idea of subsystems of language as systems that aid in the pronunciation of written words predates whole language by centuries, not just decades. In fact, this is why the alphabet was invented. The invention of the alphabet most certainly predates the invention of the term “whole language.” For an account of the history of language, literacy researchers can read the book titled “The loom of language” by Frederick Bodmer (1944). Bodmer describes the grapho-phonics subsystem of language in great detail in a chapter on the history of the alphabet and the evolution of English spelling all the way back to the dawn of Egyptian civilization and the desire of the priestly class to keep a secret code of signs and symbols. Consequently, grapho-phonic “cueing” is not a product of a modern theory of reading. The same is true of syntax, which Bodmer calls “the traffic rules of language.” Also, Bodmer (1944) analyzes the complexities of how words are structured to convey meaning in different ancient and modern languages, such as English words derived from Latin. An understanding of the history of alphabetic languages, as descendants of Latin and Teutonic roots in their evolutionary context, should lead the Indiana Department of Education to question its attribution the role of language subsystems in students’ learning of reading and writing in English to a particular approach to literacy instruction.  

Literacy Research Pioneer: Professor Kenneth S. Goodman

Professor Kenneth Goodman’s describes his Transactional Sociopsycholinguistic Model of Reading and Writing in his 1994 chapter, where he presents the history of his model and theory of reading. Prof. Goodman explains that he based his metaphor of the “psycholinguistic guessing game” on Noam Chomsky’s concept of “tentative information processing.”  Noam Chomsky proposed his theory about language’s “surface structure versus deep structure.” Consequently, when my colleagues in the literacy research community treat the term “context” as referring to something that is not, first and foremost, the whole context of language, referring to all of the subsystems of language, I become dubious about their arguments against the use of “cueing systems. The book On the Revolution of Reading (Flurkey & Xu, 2003) with a compilation of the selected writings of Ken Goodman is an excellent place to start if a person wants to explore Goodman’s contribution to our understanding of reading and writing. Professor Goodman authored 131 research articles, 26 books and monographs and 86 book chapters, along with 6 national research reports, several of which were funded by the U.S. Government.

In A. D. Flurkey, E. J. Paulson, & K. S. Goodman, 2008Ken Goodman describes his evolution as a researcher and the risks he took as an academic to pioneer a new perspective on literacy learning: 

As I read the linguists, I found a body of knowledge and a methodology for the study of language that I could use in the study of reading. Further, I had a way of dealing with language as a dynamic, living whole rather than the isolated bits and pieces with which the experimentalist researchers had concerned themselves. …I saw positive research alternatives in treating reading as language and using linguistic, psycholinguistic, and sociolinguistic concepts and methodology. That put me in a scary and lonely position and one fraught with academic danger because my research proposals, publications, and presentations usually violated the “standards” of the researchers who judged them. But I also gained an advantage. Freed from the constraints of experimental, behavioristic research, I could ask broader and more basic questions that moved away from hypothecated cause-effect and correlational studies to a broad study of reading as it was happening. Instead of testing hypotheses, my research generated hypotheses. My research led me…to develop a coherent theory of the reading process. In studying real reading, my findings helped me to integrate my research methodology and better understand my findings. (p. 5-6)

Reading is language, and I needed a sophisticated language theory to deal with the complexity I was discovering in miscue patterns. I moved from the descriptive theory of C.C. Fries (1963) to the transformative theories of Chomsky (1972) to get to the processes miscues revealed and then on to the functional-systemic theory to put miscues into text and social context. I moved from studying reading as a linguist to the stance of a psycholinguist, and finally I had a transactional sociopsycholinguistic theory. Gradually, it became clear to me that reading is a transaction between text and reader. Miscues involve as much what readers bring to the text as the writer brought to it. Every reading is a complex literacy event, and each miscue reflects the complexities of the literacy event in which it occurs. Beyond language theory, understanding reading involves theories of culture, human societies, and the place of written language in both. (p. 18)

Whole Language and the Miscue Analysis Database

As for the approach to instruction referred to as whole language, the theoretical and research base includes the following theoretical orientations: constructivism (the general concept), psycholinguistic theory, schema theory, transactional/reader response theory, metacognitive theory, and systemic-functional linguistics. Professor Kenneth Goodman himself calls what became known as whole language the Transactional Sociopsycholinguistic Theory and Model of Reading and Writing (Goodman, 2003). In his book In Defense of Good Teaching (1998), Goodman states that the theoretical model of the reading process and the pedagogical principles that it supports are inextricably linked. Here is Professor Goodman’s description in his own words of his early research on language cueing systems and miscue analysis from the Preface of the annotated bibliography of research from 1898-1995 (Brown, Goodman & Marek, 1996). 

“My research on the reading process did not start out as an analysis of reading miscues. In 1962, I set out to see if I could study reading as a language process using what little knowledge of linguistics I had at the time. My design was simple, if naïve. I decided that I would have young readers read whole texts, comparable to those that they might encounter in school. Had I known more about linguistics I would have realized how complex natural language texts are and how difficult it would be to analyze a child’s reading of even a short text. …The procedure I used then, with a few refinements over the years, is similar to most current miscue research. … There really was an “Aha!” moment of discovery with miscue analysis…. Just as young children learning to speak show their growing control over the grammar of their language by their “errors” … readers use the same process for producing their errors as they do for their accurate reading. By comparing their observed oral responses to the text with the expected responses that would appear accurate to a listener I had a continuous window on their reading. I could see the process of reading at work! … The errors my first subjects were making showed they were using their knowledge of language to make sense of the printed text. (p. v)

“But if I continued to use the term “error” teachers and other researchers might continue to think of them as simply wrong rather than resulting from the language strengths of the learners. In deciding to call them miscues, I did not coin the word “miscue.” I had already begun to talk about the cue systems of language. If language provides cues that readers use, then why not call these unexpected responses “miscues”? A simple definition of a miscue also emerged: A miscue is a point in reading where the expected response (ER) and the observed response (OR) are not the same. ER ≠ OR. Out of this work grew two assumptions that still underlie miscue analysis: 1) Miscues are never random. 2) Unexpected responses result from the same process as expected responses. Our goal was a better understanding of the reading process, a more complete theory of reading , and a taxonomy of miscue analysis appropriate to all readers.” (p. vi-vii)

Three Purposes of Miscue Analysis

There are three primary purposes for applying the systematic observation tool referred to as miscue analysis: Assessment, instruction and empirical research data collection. The effectiveness of miscue analysis must be examined in terms of its usefulness and applicability for of these purposes separate and apart from each other. An examination of miscue analysis as an assessment tool must direct attention to the ways in which educators use systematic observations of students’ oral language performance to inform their instructional planning and implementation (Cole, 2006; & 2004: Miramontes, 1990). Examination of miscue analysis as an instructional strategy entails an examination of how teachers’ theoretical orientation toward literacy learning guides their selection and implementation of instructional approaches and specific instructional strategies and practices (Y. Goodman, 2015) . An examination of miscue analysis as a research methodology, in conjunction with eye movement research and neuroscience research entails the broadest and most comprehensive evaluation of miscue analysis data base regarding its contribution to teachers’ knowledge base and theoretical orientation toward literacy instruction for all learners. Miscue analysis and eye movement research comprise a valuable empirical base for effective practices in reading and writing instruction and assessment (Flurkey, et al., 2008; Strauss, et al., 2009). This empirical research base should not be denigrated or disregarded based on the fallacious arguments of journalists in agenda-driven articles and podcasts. The data base for miscue analysis is scientific because it is based on observation and analysis of actual oral reading behaviors of students reading authentic, connected text.

Any literacy researcher who claims that observing and recording real life performance of oral reading for purposes of analysis of the match between the written text and readers’ mismatch between what the text says and readers’ rendition of the text is not science, must provide his or her definition of science. This demand is reasonable so that the claims that miscue analysis data is “unscientific” can be examined and challenged.

This analysis begins with examination of how the theoretical framework of Whole Language supports the multiple pedagogical purposes of the miscue analysis data base.

Theoretical Framework of Whole Language

It is necessary to point out the difference between “a theory” and a theoretical orientation. I refer you again to Diane Tracey and Lesley Mandel Morrow (2006). Lenses on Reading: An introduction to theories and models. Very infrequently is there a “strategy” that a teacher employs in the classroom that is not associated with a particular or multiple theoretical orientation(s), to reading instruction. These logically and theoretically related clusters of instructional strategies and related learning activities are frequently referred to as approaches. For example, Tracey and Morrow (p. 69) state that Whole Language is “associated with the following instructional strategies.”

  1. Use of real, high-quality literature for literacy learning.
  2. Use of real, meaningful contexts for literacy activities.
  3. Child-centered instruction based on children’s interests.
  4. Heavy emphasis on student choice.
  5. Use of thematic instruction.
  6. Use of active, social learning experiences.
  7. Use of “teachable moments.”
  8. Use of a variety of grouping systems.
  9. Use of large blocks of time for integrated literacy activities.
  10. Use of alternative systems of assessment, such as portfolio assessments.
  11. Use of centers in the classroom.”

These instructional strategies are all included as recommended features of curriculum and instruction in the Center for Applied Linguistics Guiding Principles for Dual Language Instruction (Howard, et al., 2018) However, rather than asking how a theory or theories are validated, we need to understand how they are formulated. This is because a theory is our best guess (hypothesis) for describing and explaining how some complex phenomenon or process works.

Models of Reading

In the current debate regarding approaches to literacy instruction, one often sees the term the “three-cueing system model” of reading. I question the rationale behind considering the linguistic descriptors of subsystems of language that researchers such as Ken Goodman used to categorize miscues as a “model” of reading. The three categories of miscues that are used in miscue analysis (grapho-phonics, semantics and syntax) are components of written language. The underlying hypothesis behind this categorization is that miscues occur because a discrepancy or mismatch between a reader’s response to the language of a text and the actual written text may be explained by a lack of knowledge of how that particular subsystem of language works. In other words, a reader’s lack of grapho-phonetic knowledge, or semantic knowledge, or syntactic knowledge, is the possible cause of the miscue. The miscue is not caused by a “cueing system” but rather, the reader’s lack of knowledge of the cueing system is the proximal cause of the miscue. The use of “cueing systems” as categories of possible sources of miscues is important to keep in mind when we are assessing the logical coherence of claims against “three-cueing” as an instructional strategy, since categories are not strategies. It is also important to remember that the data on readers’ oral reading performance reveals both successful use of cueing systems (subsystems of language) and unsuccessful or incomplete use of cueing systems, all in pursuit of deriving meaning from language represented graphically in written text. 

There is no modern research that I know of that refutes or invalidates any aspect or component of Ken Goodman’s Transactional Sociopsycholinguistic Model of Reading (1970). Critics of his research find fault with his work for ideological reasons and other reasons that have nothing to do with the value of his comprehensive body of work in enhancing our understanding of how readers make meaning from text. Goodman states this foundational principle: “The reading process cannot be fractionated into sub-skills to be taught or sub-divided into code-breaking and comprehension without qualitatively changing it.” (Flurkey & Yu, 2003:252). Those who reject a database from systematic qualitative and quantitative analysis of reading behaviors are discounting a valuable and informative source of scientific knowledge about how language and literacy work without any scientific basis (Smith, 1999).

Ken Goodman has presented and explained his theories using a complex and comprehensive theoretical model that he calls a Transactional Sociopsycholinguistic Model of Reading. The model is based on the convergence of theories about reading from several disciplines, including linguistics, psycholinguistics, constructivism, cognitive theory, sociolinguistics, systemic functional grammar, neurocognitive theories and research on eye movements (EMMA). Goodman’s greatest contribution is the data base for his model from miscue analysis procedures from hundreds of readers. His data base is a comprehensive and extensive base of systematic observations of reading behaviors. Consequently, it is not experimental research. Goodman’s theoretical model (Flurkey & Yu, 2003:55-56) describes 11 steps that are involved in a reader’s process of making meaning from text. Here is a paraphrased summary of these steps, which Goodman explains are not necessarily sequential, but that comprise a description of the processes involved in making meaning of text.

  1. Reader scans the text, left to right and down the page, line by line.
  2. Reader fixes at a point to permit eye focus, which creates a central and a peripheral field.
  3. Reader begins the selection process, picking up graphic cues, guided by prior choices and reader’s knowledge, cognitive style and learned strategies.
  4. Reader forms a perceptual image based on cues that s/he sees and expects to see.
  5. Reader searches his/her memory for related syntactic, semantic and phonological cues. Semantic analysis leads to selection of more graphic cues and to reforming the perceptual image.
  6. Reader makes a prediction or tentative choice consistent with the graphic cues.
  7. If no prediction is possible, reader checks the perceptual input to gather more graphic cues.
  8. Reader makes a decodable choice from testing for semantic and grammatical acceptability within the context (language) of the text.
  9. If the tentative choice is not acceptable grammatically or syntactically, the reader regresses to locate the point of semantic or syntactic inconsistency, or the reader moves on in the text to find cues to reconcile the inconsistencies.
  10. If the choices are acceptable, decoding is extended and assimilated into prior meaning. Expectations are formed about input and meaning that lies ahead.
  11. The cycle continues, using both long-term and short-term memory.

See Flurkey and Yu, 2003, pages 56-57 for Figure 2-1 A Flow Chart of Goodman’s Model of Reading. This figure is a visual description of the 11 steps of the reading process.

The question for Goodman’s fellow researcher and practitioners is this: Do you believe that Goodman’s data base of direct observations of reading behaviors of real readers during oral reading of real, authentic text is sufficient to support the elements of his theoretical model to render a valid explanation of how readers read to make meaning from text? Claims in the media and critics/opponents of Goodman’s research and the pedagogy that it has influenced are made that Goodman’s “theory” of reading has been “disproven” (MacPhee, et al., 2021). For this disproven-theory claim to be true, every one of the theories on which Goodman based his theoretical model and all of the steps and elements of his model would have to have been disproven. Of course, the journalists who make this claim never cite an authoritative source for their proclamation because this claim is patently false. We researchers and educators challenge our fellow researchers, as well as Emily Hanford and other journalists, who claim that the entirety of the theoretical framework of Whole Language has been “disproven” or “debunked” to provide empirical data to refute each and every step of Goodman’s Model of Reading. Otherwise, their claims must be rejected.

Eye Movement Research Validates Goodman’s Model of Reading

Ken Goodman and his colleagues used eye movement research and miscue analysis, where they used computer programming to superimpose the eye movement data to identify the relationship between eye fixation and miscues of some 500 plus readers. Then they overlayed the eye movement/fixation data onto the records of the identification of their reading miscues according to the meaning subsystem(s) that was hypothesized to be the source of the non-recognition non-comprehension of the word: grapho-phonic processing (decoding) and/or semantic processing and/or syntactic processing. This large body of data on oral reading performance of hundreds of emergent readers is known as EMMA data, an acronym for Eye Movement Miscue Analysis overlaid observational data (Strauss, Goodman & Paulson, 2009). The miscues were coded according to their amount of, and proximity to, graphic representation of the word. Ebe (2008) used this data to analyze the oral reading processes of bilingual readers using English text and Spanish text to compare their processes.

The SoR movement is using this “3 cueing approach” as a straw man argument. They have basically invented an “approach” called Structured Literacy in an attempt to discredit Ken Goodman, Frank Smith and Whole Language, the entire miscue analysis research base. Efforts to discredit a research base and basically get it banned through legislation and policy initiatives is so very dangerous. There is a vast body of research in second language acquisition and second language reading research on what is termed “lexical inferencing” that addresses how L2 learners use context (in both listening and reading) to make meaning, build vocabulary, etc. The research points out that L1 readers who are reading in their L2 have an advantage because they have already developed the metacognitive strategies for meaning-making that they can apply to both word form and semantics in comprehending unfamiliar words.

Miscue Analysis as Assessment

In a research article that describes the utility of miscue analysis for addressing reading difficulties of first grade struggling readers, McGee et al., (2015) found changes in children’s self-monitoring reading strategies over time as a result of Reading Recovery instruction. These researchers state the following: “Goodman argued that researchers and teachers could use what he called miscues as a window on students’ processing of information in text during reading. Errors or miscues deemed to be semantically or syntactically acceptable or unacceptable and/or graphophonemically similar or dissimilar to the text provide evidence of students attending to meaning, syntax, or the visual/graphophonemic information in text.” (p. 264)

First of all, psycholinguistic “guessing” is hypothesis formulation and prediction that is an aspect of language, both oral and written. Language learners form hypotheses about how the rules of language work. For example, if a child says “foots” instead of feet or “swimmed” instead of swam, we are not surprised that he “guessed” that adding s to form a plural or adding “ed” to form a past tense before he learned that these are exceptions to the rule. But his “guess” shows his teacher his understanding of the rules of the language of the text. His grammatical miscue demonstrate that he has applied them to the best of his knowledge at the stage he is at in learning language. “Guessing” or hypothesis formulation and “miscues” are a positive and useful part of the process of learning, in language learning and in reading and writing.

The empirical research base from miscue analysis provides literacy educators with pedagogical knowledge about how students’ oral reading errors and the insights that they might provide for understanding students’ strategic reading at point of difficulty. This is essential knowledge for assessment to diagnose the strengths and challenges of individual readers who encounter difficulties in decoding and comprehending text. Skilled literacy teachers use information from miscue analysis to determine students’ current use, or non-use of meaning-making cognitive strategies in order to design appropriate instruction. McGee et al. make this observation:

“Even when students already have at their disposal a more sophisticated strategy, they continue to use a variety of strategies, including less sophisticated strategies, in flexible ways until gradually, through trial and error, they discover which strategies are more accurate and efficient in solving particular problems. … Typical action chains provide evidence that students are using five strategic actions drawing on different sources of information: monitoring graphic and contextual information, self- correcting using graphic or contextual information or a combination of both, making multiple attempts that often included use of both contextual and graphic information, coordinating and organizing use of graphic and contextual in formation, and rereading to recall or elaborate on contextual information.” (p. 277)

Whole language does not fragment reading into isolated and disconnected subskills. Rather, WL emphasizes the holistic interrelationships between subsystems of language (cueing systems) that are utilized by readers themselves to construct meaning from text. We must not allow uninformed, ideologically biased journalists to make sweeping false claims and fallacious arguments with the intent of scapegoating respected researchers who have informed and enhanced literacy instruction for many of our most vulnerable students and all literacy learners through their empirical scientific research.

Kabuto (2016) offers a perspective on the use of miscue analysis as an assessment tool with multilingual learners:

Miscue analysis is a diagnostic instrument that provides both quantitative and qualitative data on readers’ oral reading performances and retellings. The in-depth procedure allows for the extensive investigation of individual oral reading miscues in conjunction with other miscues at the sentence and text levels. …Miscue analysis can be a descriptive evaluative tool that does not privilege reading in English over reading in other languages. …These findings extend the current research on miscue analysis that calls for an awareness of readers’ reading patterns and proficiencies in both languages, rather than privileging one over another. …The use of miscue analysis was a culturally relevant assessment that provided a multidimensional perspective on the ways in which these readers constructed meaning. Very few reading assessments are able to move among the languages of the text, the languages of the readers, and the languages of the social context in which the assessment is embedded. Miscue analysis is an evaluative instrument capable of doing so. (p. 38-39)

Laws and policies that ban the use of “three-cueing” as an instructional strategy, may result in the prohibition of miscue analysis as a diagnostic instruction and descriptive evaluative tool for investigating multilingual learners’ literacy acquisition in languages other than English, as well as their developing competency in reading English as a second language. Therefore, these policies violate the civil rights of these students is equitable access to literacy instruction.

Miscue Analysis as Instruction

The media’s attack on “cueing” and “three-cueing” as an instructional strategy is misinformed and misguided, despite the fact that reporters may be parroting the opinions about linguistic cueing systems of some literacy researchers. Reluctantly, I will use the term “cueing” to explain why I, as an expert in Spanish-English biliteracy instruction, object to the anti-cueing campaign of journalists such as Emily Hanford of American Public Radio and Sarah Shwartz of Education Week. What is termed “cueing” by the anti-cueing faction is identified by many different names among teachers of multilingual learners:  scaffolding, monitoring, corrective feedback, word problem-solving, transactional teaching, metalinguistic awareness instruction and lexical inferencing. Lexical inferencing is recognized as an important metalinguistic skill among multilingual literacy researchers (Ke, et al., 2023).

What is called “cueing” is transactional instruction because it is the process where a teacher listens to a student reading aloud a selected cohesive authentic text in a one-on-one setting. The teacher is analyzing the student’s oral language performance on the spot to identify the possible source of a reading miscue, which is a response to the text that does not match the exact language of the text. The theory behind this process is described above in Goodman’s theoretical model, but briefly stated, the teacher assumes that s/he is observing the student’s meaning-making (comprehension) strategies as applied to a specific text. The “context” of the text is the language of the text, which is a message from the author to the reader, which the reader must interpret.

Miscue analysis enables the teacher to observe points of difficulty that the reader encounters and to immediately and explicitly point out to the student where there is information provided in the language of the text to overcome the difficulty (Ellis, et al., 2006; MacGee, et al., 2015). Miscue analysis researchers have identified three subsystems of language that are utilized in an orchestrated meaning-making process, that contribute to the reader’s ability to comprehend the author’s message. These subsystems are graphophonics, semantics and syntax. The graphophonics subsystem is the way in which the phonology of the language of the text is represented graphically in print. The semantic subsystem is meaning at the lexical or word level primarily, which logically requires understanding of the phrase and sentence level cues to the meaning of a word. For example, most nouns can be understood at the word level, but verbs are understood at the phrase or sentence levels.

When in the interactive setting with a student one-on-one, the teacher has the option of pointing out features of print that the reader may not have fixated on sufficiently to derive meaning from a word, phrase or sentence. The teacher can, at the point of difficulty, engage in explicit print instruction (MacGee, et al., 2015; Kabuto, McGinty, et al., 2012). If the reader can sound out a word but does not have the meaning (semantics) in his/her mental lexicon, the teacher can make a reasoned judgment that the source of the miscue remedied by having the reader look for clues to the word’s meaning in the phrase or sentence. This can be accomplished either by rereading or reading ahead to find language that may implicitly define a content word. In her research using eye movement tracking with bilingual readers, Ebe (2008) found that bilingual readers fixate more often on content words, which suggests that these words carry more semantic and syntactic information. This can be through analysis of a word’s morphosyntactic features, as for example in determining the embedded subject and tense of Spanish verb conjugations (Jaichenco & Wilson, 2013).

State Mandates Against “Cueing”

State legislators have pass laws prohibiting teachers from using “cueing” or “three-cueing” for reading and writing instruction or assessment, as has occurred in three states: Arkansas, Indiana and Louisiana, as Sarah Schwartz of Education Week Reports (May 2023). For example, the Arkansas Right to Read Act has this “disqualifier” against K-2 literacy curriculum programs that apply for state approval (Arkansas Division of Elementary and Secondary Education, 2019-2020):

If the theoretical basis of any submitted program utilizes the Three Cueing Systems Model of Reading or Visual Memory as the primary basis for teaching word recognition, it shall be disqualified because cognitive science refutes use in foundational reading. (p. 4)

The fact is that there is no such thing as “the Three Cueing Systems Model of Reading” in the research literature. This is because the three “cueing systems” are simply only three subsystems of language out of six or seven commonly identified subsystems that psycholinguistic researchers have identified as the sources of miscues in oral reading performance. The miscue analysis research has led to systematic methods for categorizing reading miscues (errors) according to these language subsystems, of which there are five or six (depending upon whether or not a linguist/researcher considers orthography to be a language subsystem). Consequently, we psycholinguists believe that all language subsystems, that make up the whole or entirety of language and how meaning is made from language. None of us proposes that these three cueing systems alone comprise a theoretical model of reading. In his complete body of research, Ken Goodman (1970; 2003) proposes a theoretical model of reading and writing, which he calls a Transactional Sociopsycholinguistic Model of Reading to describe the role of “cues” within the process of reading. Miscue analysis is a data collection and analysis methodology, not a theory or a theoretical model in and of itself. No respected researcher calls his/her data collection methodology a “theory” or a “model” in a research study. Nor do we call descriptive statistics based on categories generated from data collection “models” either. A theoretical model of the reading process is based on data but is not the data itself. A Venn diagram of a database is not a theoretical model (Adams, 1998). Furthermore, raw data cannot be directly translated ia method or approach to literacy instruction. In fact, the miscue analysis database is “instruction neutral” since readers make miscues in their oral reading performance regardless of how they were taught to read. Moreover, researchers cannot identify an instructional approach or strategy as the origin of a reader’s miscues.  

Consequently, we researchers must construct hypotheses about the Arkansas legislature’s “disqualification” of commercial literacy programs: They either invented the idea of a “Three Cueing Systems Model of Reading” themselves, or someone convinced them that such a thing exists in the research literature. The most benign, but nonetheless problematic, hypothesis is that they were convinced to rename Ken Goodman’s actual model of reading, with all of its nuance and complexity, for the sake of argument. It is unclear on what scientific research base how the state of Arkansas has concluded that “cognitive science refutes use of a “model of reading” in foundational reading. Since the so-named model of reading does not exist, it is highly doubtful that “cognitive science refutes” its use in foundational reading. Since the law does not cite a direct source of this invented theoretical model, I speculate that some of the authors of the research articles they cite in their short bibliography talked them into the idea of a “model of reading” as a pretext for excluding certain publishers and authors from their list of approved literacy programs. This is a highly questionable and highly politicized practice. Woulfin and Gabriel (2022: 329) point out the following:

First, state legislators have passed bills stipulating the use of curriculum materials and appropriating funding for adopting new reading programs (Schwartz, 2022). Recent state-level initiatives sought to identify and promote SOR-aligned curricular materials, following reports that posited curricular materials in schools could be to blame for the lack of student achievement, especially in areas like reading where the nature of programs and associated materials varies so widely. (p. 329)


An obvious problem with attempts to mandate that literacy curriculum be “Science of Reading aligned” is that curriculum in most states is required to be curriculum standards aligned.” To align a curriculum to something as broadly and variably defined and interpreted as SoR research is a huge challenge, most especially when curriculum standards themselves may or may not guide implementation of a particular body of research. Alignment with research is even more challenging when a state legislature bans or prohibits literacy curriculum and literacy instruction based on claims that they are “refuted” by certain researchers, while these claims are unsupported by empirical, scientific research. The war on “three-cueing” falls into this category.

These state governments claim to be mandating the Science of Reading while banning the use of instructional approaches and strategies that have a strong and credible theoretical framework and evidence base in science. It is counter-productive for persons or entities outside of the professional context of a teacher to make pronouncements about how a teacher should interact with his/her students during oral reading based on a misinterpretation of science of reading research. This is an issue of teacher agency and teacher professional judgment within the purview of teachers alone. The use or non-use of “cueing” that is as decoding and comprehension instruction should not be second-guessed by educators and non-educators who have no knowledge of teachers’ theoretical orientation toward reading instruction and his/her knowledge base for effective teaching approaches and strategies for his/her students (Beatty & Care, 2009; Buettner, 2002; Ellis, et al., 2006; McGee, et al., 2015; Noguerón-Liu, 2020). Such attacks on teacher agency and professionalism must be opposed by the community of educators and researchers who advocate for equity for multilingual learners.

Why should teachers seek to align their classroom practices with research that contradicts what they know to be effective for the students in their classrooms? Why should teachers trust state legislators who have no knowledge or expertise in instruction of their particular demographics of students for whom they are responsible? My answer to these questions is that teachers should not seek alignment with research that lacks empirical adequacy, logical coherence and consequential validity for advancing their students’ academic achievement (Cummins, 2021). Teachers of multilingual learners must trust the knowledge base provided by scientific research that has population validity for their students and trust their own professional judgements about claims that contradict their own theoretical orientation toward literacy instruction.

Click here for access to a presentation on the transdisciplinary knowledge base for literacy curriculum and instruction for multilingual learners: La lectoescritura en español:  Reaffirming the International Scientific Research Base, La Cosecha Dual Language Conference, November 9, 2023.


Adams, M. J. (1998). The three-cueing system. In J. Osborn & F. Lehr (Eds.), Literacy for all: Issues in teaching and learning (pp. 73-99). The Guilford Press.

Beatty, L., & Care, E. (2009). Learning from their miscues: Differences across reading ability and text difficulty. Australian Journal of Language and Literacy, 32(3), 226-244.

Bodmer, F. (1944). The loom of language. W.W. Norton & Company, Inc.

Buettner, E. G. (2002). Sentence by sentence self-monitoring. The Reading Teacher, 56(1), 34-44.

Brown, J., Goodman, K. S., & Marek, A. M. (1996). Studies in miscue analysis: An annotated bibliography. International Reading Association.

Briceño, A., & Klein, A. F. (2019). A second lens on formative reading assessment with multilingual students. The Reading Teacher, 72(5), 611-621. https://doi.org/10.1002/trtr.1774

Cole, A. D. (2004). When reading begins: The teacher’s role in decoding, comprehension, and fluency. Heinemann.

Cole, A. D. (2006). Scaffolding beginning readers: Micro and macro cues teachers use during student oral reading. The Reading Teacher, 15(5), 450-459.

Davenport, R. M. (2002). Miscues, not mistakes: Reading assessment in the classroom. Heinemann.

Ebe, A. (2008). What eye movement and miscue analysis reveals about the reading process of young bilinguals. In A. D. Flurkey, E. J. Paulson, & K. S. Goodman (Eds.), Scientific realism in studies of reading (pp. 131-149). Routledge.

Ebe, A. (2010). Culturally relevant texts and reading assessment for English Language Learners. Reading Horizons: A Journal of Literacy and Language Arts, 50(3), 193-310.

Ellis, R., Loewen, S., & Erlam, R. (2006). Implicit and explicit corrective feedback and the acquisition of L2 grammar. Studies in Second Language Acquisition, 28(2), 339-369.

Flurkey, A. D., & Paulson, E. J., Goodman, K. S. (Eds.). (2008). Scientific realism in studies of reading. Routledge.

Flurkey, A. D., & Xu, J. (Eds.). (2003). On the revolution of reading. The selected writings of Kenneth S. Goodman. Heinemann.

Goodman, K. (2003). Reading, writing and written texts: A transactional sociopsycholinguistic view. In A. D. Flurkey & J. Xu (Eds.), On the revolution of reading: Selected writings of Kenneth S. Goodman (pp. 3-45). Heinemann.

Goodman, K. (1992). I didn’t found whole language. The Reading Teacher, 46(3), 188-199.

Goodman, K. S. (1967). Reading: A psycholinguistic guessing game. Literacy Research and Instruction, 6(4), 126-135.

Goodman, K., Goodman, Y. M., & Paulson, E. J. (2009). Beyond word recognition: How retrospective and future perspectives on miscue analysis can inform our teaching. In J. V. Hoffman & Y. M. Goodman (Eds.), Changing literacies for changing times: An historical perspective on the future of reading research, public policy, and classroom practices (pp. 146-161). Taylor & Francis Group.

Goodman, K., Fries, P. H., & Strauss, S. L. (2016). Reading‒The grand illusion: How and why people make sense of print. Routledge.

Goodman, Y. M. (2015). Miscue analysis: A transformative tool for researchers, teachers, and readers. Literacy Research: Theory, Method, and Practice, 64, 92-111.

Howard, E. R., Lindholm-Leary, K., Rogers, D., Olague, N., Medina, J., Kennedy, B., Sugarman, J., & Christian, D. (2018). Guiding Principles for Dual Language Education (3rd ed.). Center for Applied Linguistics.

Indiana Department of Education (2023). Education, Indiana’s Priorities for Early Literacy. Retrieved from https://media.doe.in.gov/news/indianas-priorities-for-early-literacy-final.pdf

Jaichenco, V., & Wilson, M. (2013). El rol de la morfología en el proceso de aprendizaje de la lectura en español [The role of morphology in the process of learning to read in Spanish]. Interdisciplinaria, 30(1), 85-99.

Kabuto, B. (2016). A socio-psycholinguistic perspective on biliteracy: The use of miscue analysis as a culturally relevant assessment tool. Reading Horizons, 56(1).

Ke, S. E., Zhang, D., & Koda, K. (2023). Metalinguistic awareness in second language reading development. Cambridge University Press.

MacPhee, D., Handsfield, L. J., & Paugh, P. (2021). Conflict or conversation? Media portrayals of the science of reading. Reading Research Quarterly, 56(S1), S145-S155.

McGee, L. M., Kim, H., Nelson, K. S., & Fried, M. D. (2015). Change over time in first graders’ strategic use of information at point of difficulty in reading. Reading Research Quarterly, 50(3), 263-291.

McGinty, A. S., Justice, L. M., Piasta, S. B., Kaderavek, J., & Fan, X. (2012). Does context matter? Explicit print instruction during reading varies in its influence by child and classroom factors. Early Childhood Research Quarterly, 27, 77-89.

McKenna, M. C., & Picard, M. C. (2006). Revisiting the role of miscue analysis in effective teaching. The Reading Teacher, 60(4), 378-380.

Miramontes, O. B. (1990). A comparative study of English oral reading skills in differently schooled groups of Hispanic students. Journal of Reading Behavior, 22(4), 373-394.

Noguerón-Liu, S. (2020). Expanding the knowledge base in literacy instruction and assessment: Biliteracy and translanguaging perspectives from families, communities, and classrooms. Reading Research Quarterly, 55(S1), S307-S318.

Smith, F. (1999). Unspeakable acts, unnatural practices: Flaws and fallacies in “scientific” reading instruction. Heinemann.

Stouffer, J. (2021). Seeking middle ground: Analyzing running records from the top and bottom. The Reading Teacher, 74(6), 769/784.

Strauss, S. L. (2005). The linguistics, neurology, and politics of phonics: Silent “E” speaks out. Lawrence Erlbaum.

Strauss, S. L. (2010). Neuroscience and dyslexia. In A. A. McGill-Franzen, Richard L. (Ed.), Handbook of reading disability research (pp. 79-90). Routledge Taylor & Francis Group.

Strauss, S. L., & Altweger, B. (2007). The logographic nature of English alphabetics and the fallacy of direct intensive phonics instruction. Journal of Early Childhood Literacy, 7(3), 299-319.

Strauss, S. L., Goodman, K. S., & Paulson, E. J. (2009). Brain research and reading: How emerging concepts in neuroscience support a meaning construction view of the reading process. Educational Research and Review, 4(2), 021-033.

Tracey, D. H., & Morrow, L. M. (2006). Lenses on reading: An introduction to theories and models. The Guilford Press.

Woulfin, S. l., & Gabriel, R. (2022). Big waves on the rocky shore: A discussion of reading policy, infrastructure, and implementation in the Era of Science of Reading. The Reading Teacher, 76(3), 326-332.