Instructional Models in Reading

Steven A. Stahl, David A. Hayes

• Publisher: Routledge
• Year: 2013
• ISBN:
  9781136481901
• Edition: 1
• Topic: Education
• Subtopic: Education General

This book started with a simple idea -- examine models of reading instruction that have emerged during the past 20 years. These models span a wide range of instruction representing a continuum from highly structured, task analytic instruction to child-centered and holistic instruction. Each model has its own epistemology or views on how "reading" and "instruction" are to be defined. The different epistemologies indicate different principles of instruction which, in turn, indicate different practices in the classroom. Each model is also supported by a different research base. In this volume, leading proponents of these different models discuss their ideas about reading instruction thereby encouraging readers to make their own comparisons and contrasts. The chapter authors seem to adopt the editors' eclectic approach--to some greater or lesser extent--incorporating aspects of other models into their instruction as they see other goals. Thus, models of reading instruction are complex. Complicating matters further is the fact that teachers hold their own models of reading, which may or may not be congruent with those discussed here. Although academically developed models influence college preservice and in-service instruction, teachers' own models of reading filter the information that they take from what they learn from these perspectives. By carefully examining these variables, this book makes a firm contribution toward disciplined inquiry into what it means to teach reading.

CHAPTER ONE

Instructional Models in Reading: An Introduction

Steven A. Stahl
The University of Georgia, Athens

In the 1950s, there was basically one approach to teaching reading. Basal readers dominated instruction. According to Austin and Morrison’s (1963) survey, 95% of all teachers used a basal reader to teach. Basal readers all used Betts’ (1946) Directed Reading Activity: background building and vocabulary development prior to reading, guided silent reading followed by questions and oral rereading, and a series of postquestions. This was how reading was to be taught, and the basic model was not questioned. “Reading” was thought to consist of accurate word recognition and the ability to answer questions about what was read. “Instruction” was what was done to facilitate word recognition and question answering. According to Durkin’s (1974) observations, teachers’ manuals in the basal reading program were followed closely, if not slavishly. There were dissenters to this point of view (see Y. Goodman, 1989), but they were far outnumbered by those who used basals.

In the 1990s, there are many competing models of reading instruction. The whole-language movement (K. Goodman, 1986) has had a pervasive influence in elementary classes. Direct instruction has a strong influence in special education, but is reemerging in regular education as well. Sociocultural models have a strong influence, especially in schools with multicultural populations. Response-oriented reading approaches are seen more often in secondary English classes, and emergent-literacy models are used to guide kindergartners’ development.

Our models of reading instruction guide our research and serve as a filter for us to view practice. This chapter reviews several widely held models, examining both how they view reading instruction and how they define reading. Then it discusses how these different models fit into a specific model of reading development (Stahl, 1992a). Finally, it suggests that effective instruction involves a melding of the different models. Because reading instruction involves multiple goals, different approaches are best to reach different goals.

MODELS OF READING INSTRUCTION

Garcia and Pearson (1991) divided competing approaches to reading instruction into four general approaches: direct instruction, explicit explanation, cognitive apprenticeship, and whole language. These are distinctly different from conventional, basal instruction. Taken as a group, these four approaches represent a continuum from highly teacher-directed, task-analytic approaches to more student-directed, holistic approaches—from the reliance on contrived materials designed for specific instructional purposes to the use of “natural” materials written primarily for an audience of young readers. They should not be thought of as discrete approaches, but as points on that continuum.

Direct Instruction

Although direct-instruction approaches can trace their lineage back to behavioral analyses of decoding tasks and process–product analyses of teaching (see Rosenshine & Stevens, 1984), more recent work in the direct-instruction paradigm has been directed toward teaching more complex skills. Such approaches begin with a task analysis of the target behavior, which is used to design the instruction. Students are taught each component, both singly and in combination with the other components. Teachers model the desired behavior, provide ample practice and feedback at each step, and assess whether reteaching is necessary. Unlike earlier versions of direct instruction, current use of the model also includes metacognitive explanation of the importance of the strategy; how, when, and where it is to be used; and when its use is inappropriate.

The direct-instruction model makes clear assumptions about learning, based on behavioral roots. Direct-instruction proponents assume that reading can be decomposed into identifiable subskills that, when taught directly, will improve children’s reading ability. These subskills are to be taught using a specified set of teacher behaviors. A direct-instruction phonics lesson might begin with the introduction of a letter sound, discrimination of that sound and previously taught sounds, blending that sound into words, and so on. At its most extreme, direct instruction tends to view teachers as interchangeable dispensers of instruction. By attempting to specify the exact behaviors (e.g., hand signals, pointing) that teachers perform, as well as the exact words that they say, direct-instruction proponents try to make the curriculum “teacher proof.” Many teachers resist this programming, either subverting the program by their actions or abandoning the program as soon as they have a chance (Shannon, 1987).

Direct-instruction approaches: (a) break language down into components that are taught in isolation, not in a meaningful context; (b) are highly teacher directed, allowing students little choice in what is to be learned and how it is to be learned; and (c) view the acquisition of literacy as highly “unnatural,” requiring systematic instruction, rather than absorption. This need not be so. Research shows that direct-instruction programs are most effective when combined with wide reading in tradebooks (Meyer, 1983). Indeed, direct instruction in one aspect of reading is not incompatible with allowing self-selection in other aspects of a reading curriculum (see Cunningham, 1991).

Explicit Explanation

Explicit explanation is similar to direct instruction in that they both involve explicit definitions of cognitive strategies used in reading, including discussions of their usefulness, as well as modeling and practice. In explicit-explanation lessons, there is greater emphasis on practicing the strategy in the context of reading text, although the strategy may be introduced using specially constructed materials. In explicit explanation, there is also greater concern with gradually releasing the responsibility for the execution of a strategy. In direct instruction, it is assumed that, as readers become more proficient at using a strategy, they will automatically use that strategy while reading. In explicit explanation, there is greater emphasis on leading students to make that transfer. In the beginning of the explicit-explanation instruction, the responsibility for using a strategy lies largely with the teacher; by the end, the student executes the strategy independently.

An example of an explicit-explanation program is Raphael’s work in teaching children to answer questions strategically using a question-answer relationship taxonomy (e.g., Raphael & Wonnacott, 1985). In these lessons, students are taught about the relationship between questions and their answers. After an initial exposition of the taxonomy, students are shown a passage and given questions with their answers. As a class, the students discuss the relationship between the question and the answer. In succeeding lessons, students answer questions given the relationship, provide the answer and the relationship, and generate their own questions, providing both answers and relationships. Thus, the students gradually assert “ownership” of the strategy.

Explicit explanation has been used to teach a variety of reading strategies—from using context to identify unknown words (Duffy et al., 1986) to making inferences during reading (Hansen & Pearson, 1983). It also has been used as a model for an entire school’s curriculum, involving the explicit integration of strategies in all subject matter learning from primary- to middle-school grades (Gaskins & Elliot, 1991).

Because it is so new, it is difficult to draw conclusions about the effectiveness of explicit explanation. There are some indications that, when teaching skills such as using context to identify words (e.g., skills that might have been mastered through children’s oral language development), explicit explanation makes children better able to discuss what they are doing, but does not directly improve their reading achievement (Duffy et al., 1986). Duffy et al. asserted that this greater awareness will eventually translate into higher achievement, but did not present evidence that it does. Because skills are best performed automatically, without conscious application, it is not clear that explicit explanation would be useful. For strategies such as summarization, which is applied deliberately, such explicit explanation may be highly effective.

Explicit-explanation instruction also presents a clear contrast to whole language. Although both approaches stress the importance of reading natural text for authentic purposes, explicit instruction involves isolation and instruction of subskills, and complete teacher control of the purposes of the lesson, at least in the beginning. These would be anathema to a whole-language advocate.

Cognitive Apprenticeship

A third alternative to whole language is a cognitive-apprenticeship approach. Cognitive-apprenticeship approaches attempt to set up a master–apprentice relationship between student and teacher. The teacher’s role is to scaffold the learning, withdrawing support as students are able to proceed on their own. Just as an apprentice first watches the master as the master does a skilled craft, so does the student initially observe the teacher as the teacher models the processes of comprehension. Gradually, the teacher gives more and more responsibility to the student, until it is the teacher who watches the student perform comprehension tasks.

Those who advocate the use of cognitive apprenticeships view reading as the orchestration of complex processes. Teaching these processes skill by skill creates a distorted view of reading (Brown, Collins, & Duguid, 1989). Citing Vygotsky, Brown et al. argued that cognition is socially situated—that all cognitive acts take place within a sociocultural context, and that the larger context can either support or impede cognitive activities, such as those involved in reading. This new view of cognition has been called situated cognition, or socially constructed knowledge. From this view, researchers suggest that an apprenticeship model is more appropriate for teaching than the transmittal model of direct instruction. Through the process of interacting with the knowledgeable other, students learn how an expert orchestrates the processes involved in comprehension (Garcia & Pearson, 1991).

In cognitive apprenticeships, teacher and students work together to comprehend increasingly complex text. In contrast to direct instruction, instruction is performed using authentic texts for authentic purposes. Teacher and students might read a text together, with the teacher providing as much support as necessary for the students to successfully work through the text.

The cognitive-apprenticeship model emphasizes constructing the meaning of a text through social interactions, and usually involves a reorganization of the basic organization of the class. Instead of a teacher-dominated class structure, cognitive-apprenticeship models usually involve small groups working together. Cooperative learning (Stevens, Madden, Slavin, & Farnish, 1987), reciprocal teaching (Palincsar & Brown, 1984), collaborative problem solving (Palincsar, David, Winn, & Stevens, 1991), and conversational discussion groups (O’Flahaven, 1990) all use group dynamics to scaffold or support children’s learning. For example, in reciprocal teaching, small groups of students work together with the teacher to read a text. Each person in the group takes a turn being the “teacher.” As the “teacher,” a child uses four teaching behaviors: questioning, summarizing, clarifying, and predicting. Unlike many of the training approaches discussed so far, reciprocal teaching usually takes place over a period of weeks or months, not a single class period. Initiation of reciprocal teaching might begin with direct instruction of the four teaching operations, followed by direct teacher modeling. In the early stages of reciprocal teaching, the teacher takes a dominant role, modeling the operations during his or her turn and prompting students during theirs. Such prompts might include, What is the main idea of this paragraph? or Could you make it into a question? Over time, the prompts become more general. Also, over time, students take over the groups, acting more and more like expert teachers.

By having students act as a teacher, reciprocal teaching requires students to externalize the operations of comprehension. Through this externalization, the teacher and other students are able to scaffold the development of each student’s comprehension. The learning that occurs in conversational discussion groups, where groups of students respond to stories (O’Flahaven, 1989), or cooperative learning groups, which have been used for a great many learning situations (e.g., Stevens, Madden, Slavin, & Farnish, 1987), involves similar scaffolding through group dynamic processes.

As with explicit explanation, cognitive apprenticeships are too recent to have been fully evaluated. Rosenshine and Meister (1991) performed a meta-analysis of studies evaluating reciprocal teaching, and found that reciprocal teaching was most effective when combined with direct teaching of cognitive strategies. Brown and Palincsar (1989; cited in Prawat, 1991) compared reciprocal teaching with a peer-collaboration approach (which appears to be analogous to whole language, but directed specifically toward science learning), and found that a peer-collaborative approach, combined with text materials rewritten for conceptual coherence, produced the highest quality of student discourse about science and evidence of higher level thinking in writing samples. This line of research was continued by Brown and her colleagues (Brown, 1992). Brown and Palincsar (1989; cited in Prawat, 1991) suggested that the original reciprocal-teaching approach was successful in getting poor readers to focus on comprehension, but that its structure may not be necessary for accomplished readers to develop deeper understanding of content materials.

Whole Language

As Gunderson (chap. 10, this volume) points out, whole language is difficult to define. In recent years, at least three professional journal articles (Altwerger, Edelsky, & Flores, 1987; Bergeron, 1990; Watson, 1989) and three books (Edelsky, Altwerger, & Flores, 1991; K. Goodman, 1986; Newman, 1985) have been essentially devoted to the topic of defining whole language. However, the definitions have been rather hazy. Bergeron (1990) examined articles that used the term to examine commonalities among definitions. She found that “whole language was defined differently in each of the 64 articles reviewed,” and that “little consistency was also found in the descriptions of those attributes thought to be the focus of whole language” (p. 312).

In the case of whole language, this lack of an objective definition seems deliberate. Even adherents refuse to define whole language, arguing that to do so would disempower practitioners. For example, Watson (1989) cited several different definitions of whole language, and then said: “These definitions may lack sameness, but they never go outside the boundaries of an acceptable definition of some dimension of whole language. The definitions are diverse because the personal and professional histories of the authors are different. This variety frees those who have studied and practiced whole language to generate their own definitions, then to revise their definitions again and again” (p. 132). Edelsky (1990), however, argued that the freedom to create one’s own definition is limited only to those who accept whole-language principles. Edelsky took McKenna, Robinson, and Miller (1990) to task for not understanding what whole language was, and suggested that they could not understand it unless they believed in it. Further, Edelsky argued that McKenna et al. could not propose a research agenda to evaluate whole language because they were outsiders.

Even if one cannot precisely define whole language, there are beliefs that are shared by most whole-language practitioners. Among these are that language (oral and written) is used for authentic purposes, including communication, information, and so on (K. Goodman & Y. Goodman, 1979), and that children will learn language best if it is learned for authentic purposes. In the classroom, this involves: using authentic reading and writing tasks using whole texts, not looking at parts of language (e.g., sound–symbol correspondences) for their own sake, and not using artificial tasks such as work sheets or specially adapted stories found in basal reading programs. There is also a belief in child-cent...

Table of contents

• Cover
• Halftitle
• Title
• Copyright
• Contents
• Preface
• 1 Instructional Models in Reading: An Introduction
• 2 Models of Professional Practice in Teacher Thinking
• 3 Direct-Instruction Reading
• 4 Cognitive Strategy Instruction in Reading
• 5 Reading, Writing, and Language Arts in Success for All
• 6 Creating Readers Who Read for Meaning and Love to Read: The Benchmark School Reading Program
• 7 Teaching From Theory: Decision Making in Reading Recovery
• 8 A Sociocultural Model of Reading Instruction: The Kamehameha Elementary Education Program
• 9 Fostering Literate Communities in School: A Case for Sociocultural Approaches to Reading Instruction
• 10 Whole-Language Approaches to Reading and Writing
• 11 An Emergent-Literacy Perspective on Reading Instruction in Kindergarten
• 12 Reading Instruction in an Integrated Language Perspective: Collaborative Interaction in Classroom Curriculum Genres
• 13 Response-Based Reading Instruction in the Elementary Grades
• 14 Beyond Individual Response: Toward a Dialogical Approach to Literature Instruction
• 15 Powerful Models or Powerful Teachers? An Argument for Teacher-as-Entrepreneur
• Author Index
• Subject Index