Page 172

6
Instructional Strategies for Kindergarten and the Primary Grades

The mission of public schooling is to offer every child full and equal educational opportunity regardless of the background, education, and income of the child's parents. A most fundamental and important issue facing schools is how to teach reading and writing, particularly in the early grades. Children who struggle in vain with reading in the first grade soon decide that they neither like nor want to read (Juel, 1988). Even if they do not fall into any of the recognized at-risk categories, these children soon are at risk of poor literacy outcomes.

The major prevention strategy for them is excellent instruction. The intervention considered in this chapter is therefore schooling itself; we outline the major literacy goals for kindergarten and the first three primary grades, examining evidence concerning effective methods to attain those goals.

INTRODUCTION

Previous Reviews

The issue of what constitutes optimal reading instruction has generated discussion and debate and the investment of research ef



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 172
Page 172 6 Instructional Strategies for Kindergarten and the Primary Grades The mission of public schooling is to offer every child full and equal educational opportunity regardless of the background, education, and income of the child's parents. A most fundamental and important issue facing schools is how to teach reading and writing, particularly in the early grades. Children who struggle in vain with reading in the first grade soon decide that they neither like nor want to read (Juel, 1988). Even if they do not fall into any of the recognized at-risk categories, these children soon are at risk of poor literacy outcomes. The major prevention strategy for them is excellent instruction. The intervention considered in this chapter is therefore schooling itself; we outline the major literacy goals for kindergarten and the first three primary grades, examining evidence concerning effective methods to attain those goals. INTRODUCTION Previous Reviews The issue of what constitutes optimal reading instruction has generated discussion and debate and the investment of research ef

OCR for page 172
Page 173 fort over many decades. This report builds on earlier work, yet our scope limits us only to briefly summarizing earlier efforts. We acknowledge the degree to which our report benefits from this work and draw the reader's attention to the long history of thinking about these topics. First-Grade Studies Between 1964 and 1967, the U.S. Office of Education conducted the Cooperative Research Program in First Grade Reading Instruction; this was an early and ambitious effort at large-scale evaluation of instructional approaches. The program, coordinated by Guy L. Bond and Robert Dykstra, included classroom approaches that emphasized systematic phonics instruction, meaningful connected reading, and writing; its results surpassed those of mainstream basal programs. Conceived and conducted prior to much of the psycholinguistic research on the subprocesses and factors involved in reading acquisition, these studies were not submitted to the levels of analysis characteristic of later efforts. Nonetheless, they pointed to a consistent advantage for code-emphasis approaches while indicating that one single simple method was not superior for all children and all teachers. The Great Debate Among efforts to identify factors associated with more and less effective beginning reading practices, Jeanne S. Chall's (1967) work, Learning to Read: The Great Debate, remains a classic. While producing this work, Chall visited classrooms, interviewed experts, and analyzed programs. Yet it was her review and analysis of the then-available research on instructional practices that yielded the most stunning conclusions. Chall found substantial and consistent advantages for programs that included systematic phonics, as measured by outcomes on word recognition, spelling, vocabulary, and reading comprehension at least through the third grade. Moreover, the advantage of systematic phonics was just as great and perhaps greater for children from lower socioeconomic backgrounds or with

OCR for page 172
Page 174 low-level abilities entering first grade as it was for better prepared or more privileged children. Chall also noted the need to provide children with the practice in reading that would generate reading fluency and the value of providing challenging reading material in addition to texts that enabled children to practice skills they had acquired. Chall's conclusions regarding beginning instruction were challenged by people who raised questions about the validity of the research studies available for her review and the difficulty of applying a classification system that attempted to divide programs into code- and meaning-emphasis categories (e.g., Rutherford, 1968). Although Chall did not suggest that her findings be used to endorse systematic phonics approaches, her work has been highly influential in support of those who endorse a heavy emphasis on phonics in beginning reading. Beginning to Read In 1990, Marilyn J. Adams published Beginning to Read: Thinking and Learning About Print. Like Chall, Adams synthesized available research but also included a review of the literature on the psycholinguistic processes involved in reading. She concluded that direct instruction in phonics, focusing on the orthographic regularities of English, was characteristic of good, effective reading instruction, but she noted the need for practice in reading, for exposure to a lot of reading materials as input to vocabulary learning, and for motivating, interesting reading materials. Evidence from classroom research on the advantages of incorporating a code-oriented approach to early reading instruction was interpreted by Adams in light of evidence from basic research on the cognitive processes involved in reading and evidence concerning the nature of the code itself. Adam's research synthesis was highly convergent with that of Chall, both in confirming the importance of teaching children explicitly about the code of English orthography and in noting that good readers must have access to many experiences with literacy that go beyond the specifics of phonics instruction. Adams's synthesis was especially useful in drawing together research from across several different subdisciplines of psychology,

OCR for page 172
Page 175 child development, linguistics, and education. Most importantly, perhaps, her review pointed to the critical importance not just of children's learning but also of their basic early understandings of print and how print works, and, in particular, of the scattered but already converging evidence for the key role of basic phonemic awareness in fostering alphabetic understanding. Follow Through Provoked by finding that gains made by Head Start students during preschool tended to dissipate with time, in the early 1970s the federal government sponsored another large study comparing the long-term effects of reading instructional methods. The objective of Project Follow Through was to determine which general educational approaches or models worked best in fostering and maintaining the educational progress of disadvantaged children across the primary school years. By design, the 20 models included in the project contrasted broadly in philosophy and approach and included basic skills models, emphasizing basic academic skills; cognitive-conceptual models, emphasizing process over content learning; and affective models, emphasizing self-esteem, curiosity, and persistence. Analyses of the data revealed major findings (Stebbins et al., 1977): (1) The effectiveness of each Follow Through model varied substantially from site to site. No model was powerful enough to raise test scores everywhere it was implemented. (2) Models that emphasized basic skills (language, math computation, vocabulary, spelling) succeeded better than others in helping children gain these skills. (3) Models that emphasized basic skills produced better results on tests of self-esteem than did other models, including those specifically aimed at self-esteem. (4) No model was notably more successful than the others in raising scores on cognitive conceptual skills. (5) When models emphasized cognitive areas other than basic skills, children tended to score lower on tests of basic skills than they would have without the program. The researchers concluded that "most Follow Through interventions produced more negative than positive effects on basic skills test scores" (Stebbins et al., 1977). The only notable exception to this

OCR for page 172
Page 176 trend was the Direct Instruction Model, which promoted the teaching of skills and concepts essential to reading, arithmetic, and language achievement. It emphasized the systematic teaching of phonemic and language skills and promoted academic engagement. Students who participated during four full years (kindergarten through third grade) in the direct instruction program performed close to or at national norms on measures of reading, math, language, and spelling. The national Follow Through evaluation study has been criticized for many problems of the type often associated with field research in education and social services, including nonrandom assignment of subjects, unclear definition of treatment, problems of assessing implementation, less than ideal instrumentation, misleading classification of models and outcome measures, inadequate research design, questionable statistical analyses, and the use of methodological and statistical strategies that favored some type of model over others (Stebbins et al., 1977; House et al., 1978). Perhaps because of some of these factors, intersite variation among models was larger than between-model differences (House et al., 1978). In subsequent analyses, however, much of this variation disappeared when demographic factors were properly considered in the designation of control sites and outcome aggregation (Gersten, 1984), adding confidence to Project Follow Through's positive data on the value of the Direct Instruction Model. Moreover, follow-up studies of students suggested lasting effects of direct instruction. (Recall our discussion of direct instruction and cognitively oriented preschool education models, which have some similar results as those findings on direct instruction in kindergarten through grade 3 and also some contrasting findings.) Although the Follow Through results suggest very positive effects for the program, it has not been as widely embraced as might be expected. It may be that teachers believe that direct instruction in general is only for teaching factual content to students of low ability and not for promoting problem solving or higher-level thinking (see review by Peterson et al., 1982), although confirmatory evidence is not available.

OCR for page 172
Page 177 Other Efforts The classroom observational research of Stallings et al. (1986) and Soar (1973) described and linked critical features of the Follow Through approach to student outcomes. The work of these researchers played a large role in the various syntheses of research on effective teaching written in the late 1970s and the 1980s, such as those by Brophy and Good (1984) and Rosenshine and Stevens (1986). Classroom observation by Stallings and Soar uncovered the strong correlation between children's academic engaged time and growth in achievement and certain patterns of teacher-student interaction. In addition, it indicated the importance of explicit instruction for enhancing the achievement of disadvantaged students, a conclusion reinforced by subsequent observational research (e.g., Brophy and Evertson, 1978; Good and Grouws, 1975). Given previous efforts to assess instructional practice, the committee sought to examine current research on reading instruction. The next section describes the criteria used in selecting such studies. Selection Criteria Building on the previous work on instruction, the committee examined instructional practices that were supported by convergent evidence. We sought evidence about individual differences in response to treatment. Furthermore, we were interested in studies that assessed both short- and long-term reading outcomes, although long-term outcomes were available for only a few programs. Evaluations of instructional programs in kindergarten classrooms are not numerous, yet inferences about what such programs must cover are tightly constrained by the preschool predictors of literacy success on one side and the first-grade requirements on the other. Moreover, the major instructional tension associated with kindergarten literacy objectives is less about what children should learn than how they can be helped to learn it in an appropriate manner. Similarly, we know from intense research efforts what first-grade children ought to accomplish in reading, yet intense debate contin-

OCR for page 172
Page 178 ues on what and how they should be taught. Questions of how to organize and support learning in a way that results in the best possible outcomes for the largest number of children are an urgent educational priority. In view of this and because the research base permits, the section on first grade is principally directed to evaluations and comparisons of instructional programs. Beyond first grade, the relevant issues and goals multiply as the relevant research base recedes. In the dual interest of reviewing what is known and pointing toward key unknowns, our discussion of second- and third-grade issues is taken up goal by goal. Converging evidence from experimental investigations, correlational studies, nonequivalent control-group studies, and various other quasi-experimental designs and multivariate correlational designs presented in this and other chapters led the committee to focus on particular practices and programs. Many of the classroom investigations presented in this chapter have high external validity—that is, their results are generalizable to the children and settings that we are studying—and are less robust in internal validity (i.e., experimental control of variables) because of the logistical difficulties involved in carrying out such investigations. Hence, there is a need to look for a convergence of results—not just consistency from one method. When convergence is obtained, confidence increases that our conclusions have both internal and external validity. Among the most important ways to prevent reading difficulties is classroom instruction in literacy activities, which begins in kindergarten. KINDERGARTEN The Kindergarten Challenge A kindergarten classroom typically consists of an adult and 20 to 25 students—a very different scenario from a home or preschool. The management demands of the typical kindergarten classroom necessitate a level of conformity and control of comportment that challenges many entering children, regardless of how accommodating the classroom may be to children's individual natures and needs.

OCR for page 172
Page 179 A child can no longer demand the attention or assistance of the attendant adult at will; each must learn how to solicit individual attention and to wait patiently while the teacher is attending to others. To a greater or lesser extent depending on the classroom, every kindergartner must learn to sit quietly, to listen considerately to both the teacher and other students, to communicate cooperatively, to restrain behavior to within acceptable limits, to accomplish tasks both independently and with others, to share resources, to treat others respectfully, and to try to learn and do what she or he is asked to learn and do. Meanwhile, preparing children to learn to read is the top priority on the kindergarten teacher's agenda. Fostering Literacy in the Kindergarten Classroom The delicate balance for the kindergarten teacher is thus one of realizing means of promoting literacy learning in ways that are at once developmentally sensitive and appropriately foresighted, in order to ensure that as children leave kindergarten they have the capacities needed to function well in the typical first grade. More specifically, two goals are paramount. The first is to ensure that children leave kindergarten familiar with the structural elements and organization of print. By the end of kindergarten, children should be familiar with the forms and format of books and other print resources and be able to recognize and write most of the alphabet; they should also have some basic phonemic awareness, that is, understanding of the segmentability of spoken words into smaller units. The second major goal of kindergarten is to establish perspectives and attitudes on which learning about and from print depend; it includes motivating children to be literate and making them feel like successful learners. In this section, we provide examples of materials and activities that have been used well toward these ends. Reading aloud with kindergartners has been broadly advocated. By actively engaging children with different aspects of shared books, read-aloud sessions offer an ideal forum for exploring many dimensions of language and literacy. This is especially important for children who have had little storybook experience outside school (Feitelson et al., 1993; Purcell-Gates et al., 1995). Among the goals

OCR for page 172
Page 180 of interactive storybook reading are developing children's concepts about print, including terms such as ''word" and "letter" (Holdaway, 1979; Snow and Tabors, 1993); building familiarity with the vocabulary of book language (Robbins and Ehri, 1994), as well as its syntax and style (Bus et al., 1995; Feitelson, et al., 1993); and developing children's appreciation of text and their motivation to learn to read themselves. Effective practices for fostering these goals include encouraging children to ask their own questions about the story; to respond to others' questions; to follow the text with movement, mime, or choral reading; and to notice the forms and functions of print features (words, punctuation, letters, etc.). In addition, children's learning from and about storybooks is enhanced by repeated readings (Martinez et al., 1989). Recall from Chapter 4 that many of the outcomes of reading aloud as measured in kindergarten are significantly associated with reading achievement outcomes in first through third grades. In recent years, parents and teachers have been increasingly encouraged to share nonfiction as well as fiction with youngsters. To explore the educational impact of these recommendations, Mason et al. (1989) asked several kindergarten teachers to read three different types of selections: storybooks, informational texts, and easy-to-read picture books. They found that, depending on the type of text with which they were working, these teachers spontaneously but consistently and dramatically shifted the focus and nature of the accompanying discussion and surrounding activities. Not only the instructional emphases but also the complexity and nature of the language produced by both the teacher and the students appeared to change distinctively across these types of reading situations. Before reading the storybook aloud, the teachers initiated discussions about its author, central characters, and concepts; during story reading, they clarified vocabulary and engaged the students in making predictions and explaining motives and events; afterward they asked them to reflect on the meaning and message of the story. Given the science text, in contrast, teachers engaged the children in activities designed to help them relate the text to their everyday experiences. Socratically probing their responses, teachers led stu-

OCR for page 172
Page 181 dents to predict and explain, to deduce and test causes, and to discern necessary from sufficient conditions. In addition, vocabulary tended to be handled through rather elaborate concept development instead of definition. Finally, given easy-to-read picture books, discussion was more limited but firmly focused on the print and the words on each page. In short, the potential value of reading different genres with children extends well beyond any properties of the texts themselves. Moreover, the kinds of activities and discussion associated with each genre make distinctive contributions toward developing children's appreciation of the nature, purposes, and processes of reading. The sheer availability of books has been suggested as an important catalyst for children's literacy development (Gambrell, 1995; Gambrell and Morrow 1996; Krashen, 1996). But the impact of books on children's literacy development depends strongly on how their teachers make use of them. Demonstration of the effects of books, augmented with materials, training, and home involvement to stimulate oral interaction around books, with Spanish-speaking kindergartners can be found in Goldenberg (1994). A good kindergarten program should also prepare children to read by themselves. Few kindergartners are developmentally ready for real reading on their own. However, a variety of print materials have been especially designed to support early ventures into print. By way of example, we describe three: big books, predictable books, and rebus books. Big books are nothing more than oversized storybooks. As such, they offer opportunity for sharing the print and illustrations with a whole group of children in the ways that one might share a standardsized book with just a few (Holdaway, 1979). A common classroom activity with big books, for example, is fingerpoint reading: as the teacher points to the words of a familiar text or refrain in sequence, the children are challenged to recite the words in time. Beyond leading children to internalize the language of a story, fingerpoint reading is useful for developing basic concepts about print, such as directionality. Slightly more advanced children can be led to discover the visual differences between one word and two words or between long words and short words. Repeated words may be

OCR for page 172
Page 182 hunted down with the goal of establishing them as sight words, and rhyming texts may be well suited to introducing a basic notion of letter-sound correspondences. Patterned or predictable books, as their name suggests, are composed of text that is at least semirepetitive or predictable. The classic in this category is the story by Bill Martin, Jr., Brown Bear, Brown Bear, What Do You See? (1992). The first page of the story vividly depicts a red bird along with a printed answer for the bear, "I see a red bird looking at me." The second page restates the initial question as "Red bird, red bird, what do you see?" and answers with reference to a third animal. Each successive page varies only the name of the creature that is pictured and named. By perusing patterned and predictable books, children learn how to use predictions and picture cues to augment or reinforce the text, even as they develop basic book-handling habits. In rebus books, words or syllables of words that are beyond the children's reading ability are represented in the text itself by little pictures, or rebuses, of their referents. An example of a sentence in a rebus book is presented in Box 6-1. Entry-level rebus books are often designed to build a basic sight repertoire of such short and very frequent function words as "the," "of," "is," and ''are." As the child's skill in word recognition progresses, the number of different printed words is increased. Several studies have demonstrated that the use of rebus books at entry levels can measurably ease children's movement into real reading (Biemiller and Siegel, in press; MacKinnon, 1959). Variations of the language experience approach offer yet another way to ease children into reading. The objective of this approach is to impart the understanding that anything that can be said can be

OCR for page 172
Page 215 children ranged from virtual nonreaders to those who could read comfortably at the fourth-grade level; of the 230 children in all 14 classrooms, 120 were reading at or above grade level. In these classrooms, the teachers introduced each new basal selection by reading it aloud. The discussion following the reading of the selection was complemented with teacher- and student-generated questions and vocabulary work. In addition, the selection was explored more analytically with the help of a variety of organizational frames such as story maps, plot charts, and Venn diagrams. Children in need of extra help were pulled aside for echo reading: each paragraph was read first by the teacher and then by the students. That evening, each student was to read the selection again at home, preferably aloud to a parent. The next day, students paired up, taking turns reading each page or paragraph to each other. The partner reading routine was pursued for three reasons. First, reading with another was useful in keeping children engaged and on task. Second, the teacher could easily monitor progress and performance by moving around the classroom and listening. Third, following Chall's (1983) recommendation, the researchers sought to increase students' amount of oral reading. For further reinforcement, a variety of other options were adopted from time to time, such as having each child practice reading one part of the selection for performance; students still having difficulty were asked to reread the selection at least one more time at home. Each selection was also reviewed by completing journals in pairs or as a class. In addition to this work with the basal selections, children were asked to read books of their own choice, both during each school day for 15-20 minutes and at home. In short, the program was set up to promote comprehension growth while encouraging a great deal of reading and rereading for building reading fluency. Responses to the program were strongly positive from both teachers and students. Oral reading growth was assessed by asking a subsample of 89 students to read aloud both familiar and previously unseen excerpts from their basal reader in November, January, and May. Growth was most pronounced for children who had been reading at or above

OCR for page 172
Page 216 the primer level at the start of the year, and it was fastest between November and January. Due to ceiling effects, improvement among the children who began the year above the second-grade level could not be measured. The group that started the year below the primer level never caught up; their readings of the basal passages continued to be slow and error prone. Impact of the intervention was also measured by using Leslie and Caldwell's (1988) Qualitative Reading Inventory. This test consists of graded passages for oral reading, each accompanied by comprehension questions. Growth across the school year averaged 1.88 and 1.77 grade levels for the 4 and 10 classrooms that respectively participated in the first and second years of the study. Of the 190 students who started second grade at the primer level or above, only 5 were still unable to read at the second-grade level by spring. For 20 who could not read even the primer on entry, 9 reached or surpassed the second-grade level by spring, and all but one could read at least at the primer level. Thus, although about 10 percent of the children were still performing below grade level, and although results are measured against expectable gains rather than against the performance of a control group, the outcomes of the study are impressive. It was also longer in duration and broader in scope than most other second-grade reading interventions. In particular, its scope embraced both fluency and comprehension support; children need both. Comprehension and Word Knowledge Mature readers construct meaning at two levels. One level works with the words of the text for a literal understanding of what the author has written. However, superior word recognition abilities do not necessarily translate into superior levels of reading achievement (Chall et al., 1990). Productive reading involves, in addition to literal comprehension, being able to answer such questions as: Why am I reading this and how does this information relate to my reasons for so doing? What is the author's point of view, what are her or his underlying assumptions? Do I understand what the author is saying and why? Do I know where the author is headed? Is the text

OCR for page 172
Page 217 internally consistent? Is it consistent with what I already know or believe? If not, where does it depart and what do I think about the discrepancy? This sort of reflective, purposive understanding goes beyond the literal to the underlying meaning of the text. For purposes of discussion, the development of productive reading comprehension can be considered in terms of three factors: (1) concept and vocabulary development, (2) command of the linguistic structures of the text, and (3) metacognitive or reflective control of comprehension. Written text places high demands on vocabulary knowledge. Even the words used in children's books are more rare than those used in adult conversations and prime-time television (Hayes and Ahrens, 1988). Learning new concepts and the words that encode them is essential for comprehension development. People's ability to infer or retain new words in general is strongly dependent on their background knowledge of other words and concepts. Even at the youngest ages, the ability to understand and remember the meanings of new words depends quite strongly on how well developed one's vocabulary already is (Robbins and Ehri, 1994). Can children's word knowledge and reading comprehension be measurably improved through instruction? The answer is yes, according to a meta-analysis of relevant research studies by Stahl and Fairbanks (1986). First, vocabulary instruction generally does result in measurable increase in students' specific word knowledge. Sometimes and to some degree it also results in better performance on global vocabulary measures, such as standardized tests, indicating that the instruction has evidently enhanced the learning of words beyond those directly taught. Second, pooling across studies, vocabulary instruction also appears to produce increases in children's reading comprehension. Again, although these gains are largest where passages contain explicitly taught words, they are also significant given general standardized measures. Looking across studies, Stahl and Fairbanks (1986) noted differences in the effectiveness of vocabulary instruction as well. Methods providing repeated drill and practice on word definitions resulted in significant improvement with the particular words that had been taught but no reliable effect on reading comprehension scores. In

OCR for page 172
Page 218 contrast, methods in which children were given both information about the words' definitions and examples of the words' usages in a variety of contexts resulted in the largest gains in both vocabulary and reading comprehension. An important source of word knowledge is exposure to print and independent reading. As noted above, books introduce children to more rare words than conversation or television does. So educational approaches that encourage children to read more both in school and out should increase their word knowledge (Nagy and Anderson, 1984) and reading comprehension (Anderson et al., 1988). However, several efforts to increase the breadth of children's reading have produced little measurable effect on their reading ability (Carver and Liebert, 1995; see review in Taylor et al., 1990), perhaps because books selected for free reading tend to be at too easy a level for most children (Carver, 1994). Alternately, perhaps children who are doing poorly are less likely to profit from extensive exposure to print than children who are already progressing quite well. One group of researchers reviewed interactions among print exposure, word knowledge, and comprehension, teasing apart the relations among prior ability and increased reading (Stanovich et al., 1996). They concluded (p. 29): ''In short, exposure to print is efficacious regardless of the level of the child's cognitive and comprehension abilities. Even children with limited comprehension skills will build vocabulary and cognitive structures through immersion in literacy activities.   An encouraging message for teachers of lowachieving children is implicit here. We often despair of changing 'abilities,' but there is at least one partially malleable habit that will itself develop 'abilities'—reading." The relation between print exposure and comprehension need not be limited to the child's own reading in school. Cain (1996) studied the home literacy activities of 7- and 8-year-olds whose word reading accuracy was appropriate for their chronological age but who differed in their comprehension ability. She reports the following contrasts: "The children who were skilled comprehenders reported reading books at home more frequently than the less skilled children, and their parents reported that they were more likely to read story books. The skilled comprehenders also reported that they

OCR for page 172
Page 219 were read to more frequently at home by their parents than the less skilled group and this was confirmed by their parents' responses. . . . The skilled children were significantly more likely to read books with their parents than were the less skilled children and also tended to talk about books and stories more frequently than did the less skilled comprehenders." (Cain, 1996:189) It might be assumed that reading aloud with a child loses its value once children have attained independent accuracy in reading words, but Cain's findings raise the possibility that being read to promotes skilled comprehension at ages 7 and 8, although she points out that no causal link has yet been demonstrated. Comprehension and Background Knowledge The breadth and depth of a child's literacy experiences determine not only how many and what kinds of words she or he will encounter but also the background knowledge with which a child can conceptualize the meaning of any new word and the orthographic knowledge that frees that meaning from the printed page. Every opportunity should be taken to extend and enrich children's background knowledge and understanding in every way possible, for the ultimate significance and memorability of any word or text depends on whether children possess the background knowledge and conceptual sophistication to understand its meaning. A program designed to enhance background knowledge and conceptual sophistication among third graders is Concept Oriented Reading Instruction (CORI). The emphasis of the program is on the comprehension of interesting texts. The program is designed around broad interdisciplinary themes, exploiting real-world experiences, a range of cognitive strategies, and social groupings to promote self-direction. Designed for third graders in high-poverty schools with a history of low achievement, it has been successfully used at both the classroom and the whole-school level. The third-grade students have ranged in reading levels from first to fourth grade, and students with limited English proficiency are mainstreamed and included in the classroom. The program has effectively increased narrative text comprehension, expository text comprehension, and other language

OCR for page 172
Page 220 arts skills on standardized tests, as well as increasing students' performance on the Maryland School Performance Assessment Program (MSPAP) (Guthrie et al., 1996). Compared to control students, students in the program improved significantly on reading, writing, science, social studies, and language use but not in math, which was not taught in the program. CORI has also been shown to increase the amount and breadth of independent reading and volitional strategies for maintaining engagement in reading activities. Structures, Processes, and Meta-Processes in Comprehension Instruction Research on comprehension among young readers has not resolved questions about the nature and separate identity of the difficulties they encounter as they attempt to understand texts.  It is difficult to tease apart the effect of stores of word knowledge and background knowledge from the effect of processes (e.g., identifying words quickly and accurately, constructing mental representations to integrate information from the text) and meta-processes (making inferences, monitoring for inconsistencies) (Cornoldi and Oakhill, 1996). Instruction for comprehension, however, generally focuses on understanding complete connected text in situations in which many of the possible difficulties appear bound together and often can be treated as a bundle to good effect. Many comprehension instruction techniques used in schools today are described as meta-cognitive. A meta-analysis of 20 meta-cognition instruction programs found a substantial mean effect size of .71 (Haller et al., 1988). Instructional programs focusing on self-questioning and identifying text consistencies were found to be most effective. A meta-analysis of 10 studies related to a technique called reciprocal teaching found a median effect size of .88 (Rosenshine and Meister, 1994). For most active comprehension instruction, whether considered meta-cognitive or not, two pedagogic processes are intermingled: traditional instruction in basic stores of knowledge (the background for the text and for particular words) and instruction in particular comprehension strategies complemented by the active skilled reading

OCR for page 172
Page 221 of the text by an expert (the teacher) done in such a way that the ordinarily hidden processes of comprehension are displayed (see Kucan and Beck, 1997; Beck and McKeown, 1996). The children have an opportunity to learn from the joint participation (a form of cognitive apprenticeship) as well as from the particulars in the instructional agenda. As Baker (1996) notes, it is an open question whether direct instruction or observational learning provides the greater contribution to student progress. Reciprocal teaching is a particularly interesting approach to consider in detail both because of its apparent effectiveness and because it illustrates the mixed instructional agenda and pedagogical strategies. Reciprocal teaching provides guided practice in the use of four strategies (predicting, question generating, summarizing, and clarifying) that are designed to enhance children's ability to construct the meaning of text (Palincsar et al., 1993).  To engage in reciprocal teaching dialogues, the children and their teacher read a piece of common text. This reading may be done as a read-along, a silent reading, or an oral reading, depending on the decoding abilities of the children and the level of the text. The children and the teacher take turns leading the discussion of segments of the text, using strategies to support their discussion. The ultimate purpose of the discussion, however, is not practice with the strategies but the application of the strategies for the purpose of coming to a shared sense of the meaning of the text at hand. The tenets of reciprocal teaching include (a) meaningful use of comprehension-monitoring and comprehension-fostering strategies; (b) discussion for the purpose of building the meaning of text; (c) the expectation that, when children are first beginning these dialogues, they will need considerable support provided by the teacher's modeling of the use of the strategies and guiding students' participation in the dialogues; (d) the use of text that offers appropriate challenges to the children (i.e., there is content worth discussing in the text and the text is sufficiently accessible to the children); and, finally, (e) the use of text that is thematically related so that children have the opportunity to build their knowledge of a topic or area over time. Reciprocal teaching was designed as both an intervention to be used with youngsters who were experiencing language-related diffi-

OCR for page 172
Page 222 culties and as a means of prevention given the hypothesis that young children should experience reading as a meaningful activity even before they are reading conventionally. It has been investigated principally with children who come from high-poverty areas, children being served in developmental and remedial reading programs, and children identified as having a language or learning disability. Research on reciprocal teaching with young children in first and second grades indicates statistically significant improvement in listening comprehension (which assessed ability to recall information, summarize information, draw inferences from text, and use information to solve a novel problem) and fewer referrals to special education or remedial reading programs. In addition, teachers reported that, as a result of their experiences in reciprocal teaching dialogues, their expectations regarding these children were raised. In other words, children who appeared to have a disability on the basis of their participation in the conventional classroom dynamic appeared quite able in the context of reciprocal teaching dialogues. Training studies on inferences and comprehension monitoring with 7- and 8-year-olds show that children identified specifically as poor comprehenders profit differentially from certain kinds of instruction. Yuill and Oakhill (1988) compared the effect on skilled and less skilled comprehenders (matched for age and reading accuracy) of a program that lasted for seven 30-minute sessions spread over about two months. The treatment group worked on lexical inferences, question generation, and prediction. One control group read the same texts and answered questions about them in a group discussion format. A second control group read the same texts and practiced rapid word decoding. There appears to have been an interaction between aptitude and treatment. Analyses of post-test results showed that the less skilled comprehenders benefited more from the experimental treatment than did the more skilled, that the less skilled comprehenders derived more benefit from the comprehension training than they did from the rapid decoding condition, but that the more skilled benefited more from the decoding training than from the comprehension training. Yuill (1996) worked with a similar set of subjects (matched for age and reading accuracy, differing on comprehension ability) to

OCR for page 172
Page 223 train for the ability to recognize that texts could have more than a single obvious interpretation by using the genre of riddles, which depend on ambiguity and its resolution. The treatment condition focused the children on alternative interpretations in texts by training them to explain the ambiguity in riddles; the control group children also read amusing texts but focused on sublexical awareness activities rather than on meta-comprehension activities. At the end of the two-month period, the experimental treatment group performed significantly better on the post-test in comprehension than the control group did, but there was no significant interaction between skill group and training. SUMMARY The nature and quality of classroom literacy instruction are a pivotal force in preventing reading difficulties in young children. Adequate initial reading instruction requires a focus on using reading to obtain meaning from print; understanding the sublexical structure of spoken words; exposing the nature of the orthographic system; practice in the specifics of frequent, regular spelling-sound relationships; and frequent and intensive opportunities to read. Adequate progress in learning to read English beyond the initial level depends on having established a working understanding of how sounds are represented alphabetically, sufficient practice in reading to achieve fluency with different kinds of texts written for different purposes, instruction focused on concept and vocabulary growth, and control over procedures for monitoring comprehension and repairing misunderstandings. Activities designed to ensure these opportunities to learn include practice in reading (and rereading), writing as a means of word study and for the purpose of communication, invented spelling as a way to explore letter-sound relationships, and spelling instruction to enhance phonemic awareness and letter-sound/sound-letter relationships. The context of the instruction varied considerably across the interventions considered in this chapter. Although the materials used ranged widely, a significant shared feature was attention to the

OCR for page 172
Page 224 use of continuous text. The characteristics of the texts used include predictability, the opportunity the text provides to use spelling patterns that have been studied, what Juel (1991) refers to as "phonologically protected" text. Effective instruction includes artful teaching, a thing that transcends—and often makes up for the limitations of—specific instructional strategies (see Box 6-5). Although in this report we have not incorporated lessons from exceptional teaching practices with the same comprehensiveness as other topics in the research on reading, we acknowledge their importance in conceptualizing effective reading instruction. Classroom instruction is not the only method of intervention used to prevent reading difficulties. In Chapter 5, we reviewed efforts that can take place in the preschool years. In the next two chapters on prevention and intervention strategies to preventing reading difficulties, we review organization strategies in kindergarten and the primary grades and research on providing extended time in reading-related instruction. In the next chapter, we review institutional responses to the prevention of reading problems.

OCR for page 172
Page 225 BOX 6-5 Teaching Children Versus Teaching a Curriculum Language Arts:You come down solidly advocating that educators need to teach children rather than to teach a curriculum. And you have also stated that the wars between whole language advocates and phonics advocates "are based more on educator identities than on children's needs." Would you talk about that a bit? Lisa Delpit:I continue to be astounded that folks seem to put themselves into a political and ideological camp and indicate, "I'm going to stay in this camp come hell or high water." I view teaching a little differently. I don't place myself as a teacher in a camp. I see myself as responder to the needs of children. Some children will need to learn explicitly certain strategies or conventions: some children will not need that because they've gotten it through the discourse that they learned in their homes. In California I saw a black child who was in a class where the kids were supposed to read a piece of literature and then respond to it. The child clearly couldn't read the selection. When asked about the situation, the teacher said, "Oh, he can't read it, but he'll get it in the discussion." Perhaps it's good that he will be able to get it in the discussion, but at the same time nobody is spending time teaching him what he also needs to learn—how to read for himself. So, we can lose track of the fact that children may need different kinds of instruction, depending on their knowledge and background. Sometimes we have the best intentions but actually end up holding beliefs that result in lower expectations for certain students. We are content that the students are just becoming fluent in writing, so we don't push them to edit their pieces into final products that can be published. We don't do the kind of pushing necessary to get students to achieve at the level that they might be capable of. SOURCE: An excerpt from "A Conversation with Lisa Delpit" by Language Arts  (1991:544-545).