Review of Converging Evidence
Learning to read words is anything but natural. On the contrary, it requires integration of numerous complex processes. While successful acquisition and application of these processes is incidental for many children, others require systematic and planful teaching. From our synthesis of beginning reading research (Chard, Simmons, & Kameenui, 1994), four areas of convergence bear implications for word recognition for children with diverse learning needs.
It is important to note that our focus is on learning to read words independently does not pre-empt or preclude simultaneous attention to other reading processes. Rather, throughout the process of learning to read, storyreading and demonstrations of the role of reading for information should be integrated with learning to master the code. Once early readers learn some letter-sound correspondences, they can learn to blend those sounds into simple words. Similarly, as children begin to blend sounds into words, the words can be put into sentences to form connected texts.
In the next section, we describe big ideas from our review that are instructionally important and empirically validated. We attempt to connect research and practice by responding to two focal questions: (a) What are the research-based instructional priorities, or "big ideas," in word recognition? and (b) For the instructional priorities of word recognition, what is the existing research evidence regarding curriculum design?
The four main areas of convergence identified from the review of reading research by Chard, Simmons, and Kameenui (1994) are captured by three big ideas that serve as a framework for designing instruction in word recognition for diverse learners:
1. Teach prerequisite skills in combination with word recognition instruction.
2. Teach alphabetic understanding (i.e., letter-sound correspondence) in combination with word recognition.
3. Teach blending of words and encourage readers to make sense of the words they blend.
The prerequisites to word recognition of print awareness and phonological awareness are addressed in detail in two other syntheses (Gunn, Simmons, & Kameenui, 1995; Smith, Simmons, & Kameenui, 1995) and, therefore, will not be addressed in this discussion. In the following section, we discuss Big Ideas 2 and 3 in relation to a framework of curriculum design principles that include conspicuous strategies , strategic integration , mediated scaffolding , primed background knowledge , and judicious review to render the implications more explicit and employable.
In the following section, we discuss these procedural curriculum design principles in combination with the big ideas of teaching (a) word recognition in combination with prerequisite skills, (b) alphabetic understanding, and (c) phonological recoding to illustrate how to translate research into practice. Because recent research has not focused on reading acquisition, but rather on reading and writing at higher levels (Juel, 1991; Pearson, 1993), we found conceptual convergence that students need the big ideas of word recognition, but procedural paucity on how to teach the big ideas. In other words, the procedural details were often scant despite strong support for the underlying concepts (i.e., phonological recoding). The following sections should not be viewed as a prescription, but rather as reasonable extensions of areas of convergence.
Evidence of Curriculum Design in Word Recognition
Alphabetic understanding . The first step in a conspicuous strategy for alphabetic understanding is to teach individual letter-sound correspondences. However, before offering an example of a conspicuous strategy, it may be helpful to describe instruction of letter-sound correspondences that is not conspicuous. For example, it would not be conspicuous to teach letter-sound correspondence within the context of words. Approaches such as this place extraordinary responsibility on the learner to isolate the letter-sound correspondence being taught from other letters in proximity. Some commercial reading programs attempt to teach letter-sound correspondences within words to avoid decontextualized language as when one pronounces phonemes in isolation (Adams, 1990). Still other programs avoid a conspicuous approach to teaching letter-sound correspondences by not addressing letter-sound correspondences at all (Adams, 1990). Evidence suggests that neither approach leads students efficiently to the instructional goal--alphabetic understanding.
In contrast to inconspicuous strategies, an explicit and unambiguous approach to teaching letter-sound correspondences would begin when the teacher presents a letter symbol and models the corresponding sound. For example, the teacher points to the letter m on the board or in a book and says "mmmmmmm." Next, the teacher tests student acquisition of the correspondence by having the group produce the sound corresponding to the letter m . Finally, once the teacher is certain that students know the correspondence, individuals can be tested.
This strategy is simple and conspicuous. Each letter-sound pair is presented in isolation without the distractions of other correspondences. The approach benefits the reader by ensuring that the connection between letter and corresponding sound is salient, and with frequent exposure the connection becomes automatic for students. Such automaticity is critical for poor readers to begin to sound out and blend words. Despite converging evidence implying the importance of a conspicuous strategy for alphabetic understanding, no primary studies in this review were identified that tested this principle explicitly.
Phonological recoding . Across models of word recognition, researchers agreed on the importance of phonological recoding , which refers to the process of identifying the sounds in a word, blending the sounds together, and searching for the meaning of the word in memory. While phonological recoding need only occur a few times for typical readers to be able to read with ease and speed, diverse learners must be encouraged to look carefully at spellings and sounds and to repeatedly sound out and blend words (Reitsma, cited in Adams, 1990). Phonological recoding can be taught as an explicit strategy once students master a few letter-sound correspondences. The next step is to sound out the letters of words and blend them together. Blending the sounds in words is familiar to students who have practiced auditory blending tasks as part of phonemic awareness.
A conspicuous strategy for sounding out and blending words could be taught with the teacher first pointing to each letter of a word and then saying its sound without stopping between the sounds. Second, the teacher could sound out the word again with the students. Careful monitoring of the children's sounding out informs the teacher whether the children are automatic with the letter-sound correpondences. Next, the students can sound out the word as the teacher points to the sounds. Finally, students should be encouraged to blend the word at a normal speed. The teacher should take care to test individual students.
The strategy described above is required as it represents the first step in teaching phonological recoding. As learners master letter-sound correspondences and become more facile with sounding out and blending, similar strategies can be taught that emphasize frequent sounding out and blending of word parts (e.g., "an" or "it") and common spelling patterns (e.g., "-th" or "-ing") that can be preceded by a host of beginning sounds. Research evidence strongly supports that for skilled readers word recognition is effortless and automatic because they repeatedly sound out and blend familiar words (see Adams, 1990, for detailed discussion). Therefore, frequent opportunities to sound out and blend words is particularly critical for diverse learners. Practice at blending common word parts and letter patterns promotes automaticity as students no longer approach each word as a new string of letters.
Strategic Integration
Alphabetic understanding . The simplicity of teaching letter-sound correspondences belies the complexity of integrating such correspondences with other correspondences and within words. The first point of integration is to provide students the opportunity to discriminate new letter-sound correspondences from previously learned correspondences . Being able to discriminate letter-sound correspondences is important for beginning readers to understand the nature of our alphabetic language (Adams, 1990). Toward that goal, teachers could present newly introduced letters in the proximity of earlier introduced and mastered letters, with new letters appearing more often than previously learned letters.
Since the goal of word recognition instruction is to teach learners to read connected text as quickly as possible, it is necessary to choose strategically the order in which to teach letter sound correspondences. The following guidelines may help:
These instructional strategies will enable learners to generalize sounds across a number of new words and will make meaningful texts more accessible.
Phonological recoding . It is imperative that learners begin reading words before they learn all letter-sound correspondences. Therefore, as soon as learners have mastered a few letters, they should be used to construct decodable words. Thus, a second point of integration: Learners should be taught how to blend the sounds to form words . Initially, students may struggle with the sounding-out and blending process. With planfully integrated instruction, however, the number of mastered letter-sound correspondences will increase gradually, the process of sounding out and blending will become more automatic, and the numbers of words being recoded will expand.
A third point of strategic integration is the use of words in context . As students learn the strategy for sounding out and blending words, it is crucial that they read these decodable words in stories. Although it is extremely difficult to create interesting stories with highly stipulated decodable vocabularies, there are a number of ways to move learners from reading only isolated words to reading words in context. For example, although the list of decodable words is short, teachers can systematically point out decodable words and have learners read them in the story. This requires careful selection of stories containing words that learners are able to read.
The following steps should be considered in the strategic integration of phonological recoding:
Diverse learners may not benefit fully from traditional reading curricula. Although reading research has consistently supported systematic phonics for helping diverse learners read successfully, most reading curricula do not contain a systematic phonics component. Diverse learners' need for systematic instruction can be accommodated by using mediated scaffolds.
Mediated scaffolds are external supports provided by the teacher, tasks, and materials during initial learning of word recognition. As such, mediated scaffolds provide learners with guidance and support toward independence. Beginning readers' acquisition of letter-sound correspondences can be mediated in a number of ways, so that students move successfully and efficiently from unknowing to knowing. Mediated scaffolding can come in the form of (a) teacher assistance, (b) peer assistance, and (c) task sequence and selection. The scaffolds should meet the needs of the beginning reader and should be diminished as the reader grows more facile with the prerequisites of reading, letter-sound correspondences, and phonological recoding.
Alphabetic understanding . For diverse learners, teacher-mediated scaffolding must take the form of prompt, direct error correction and teacher modeling of newly introduced letter-sound correspondences. Knowledge of letter-sound correspondences must be automatic. Therefore, student errors must be identified and corrected before students have opportunities to practice incorrect responses. In addition, new letter-sound correspondences must be modeled explicitly for beginning readers. Unfortunately, many commercial reading programs replace modeling new information with testing. For example, instead of modeling /m/ while pointing to the letter, directions are common in which the teacher "asks students what sound the letter m makes" before any initial instruction on the correspondence. Omitting the modeling step in teaching letter-sound correspondences is similar to missing the first rung on the ladder of alphabetic understanding. Without modeling, the burden is placed on the learner to intuit the corresponding sound, reducing the probability that the learner will learn the correspondence successfully.
Choosing which letter-sound correspondences to teach and which order to teach them in serves as another form of mediated scaffolding. Although there are no explicit empirically validated guidelines for ordering the introduction of letter-sound correspondences, some logical principles may be considered.
In sum, mediated scaffolding systematically structures the load and distribution of print and sound stimuli that the beginning reader has to consider, while providing ample time to acquire and practice newly learned letter-sound correspondences.
Phonological recoding . When students begin to sound out and blend words, teacher scaffolding is crucial. Teacher demonstration must illustrate that sounds should be connected without stops between individual sounds. The difficulty of teaching children to connect sounds during phonological recoding should not be underestimated. For example, students must coordinate saying one sound while identifying the corresponding sound of the next letter. Once the process has been modeled, a second scaffold involves the teacher sounding out with students before expecting them to recode independently. In a third form of scaffolding, the teacher points to each sound as students sound out words. This focuses attention on the importance of each individual sound (Adams, 1990).
Task scaffolds differ from teacher scaffolds by being embedded in the tasks and designed to promote reading by reducing the information students must generate independently. For example, in initial attempts to sound out and blend words children should focus on only two or three words. A reduced number of words allows students to focus on the goal of the initial task, namely, sounding out and blending words. Similarly, tasks can be sequenced such that only firm letter-sound correspondences are included in the phonological recoding tasks. Also, new letter-sound correspondences should be reviewed and mastered before appearing in the sounding out and blending tasks.
Initially, phonological recoding tasks are difficult due to the awkwardness of the process. As students learn the strategy, they should be encouraged to sound out and blend the words "in their head" and read the word as a whole. This progression of tasks approximates the end goal of word reading.
Alphabetic understanding . Teaching letter-sound correspondences as the key to alphabetic understanding without connection to words and text may seem like teaching in a vacuum. However, if carefully linked to the beginning reader's background knowledge, letter-sound instruction provides students with information about familiar sounds and symbols that they know are important to the message in text.
To be more specific, connecting letters to sounds first seems to be a clear extension of learning the names and shapes of letters. Adams (1990) discussed in detail the importance of teaching letter names and shapes as a precursor to letter sound correspondence. Adams argued that most children have a firm grasp of letter names before entering school because their parents teach them the "Alphabet Song" (to the tune of "Twinkle, Twinkle Little Star") without showing them the letters. After that, many children are introduced to the letter shapes and the opportunity to trace or write the letters. This progression provides a strong platform from which to leap into letter sounds.
As a result, consideration should be given to what children know about letters and sounds before letter-sound correspondences are taught. Indeed, some researchers suggest that children who are not facile at identifying the names of letters when presented the letter symbol should be pretaught the letters and their names. It is argued that primed background knowledge of letter names will serve as an anchor for the more complex letter-sound correspondences to follow.
Additionally, phonological awareness exercises that introduce children to a wide range of phonemes are a critical precursor to linking letters to sounds. Logically, it makes more sense to link the letter s with /s/ if students have already learned and practiced /s/ in phonological exercises such as segmenting the word sam into
s - a - m .
Phonological recoding . Prerequisite knowledge for sounding out and blending words (i.e., print awareness, phonological awareness, letter sound correspondences) must be primed prior to engaging students in tasks involving phonological recoding. This means sequencing the prerequisites in such a manner that learners are not expected to perform tasks for which they are not adequately prepared. For example, before students are taught to sound out and blend words, it is important that they have learned how to segment and blend words auditorily (i.e., in the absence of print). Background knowledge of segmenting and blending facilitates a smooth transition to sounding out and blending words in print.
Similarly, care should be taken to teach letter-sound correspondences to mastery before they appear in words to be sounded out and blended. Carefully sequenced instruction prevents students and teachers from becoming frustrated with tasks that involve phonological recoding. Very simply, providing a foundation of background knowledge builds strong word recognition skills.
Judicious Review
What to review . Review activities must focus on content that is essential for improving word recognition. Thus, review should include letter-sound correspondences, spelling patterns, and exercises that (a) occur frequently in oral and written language, (b) are essential to a majority of words, and (c) are of high quality. That is, time should be spent on activities that generalize across most words that students will encounter. Because the content of review will differ from student to student, informal progress monitoring helps discriminate what students have yet to master and what can be reviewed less frequently.
When to review . Research on review has shown that repeated, shorter review distributed over time is more effective than review that is presented all at once (e.g., Dempster, 1991). Word recognition activities are particularly well suited for distributed review. For example, letter-sound correspondences can be introduced and later reviewed in groups. Newly learned information should be reviewed frequently, but as the information is mastered, greater amounts of time can be interjected between review sessions. It is not advisable to remove information from review activities based on initial success, however. If students are expected to maintain mastery of items, as is the case with many components of word recognition, teachers must ensure that high-utility information and skills are reviewed cumulatively.
Scheduling review the way we have proposed here seems logistically complex. Indeed, the complexity is real. This is primarily because the skills and information learned in beginning reading are likely to affect students' long-term academic success. Care should be taken at this early stage to assist students in maintaining the critical components of word recognition. Fortunately, many studies have documented the effectiveness of repeated reading of favorite books and passages as a system of review once readers become facile with phonological recoding (e.g., Sindelar, Monda, & O'Shea, 1990; Weinstein & Cooke, 1992).
Designing review activities . Contrary to popular belief, review activities need not be rote rehearsal or what many people refer to as "drill and kill." Rather, it is incumbent on teachers to develop flexible and creative methods of helping students maintain knowledge. The only requirement is that review activities include important knowledge and provide opportunities to assess students' understanding.
The following guidelines should be considered in designing review activities for developing strong word recognition skills:
We would be remiss if we were to leave the reader with the impression that teaching word recognition should sacrifice attention to building comprehension skills. Indeed, this is not the case. Research suggests that comprehension and word recognition have a reciprocal effect on one another. That is, comprehension is dependent on word recognition, and strong word recognition, while not dependent on comprehension, is enhanced by readers' interest in the content of text.
Keeping in mind the importance of integrating word recognition and beginning comprehension instruction, we must focus on the needs of diverse learners. Diverse learners are at risk of failure in reading. Based on our review of word recognition research, it appears that many diverse learners require a qualitatively different form of instruction; one that is more systematic and planful. Primary emphasis must be placed on helping diverse learners acquire the "deep and thorough knowledge of letters, spelling patterns, and words, and of the phonological translations of all three" (Adams, 1990, p. 416) that are critical to later reading success. Towards this end, we have proposed three big ideas for word recognition instruction:
These big ideas are supported by the proposed framework of instructional design principles including conspicuous strategies, mediated scaffolding, strategic integration, primed background knowledge, and judicious review. Primary emphasis on this framework and planful integration with other literacy activities may reduce the risk faced by diverse learners in beginning reading as it now is commonly taught.
Adams, M. (1990). Beginning to read: Thinking and learning about print. Cambridge, MA: MIT Press.
Chard, D. J., Simmons, D. C., & Kameenui, E. J. (1995). Understanding the primary role of word recognition in the reading process: Synthesis of research on beginning reading . (Tech. Rep.). Eugene, OR: National Center to Improve the Tools of Educators.
Dempster, F. N. (1991). Synthesis of research on reviews and tests. Educational Leadership , 71-76.
Gunn, B., Simmons, D. C., & Kameenui, E. J. (1995). Emergent literacy: Curricular and instructional implications for diverse learners. (Tech. Rep.). Eugene, OR: National Center to Improve the Tools of Educators.
Juel, C. (1991). Beginning reading. In R. Barr, M. L. Kamil, P. B. Mosenthal, & P. D. Pearson (Eds.), Handbook of reading research (Vol. 2, pp. 759-788). New York: Longman.
Pearson, P. D. (1993). Reading. In The encyclopedia of educational research (Vol. 3, pp. 1075-1085). New York: Macmillan.
Sindelar, P. T., Monda, L., & O'Shea, L. (1990). Effects of repeated readings on instructional- and mastery-level readers. Journal of Educational Research, 83 , 220-226.
Smith, S. B., Simmons, D. C., & Kameenui, E. J. (1995). Phonological awareness: Curricular and instructional implications for diverse learners. (Tech. Rep.). Eugene, OR: National Center to Improve the Tools of Educators.
Weinstein, G., & Cooke, N. L. (1992). The effects of two repeated reading interventions on generalization of fluency. Learning Disability Quarterly, 15 , 21-28.