Teaching word identification

Introduction

Over the last decade, much has been learned about how children acquire the ability to identify words. Nevertheless, substantial disagreement remains as to the form that beginning reading instruction should take. Part of the difficulty results from the indefensible position that beginning reading programs must emphasize either meaning or skills.

The position taken in this paper is that beginning reading instruction must be meaning-based, involve students in frequent reading of informative and entertaining texts, and provide clear, explicit instruction of important word-identification strategies and skills. In an effective program of reading instruction, there must be an appropriate balance between teaching skills and strategies and reading and responding to a wide range of texts. After a comprehensive review of the research on beginning reading instruction, Adams (1990) concluded that whenever that “balance” is lost — when reading instruction becomes so skills-oriented that meaning and the joy of reading are lost, or when literature is emphasized to the point that important skills and strategies are not taught — students are likely to encounter difficulty in learning to read.

Definitions of Reading and Word Identification

Establishing a clear definition of reading provides an important perspective for evaluating approaches to teaching word-identification skills. Most educators would agree that the major purpose of reading should be the construction of meaning — comprehending and actively responding to what is read. Two of the most widely cited and agreed-upon definitions of reading are the following:

    Reading is the process of constructing meaning from written texts. It is a complex skill requiring the coordination of a number of interrelated sources of information (Anderson et al., 1985).

Reading is the process of constructing meaning through the dynamic interaction among: (1) the reader’s existing knowledge; (2) the information suggested by the text being read; and (3) the context of the reading situation (Wixson, Peters, Weber, & Roeber, 1987, citing the new definition of reading for Michigan).

Older, mechanistic definitions of reading as the translation of printed symbols into oral language equivalents are incomplete, given the progress made in understanding the nature of the reading process. There is widespread agreement that without the activation of relevant prior knowledge by a cognitively active reader and the melding of that prior knowledge with the text information, there can be no reading of text.

Even definitions of reading that emphasize meaning indicate that reading is activated by print. The reader must be able to translate the written words into meaningful language. Virtually all four- and five-year-old children can communicate with and learn from oral language, but very few can read, because they lack the ability to identify printed words. While simply being able to recognize or “say” the printed words of text without constructing the meaning of that text is not reading, constructing meaning from written text is impossible without being able to identify the words.

The terms word identification, word recognition, and decoding are frequently used interchangeably. The new Literacy Dictionary (Harris & Hodges, 1995) defines both word recognition and word identification as “the process of determining the pronunciation and some degree of meaning of an unknown word” (pp. 282-283). For words that are in a reader’s meaning vocabulary, unlocking the pronunciation leads to the word’s meaning. If a printed word is not in a reader’s meaning vocabulary, word-identification skills may allow access to the word’s pronunciation, but not its meaning. Being able to arrive at the pronunciation of a printed word constitutes word identification in the most minimal sense; however, if the reader is unable to attach meaning to the word, then he or she has not read the word, since reading must end in meaning construction.

Emerging Reading and Word-Identification Skills

Most children in kindergarten cannot read and write, although clearly some can. Most children who cannot read have nonetheless developed many language and cognitive skills that form the foundation for learning to read and write. These children tend to come from families that have provided them with many rich, early experiences with literacy. Other children, often referred to as “at risk,” have had fewer such experiences; these youngsters depend heavily on teachers and schools to help them become readers and writers.

A full discussion of emerging literacy skills would need to consider many topics. However, since the subject of this paper is word-identification instruction, it seems important to focus on two areas specifically related to success in that area: familiarity with print and phonemic awareness.

Familiarity with Print
Concepts of Print. Children who have had many experiences with language, especially the experience of having someone read to them regularly, may have some concept of what printed words and letters are. They may realize that the print on a page is the source of the text information needed for reading or know that a reader looks at print from left to right. These concepts, referred to as concepts of print, are important for success in learning to read (Adams, 1990), and children who have had limited preschool experiences with printed language will need to be taught them.

Teaching concepts of print should be approached systematically. After reading a favorite book, such as a Big Book version of Robert Kalan’s Rain, what could be more natural than for a teacher to point to the title and ask how many words are in the title, and to contrast the number of words in that title with the number in Mary Serfozo’s Who Said Red? As the teacher rereads the book, what would be more effective than to point to the words as he or she says them, moving from left to right across the page? Instruction in important concepts of print would be taking place but within the context of entertaining reading activities.

Letter Names. One of the strongest research findings in the field of reading is the high correlation between knowledge of letter names and success in learning to read (Adams, 1990; Adams & Pikulski, 1996; Durrell, 1980; Ehri 1983; Venezky, 1975). Young children need to develop the concept that printed words are composed of letters; they then can be taught letter names if they don’t come to school knowing them. While teaching children letter names does not in itself result in success in learning to read (Jenkins, Bausell, & Jenkins, 1972), it can facilitate memory for the forms or shapes of letters and can serve as a mnemonic for letter-sound associations or phonics (Adams, 1990).

Most kindergarten children learn letter names without difficulty. Many teachers introduce letter names by teaching emerging readers to sing the alphabet song. Thus, children often learn the names first and then attach them to the letter forms. There is now a wide variety of high-quality alphabet books such as ABC and You (Fernandez, 1990) or Annie, Bea, and Chi Chi Dolores: A School Day Alphabet (Maurer, 1993).

While knowing letter names appears to facilitate the development of word-recognition skills, it would be inappropriate to delay introducing other literacy activities (language expansion, shared reading, beginning writing activities, and so on) to children who have not yet learned letter names.

Phonemic Awareness
Children in the beginning stages of learning to read need to learn that spoken words are composed of a limited number of identifiable, individual sounds or phonemes. This understanding, often referred to as phonemic awareness, is a very important factor in success in learning to read (Juel, 1988). Phonics involves building associations between written letters and speech phonemes. If a child has no concept of what a speech sound is, building these associations will be difficult, if not impossible. A growing body of research shows that phonemic awareness is not only the most powerful predictor of success in beginning reading, but also, for most children, a necessary prerequisite for learning to read (Bradley & Bryant, 1983, 1985; Juel & Leavell, 1988; Stanovich, 1993-1994; Tunmer, Herriman, & Nesdale, 1988). Fortunately, there is also a substantial body of research that suggests that emerging readers can be taught phonemic awareness skills and that learning such skills leads to significantly greater success in learning to read (for example, Ball & Blachman, 1991; Bradley & Bryant, 1983; Lundberg, Frost, & Petersen, 1988). Griffith and Olsen (1992) and Yopp (1992, 1995) offer helpful suggestions and activities for developing phomenic awareness.

Adams (1990) suggests rhyming activities for initiating phonemic awareness. The reading and rereading of books with clear, simple rhymes, such as Nancy Shaw’s Sheep in a Jeep, offer abundant and fun opportunities for direct instruction in rhyming and the beginnings of phonemic awareness. The very title of the book Whistle for Willie (Keats, 1964) draws attention to beginning sounds that can be further developed through instruction.

Research also suggests that young children’s awareness of onsets (the initial consonant of a word or syllable) and rimes (everything after the initial consonant in a one-syllable word or in syllables, traditionally referred to as phonograms or word families) is related to success in beginning reading (Goswami, 1988, 1990; Goswami & Bryant, 1994). Therefore, children should be taught to identify and manipulate these sound units.

Children in kindergarten should be introduced to common phonograms. In addition to building phonemic awareness, providing instruction with phonograms also prepares children for reading words by analogy. For example, a child who never saw the word rug in print but who knows initial consonant sounds and how to read the word bug can very likely identify the new word if he or she has had practice manipulating onsets and rimes.

Encouraging young children to engage in writing using temporary (also called invented) spelling is another excellent way to foster phonemic awareness. As children learn to use letters to represent words in writing, they naturally need to think about the sounds that compose the words. Mason, Herman, and Au (1991) and Griffith and Olson (1992) offer a fuller explanation of what phonemic awareness is, review evidence that it can be taught, and give suggestions of how such awareness can be taught through meaningful language activities.

Word Recognition Skills and Strategies

The following are widely acknowledged as skills that readers use to identify printed words.

Instant Recognition
Mature readers identify words with remarkable speed and accuracy. Indeed, fluent word identification appears to be a prerequisite for comprehending text. If a reader must slowly analyze many of the words in a text, memory and attention needed for comprehension are drained by word analysis.

Beginning readers recognize very few words instantly. Through repeated exposure to the same words, instant recognition vocabulary grows. It is particularly important that developing readers learn to recognize those words that occur very frequently in print. A mere 100 words make up a full 50 percent of the words read, even by adults. The, and, to, you, he, it, and said are examples of these high-frequency words. Developing readers also need to learn to recognize high-frequency words instantly because many of them are not phonically regular. Based on phonics generalizations, to should rhyme with go, said should rhyme with paid, and so on.

Children’s ability to recognize words can be developed by teachers’ pointing out the words, by a variety of game-like activities, and by writing those words. However, it appears that instant recognition of words, especially high-frequency words, develops best when students read large amounts of text, particularly text that is relatively easy for the reader (Cunningham, 1995).

Context Clues
There is a good research base for concluding that students can use meaning or context clues to help identify words and that instruction can help improve their use of such clues (Johnson & Baumann, 1984).

Three different types of context clues are frequently distinguished:

  • Semantic or Meaning Clues. There are general semantic clues. For example, when reading a story about cats, good readers develop the expectation that it will contain words associated with cats, such as tail, purr, andwhiskers. Sentence context clues are more specific. In the sentence “My cat likes to _____,” given the sentence context and what most of us know about cats, words like play, jump, and scratch seem reasonable.
  • Syntactic or Word Order Clues. In the previous example, the order of the words in the sentence indicates that the missing word must be a verb. Other parts of speech, such as adjectives (nice, brown) or nouns (man, fence), make no sense or don’t result in what sounds like a real sentence.
  • Picture Clues. Illustrations can often help with the identification of a word. In the example, if a picture of a cat leaping through the air accompanies the text, jump seems a very good possibility.

Context clues are often helpful, but they often are not specific enough to predict the exact word. Often several choices are possible, as in the example given. However, when context clues are combined with other clues such as phonics and word-part clues (for example, the sounds associated with j and mp), accurate word identification is usually possible.

Context clues allow readers to “crosscheck” their identification of words. For example, a reader encountering the word scratch for the first time should look carefully at the letters of the word, apply what he or she knows about phonics and word parts, and check to be sure that an attempted pronunciation matches the letter clues. In addition, the reader should always crosscheck to be sure that the word makes sense in terms of syntactic and semantic cues. Cunningham (1995) offers examples of activities that build and extend children’s crosschecking activities.

Word Structure Clues
There are many groups of letters that occur frequently in words. These are generally perceived by more mature readers as clusters of letters. Among these letter groups are prefixes (un-, re-, in-), suffixes (-ful, -ness, -est), and inflectional endings (-ed, -ing, -es). Common prefixes, suffixes, and inflectional endings should be pointed out to students. Being able to associate sounds with a cluster of letters leads to more rapid, efficient word identification.

Analogy Clues
As readers build an increasing store of words that they can recognize with little effort, they use the words they know to help them recognize words that are unfamiliar. For example, a child who has seen the word will many times and who knows the sound associated with the consonant f will probably have little difficulty recognizing the word fill. Building phonemic awareness for onsets and rimes builds a foundation for being able to identify simple words and syllables by analogy. Many teachers encourage developing readers to use analogy strategies by engaging students in word family (man, ran, pan) and initial consonant substitution (“What word would I have if I changed the min man to an r?”) activities. One clear advantage to the use of analogy strategies is that vowels, which can be variable in the sounds they represent, are much more stable within rimes (-eam).

The Role of Phonics

Phonics, another critical word-identification skill, is given more extensive treatment in this paper than other skills because of the controversy surrounding its use in programs for beginning reading instruction. Indeed, few topics in the field of education have engendered more emotional response. In 1955 Rudolph Flesch attracted national attention with the publication of Why Johnny Can’t Read, which suggested that virtually all reading problems in the United States were the result of a conspiracy on the part of educators and publishers to withhold phonics instruction from students. Widespread debate again took place in educational circles after the publication of Learning to Read: The Great Debate (Chall, 1967), a scholarly attempt, by a very respected educator, to review all published research on the effects of various approaches to teaching beginning reading. A much more recent research review (Adams, 1990) created a similar tumult. Clearly, the debate has not ended (Stahl, 1992).

Unfortunately, much of the debate seems based on emotion, not on the available evidence. Much of the furor seems a reflection of the indefensible position that beginning reading must be either phonics-based or meaning-based, often expressed as the “phonics vs. whole language” debate. No convincing evidence exists to show that a meaning-based, literature-based reading program cannot teach word identification, including the use of phonics skills. On the contrary, it seems that a meaning-based program can and must include systematic instruction in phonics, combined with other word-identification skills that form an effective, efficient, and balanced word-recognition strategy.

The position that instruction in phonics and word-identification skills is unnecessary because children can learn to read naturally without such instruction, simply by being read to and by being encouraged to read good books, is indefensible. Unquestionably, some children come to school knowing how to read, and a few children seem to learn with little systematic instruction. However, such early readers constitute only about one percent of the population of beginning readers (Pikulski & Tobin, 1988); it is unreasonable to generalize from them to the other 99 percent. Major literature reviews on the topic — such as those by Chall (1967); Johnson and Baumann (1984); Anderson et al. (1985); Adams (1990); Ehri (1991); and Mason, Herman, and Au (1991) — consistently conclude that early and systematic instruction in phonics skills results in superior reading achievement. After a careful review of the research, Anderson et al. concluded, in Becoming a Nation of Readers (1985): “Thus, the issue is no longer, as it was several decades ago, whether children should be taught phonics. The issues now are specific ones of just how it should be done” (p. 37).

The conclusion that Adams (1990) reached after her extensive research review offers some suggestions as to how phonics instruction should be approached:

    Phonics instruction by itself is not enough, however. To support skillful reading, the information in all the processors must be richly interconnected. To learn to read skillfully, children need practice in seeing and understanding decodable words in real reading situations and with connected text. The purpose of word-identification instruction is to establish paths from the print to spelling, speech, meaning, and context. This can best be done when phonics instruction is part of a reading program that provides ample practice in reading and writing. Encouraging children with connected text can also show them the importance of what they are learning and make the lessons in phonics relevant and sensible (pp. 93-94).

Adams is a cognitive psychologist, not a teacher of reading, so she sometimes uses terms, such as connected text, that seem a bit foreign. However, it seems reasonable to interpret the term connected text to mean “real books” or even “good literature.” Her conclusions call for a balance, teaching phonics in a meaningful context so that students can see the value of phonics for learning to read, along with abundant opportunities to practice the skills and strategies they are learning in real reading and writing activities.

Based on large-scale reviews of research related to beginning reading instruction (Adams, 1990; Anderson et al., 1985; Chall, 1967; Johnson & Baumann, 1984), it appears that if phonics instruction is to be maximally effective, it should be early, systematic, clear and direct, frequently practiced and applied, meaningful, and integrated with other word-identification skills into an effective word-recognition strategy. (See Tractenburg, 1990, for many examples of how to teach phonics meaningfully through children’s literature.)

Early
As noted earlier, familiarity with print and phonological awareness are necessary foundations for success in reading and should be taught and developed in kindergarten. Children can also begin to learn letter-sound associations while in kindergarten. Most of the useful phonics skills can be taught in first grade and then reviewed, refined, and extended in second grade and beyond.

Systematic
While there is no evidence to suggest a precise order in which letter-sound associations should be taught, a number of dimensions allow for systematic instruction. There should be a clear, logical progression from emerging literacy skills to actual phonics skills. When phonics skills are taught, consonant sounds should be taught first, since they are more reliable in their letter-sound associations. More complex consonant forms such as consonant clusters (br, gl, sw) and digraphs (sh, ch) can follow.

Vowel-sound associations are much more variable than consonants, so some stability can be achieved by introducing them as part of a phonogram or rime. Since words containing short vowel sounds are more frequent in beginning reading materials, it makes sense to introduce these first.

By second grade, much more attention can be focused on common, helpful prefixes and suffixes such as re-, un-, and -ful. At third grade and beyond, instruction should increasingly focus on decoding longer words and words whose meanings may be unknown.

Clear and Direct
While some precocious readers seem to become aware of letter-sound associations on their own as they read, most beginning readers do not. They need to be directly taught the sounds most frequently associated with letters. For example, after a shared reading of a Big Book version of Atsushi Komori’s Animal Mothers, it might be appropriate to introduce children to the sound associated with the letter m. Some children, after being introduced to the words me, my, and mother, might on their own infer that the letter m stands for the m sound. But research indicates that most students become independent readers faster if the teacher refers them directly to the word mother, has them listen for the beginning sound, and then has them practice saying the sound. Research (Adams, 1990) also suggests that children’s memory for letter-sound associations is enhanced if they are presented with a picture of an object in the shape of the letter (for example, a monster whose face is in the form of an m). Thus, beginning readers are given helpful information for learning to read rather than being expected to discover such information on their own.

Frequently Practiced and Applied
Most beginning readers need many opportunities to practice and apply the phonics skills they are learning to the reading and writing of words. They need to read stories that have words to which the phonics skills apply; they need to use the phonics skills as they write and spell words. They can also practice letter-sound associations in game-like activities after being taught the sound for a letter.

For the letter m, for instance, children might reread Animal Mothers, noting the sound at the beginning of words like mother and mouth. High-frequency words such as my and me might be pointed out as they are found in other selections that students read. Children can be reminded to think about the sound for m and other letter sounds as they label pictures or engage in beginning writing and use temporary spellings. They can work in pairs to sort pictures into those whose names do and do not begin with the sound for m.

Meaningful
The point was made earlier but bears repeating: If beginning readers have opportunities to apply their reading skills to many meaningful, informative, and enjoyable reading and writing experiences, they will see the utility of the phonics skills they are learning and grow as independent readers.

Building an Effective Strategy

Although a review of the available research makes it clear that failure to include phonics instruction as part of a beginning reading program results in poorer reading achievement, a full discussion of word recognition such as that by Ehri (1991) points out that phonics is but one aspect of word identification. Recognition of words whose printed form is not familiar to a reader may be facilitated through the use of meaning or context clues, through identification of word parts, or through analogy. Research (Johnson & Baumann, 1984) makes it clear that different readers use different combinations of these word-identification clues and that over-reliance on any one cuing system, such as meaning clues or phonics clues, results in problems for the reader.

Strategic combining of the various sources of information results in effective, efficient word identification. Consider the following example from Byron Barton’s retelling of the folktale “The Little Red Hen.” One sentence in this story is “Who will help me plant these seeds?” Assume that a child who is in mid-first grade can recognize, with little effort, the high-frequency words who, will, help, and me, but hasn’t seen the printed word plant before. If the child had been reading the previous page of text for meaning, he or she will remember that the hen had just found some seeds, and that she is now approaching her animal friends (the general context). The syntax (grammar) and the meaning (semantics) of the sentence cue the reader that the hen is asking (cued by who) for help to do something. By thinking about the sounds for pl, a, and nt, the probability is good that the child might think of the word plant. Indeed, the child may have perceived the pl, the an (since an is a frequent letter combination in beginning reading), and the t. Reading the rest of the sentence, plant the seeds makes good sense and serves to crosscheck the word plant as the correct choice.

Even this brief example shows the complexity of the process of identifying words and points to the fact that different children, based on experiences in reading and instruction, may use different routes (for example, an and t versusa and nt versus ant) to arrive at a word’s pronunciation. It also points out how phonic, syntactic, semantic, and word-part clues interact and strategically complement one another. As a result of frequent opportunities to identify words independently, children learn effective strategies for recognizing new words.

Relating Word Recognition to Writing and Spelling

Consider how word-recognition skills might be interrelated with listening, speaking, reading, writing, and spelling. Return to Byron Barton’s retelling of the “The Little Red Hen.” Certainly, if some children didn’t know the meaning of the word hen when they heard it, it would be appropriate to develop the meaning of that word orally. Some children might be familiar with the words chicken or mother chicken, but not the word hen. Through listening and speaking activities and the use of story illustrations, meaning can be developed for the text that is to be read.

One of the important criteria for effective phonics instruction is that it should be meaningful. Given the title “The Little Red Hen,” it would be very meaningful to focus word-recognition instruction on the short e vowel (found in bothred and hen) and phonograms with short e, certainly -ed and -en. In order to expand children’s skill base, common phonograms like -et and -est might be added.

The sound of the short e and the short e phonograms should be directly, explicitly, and clearly taught. Children can practice working with onsets and rimes by adding various consonants to the phonograms with short e. They can return to the story of “The Little Red Hen” to search for other short e words (application).

It would also be beneficial to encourage children to write sentences and words with short e and short e phonograms and to provide spelling instruction, practice, and assessment that focused on this same skill. In this way, the language arts can be interrelated so that they are mutually reinforcing.

Children who are developing independent reading skills do not rely solely on any one of these approaches to word identification but, instead, strategically use balanced combinations of these skills depending on the word to be identified and the context in which it is identified. The word undo, for example, lends itself to recognition of word parts, while the word skip does not.

Meaningful Practice for Word-Identification Skills

Unfortunately, in some settings the quality of instruction in word-recognition skills has become equated with the number of workbook pages or ditto sheets that students complete. Research indicates, however, that there is no relationship between the number of worksheets that students do and their achievement in reading (Anderson, Wilson, & Fielding, 1988). The best practice for helping students to gain proficiency in word-recognition skills is real reading and writing activities.

As beginning readers write using temporary spelling, they have many opportunities to use the word-identification skills they are developing. As they construct sentences of their own in writing, they are developing a greater sensitivity to meaning or context clues. As they utilize invented spellings, they are reviewing and applying what they know about letter-sound associations.

Children also have numerous opportunities to use their developing word-identification skills as they read on their own. By using some simple texts that have very limited word-recognition demands, teachers have an opportunity to check the degree of independence that students are developing in word recognition; the use of these books also has a very positive effect on students’ concepts of themselves as readers. Clearly, meaningful reading and writing activities are by far the best ways for students to practice and extend their word-recognition skills.

Word Identification Beyond Second Grade

Based upon recommendations in publications such as Becoming a Nation of Readers (Anderson et al., 1985) and common practice, the foundations for word recognition are laid down in kindergarten, the most important phonics skills and many word-part skills are taught in first grade, and the major word-identification skills discussed in this paper are reviewed and extended in second grade.

By third grade, however, word identification demands change as students begin to encounter greater numbers of longer, multisyllabic words. Even if students can arrive at a pronunciation of some words, their meaning remains unknown. At third grade and beyond, students can be introduced to some simple rules for dividing longer words into pronounceable units; however, syllable generalizations are very unreliable, so teachers need to model a flexible approach to breaking larger words into chunks. Students also need to be taught how to use a dictionary and its phonetic respellings to arrive at accurate pronunciations of unknown words.

Students need guidance in modifying their word-identification strategies to use the various context clues discussed earlier, along with word parts that carry meaning (prefixes, suffixes, base words), to infer the meaning of unknown words. Beyond the earliest grades, considerable emphasis must be placed on strategies that unlock word meaning as well as word pronunciation.

Conclusions

Virtually all of the debate about teaching word recognition seems unnecessary. The results of research, teaching experience, and common sense all point in the same direction — a need for balance and moderation. Reading divorced from meaning simply is not reading. Reading involves the construction of meaning, and that meaning is constructed from two primary sources — the print that has been created by an author and the various types of knowledge that the reader brings with him or her. A complete and effective program of beginning reading instruction must provide students with the strategies and skills needed to identify the printed words as a prerequisite to actively, dynamically, and critically constructing meaning.