CHAPTER 14

Psychology of Literacy and Literacy Instruction

MICHAEL PRESSLEY

333

EMERGENT LITERACY DURINGTHE PRESCHOOL YEARS 334Emergent Literacy 334Phonemic Awareness 336

FIRST GRADE AND THE PRIMARY YEARS 337Word Recognition 337Teaching Primary-Level Students to

Sound Out Words 338Reading Recovery 339Studies of Exceptional Primary-Level Teachers 340Summary 342

COMPREHENSION 343Fluent Word Recognition 343

Vocabulary 343Comprehension Strategies 343Summary 344

WRITING 344ENCOURAGING ADULT LITERACY 346

Basic, Word-Level Difficulties 346Comprehension Difficulties 346Writing Difficulties 347Summary 348

CLOSING COMMENTS 348REFERENCES 348

When first asked whether I could prepare a chapter summa-rizing literacy research, my initial response was that therequest was impossible. What came to mind immediatelywere the three volumes of the Handbook of Reading Re-search (Barr, Kamil, Mosenthal, & Pearson, 1991; Kamil,Mosenthal, Pearson, & Barr, 2000; Pearson, Barr, Kamil, &Mosenthal, 1984), the most prominent compendiums of read-ing research, which collectively include 3,000 pages to sum-marize just reading research (although some writing researchfound its way into those volumes).

Even more daunting than just the volume of research, how-ever, is its diversity. From a methodological perspective, thereare experimental and correlational traditions in literacy studies.In recent years, however, such traditional and quantitative ap-proaches have been supplanted largely by more qualitativemethods, including ethnographies (Florio-Ruane & McVee,2000), verbal protocol analyses (Afflerbach, 2000; Pressley &Afflerbach, 1995), narrative approaches (Alvermann, 2000),and single-subject designs (Neuman & McCormick, 2000).

Conceptually, literacy at one time was primarily seen froma behavioral perspective, with such behaviorism yielding tocognitivism in the 1970s and 1980s. Although there is stillmuch cognitive study of reading, sociocultural emphasis inthe field has been increasing, beginning in the 1990s and mov-ing into the twenty-first century (Gaffney & Anderson, 2000).

Literacy is also a decidedly international field of study;exciting ideas have come from Australia and New Zealand(Wilkinson, Freebody, & Elkins, 2000), the United Kingdom(Harrison, 2000), Latin America (Santana, 2000), and in-creasingly from former Iron Curtain countries (Meredith &Steele, 2000). Although much of literacy instruction hasbeen and remains focused on kindergarten through Grade 12instruction, in recent decades a great deal of work has beendone on literacy development during the preschool years(Yaden, Rowe, & McGillivray, 2000) as well as research ex-tending into the college years (Flippo & Caverly, 2000) andbeyond (Kirsch, Jungeblut, Jenkins, & Kolstad, 1993). Also,there has been a clear shift away from thinking about literacyas a development that occurs purely in the schools; it is nowconceived as more an acquisition that occurs in families,(Purcell-Gates, 2000) in the workplace, and in the larger, in-creasingly technological community (Reinking, McKenna,Labbo, & Kieffer, 1998).

Of course, one way to deal with this enormous and multi-dimensionally expanding literature would be to focus only onthe parts that are decidedly psychological because much of lit-eracy research was not carried out by psychologists and seemsrather far afield from psychological issues; in fact, that is atactic taken in this chapter. The downside of this approach isthat some of the most interesting and cutting-edge directions

334 Psychology of Literacy and Literacy Instruction

are neglected. Some ideas that might start psychologiststhinking about new directions they might pursue are not putbefore readers’ eyes. The serious scholar in literacy—or any-one who wants to have a broadly informed opinion—will (ata minimum) spend much time with the 3,000 Handbook pagesnow available at the beginning of this millennium.

Another tactic that I employ here is to focus on primary andsignificant issues and questions—ones that have been of con-cern for a very long time. This approach in particular makessense because it does lead to some answers—that is, a numberof important issues in reading and writing have been studiedlong enough that replicable findings have emerged. This em-phasis on replicable findings—on the surface at least—makesthis chapter consistent with the approach of the National Read-ing Panel (2000). I am inconsistent with the National ReadingPanel, however, in that I am willing to consider a greater di-versity of methods than that group was. That group generallylimited itself to experimental studies; it admitted only the oc-casional quasi-experimental study and distanced itself fromqualitative approaches entirely. This chapter certainly doespresent much coverage of outcomes produced in true experi-ments and approximations to experiments, but these out-comes are complemented by other scientific findings as well.In particular, descriptive methods, including ethnographies,have provided rich understandings about the complexities ofsome important instructional approaches—understandingsthat never would be produced in true experiments or repre-sented in the write-ups of experimental studies.

This chapter could have been organized in a number of dif-ferent ways; I have decided to organize this one along devel-opmental lines. In fact, there have been studies of literacydevelopment beginning in late infancy and proceeding throughadulthood. Of course, what develops varies with each develop-mental period; the development of general language compe-tencies is particularly critical during the preschool years.Although beginning reading instruction during the early ele-mentary school years focuses on the development of letter- andword-level competencies in reading and writing, this focuseventually gives way to the development of fluent reading as agoal and increasing concerns with comprehension and compo-sition in the later elementary and middle school grades. Byhigh school and college, much of the emphasis is on honing lit-eracy skills in the service of the learning demands of secondaryand postsecondary education. Researchers interested in adultliteracy have often focused on adults who did not develop lit-eracy competencies during the schooling years; such researchgenerally attempts to develop interventions to promote literacyin these populations, whose members often suffer socio-economic and personal disadvantages directly attributable totheir reading problems.

EMERGENT LITERACY DURINGTHE PRESCHOOL YEARS

What happens to children during the preschool years relatesto later literacy development. Many developmentalists inter-ested in literacy have focused on what is known as emergentliteracy, which is the development of the language skills un-derlying literacy through interactions with the social world.Other developmentalists who have been interested in chil-dren’s beginning letter-level and word-recognition skills havefocused more on a competency known as phonemic aware-ness, which is the awareness that words are composed ofsounds blended together.

Emergent Literacy

One of the more heavily researched topics by developmentalpsychologists is the nature of mother-infant attachment.When interactions between the principal caregiver and an in-fant are constructive and caring, the attachment that developscan be described as secure (Bowlby, 1969). In particular,when parents are responsive to the child and provide for itsneeds, secure attachment is more likely. The securely at-tached baby interacts with the world comfortably in the care-giver’s presence and responds favorably to the caregiver aftera period of caregiver absence.

Matas, Arend, and Sroufe (1978) made a fundamentallyimportant discovery. Children who experience secure at-tachment during infancy engage in more effective problemsolving with their parents during the preschool years. Whenparents are securely attached to their children, they are morelikely to provide appropriate degrees of support as their chil-dren attempt to solve problems (Frankel & Bates, 1990;Matas et al., 1978).

A related finding is that when parents and preschoolers aresecurely attached, they interact more productively in situa-tions involving literacy. Bus and van IJzendoorn (1988)observed both securely attached and insecurely attachedmother-child pairs as they watched Sesame Street together,read a picture book, and went through an alphabet book. Theinteractions involving securely attached parents and childrenwere much more positive than were the interactions betweeninsecurely attached parents and children. Securely attachedpreschoolers were more attentive and less easily distractedduring interactions, and much more literate activity was ob-served in the interactions of securely attached pairs comparedto those of insecurely attached pairs. Storybook reading wasmore intense with the secure pairs than with the insecurepairs; the secure parent-child pairs talked more about thestory than did the insecure pairs. An especially interesting

Emergent Literacy During the Preschool Years 335

finding was that securely attached parents and their 3-year-old children reported doing more reading together (Bus &van IJzendoorn, 1995).

That storybook reading brings greater rewards when at-tachment security is greater is an important finding becausehigh-quality storybook reading during the preschool yearsclearly promotes literacy development. There are clear corre-lations between the amount of storybook reading during thepreschool years and subsequent language development, chil-dren’s interest in reading, and their success as beginningreaders (Sulzby & Teale, 1991); this is sensible becausestorybook reading at its best is a rich verbal experience, withmuch questioning and answering by both reader and child.Storybook reading permits practice at working out meaningfrom words in text and pictures, as well as opportunities forthe child to practice relating ideas in stories to their own livesand the world as they understand it (Applebee & Langer, 1983;Cochran-Smith, 1984; Flood, 1977; Pelligrini, Perlmutter,Galda, & Brody, 1990; Roser & Martinez, 1985; Taylor &Strickland, 1986). As a child matures and gains experiencewith storybook reading, the conversations between reader andchild increase in complexity (Snow, 1983; Sulzby & Teale,1987). Older preschoolers who have had much storybookreading experience are much more attentive during such read-ing than are same-age peers who have had relatively little op-portunity to experience books with their parents or other adults(Bus & van IJzendoorn, 1988). Many correlational data sup-port the hypothesis that storybook reading is beneficial forchildren’s cognitive development—that it stimulates languagedevelopment and sets the stage for beginning reading.

This body of evidence in the context of storybook read-ing is complemented by other data substantiating strikingconnections between the richness of preschoolers’ verbalworlds and subsequent language development. One of themost ambitious and most cited analyses was made by Uni-versity of Kansas psychologists Hart and Risley (1995).They observed 42 families for 2.5 years, beginning in thesecond semester of a child’s life. During these observations,they recorded all actions and interactions. The first im-portant finding was that there were significant differencesbetween families in both the quality and the extensivenessof verbal interactions. The quality of interactions in terms ofcompleteness and complexity of language was greater inprofessional homes than in working-class homes, and lan-guage complexity in working-class homes was greater thanin welfare homes—that is, in homes of higher socioeco-nomic status, parents listened more to their children, theyasked their children to elaborate their comments more, andthey taught their children how to cope verbally when con-fronted with ideas that were challenging for the children to

communicate. Quantitatively, the differences in verbal inter-actions were really striking: Whereas a child in a profes-sional home might experience 4 million verbalizations ayear, a child in a welfare family could be exposed to only250,000 utterances. Did these vast differences in experiencetranslate into later performance differences? There was nodoubt about it; superior language was detected by age 3 inthe children raised in professional families compared tochildren in working-class and welfare families.

Of course, the problem with correlational data is thatcausality is never clear. Yes, it could be that the richer experi-ences promoted language development, or it could be thatmore verbal children stimulated richer language interactionsduring storybook reading and throughout their days. Fortu-nately, complementary experimental studies establish moredefinitively that high-quality verbal interactions result inlinguistic advances in children.

Grover Whitehurst and his colleagues (Whitehurst et al.,1988) hypothesized that if parents were coached in order toimprove their verbal interactions with their children duringstorybook reading, the language functioning of the childrenwould improve. Whitehurst et al. worked for a month withthe parents of 14 children between the ages of 1.5–3 years. Inparticular, the parents were taught to use more open-endedquestions as they read storybooks with their children; theywere also taught to ask more questions about the functionsand attributes of objects in stories. Whitehurst et al. (1988)also taught the parents to elaborate and expand on commentsmade by their children during reading. In short, the parentswere taught the tricks of the trade for stimulating productiveand verbally rich conversations with young children. Incontrast, parents and children in a control condition simplycontinued to read together for the month corresponding totreatment for the experimental participants.

First, the intervention worked in that it did increase theverbal complexity and extensiveness of communicationsbetween parents and children. Although experimental andcontrol parent-child interactions were similar before thestudy, the experimental group conversations during bookreading were much richer following the intervention. More-over, clear differences appeared in the language functioningof the experimental group children following the interven-tion, reflected by performance on standardized tests of psy-cholinguistic ability and vocabulary. These effects have beenreplicated several times, both by Whitehurst’s associates(Valdez-Menchaca & Whitehurst, 1992; Whitehurst et al.,1994) and by others (Crain-Thoresen & Dale, 1995; Dickinson& Smith, 1994; Lonigan, Anthony, & Burgess, 1995).

In short, evidence suggests that preschool verbal experi-ences promote language development, potentially in ways

336 Psychology of Literacy and Literacy Instruction

promoting subsequent development of reading. Whetherthese effects are great enough to inspire enthusiasm, how-ever, depends on the eye of the observing scientist; some sci-entists see large and important effects (Bus, van IJzendoorn,& Pelligrini, 1995; Dunning, Mason, & Stewart, 1994;Lonigan, 1994), whereas others who examine the same out-comes see small effects that might be explained away as dueto factors other than verbal stimulation (Scarborough &Dobrich, 1994). I tend to favor the former rather than the lat-ter conclusion; the experimental work of Whitehurst and hiscolleagues especially affects my thinking on this matter. Ingeneral, my optimism is consistent with the general optimismof the field that rich early language experiences affect lan-guage development in ways that should affect later readingdevelopment (Sulzby & Teale, 1991; Yaden et al., 2000).

Phonemic Awareness

In recent years, no prereading competency has received asmuch attention from researchers and practitioners as phone-mic awareness has. Understanding that words are composedof blended sounds seems essential for rapid progress inlearning letter-sound associations and learning to use thoseassociations to sound out words (Adams, 1990; Pennington,Groisser, & Welsh, 1993; Stanovich, 1986, 1988). This is notan all-or-none acquisition, however; Adams (1990) provides aconceptualization of phonemic awareness subcompetencies,listed as follows from most rudimentary to most advanced:(a) sensitivity to rhymes in words, (b) being able to spot wordsthat do not rhyme (e.g., picking the odd word out if given can,dan, sod), (c) being able to blend sounds to form words (e.g.,blending the sounds for M, short A, and T to produce mat),(d) being able to break words down into sound components(e.g., sounding out mat to indicate awareness of M, short A,and T sounds), and (e) being able to split off sounds fromwords (e.g., dropping the M sound from mat to say at; drop-ping the T sound from mat, producing ma).

Why is there such great interest in phonemic awareness?When phonemic awareness is low at ages 4–5, there is in-creased risk of difficulties in learning to read and spell (Bowey,1995; Griffith, 1991; Näsland & Schneider, 1996; Pratt &Brady, 1988; Shaywitz, 1996; Stuart & Masterson, 1992).Perhaps the best-known study establishing linkage betweenphonemic awareness at the end of the preschool years and laterreading achievement was Juel (1988). She studied a sample ofchildren as they progressed from first through fourth grade.Problems in reading during Grade 1 predicted problems inreading at Grade 4—that is, problem readers in first grade donot just learn to read when they are ready! Rather, they never

seem to learn to read as well as do children who were strongreaders in Grade 1. More important to this discussion is thatlow phonemic awareness in Grade 1 predicted poor readingperformance in Grade 4, a result generally consistent withother demonstrations that low phonemic awareness between4 and 6 years of age predict later reading problems (Bowey,1995; Griffith, 1991; Näsland & Schneider, 1996; Pratt &Brady, 1988; Shaywitz, 1966; Stuart & Masterson, 1992).

Given that phonological awareness is so critical in learningto read, it is fortunate that phonological awareness has proventeachable; when taught, it influences reading performancepositively. Perhaps the best known demonstration of the po-tency of phonemic awareness instruction is that provided byBradley and Bryant (1983). They provided 5- and 6-year-oldswith 2 years of experience categorizing words on the basis oftheir sounds, including practice doing so with beginning, mid-dle, and ending sounds. Thus, given the words hen, men, andhat with the request to categorize on the basis of initial sound,hen and hat went together; in contrast, hen and men was thecorrect answer when the children were asked to categorize onthe basis of middle or ending sound. The students in the studyfirst read pictures and made their choices on the basis ofsounds alone; then they were transferred to words and couldmake their choices on the basis of letter and orthographicfeatures as well as sounds.

The training made a substantial impact on reading mea-sured immediately after training, relative to a control condi-tion in which students made judgments about the conceptualcategory membership of words (e.g., identifying that cat, rat,and bat go together as animals). Even more impressive wasthat the trained participants outperformed control participantsin reading 5 years after the training study took place (Bradley,1989; Bradley & Bryant, 1991).

Bradley and Bryant’s work was the first of a number ofstudies establishing that phonemic awareness could be de-veloped through instruction and influence reading perfor-mance (Ball & Blachman, 1988, 1991; Barker & Torgesen,1995; Blachman, 1991; Byrne & Fielding-Barnsley, 1991,1993, 1995; Cunningham, 1990; Foster, Erickson, Foster,Brinkman, & Torgesen, 1994; Lie, 1991; Lundberg, Frost, &Peterson, 1988; O’Connor, Jenkins, & Slocum, 1995; Tangel& Blachman, 1992, 1995; Vellutino & Scanlon, 1987;Williams, 1980; Wise & Olson, 1995). Although the instruc-tional procedures varied somewhat from study to study, ingeneral, phonemic awareness training has included at leastseveral months of exercises requiring young children to attendto the component sounds of words, categorizing and dis-criminating words on the basis of sound features. Thus, some-times children were asked to tap out the syllables of words,

First Grade and the Primary Years 337

sometimes asked to say the word with the last sound deleted,and sometimes requested to identify the odd word out whenone does not share some sound with other words in a group.

Bus and van IJzendoorn (1999) provided especially com-plete and analytical review of the phonemic awareness in-structional data. Collapsing data over 32 research reports, allof which were generated by U.S. investigators, Bus and vanIJzendoorn (1999) concluded that there was a moderate rela-tionship between phonemic awareness instruction and laterreading. When long-term effects (i.e., 6 months or more fol-lowing training) were considered, however, the phonemicawareness instruction had less of an impact on reading—asmall impact at best. Thus, although delayed effects of phone-mic awareness training can be detected, they are not huge.

All scientifically oriented reviewers of the early readingliterature have concluded that phonemic awareness is impor-tant as part of learning to read (e.g., Adams, 1990; Adams,Treiman, & Pressley, 1998; Goswami, 2000; National Read-ing Panel, 2000; Snow, Burns, & Griffin, 1998). The availablecorrelational and experimental data converge on the conclu-sion that phonemic awareness is probably an important pre-requisite for learning to read words. After all, if a child doesnot understand that words are composed of sounds blendedtogether, why would reading instruction emphasizing thecomponent sounds of words make any sense to the child? Ofcourse, the answer is that it would not, which explains whyphonemic awareness is so critical for a child to learn to read(e.g., Fox & Routh, 1975). Acquiring phonemic awareness isjust a start on word recognition competence, which is a criti-cal task during the primary grades.

In summary, much progress in literacy development canand does occur before Grade 1, which has traditionally beenviewed as the point of schooling for beginning reading in-struction. Much of it is informal—the learning of language ina language-rich environment that can include activities suchas storybook reading with adults. Increasingly, high-qualitykindergarten programs include activities explicitly intendedto develop phonemic awareness.

FIRST GRADE AND THE PRIMARY YEARS

There has been tremendous debate in the past quarter centuryabout the best approach to primary-grades reading educa-tion. This debate somewhat reflects a much longer debate (i.e.,one occurring over centuries to millennia) about the natureof beginning reading instruction (see Pressley, Allington,Wharton-McDonald, Block, & Morrow, 2001). In recentyears, at one extreme have been those who have advocated an

approach known as whole language, which posits that chil-dren should be immersed in holistic reading and writing tasksfrom the very start of schooling—that is, reading trade booksand composing their own stories. At the other extreme arethose who argue that skills should be developed first. Theskills-first advocates particularly favor phonics as an ap-proach to developing word-recognition abilities; they arguethat if students learn letter-sound associations and how toblend the component sounds in words to recognize words,their word recognition will be more accurate and more certain.

Word Recognition

Even preschoolers can read some words, such as McDonald’swhen in the context of the company’s logo, Coca-Cola whenencountered on a bottle or aluminum can, and Yankees whenscripted across a ballplayer’s chest. Young children learn torecognize such logographs from their day-to-day experiences.When presented the words McDonald’s, Coca-Cola, andYankees out of their familiar contexts, preliterate childrencannot read them. Even so, encountering words as logographssomehow seems to make it easier for preschoolers to learnwords out of context. When Cronin, Farrell, and Delaney(1995) taught preschoolers words as sight words, previouslyencountered logographs were learned more easily than werecontrol words never encountered as logographs. At best, how-ever, logographic reading is just a start on word-recognitionskills and is very different from most of word recognition.

Well before children can sound out words using all the let-ters of a word, they sometimes can read words based on a fewletters, a process Ehri (1991) referred to as phonetic cue read-ing. Thus, as a little boy, I learned the very long word ele-mentary because I encountered it often during first grade. Asa consequence, I could read elementary wherever I encoun-tered the word. The problem was that I was reading the wordbased on a couple of cues (probably the beginning e and thefact that it was a long word) shared by other words. Thus, forquite a while, I thought that label on the escape hatch in theschool bus was labeled elementary door, when in fact it wasan emergency door! Such mistakes are common in childrenwho are 5–6 years old (Ehri & Wilce, 1987a, 1987b; Gilbert,Spring, & Sassenrath, 1977; Seymour & Elder, 1986).

Many children do reach the kindergarten doors knowingthe alphabet. One reason is that as a society, we decided toteach the alphabet to preschoolers—for example, through ef-forts such as those in Sesame Street; it is clear from the earliestevaluations that such environmental enrichment did affect ac-quisition of alphabetic knowledge (e.g., Anderson & Collins,1988; Ball & Bogatz, 1970; Bogatz & Ball, 1971). It is now

338 Psychology of Literacy and Literacy Instruction

known that Sesame Street contributes to alphabetic learningover and above the contributions made by family and others(Rice, Huston, Truglio, & Wright, 1990).

Knowing letter names and letter-sound associations alonedoes not result in word recognition competence, however.Children must also learn the common blends (e.g., dr, bl) anddigraphs (e.g., sh, ch). In general, primary education includeslots of repetition of the common letter-sound associations,blends, and digraphs—for example, through repeated readingof stories filled with high-frequency words. Walk into anyGrade 1 classroom: It will be filled with many single-syllablewords, including lists of words featuring the common di-graphs and blends. Word families also will be prominent (e.g.,beak, peak, leak). Grade 1 teachers spend a lot of time model-ing for their students how to sound out words by blending thecomponent sounds in words and using common chunks; theyalso spend a lot of time encouraging students to sound outwords on their own, including doing so to write words in theircompositions (Wharton-McDonald, Pressley, & Hampston,1998).

The students most likely to make rapid progress in learn-ing to sound out words are those who already have phonemicawareness and know their letter-sound associations (Tunmer,Herriman, & Nesdale, 1988). Even so, a large body of evi-dence indicates that teaching students to sound out words byblending components’ sounds is better than alternative ap-proaches with respect to development of word-recognitionskills.

Teaching Primary-Level Students to Sound Out Words

One of the most important twentieth-century contributions toreading research was Jeanne Chall’s (1967) Learning toRead: The Great Debate. After reviewing all of the evidencethen available, Chall concluded that the best way to teach be-ginning reading was to teach students explicitly to sound outwords—that is, she felt that early reading instruction shouldfocus on teaching letter-sound associations and the blendingof letter sounds to recognize words, an approach she referredto as synthetic phonics. Based on the available research,Chall concluded that synthetic phonics was superior to otherapproaches regardless of the ability level of the child, al-though synthetic phonics seemed to be especially beneficialto lower-ability children. After the publication of the firstedition of the Chall book, there was a flurry of laboratorystudies of phonics instruction, and most researchers foundsynthetic phonics to be better than alternatives (Chall, 1983,Table I-2, pp. 18–20).

The next book-length treatment of the scientific founda-tions of beginning reading instruction was Marilyn Adams’

(1990) Beginning to Read. By the time of that publication, agreat deal of conceptualization and analysis of beginningreading had occurred. Adams reviewed for her readers the ev-idence permitting the conclusion that phonemic awareness isa critical prerequisite to word recognition. So was acquisitionof the alphabetic principle, which is the understanding thatthe sounds in words are represented by letters. Researchersinterested in visual perceptual development had made thecase that children gradually acquire understanding of thedistinctive visual features of words, gradually learning todiscriminate Rs from Bs and Vs from Ws (Gibson, Gibson,Pick, & Osser, 1962; Gibson & Levin, 1975). Consistent withChall (1967, 1983), Adams also concluded that instructionin synthetic phonics promoted beginning word-recognitionskills.

Since Adams’ (1990) book, a number of demonstrationshave shown that intensive instruction in synthetic phonicshelps beginning struggling readers. For example, Foorman,Francis, Novy, and Liberman (1991) studied urban first-gradestudents who were enrolled either in a program emphasizingsynthetic phonics or in a program downplaying phonics inword recognition in favor of whole language. By the end ofthe year, the students in the synthetic phonics program werereading and spelling words better than were students in theother program. Foorman, Francis, Fletcher, Schatschneider,and Mehta (1998) reported a similar outcome; a program em-phasizing synthetic phonics produced better reading after ayear of instruction than did three alternatives that did notprovide systematic phonics instruction. Maureen Lovetttreats 9- to 13-year-olds who are experiencing severe readingproblems; she and her colleagues have presented consider-able evidence that systematic teaching of synthetic pho-nics improves the reading of such children (Lovett, Ransby,Hardwick, Johns, & Donaldson, 1989; Lovett et al., 1994).Similar results have been produced in a number of well-controlled studies (Alexander, Anderson, Heilman, Voeller, &Torgesen, 1991; Manis, Custodio, & Szeszulski, 1993; Olson,Wise, Johnson, & Ring, 1997; Torgesen et al., 1996; Vellutinoet al., 1996), permitting the clear conclusion that intensive(i.e., one-on-one or one teacher to a few students) syntheticphonics instruction can help struggling beginning readers.

In recent years, a popular alternative to synthetic phonicshas been teaching students to decode words by recognizingcommon chunks (or rimes) in them (e.g., tight, light, andsight include the -ight chunk). Use of such chunks to decode,however, requires that students know something about lettersand sounds and about blending (Ehri & Robbins, 1992;Peterson & Haines, 1992) because word recognition requiresblending the sounds produced by individual letters with thesounds produced by a chunk (e.g., tight involves blending the

First Grade and the Primary Years 339

t and ight sounds; Bruck & Treiman, 1992). In evaluations todate, when struggling readers have been taught to use com-mon word chunks to decode words they have not seen before,this approach has been successful relative to controls who re-ceive conventional instruction not emphasizing word recog-nition (e.g., Lovett et al., 2000). Students taught to use wordchunks have fared as well after several months of such in-struction as students taught to use synthetic phonics (Walton,Walton, & Felton, 2001). Thus, available data indicate thatyoung children can learn to use both chunks and the soundingout of individual sounds as they learn to recognize words(Goswami, 2000). Perhaps most striking in the Walton et al.(2001) report was that weak first-grade readers tutored eitherto use chunks to decode or to sound out words using phonicscaught up with good first-grade readers who continued toreceive conventional reading instruction that emphasizedneither use of chunks during reading nor synthetic phonics.These are powerful procedures for remediating the mostsalient problem in beginning reading, which is difficulty inrecognizing words. Even so, they have not been the mostpopular procedures in recent years for remediating troubledbeginning readers.

Reading Recovery

Reading Recovery™ is a widely disseminated approach tobeginning reading remediation (Lyons, Pinnell, & DeFord,1993). Typically, students in Reading Recovery are in Grade 1and making slow progress in learning to read in the regularclassroom. The intervention supplements classroom instruc-tion and involves daily one-teacher-to-one-child lessons;each lesson lasts about a half hour, and lessons continue foras long as a semester.

A typical Reading Recovery lesson involves a series ofliteracy tasks (Clay, 1993; Lyons et al., 1993). First, the childreads a familiar book aloud to the teacher. Often, this taskis followed by reading of another book that is not quite asfamiliar—one introduced to the child the day before. Duringthis reading of yesterday’s new book, the teacher makes arunning record, noting what the child does well during read-ing and recording errors. Information gleaned by the teacheras the child reads is used to make instructional decisions, andthe teacher attempts to determine the processes being used bythe child during reading.

For example, when the child makes an error during reading,the teacher notes whether the child relied on meaning clues toguess the word, syntactic cues, or visual cues; this analysis ofprocessing informs instructional decision making. Thus, if thechild misreads bit as sit, the teacher might focus the child’sattention on the it chunk in the word and prompt the child

to blend the s sound and the sound made by it. After the read-ing, the teacher continues the lesson by asking the student toidentify plastic letters or by having the child make and breakwords with plastic letters. For example, the teacher mightfocus on words with the it chunk, prompting the child to formnew words with the it chunk, using magnetic letters to con-struct the words (e.g., bit, fit, mit, pit, etc.). Then the childmight break these words to see that bit is b plus it, fit is f plusit, and so on. Then the child might do some writing in responseto the story, with the teacher providing assistance as the childworks on writing (e.g., writing a sentence about the story, suchas The dog sits down). During writing, the teacher encouragesthe child to listen for the sounds in words in order to spellout the word in writing. Then, the teacher writes the sentenceconstructed by the student on a paper strip, using conven-tional spelling to do so, then cutting up the strip into individ-ual words. The child reassembles the sentence and reads itfor the teacher. The Reading Recovery lesson concludes withthe teacher’s introducing a new book to the student, whoattempts to read the book for the teacher. Homework involvestaking home the books read during the lesson and readingthem to a parent.

Reading Recovery is all about children’s reading strategiesand the teaching of strategies to struggling readers (Clay,1993; Lyons et al., 1993). Throughout a Reading Recoverylesson, the teacher attempts to determine how the child is pro-cessing during reading and writing and the what reading andwriting strategies are used by the child. Specifically, theteacher attempts to determine the reader’s directions of pro-cessing (i.e., whether reading is left to right, from the topof the page down; whether writing is left to right, from thetop to bottom of the page). The teacher also attempts to dis-cern whether the child is processing individual words in asequence—for example, whether the child is noticing thespaces between words read and putting spaces between wordswritten. The teacher notes whether the child is monitoringreading and writing—for example, going back and attemptingto reread a misread word or asking for help during writing re-garding spelling an unknown word. The Reading Recoveryteacher focuses on the nature of reading errors—whether theyreflect attempts to sound out a word, a reliance on meaning orsyntactic cues, or dependence on visual similarity of the at-tempted word with a word known by the child. In short, theassumption in Reading Recovery is that the struggling readeris attempting to problem-solve when reading and writing,and that the child’s errors are particularly revealing about heror his reading and writing strategies.

The teacher’s knowledge of the child’s strategies is used toguide teaching, and the teacher’s role is to stimulate use ofstrategies during reading and writing that are more effective

340 Psychology of Literacy and Literacy Instruction

than the ones currently being used by the young reader (Clay,1993; Lyons et al., 1993). For example, to encourage the de-velopment of directionality, the teacher prompts the child toRead it with your finger, pointing to each word as it is encoun-tered in text. At first, this can require the teacher actually hold-ing and directing the child’s hand, but eventually the childinternalizes the left-to-right and top-to-bottom movementsduring reading. In order to increase the child’s understandingof the concept of individual words, the teacher prompts thechild to write words with spaces between them, using the strat-egy of putting a finger space between written words. Theteacher teaches the child to sound out words by saying themslowly, breaking words into discrete sounds (e.g., cat into theC, short A, and T sounds). Consistent with the demonstrationby Iversen and Tunmer (1993) that Reading Recovery is moreeffective when it includes systematic teaching of chunks andhow they can be blended with letter sounds as part of reading,Reading Recovery now includes more making and breakingof words that share chunks (e.g., bake, cake, lake, make, take,etc.) to highlight blending of individual sounds and spellingpatterns. The Reading Recovery teacher also teaches theyoung reader to check decodings by determining whether thereading of a word makes sense in that semantic context. Inshort, the Reading Recovery teacher instructs the strugglingreaders in the strategies that effective young readers use; theultimate goal of Reading Recovery is the development ofreaders who use effective reading processes in a self-regulatedfashion (Clay, 1991).

As is the case for many forms of strategy instruction(Duffy et al., 1987; Palincsar & Brown, 1984; Pearson &Gallagher, 1983; Pressley, El-Dinary, et al., 1992), there is agradual release of responsibility during Reading Recovery;the teacher is more directive and explicit at first, and the childtakes over as lessons proceed and competence develops—thatis, the strategy instruction is scaffolded (Wood, Bruner, &Ross, 1976). The teacher provides just enough support so thatthe child can complete the task; then the teacher reduces thesupport as the child becomes more competent and able to as-sume greater responsibility for reading. Of course, the intentof such an instructional approach is to develop self-regulationin the child—first by permitting the child to tackle a task thatis beyond her or him and then by allowing self-controlledfunctioning as the child becomes equal to the task.

Also, as is the case with many forms of strategy instruction,evidence indicates that scaffolded teaching of processes wellmatched to the target task is effective—that is, a large propor-tion of children who experience Reading Recovery improve asreaders, and improvement is greater than that occurring whencomparable children do not receive Reading Recovery, at leastwhen reading achievement is measured immediately afterReading Recovery occurs (see Pinnell, 1997). An important

distinction is between Reading Recovery students who gradu-ate and those who do not make enough progress in the pro-gram to graduate—that is, Reading Recovery does not alwayswork; when it does work, however, it seems to produce sub-stantial improvement (Elbaum, Vaughn, Hughes, & Moody,2000). As is the case with many early childhood interventions,if students are simply returned to the classroom without addi-tional support, however, the advantages of Reading Recoveryfade, and such Reading Recovery students are often not dis-cernibly different in reading achievement measured severalyears after the completion of the treatment (Hiebert, 1994).

Studies of Exceptional Primary-Level Teachers

Phonemic awareness instruction, phonics, and ReadingRecovery are theory-driven educational interventions—thatis, based on theory, researchers devised instruction they feltwould promote beginning reading, and their instructional stud-ies served as tests of the theories that inspired the interven-tions. There is another way to discover effective instruction,however, which is to find very good reading teachers and not-so-good ones and document what occurs in effective versus in-effective classrooms. Pressley and his colleagues have doneexactly that with respect to Grade 1 in particular.

In both Wharton-McDonald et al. (1998) and Pressley et al.(2001), the researchers observed first-grade classrooms overthe course of an academic year. In some classrooms, engage-ment and achievement was better than in other classrooms.For example, in some classrooms, a higher proportion of stu-dents were reading more advanced books than was observedin more typical classrooms; in some classrooms, studentswere writing longer, more coherent, and more mechanicallyimpressive stories (i.e., stories with sentences capitalized,punctuation, correctly spelled high-frequency words, sensibleinvented spellings of lower-frequency words) than were stu-dents in other classrooms. Most striking was that the moreengaged classrooms also tended to be the ones with more ad-vanced reading and better writing.

What went on in the really impressive classrooms?

• There was a lot of teaching of skills, and this instructionwas very consistent. Much of this instruction was in re-sponse to student needs, however, with many minilessonson skills.

• Fine literature was emphasized; students read excellentliterature and heard it during teacher read-alouds.

• The students did a lot of reading and writing.

• Assignments were matched to students’ abilities, and thedemands were gradually increased as students improved.Such matching requires different assignments for differ-ent students (e.g., one student being urged to write a

First Grade and the Primary Years 341

two-page story and another a two-sentence story, with thedemand in each case for a little more than the child pro-duced previously).

• Self-regulation was encouraged; the message was consis-tent that students were to make choices for themselves andwere to keep themselves on task.

• Strong connections were made across the curriculum; sci-ence and social studies occurred in the context of readingand writing, and science and social studies units werefilled with good literature and composing.

• The class was positive and very reinforcing, with muchcooperation between students and between teachers, otheradults, and students.

• The teacher’s classroom management was so good that itwas hardly noticeable at all, with little apparent need fordisciplining of students.

How different the effective classrooms were really be-came apparent in analyses that contrasted the effective andineffective classrooms explicitly—analyses designed to iden-tify what was very different in the excellent compared to thenot-so-excellent classrooms:

• Many more skills were covered during every hour of in-struction in the most effective compared to the least effec-tive classrooms.

• Word-recognition instruction involved teaching multiplestrategies (i.e., using phonics, noting word parts, lookingat the whole word, using picture clues, using semanticcontext information provided earlier in the sentence orstory, using syntactic cues).

• Comprehension strategies (e.g., making predictions, men-tal imagery, summarizing) were explicitly taught.

• Students were taught to self-regulate.

• Students were taught to plan, draft, and revise as part ofwriting.

• Extensive scaffolding (i.e., coaching) took place duringwriting—for example, with respect to spelling and elabo-rating on meanings in text.

• Printed prompts for the writing process (e.g., a cardabout what needs to be checked as part of revision) wereavailable.

• By the end of the year, high demands to use writingconventions (e.g., capitalizing, using punctuation marks,spelling of high frequency words) were placed on students.

• Tasks were designed so that students spend more timedoing academically rich processing (i.e., reading andwriting) and relatively little time on nonacademic process-ing (e.g., illustrating a story).

• The class wrote big books, which were on display.

In short, excellent first-grade classrooms are very busy—filled with teaching of skills and demands but also filled withsupport and opportunities for rich intellectual experiences.Although phonics is taught as skills advocates would have itbe taught, it is only part of an enormously complex curricu-lum enterprise that includes many holistic experiences—thatis, systematic skills instruction does not happen first beforegetting to literature and writing in effective first-grade class-rooms; rather, skills are learned largely in the context ofreading literature and writing. Although literature and writ-ing are emphasized as the whole language theorists wouldhave it, holistic experiences are constantly intermixed withthe systematic and opportunistic instruction of specific skills,and skills were much more an emphasis than many wholelanguage theorists would consider appropriate. Excellentprimary-level classrooms—ones in which growth in readingand writing is high—cannot be reduced to a very few in-structional practices; rather, they are a complex, articulatedmix of practices and activities.

The most recent work of Pressley and colleagues (Raphael,Bogner, Pressley, Masters, & Steinhofer, 2000) has taken a de-cided psychological turn. They observed first-grade class-rooms with the goal of determining how excellent first-gradeteachers motivate their students to participate in literacy-promoting activities. In part, this research was stimulatedby the engagement perspective, which posits that literacyachievement depends on instruction that motivates literacyengagement (e.g., Guthrie & Alvermann, 1999). Such engage-ment is promoted when classrooms emphasize learning ratherthan student competition, meaningful interactions betweenstudents and ideas, student autonomy and self-regulation,interesting content, teaching of useful strategies, praise con-tingent on literacy engagement and progress, teacher involve-ment with students, and evaluations that make sense tostudents (Guthrie & Wigfield, 2000). From this perspective,it was expected that classrooms loaded with mechanismspromoting literacy engagement in fact would be classroomshigh in student literacy engagement.

What Raphael et al. (2000) found was that first-gradeteachers who had students who were highly engaged in read-ing and writing constructed classrooms filled with positivemotivational mechanisms compared to teachers overseeingclassrooms in which engagement was not as certain. Thus, inclassrooms where engagement was high, the following moti-vational mechanisms were observed:

• Much cooperative learning took place.

• Individual accountability (i.e., students were rewarded fordoing well and held accountable when they did not) wasdemonstrated.

• As they worked, students received much coaching.

342 Psychology of Literacy and Literacy Instruction

• Strong library connections were maintained.

• Students were encouraged to be autonomous and givenchoices.

• The teacher was gentle, caring, and inviting.

• Much one-to-one interaction took place between teachersand students.

• Strong home-school connections were maintained.

• Many opportunistic minilessons were taught.

• Deep connections with students were maintained.

• Appropriate risk taking was supported.

• The classroom was fun.

• Strong connections to other classes in the school weremaintained.

• The teacher encouraged creative and independent thinking.

• The teacher encouraged rich and detailed learning.

• The class took a clear positive tone.

• Assignments were appropriately challenging.

• Students produced meaningful products (e.g., stories).

• Depth in coverage was favored over breadth in coverage.

• Assignments and units matched student interests.

• Abstract content was made more personal and concrete.

• The teacher encouraged curiosity and suspense.

• Learning objectives were clear.

• Praise and feedback were effective.

• The teacher modeled interest and enthusiasm.

• The teacher modeled thinking and problem–solving.

• The teacher communicated that academic tasks deserveintense attention.

• The teacher inserted novel material into instruction.

• The teacher provided clear directions.

• The teacher made apparent the relevance of learning toreal life.

• The teacher encouraged persistence.

• The teacher encouraged cognitive conflict.

• The teacher communicated a wide range of strategies foraccomplishing academic tasks.

• The teacher encouraged self-reinforcement by studentswhen they did well.

• The teacher provided immediate feedback.

• The teacher urged students to try hard.

• The teacher expressed confidence in students.

• The teacher encouraged students to attribute their successesto hard work and their failures to a need to work harder.

• The teacher had realistic ambitions and goals for students.

• The teacher encouraged students to think they can getsmarter by working hard on school work.

• Classroom management was good.

• The teacher provided rewards that stimulate students pos-itively (e.g., gift book).

• The teacher monitored the whole class.

• The teacher monitored individual students carefully.

In short, consistent with the engagement perspective, engag-ing classrooms were filled with positive motivational mecha-nisms; less engaging classrooms showed many fewer of thesemechanisms.

That is not to say that the teachers in the less engagingclassrooms did not try to motivate their students. In fact, theydid. In less engaging classrooms, however, teachers weremuch more likely than were those in the more engagingclassrooms to use negative approaches to motivations—emphasizing competition between students; giving studentstasks that were very easy, boring, or both; providing negativefeedback; making students aware of their failures; scapegoat-ing students; threatening students; and punishing students.Such negative approaches to motivation were almost neverobserved in the most engaged classrooms.

Summary

Many psychologists have been at the forefront of efforts to de-velop effective beginning reading instruction. One reason isthat learning to read is a salient event in the life of the devel-oping child—an event that is decidedly psychological in na-ture. There are huge cognitive conceptions to acquire, such asphonemic awareness and the alphabetic principle, which de-velop in the context of much associative learning (i.e., learn-ing letter-sound associations and chunk-sound associations)and development of subtle perceptual discriminations (e.g.,the visual identity of each letter, both upper- and lowercaseversions). An important hypothesis among psychologists isthat beginning reading skills can be taught directly. In fact,quite a bit of evidence has accumulated making clear that di-rect teaching of synthetic phonics does in fact make develop-ment of word-recognition skills more certain. In recent years,there have also been validations of teaching involving empha-sis on word chunks and blending of word parts in sounding outof words; this approach is now part of the prominent remedialapproach to beginning reading known as Reading Recovery.Although Reading Recovery teachers are highly trained fortheir work, it is auspicious that even college students can tutorbeginning struggling readers with substantial gains (Elbaumet al., 2000) because the need in the nation for tutoring

Comprehension 343

primary-level readers in beginning reading skills is very, verygreat. This work on primary-level reading is an excellentexample of how psychological theory and research can informmeaningful educational practice.

That said, the psychological theory related to beginningword recognition seems simple relative to the complexity ofexcellent first-grade instruction that can be observed in many(although certainly not all) classrooms. Although instructionto promote phonemic awareness, phonics, and word recogni-tion in general is prominent in such classrooms, it occurs in acontext that attends to student motivation and excellent holis-tic experiences, including the reading of much good literatureand extensive writing.

COMPREHENSION

Developing students who can understand what they read isa primary goal of reading instruction. This goal should beprominent beginning with the introduction to stories andbooks in the preschool years. Even so, it definitely becomes amore prominent purpose for literacy instruction during themiddle and upper elementary grades, with a number of aspectsof reading that can be stimulated to improve comprehension(Pressley, 2000).

Fluent Word Recognition

When a reader cannot decode a word, it is impossible for thereader to understand it (Adams, 1990; Metsala & Ehri, 1998;Pressley, 1998, chap. 6). When young readers are first learningto recognize words—either by blending individual sounds orblending sounds and chunks—such decoding takes a lot ofeffort, and hence it consumes much of the reader’s attention.This situation is a problem because human beings can only at-tend to a limited number of tasks at once (Miller, 1956). If thatattention is totally devoted to word recognition, nothing is leftover for comprehending the word, let alone the higher-orderideas encoded in sentences, paragraphs, and whole texts(LaBerge & Samuels, 1974). Thus, for comprehension to behigh, not only must young readers learn how to recognizewords, but they also must become fluent in word recognition(National Reading Panel, 2000). Although not every analysishas confirmed that comprehension improves as word re-cognition fluency improves (Fleisher, Jenkins, & Pany, 1979;Samuels, Dahl, & Archwamety, 1974; Yuill & Oakhill, 1988,1991), some recent and especially well-done analyses haveproduced data in which fluency and comprehension havecovaried (Breznitz, 1997a, 1997b; Tan & Nicholson, 1997).Unfortunately, little is known about how to develop fluency

beyond the fact that fluency generally increases with addi-tional practice in reading (National Reading Panel, 2000).

Vocabulary

People with more extensive vocabularies understand text bet-ter than do individuals with less well-developed vocabularies(Anderson & Freebody, 1981; Nagy, Anderson, & Herman,1987). In fact, some experimental studies have even suggestedthat the development of vocabulary knowledge resulted in im-proved comprehension (Beck, Perfetti, & McKeown, 1982;McKeown, Beck, Omanson, & Perfetti, 1983; McKeown,Beck, Omanson, & Pople, 1985). Although vocabulary isoften taught extensively in school, for the most part, vocabu-lary is acquired incidentally as a by-product of encounteringwords in text and in real-world interactions (Sternberg, 1987).There have been a number of demonstrations that vocabu-lary knowledge increases with how much a reader reads(Dickinson & Smith, 1994; Elley, 1989; Fleisher et al., 1979;Pellegrini, Galda, Perlmutter, & Jones, 1994; Robbins &Ehri, 1994; Rosenhouse, Feitelson, Kita, & Goldstein, 1997;Valdez-Menchaca & Whitehurst, 1992; Whitehurst et al.,1988).

Comprehension Strategies

When mature readers are asked to think aloud as they read,they report using many strategies before, during, and afterthey read as part of processing the text. These processes in-clude predicting what will be in the text based on prior knowl-edge and ideas encountered in the text already, constructingmental images of ideas expressed in the text, seeking clarifi-cation when confused, summarizing the text, and thinkingabout how ideas in the text might be used later (Pressley &Afflerbach, 1995). Because good readers consciously usesuch strategies, it was sensible to teach such strategies toyoung readers, with the hypothesis that the reading compre-hension of young readers would improve following suchinstruction; that is exactly what happens.

There were many studies in the 1970s and 1980s in whicha particular strategic process was taught to students in theelementary grades with comprehension and memory of textsthat were read and then tested. These studies included those inwhich students were encouraged to activate prior knowledge(Levin & Pressley, 1981), generate questions as they read(Rosenshine & Trapman, 1992), construct mental images(Gambrell & Bales, 1986; Gambrell & Jawitz, 1993; Pressley,1976), summarize (Armbruster, Anderson, & Ostertag, 1987;Bean & Steenwyk, 1984; Berkowitz, 1986; Brown & Day,1983; Brown, Day, & Jones, 1983; Taylor, 1982; Taylor &

344 Psychology of Literacy and Literacy Instruction

Beach, 1984), and analyze stories into component parts (Idol,1987; Idol & Croll, 1987; Short & Ryan, 1984). In general,all of these strategies proved to improve comprehension andmemory of texts when taught to elementary readers who didnot use such approaches on their own.

The problem with single-strategy instruction, however, isthat good readers do not use single strategies to understandtext; rather, they use a repertoire of strategies (Pressley &Afflerbach, 1995). Thus, in the early to middle 1980s, re-searchers began experimenting with teaching repertoires ofstrategies to elementary-level readers. Perhaps the bestknown of these efforts was reciprocal teaching, developed byPalincsar and Brown (1984). Small groups of students met to-gether to practice four strategies to read text: They predictedwhat would be in the text, asked questions about the contentof the text, sought clarification when confused, and summa-rized the text. Although at first the teacher modeled the strate-gies and led the group in applying them to text, control of thestrategies was quickly transferred to the members of thegroup; the members took turns leading the group as they read.The leader made predictions, asked questions, and attemptedsummaries; the leader also asked for clarification questionsfrom group members and for predictions about what might becoming next in the text. The assumption was that by partici-pating approximately 20 sessions of reciprocal teaching, stu-dents would internalize the reciprocal teaching strategies andcome to use them when they read on their own.

Reciprocal teaching did increase use of the cognitiveprocesses that were taught (i.e., prediction, questioning,seeking clarification, summarization). With respect to perfor-mance on standardized tests, the approach produced moremodest benefits. In general, reciprocal teaching was moresuccessful when there was more up-front teaching of the fourcomponent strategies by the teacher (Rosenshine & Meister,1994).

In general, when researchers directly taught elementarystudents to use repertoires of comprehension strategies, stu-dents have shown increases in comprehension. Teachers whoteach comprehension strategies effectively begin by explain-ing and modeling the strategies for their students (Roehler &Duffy, 1984)—typically by introducing a repertoire of strate-gies over the course of several months or a semester (e.g.,introducing previewing, then connecting to prior knowledge,generating mental images about text meaning, askingquestions, seeking clarification when confused, and summa-rizing). Often, these strategies are practiced in small groups ofreaders, and the students choose which strategies to carry outand when to do so. Thus, as students read a story aloud, theyalso think aloud about which strategies they are employing tounderstand the text. Sometimes other students in the group

react—perhaps coming up with a different mental image fromthat reported by the reader or perhaps using a different strategyaltogether. Such discussions result in readers’ getting a greatdeal out of a reading; they learn the literal meaning of the storybut also have a chance to reflect on alternative interpretationsof the story (Brown, Pressley, Van Meter, & Schuder, 1996).By practicing such strategies together, the individual membersof the reading group gradually internalize the comprehen-sion processes that are modeled and discussed (Pressley,El-Dinary, et al., 1992). In general, reading comprehensionimproves as a function of such teaching (Anderson, 1992;Brown et al., 1996; Collins, 1991). This form of teachinghas become known as transactional strategies instruction(Pressley, El-Dinary, et al., 1992) because it encourages readertransactions with text (Rosenblatt, 1978), interpretationsconstructed by several readers interacting (transacting) to-gether (Hutchins, 1991), and teachers and group membersreacting to each others’ perspectives (i.e., interactions weretransactional; Bell, 1968).

Summary

High comprehension involves both word-level processes andprocesses above the word level. Fluent reading of words andextensive vocabulary are critical for readers to be able to un-derstand demanding texts. Good readers, however, do muchmore than read words. They predict what will be in text, relateinformation in text to their prior knowledge, ask questions,summarize the big ideas in a text, and monitor whether theyare understanding text. In short, good readers are very activeas they make sense of text. The way to develop good compre-hension in students is to encourage a great deal of reading toincrease fluency, develop the readers’ vocabulary, and teachthem to use the comprehension strategies that good readersuse. All of these competencies can be developed beginning inthe early to middle elementary years.

WRITING

In recent decades, writing instruction in school has becomecommonplace, stimulated in large part by a language arts cur-riculum reform movement that argued for a broader view of lit-eracy than simply reading (e.g., Atwell, 1987; Calkins, 1986;Graves, 1983). Yes, elements of writing such as grammar andspelling have been taught in school since the beginning of theinstitution; the thrust in recent decades, however, has been toencourage students, beginning in kindergarten, to develop andwrite whole pieces—both stories and expositions. One as-sumption is that a lot of learning of lower-level mechanics canoccur in the context of writing real stories and essays.

Writing 345

Young children have much to learn about writing as com-posing. Many K–12 writers do not formulate a clear writinggoal before they begin writing (i.e., they do not know whatthey want to say; Langer, 1986, chap. 3). Also, young writersoften do not take into consideration the perspective of poten-tial readers (e.g., Bereiter, 1980). These failures in planningare compounded by failures to revise first drafts; a K–12 stu-dent’s first draft of a story or essay is often the final draft aswell (Bereiter & Scardamalia, 1987; Fitzgerald, 1987). Thecoherence of writing does indeed improve with age during theK–12 years (Langer, 1986; Stahl, 1977); still, much so-calledknowledge telling at the end of the elementary years continuesinto the secondary school years, with young writers simplyadding ideas to essays willy-nilly as the ideas come to mind(Bereiter & Scardamalia, 1987; Emig, 1971; Pianko, 1979;Scardamalia & Bereiter, 1986). Even the best of high schoolwriters tend to produce essays with a simple structure: Often,high school essays consist of a thesis statement, followedby several paragraphs, each of which makes one point. Thewriter then closes with a single-paragraph summary and con-clusion (Applebee, 1984; Durst, 1984; Marshall, 1984).

Scholars interested in the development of compositionskills reasoned that it was the attention to mechanics ratherthan to holistic composition that was the culprit behind theunimpressive writing typically occurring in school—that is,holistic composition skills were typically not taught beforethe 1980s. Moreover, the most frequent type of writing as-signment in school did not demand much in the way of plan-ning or revision; rather, it encourages students simply to dumpknowledge—that is, the most typical writing assignment inschool is to write a few sentences in reaction to short-answerquestions on study guides or tests, with the evaluation of an-swers based on content rather than form (Applebee, 1984;Langer & Applebee, 1984, 1987; Marshall, 1984). When formdoes matter, typically spelling and punctuation count morethan does overall organization of the writing, which does notencourage students to be attentive to the higher-order organi-zational aspects of writing (Langer, 1986).

An important analysis of composition was carried out byFlower and Hayes (1980, 1981). They directed college-level writing teachers and college freshmen to think aloud asthey wrote. The most striking and important finding in thestudy was that excellent writers viewed writing as problem–solving; planning, drafting, and revision were the threeprocesses required to solve the problem of creating a composi-tion. Moreover, Flower and Hayes observed that good writersdid not simply cycle through these processes in a linear fashion;rather, they used the processes recursively—some planning,then some drafting, followed by more planning, some drafting,and then some revision, which makes it clear that still more

planning and drafting are required, and so on until the writer issatisfied with the product. In contrast to the college writingteachers, freshmen were much less likely to have clear goalsbefore beginning to write; they did less planning and revisionthan did the teachers; knowledge telling was more prominent inthe student writing than in the teachers’ writing. Attention tomechanics was prominent throughout writing for the students;in contrast, college writing teachers only worried about me-chanics as they were nearing the end of writing, seeing it as partof the polishing process. The Flower and Hayes work providedboth a clear vision of the nature of excellent writing and a vividunderstanding about how the writing of beginning collegestudents falls far short of the expert ideal.

Curriculum developers took notice of the Flower andHayes’ work. In particular, scholars identifying with the wholelanguage approach to beginning language arts began to en-courage much writing every day (Atwell, 1987; Calkins,1986; Graves, 1983). The role of the teacher in this effort waslargely to coach students during revision—providing promptsto student writers to revise spelling, grammar, and capitaliza-tion, and of course minilessons on these topics when theywere required. Even so, instruction to plan, draft, and revisewas less prominent in the whole language efforts than it wasin other approaches to teaching of writing.

One notable approach was dubbed cognitive strategy in-struction in writing (CSIW) by its creators (Englert, Raphael,Anderson, Anthony, & Stevens, 1991). Englert et al. particu-larly focused on expository writing. There was a great deal ofteacher explanation about writing structures for conveyingideas (e.g., teaching of compare-and-contrast essay struc-tures). Teachers often would share examples of good and pooressays with students, thinking aloud as they worked on revis-ing such essays. Such thinking aloud was central as the teach-ers modeled the construction and revision of expositions.Thus, during planning, the teacher modeled the use of a seriesof questions that should be on the mind of anyone preparingto write an essay: Students saw the teacher reflecting on thequestions Who am I writing for?, Why am I writing this?,What do I know?, How can I group my ideas?, and How will Iorganize my ideas? If such direct explanation and modelingof strategies seems familiar, it should, since Englert et al.(1991) were very much influenced by their Michigan Statecolleague Gerry Duffy, who developed the direct explanationand modeling approach to comprehension instruction coveredearlier in this chapter. Just as such direct explanation of strate-gies improved comprehension, it also improved essay writingin Englert et al. (1991) relative to students not receiving suchinstruction, with the study taking place over an entire schoolyear. In particular, the essays of students (both regular-education students and those with reading disabilities) taught

346 Psychology of Literacy and Literacy Instruction

to plan, draft, and revise were judged to convey their mes-sages better overall. In recent years, a number of replicationshave supported the general finding that teaching elementarystudents to plan, draft, and revise improves writing (Harris &Graham, 1996).

Most striking, effective writing instruction for elementarystudents provides detailed guidance and support about how toplan, draft, and revise (e.g., De La Paz & Graham, 1997;Graham, 1997). Thus, one effective instruction for stimulat-ing story writing involved providing prompts for each part ofa story. Thus, young writers were taught to respond to the fol-lowing questions as they wrote (Harris & Graham, 1996):Who is the main character? Who else is in the story? Whendoes the story take place? Where does the story take place?What does the main character do or want to do? What do othercharacters do? What happens when the main character does ortries to do it? What happens with the other characters? Howdoes the story end? How does the main character feel? Howdo other characters feel?

In summary, psychological research has greatly informedthe teaching of writing in elementary schools; a substantialbody of research validates the plan, draft, and revise model.Most impressively, writing researchers have been able todemonstrate consistent benefits for children experiencing greatproblems with writing, including those classified as readingand writing disabled (Harris & Graham, 1996).

ENCOURAGING ADULT LITERACY

Adults in need of literacy instruction vary greatly. Some re-quire basic, word-level instruction, whereas others can readwords but do not understand very well what they read. Manyadults have not learned to compose well enough to expressthemselves well in writing.

Basic, Word-Level Difficulties

Many adults cannot read at all. Although the problem is espe-cially acute in many developing countries, adult illiteracy inthe industrialized world is common as well; persons who are il-literate suffer economically and psychologically because oftheir condition. A number of countries and their political lead-ers view literacy development as a key to their economic de-velopment and general betterment; hence, national literacycampaigns have been common in developing countries (Bhola,1999; Wagner, 1999; Windham, 1999). Some who particularlyidentify with the masses in developing countries conceive ofliteracy development as a powerful political tool—one withthe potential to empower the masses (Freire & Macedo, 1987).

Religious groups have also been interested in developingliteracy in many underdeveloped regions as part of their evan-gelization efforts, recognizing that people who can read reli-gious texts are more likely to become converts than are peoplewho cannot (Venezky, 1999).

In some U.S. locales, for example, as much as 10–20% ofthe population lacks the most basic literacy skills (NationalInstitute for Literacy, 1998). In recent years, the negative eco-nomic impact of these illiterate citizens has been emphasizedas a motivation for addressing problems of adult literacy inAmerica (Hull & Grubb, 1999). Most conspicuously, illiter-ate adults are much less employable than are people who canread and write, and they are also certainly less able to meetthe demands of an ever more technological world.

Unfortunately, many adults who are illiterate have low psy-chometric intelligence, and no dramatic advances have beenmade in understanding how to develop reading and writingskills in illiterate adults who have low intelligence. What is of-fered to most adult illiterates—ones who cannot read at all—isvery basic instruction in word recognition skills; groups suchas Literacy Volunteers of America (LVA) and Laubach havebeen prominent in these efforts. This approach makes concep-tual sense: The core problem for most illiterate adults—if theyhave at least average intelligence—is their ability to recognizewords, and they have great difficulties in mapping the letters inwords to component sounds and blending them (Bell &Perfetti, 1994; Elbro, Nielsen, & Petersen, 1994; Greenberg,Ehri, & Perin, 1997). Thus, there is good reason to believe thatmost normally intelligent adults who cannot read words couldlearn to do so with systematic instruction in the sounding outof words (e.g., Vellutino et al., 1996), if only these nonreaderswould be willing to stay the course of basic letter- and word-level instruction—that is, many adult education programsfocusing on basic reading have difficulty keeping students en-rolled because the instruction is not very motivating (i.e.,letter-sound drills and drilling on sight vocabulary bores manyadults). Recent efforts to make such instruction attractive toadult learners have focused on use of technology in instruc-tion, although much of the technology now available isrunning skill and drill routines much like the human skill anddrill instruction that has failed previously to hold adults inbasic reading education (Askov & Bixler, 1998).

Comprehension Difficulties

Although adults who cannot read words at all are salientlyilliterate, many other adults can read words but do notcomprehend or remember very well what they read. Otheradults have great difficulty expressing themselves in writing.Adults who have difficulties with comprehension and writing

Encouraging Adult Literacy 347

are especially challenged by the demands of higher education;hence, it is during higher education that their problems arebeing addressed. Most contemporary institutions of highereducation offer remedial reading, writing, and study skillscourses. Although such efforts have more than a century ofprecedence in higher education, their prevalence increasedthroughout the twentieth century (Stahl & King, 2000), so thateven the most elite of colleges and universities now offer suchinstruction. Such instruction early in the college career isimportant for many students because college requires readingof textbooks that are more demanding than are those encoun-tered in high school, reading of genres never encounteredbefore (e.g., journal articles), and—increasingly—interactionwith electronic sources of information (Pugh, Pawan, &Antommarchi, 2000).

College reading, writing, and study skills courses—at theirbest—are informed by a substantial research literature on theimprovement of reading, writing, and study skills in adults.Much of college-level remedial reading involves teaching stu-dents classic comprehension strategies (Nist & Holschuh,2000). Thus, because most students are confronted with text-books containing many related ideas, study skills courses typ-ically teach students how to construct concept maps oroutlines (Caverly, Orlando, & Mullen, 2000; Nist & Holschuh,2000). The students are taught to devise maps or outlines thatindicate relationships between ideas in the text, including hi-erarchical ideas as well as cause-and-effect relationships, se-quences, and simple listings of ideas. Although such mappinghas the benefit of forcing students to attend to relationshipsspecified in text (Lipson, 1995), it can be very challenging tosome students to identify the relationships that need to bemapped (Hadwin & Winne, 1996)—that is, concept mappingcan make ideas clearer and more memorable, but if a studenthas real problems with text comprehension, she or he may notbe able to produce much of a concept map. Students in studyskills courses are also taught to underline, highlight, and anno-tate text selectively; again, however, doing so effectively re-quires understanding the material (e.g., Caverly et al., 2000;Nist & Kirby, 1989).

There are also a variety of strategies that require readers towork actively with text. Thus, elaborative rehearsal requiresthe reader to restate the text as she or he would if teaching aclass (Simpson, 1994). Readers can be taught to self-test them-selves on material they have read (Weinstein, 1994), which re-quires both rehearsal of material and confronting ideas that arenot yet known (see Pressley, Borkowski, & O’Sullivan, 1984,1985, for coverage of how testing increases awareness of whatis known and unknown). Readers can also be taught to figureout why ideas and relationships in text make sense, rather thanpassively accepting facts and relationships as stated—an

approach known as elaborative interrogation (see Menke &Pressley, 1994; Pressley, Wood, et al., 1992). Perhaps the mostwidely disseminated study skills approach is SQ3R (Robinson,1946), which stands for survey the text, ask questions aboutwhat might be in the text, read the text, recite it, and review it.

Do such reading strategies work? Each of the strategiesworks under some circumstances, with some types of readers,and with some types of texts. Some require extensive instruc-tion in order for students to learn them, such as the complexSQ3R (Caverly et al., 2000; Nist & Kirby, 1989). Most stud-ies skills experts recommend teaching such strategies notalone, but rather in conjunction with other procedures in-tended to keep students on task, such as time managementtechniques. In addition, many studies skills programs also in-clude teaching of vocabulary to increase comprehension, rec-ognizing that many struggling college readers do not knowthe words they need to know in order to comprehend readingsencountered in college (Simpson & Randall, 2000). Forexample, students are often taught the important Latin andGreek root words as an aid to understanding new vocabulary;also, they can be taught how to make use of context clues insentences and paragraphs (Simpson & Randall, 2000).

Of course, a key ingredient in any program to enhancecomprehension has to be reading itself. Like all skills, readingimproves with practice. So do component competencies ofreading. Thus, a great deal of incidental learning of vocabularyoccurs during reading (Nagy, 1988; Sternberg, 1987), and thisincidental learning makes future comprehension easier.

Writing Difficulties

With respect to writing, many college-age writers do not plan,draft, and revise (Flower et al., 1990). More positively, atleast at selective universities, such as Carnegie-Mellon (i.e.,where Flower et al.’s 1990 work was carried out), a sizableportion—perhaps 40% of students—do at least some plan-ning before they write; they think about the goal of writingand how that goal can be accomplished as well as the infor-mation they need to accomplish the goal. That the majority ofstudents at elite schools and a much higher proportion at lessselective universities (see Rose, 1989, chap. 7) need instruc-tion in all aspects of the composing process has stimulated thedevelopment of college writing programs that teach studentshow to plan, draft, and revise as part of composition, and oneof the best developed of these was devised at Carnegie-Mellon (Flower, 1997). A variety of alternative approaches toteaching of college writing are available to students in needof assistance (Valeri-Gold & Deming, 2000), although fewwell-controlled evaluations of these programs are currentlyavailable.

348 Psychology of Literacy and Literacy Instruction

Summary

Although some research has examined how to improve word-level problems in adults as well as their comprehension andwriting difficulties, much remains left to learn. With the ex-pansion of instructional opportunities for adults in need of lit-eracy instruction, the need is greater than ever for research onadult literacy and how it can be enhanced. Society is willing toprovide the resources for adult literacy instruction; researchmust provide interventions worth delivering to adults whoneed to improve their reading and writing skills.

CLOSING COMMENTS

Much has been learned about reading and writing and how itcan be enhanced, beginning with infancy and extending intoadulthood. That said, enormous gaps still remain in under-standing literacy. For example, more is known about teachingword recognition skills to struggling young readers than isknown about how such instruction affects normal and giftedreaders. Finding out what difference word recognition instruc-tion makes to such populations is important because societyand the institution of schooling increasingly favors extensive,explicit decoding instruction for all primary-level students.Similarly, although much has been learned about how to in-crease comprehension in elementary students, we still do notknow how to develop teachers who can deliver such instruc-tion well and who will deliver it faithfully. What we do knowis that such instruction is very challenging for many teachers(Pressley & El-Dinary, 1997). The many research successes inthe area of literacy research and instruction should go far instimulating a great deal of additional research in the next quar-ter century; such work is necessary because the research of thetwentieth century permitted much progress in understandingliteracy without providing definitive understanding about howto prevent literacy difficulties and failures. Many children andadults continue to struggle to be readers and writers, which isan increasingly serious situation because our technologicallydriven society demands greater literacy competencies in everynew generation.

REFERENCES

Adams, M. J. (1990). Beginning to read. Cambridge, MA: HarvardUniversity Press.

Adams, M. J., Treiman, R., & Pressley, M. (1998). Reading, writing,and literacy. In I. Sigel & A. Renninger (Eds.), Handbookof child psychology: Vol. 4. Child psychology in practice(pp. 275–355). New York: Wiley.

Afflerbach, P. (2000). Verbal reports and verbal protocol analysis. InM. L. Kamil, P. B. Mosenthal, P. D. Pearson, & R. Barr (Eds.),Handbook of reading research (Vol. 3, pp. 163–179). Mahwah,NJ: Erlbaum.

Alexander, A., Anderson, H., Heilman, P. C., Voeller, K. S., &Torgesen, J. K. (1991). Phonological awareness training andremediation of analytic decoding deficits in a group of severedyslexics. Annals of Dyslexia, 41, 193–206.

Alvermann, D. E. (2000). Narrative approaches. In M. L. Kamil,P. B. Mosenthal, P. D. Pearson, & R. Barr (Eds.), Handbook ofreading research (Vol. 3, pp. 123–139). Mahwah, NJ: Erlbaum.

Anderson, D. R., & Collins, P. A. (1988). The impact on children’seducation: Television’s influence on cognitive development(Office of Research Working Paper No. 2). Washington, DC:U.S. Department of Education, Office of Educational Researchand Improvement.

Anderson, R. C., & Freebody, P. (1981). Vocabulary knowledge. InJ. T. Guthrie (Ed.), Comprehension and teaching: Researchreviews (pp. 77–117). Newark, DE: International ReadingAssociation.

Anderson, V. (1992). A teacher development project in transactionalstrategy instruction for teachers of severely reading-disabledadolescents. Teaching and Teacher Education, 8, 391–403.

Applebee, A. N. (1984). Contexts for learning to write. Norwood,NJ: Ablex.

Applebee, A. N., & Langer, J. A. (1983). Instructional scaffolding:Reading and writing as natural language activities. LanguageArts, 60, 168–175.

Armbruster, B. B., Anderson, T. H., & Ostertag, J. (1987). Does textstructure/summarization instruction facilitate learning fromexpository text? Reading Research Quarterly, 22, 331–346.

Askov, E. N., & Bixler, B. (1998). Transforming adult literacy in-struction through computer-assisted instruction. In D. Reinking,M. C. McKenna, L. D. Labbo, & R. D. Kieffer (Eds.), Handbookof literacy and technology (pp. 167–184). Mahwah, NJ: Erlbaum.

Atwell, N. (1987). In the middle: Reading, writing, and learningfrom adolescents. Portsmouth, NH: Heinemann.

Ball, E. W., & Blachman, B. A. (1988). Phoneme segmentation train-ing: Effect on reading readiness. Annals of Dyslexia, 38, 203–225.

Ball, E. W., & Blachman, B. A. (1991). Does phoneme segmentationtraining in kindergarten make a difference in early word recogni-tion and developmental spelling? Reading Research Quarterly,26, 49– 66.

Ball, S., & Bogatz, G. A. (1970). The first year of “Sesame Street”:An evaluation. Princeton, NJ: Educational Testing Service.

Barker, T. A., & Torgesen, J. K. (1995). The evaluation of computer-assisted instruction in phonological awareness with below aver-age readers. Journal of Educational Computing Research, 13,89–103.

Barr, R., Kamil, M. L., Mosenthal, P. B., & Pearson, P. D. (Eds.).(1991). Handbook of reading research (Vol. 2). New York:Longman.