Becoming a fluent and automatic reader ... - web.macam.ac.ilweb.macam.ac.il/~offprint/79934.pdfautomática en los primeros años de escuela primaria y b) determinar si las habilidades

Reading Research QuarterlyVol. 41, No. 4

October/November/December 2006© 2006 International Reading Association

(pp. 496–522)doi:10.1598/RRQ.41.4.4

T

Becoming a fluent andautomatic reader in the earlyelementary school yearsPAULA J. SCHWANENFLUGEL, ELIZABETH B.MEISINGER, JOSEPH M. WISENBAKERUniversity of Georgia, Athens, USA

MELANIE R. KUHNRutgers University, New Brunswick, New Jersey, USA

GREGORY P. STRAUSSUniversity of Nevada–Las Vegas, USA

ROBIN D. MORRISGeorgia State University, Atlanta, USA

he development of fluent and automatic reading skills is considered a primary edu-cational goal for elementary school–age children. Although there is no single defin-ition of reading fluency, there is general agreement that fluent reading incorporatesthe ability to read quickly, accurately, and, when oral reading is considered, withexpression (National Institute of Child Health and Human Development[NICHD], 2000). In fact, a recent study indicates that second- and third-gradechildren who read quickly and accurately also tend to read with expression, sug-gesting that this definition of reading fluency is quite appropriate (Schwanenflugel,Hamilton, Kuhn, Wisenbaker, & Stahl, 2004). In general, fluent reading emergesin most children between first and third grade, when decoding skills are confirmedthrough practice (Kuhn & Stahl, 2003). Reading fluency, as measured either byquick and accurate word reading or by text-reading rate, continues to develop be-yond this period (Hasbrouck & Tindal, 1992; Horn & Manis, 1987).

496

497

The goals of this study were to (a) develop an empirically based model regarding the development of fluent andautomatic reading in the early elementary school years and (b) determine whether fluent text-reading skills provid-ed benefits for reading comprehension beyond those accounted for by fluent word decoding. First-, second-, andthird-grade children completed a series of reading tasks targeting word and nonword processing, text reading,spelling knowledge, autonomous reading, and reading comprehension. Structural equation modeling was carriedout to evaluate how these skills operated together to produce fluent text reading and good comprehension. Evidencesupported a simple reading fluency model for the early elementary school years suggesting that fluent word and textreading operate together with autonomous reading to produce good comprehension.

Becoming afluent andautomatic readerin the earlyelementaryschool years

Los objetivos de este estudio fueron: a) desarrollar un modelo de base empírica del desarrollo de la lectura fluida yautomática en los primeros años de escuela primaria y b) determinar si las habilidades de lectura fluida de textos pro-porcionan beneficios para la comprensión más allá de los ya documentados que están implicados en la decodifi-cación fluida de palabras. Niños de primero, segundo y tercer grado realizaron una serie de tareas de lectura queevaluaban el procesamiento de palabras y no palabras, la lectura de textos, los conocimientos ortográficos, la lec-tura autónoma y la comprensión lectora. Mediante un modelo de ecuaciones estructurales se evaluó de qué modoestas habilidades operaban en conjunto para producir lectura fluida de textos y buena comprensión. La evidenciaapoya un modelo simple de fluidez lectora en los primeros años de la escuela primaria, lo cual sugiere que la lec-tura fluida de palabras y textos opera en conjunto con la lectura autónoma para producir buena comprensión.

Cómoconvertirse en unlector fluido yautomático en losprimeros años de escuelaprimaria

Die Ziele dieser Studie waren es, (a) ein empirisch basiertes Modell bezüglich der Entwicklung von fließendemund automatischem Lesen in frühen Grundschuljahren zu entwickeln, und (b) zu bestimmen, ob fließendeLesefertigkeiten von Text zu Verbesserungen im Leseverständnis führten, die über jene hinausgingen, die in fließen-der Wortentzifferung mit einbezogen waren. Kinder der ersten, zweiten und dritten Klasse komplettierten eine Serievon Leseaufgaben, durch Angriff von Wort- und Nichtwortverarbeitung, Lesen von Texten, Kenntnisse imBuchstabieren, autonomes Lesen, und Leseverständnis. Strukturelle Gleichungsmodelle wurden ausgeführt, umzu bewerten, wie diese Fertigkeiten gemeinsam funktionierten, um fließendes Lesen von Text und gutes Verstehenzu produzieren. Als Beweisunterstützung diente ein einfaches Modell zum fließenden Lesen für die frühenGrundschuljahre, indem es vorschlägt, dass fließendes Lesen von Wort und Text mit autonomem Lesen zusammenoperieren, um gutes Verständnis hervorzubringen.

Werdegang zumfließenden undautomatischenLesen in den frühenGrundschuljahren

ABSTRACTS

498

Cette étude avait pour buts de (a) développer un modèle empriquement fondé du développement de la lecturecourante et automatisée pendant les premières années d’école élémentaire, et (b) de déterminer si les compétencesde la lecture courante de textes apportent des bénéfices pour la compréhension de la lecture allant au-delà de ceuxqui sont impliqués dans le décodage rapide des mots. Des enfants de première, seconde et troisième année ont ef-fectué une série de tâches de lecture ciblant le traitement de mots et de non-mots, la lecture de textes, la connaissancede l’orthographe, la lecture autonome, et la compréhension de la lecture. On a utilisé un modèle d’équation struc-turale pour évaluer comment ces compétencs opèrent ensemble pour produire une lecture courante de textes etune bonne compréhension. Les données permettent de défendre un simple modèle de lecture courante pour les pre-mières années d’école élémentaire, suggérant que le mot lu couramment et la lecture de textes opèrent ensemble avecla lecture autonome pour produire une bonne compréhension.

Lire couramment,de façon

automatiséedans les

premièresannées d’école

élémentaire

ABSTRACTS

The development of reading fluency is viewedas important because of its relationship with im-proved comprehension (Fuchs, Fuchs, Hosp, &Jenkins, 2001). LaBerge and Samuels’s (1974) now-classic article presenting an automaticity theory ofreading argued that proficient word-recognitionskills underlie fluent reading and adequate compre-hension of text. According to the model, fluent read-ers are characterized by the ability to read quicklyand without conscious effort (Logan, 1997).Dysfluent beginning readers, by contrast, are identi-fied by their excessively slow, laborious reading,which, in turn, impairs comprehension.

Automaticity is an important component ofskilled reading. Whether considered in terms ofreading or any other skill, automaticity is identifiedby a number of characteristics, some of which con-cern us here. The first of these are speed and accura-cy, which seem to emerge simultaneously withpractice (Logan, 1988). Practice strengthens connec-tions between word and letter patterns in long-termmemory (e.g., LaBerge & Samuels, 1974), unitizesthese letter patterns in memory so that they can beprocessed as whole units (Anderson, 1987), and pro-liferates the availability of instances of these wordand letter unit representations in long-term memorywith every encounter of them during reading(Logan, 1997).

Another characteristic of automatic skill is au-tonomy, or the ability to initiate a task without ac-tively attending to it. The automatic reader cannothelp but process print, even when he or she may in-tend to avoid doing so. Skilled readers find them-selves reading the conflicting “crawl” at the bottomof a television news story, even when they try to payattention to the main story (Bergen, Grimes, &Potter, 2005). However, the automatic nature ofskilled word recognition is usually helpful because itorients our attention to the reading. Occasionally, asnoted above, this can get in the way of our goals.Experimentally, the autonomous nature of automaticword reading has been measured through the use ofthe Stroop task (Stroop, 1935) and its variants. Inthe Stroop task, the reader is typically asked to namea picture or color without reading distracting wordsthat are embedded. Autonomy is indicated by the in-terference that the distracting print causes to the pic-ture or color naming. Thus, a child who is anautomatic reader will be slowed in the naming a pic-ture of an apple with the word doll written on itcompared with naming a picture without the dis-tracting print.

Finally, automaticity allows cognitive resourcesto be used to benefit larger goals. For reading, as au-

tomaticity develops for skills like word recognition,attentional resources become available for compre-hension. Thus, children move from relying on theslow letter-by-letter (or unit-by-unit) decoding topainlessly retrieving cued words from long-termmemory (Logan, 1997). This means that childrenwho have efficient word-recognition skills should beable to read connected text fluently and better un-derstand what they read.

The three aspects of automaticity (speed, au-tonomy, and resource use) are thought to developconcomitantly with one another, but how closelytheir development coincides is uncertain (Logan,1985; Paap & Ogden, 1981). Clearly, at the verybeginning of skill development, word reading is ex-tremely slow, very effortful, and resource intensive(i.e., requiring a great deal of attention). Once achild has attained a high level of skill, word and textreading is very quick, occurs without intention, andensures that plenty of resources are left available forcomprehension. In the middle of skill development,it is unclear how the autonomy feature of auto-maticity relates to speed and resource aspects. Thus,it is uncertain whether the extent to which youngreaders experience Stroop interference, for example,is relative to their reading fluency or to improvedcomprehension.

How quickly automaticity develops dependson the underlying mechanisms and how they maywork. For example, in some views, this buildup toautomatic processing is seen as slow and laborious,occurring over hundreds of decoding attempts(Schneider, Dumais, & Shiffrin, 1984), whereas inothers automaticity can develop after correctly recog-nizing a word only once or twice (Logan, 1997;Logan & Klapp, 1991). If the latter is true, thenStroop interference should develop quickly as chil-dren develop the ability to read. Moreover, the bene-fits for reading fluency and comprehension mightemerge even for very young readers whose wordrecognition during reading is better than that oftheir peers. Despite the popularity of this automatic-ity account of the development of reading skill,model building to assess how these aspects of auto-maticity in early reading operate together is rare.

The purposes of the current study were two.First, we wished to provide a more comprehensiveaccount regarding how fluency, autonomy, and freedresources (in terms of improved comprehension) op-erate together in early reading skill development.Thus, children completed reading tasks targeted ateach aspect of automaticity, and the interrelation-ships among them were examined.

Authenticity, fluency, and early reading 499

Second, we wished to determine the relativeadequacy of two potential variants of this automatic-ity view of reading fluency, depicted in Figures 1aand 1b. In one view, which we call the text reading asmediator model, text-reading fluency is seen asuniquely relevant for comprehension (Jenkins,Fuchs, van den Broek, Espin, & Deno, 2003; Kuhn& Stahl, 2003) in contrast with other indicators ofreading fluency. Young readers may or may not exe-cute a variety of cognitive processes relevant to thecomprehension of text if they have the cognitive re-sources available to generate them, such as inferenc-ing to fill in the gaps in the text, creating a mentalmodel of the world described by the text, and pro-cessing sentence grammar. At minimum, additionalcognitive resource benefits may derive from fluenttext reading through the contextual activation ofword meanings (Stanovich, 1980) and the automaticlinguistic segmentation of text into major syntacticunits (Young & Bowers, 1995). This view is support-ed by studies showing that the fluent reading of textaccounts for additional variance in reading compre-hension beyond that accounted for by isolated wordreading alone (Jenkins et al., 2003). In Figure 1a,this partial mediating effect of text reading is shownby the indirect path leading from GORT–3 Text-Reading Rate and to WIAT ReadingComprehension.

Another view, which we call the simple readingfluency model, states that reading fluency is a generalskill that includes skills related to word-reading flu-ency as well as those related to text-reading fluency.In this model, it is assumed that, early in the devel-opment of reading skills, word-recognition skills arethe limiting factor in reading comprehension andthat the cognitive resources gained from automatici-ty derive mainly from quick and accurate word read-ing. Moreover, the skills that allow children tocoordinate the benefits from fluent word and textreading may not yet be fully developed. This view issupported by studies showing that reading skill inyoung elementary school readers can be described asa single factor consisting of word-recognition skill,text-reading fluency, and reading comprehension(Shinn, Good, Knutson, Tilly, & Collins, 1992).Further, the texts that children read increase in com-plexity as they proceed in school. These changes incomplexity may include, among others, greater syn-tactic complexity (Harber, 1979) and fewer repeti-tions of specific words within texts (Hiebert &Fisher, 2005). If text complexity increases faster thanchildren’s text fluency resources can be expended tocapitalize on it, children may not experience the ben-efits gained from fluent text reading to support in-

creases in comprehension during the early stages offluent reading. Thus, children’s word-recognitionskills may need to become yet more automatic tomanage these more complex text features given theresources children have available. In Figure 1b, thisgeneral influence of word-reading fluency is illustrat-ed by showing GORT–3 Text-Reading Rate as mere-ly another indicator of the latent variable ReadingFluency.

In what follows, we describe our rationale forincluding specific tasks and assessments as indicatorsof model features in tests of these larger models. Webegin with our discussion of tasks and assessmentsthat should be most closely related to reading fluen-cy per se. Then, we describe research on tasks de-signed to assess the autonomy component and ourassessment of reading comprehension.

Reading fluency component

Word-reading fluencyIt is often asserted that the development of effi-

cient word-recognition skills is the sine qua non ofskilled reading in the early stages of learning to read(Adams, 1990; Lyon, Shaywitz, & Shaywitz, 2003;NICHD, 2000; Olson, Gillis, Rack, DeFries, &Fulker, 1991; Stanovich, 1985; Stanovich, Nathan,& Vala-Rossi, 1986; Stanovich, Nathan, & Zolman,1988). It is the core skill around which reading flu-ency is built and an important skill for predictingreading comprehension (Gough, 1996; Perfetti &Hogaboam, 1975; Schwanenflugel et al., 2004). Itdistinguishes skilled from less skilled readers in amanner that is remarkably stable from first to fourthgrade (Juel, 1988). It is often considered one, if notthe key, bottleneck on the way to good reading com-prehension (LaBerge & Samuels, 1974; Lyon, 1995;Nicholson, 1999; Perfetti, 1985).

A number of word features seem to enter intothe skilled reading of words in young children.High-frequency words tend to be named faster thanlow-frequency words (Gottardo, Chiappe, Siegel, &Stanovich, 1999; Waters, Seidenberg, & Bruck,1984). Orthographically irregular words (wordswhose spellings do not follow typical letter–soundcorrespondences) tend to be read more slowly andinaccurately than orthographically regular words inyoung readers (Backman, Bruck, Hebert, &Seidenberg, 1984; Nation & Snowling, 1998;Waters et al., 1984). Semantically difficult words(words rated low in imageability; Laing & Hulme,1999; Nation & Snowling; Schwanenflugel & Akin,1994; or rated context availability; McFalls,

500 Reading Research Quarterly OCTOBER/NOVEMBER/DECEMBER 2006 41/4


FIGURES 1A AND BTHEORETICAL MODELS FOR TWO AUTOMATICITY VIEWS OF READING FLUENCY

FIGURE 1A: TEXT READING AS MEDIATOR MODEL

SpellingRT

TOWRESWE

TOWREPDE

GORT–3Text-ReadingRate

Word-reading fluency

Autonomous reading

interference

WIAT Reading Comprehension

Spellingaccuracy

Vowel doublet RT

Vowel doublet accuracy

Single-word–naming accuracy

Single-word–naming WPM

Consonantdoublet RT

Consonantdoublet accuracy

High-GPCHigh-Rime

High-GPCLow-Rime

Low-GPCHigh-Rime

Nonword

CTOPPRapid Object Naming

FIGURE 1B: SIMPLE READING FLUENCY MODEL

SpellingRT

TOWRESWE

TOWREPDE

GORT–3Text-ReadingRate


Autonomous reading

interference

WIAT Reading Comprehension

Spellingaccuracy

Vowel doublet RT

Vowel doublet accuracy


Single-word–namingWPM

Consonantdoublet RT

Consonantdoublet accuracy

High-GPCHigh-Rime

High-GPCLow-Rime

Low-GPCHigh-Rime

Nonword

CTOPPRapid Object Naming

Schwanenflugel, & Stahl, 1996; Schwanenflugel &Noyes, 1996) are read more slowly and inaccuratelythan semantically easy words in young readers.

In the current study, we assessed children’sword-reading fluency in two ways. First, we assessedchildren’s automaticity for sight words using theSight Word Efficiency subtest of the Test of WordReading Efficiency (TORP; 1999), a speeded sight-word–reading test. Second, we assessed children’sspeed of processing for a potentially wider range ofword types by presenting them with a mixed varietyof words in a timed computerized single-word–naming task.

Nonword readingNonword reading is often used as an indicator

of children’s understanding of the relationships be-tween letters, letter strings, and larger units such asrime units and speech sounds (Coltheart & Leahy,1992; Treiman, Goswami, & Bruck, 1990).Nonword reading is used as a way of assessing chil-dren’s ability to read unfamiliar words (Gottardo etal., 1999; Rack, Snowling, & Olson, 1992).Nonword-reading speed is widely considered an in-dicator of the automaticity of essential phonics andblending skills and is usually highly correlated withword-recognition skills (Torgesen, Wagner, &Rashotte, 1999). In the current study, we assessedchildren’s nonword-reading skills using thePhonemic Decoding Efficiency subtest from the Testof Word Reading Efficiency (Torgesen et al.), aspeeded nonword-naming task.

Rapid object namingAssessment of rapid object naming involves

measuring how rapidly a child can carry out rapidnaming of series of pictures. Although not obviouslyrelated to reading fluency, the speed with which chil-dren can carry out the rapid naming of pictures, aswell as numbers, letters, and colors, has been shownto be associated with the development of readingskill in general (Manis, Seidenberg, & Doi, 1999;Morris et al., 1998; Scarborough, 1998; Wolf, Bally,& Morris, 1986; Wolf & Bowers, 1999) and poten-tially involved in the orchestration of processes in-volved in reading fluency (Wolf, Bowers, & Biddle,2000). Rapid object-naming speed shares consider-able independent variance with word identification(Bowers & Swanson, 1991). For the purposes of thepresent study, naming-speed deficits have also beenshown to affect the rate with which individuals areable to read connected text (Bowers, 1993; Breznitz

& Berman, 2003; Katzir, Shaul, Breznitz, & Wolf,2004; Stage, Sheppard, Davidson, & Browning,2001; Young & Bowers, 1995) and so may have par-ticular implications for the development of readingfluency. Thus, in the development of reading, naming-speed problems result in the “slower accessto lexical and sublexical information that may im-pede the development of fluency in reading” (Wolfet al., 2000, p. 401).

Orthographic processingOrthographic processing can be defined as “the

ability to represent the unique array of letters that de-fines a printed word, as well as general attributes ofthe writing system such as sequential dependencies,structural redundancies, and letter position frequen-cies” (Vellutino, Scanlon, & Tanzman, 1994, p. 314).Orthographic coding tasks are designed to assess chil-dren’s knowledge of visual spelling patterns in a waythat distinguishes them from the simple retrieval ofletter sounds. Orthographic processing differencesamong children may account for additional variancein word-reading skill independent of that accountedfor by phonological processing, such as nonwordreading (Cunningham & Stanovich, 1990), but thisis not always found (Manis, Custodio, & Szeszulski,1993). Low-skilled readers may use this knowledge ofvisual spelling patterns to compensate for otherwisepoor phonics skills as an alternative route to wordidentification (Pennington, Lefly, Van Orden,Bookman, & Smith, 1987; Stanovich & Siegel,1994), but this point is also controversial (seeFoorman, Francis, Fletcher, & Lynn, 1996).

In the current study, we assessed orthographicknowledge in two ways. First, we used theExperimental Spelling Task, developed by Olson,Kliegl, Davidson, and Foltz (1985), to assess chil-dren’s orthographic processing. In this task, childrenare asked to decide which of two letter strings repre-sents the correct spelling of a real word, such as raneversus rain. Because both strings obey the spelling–sound correspondences of English and both soundlike a real English word when sounded out, childrenmust retrieve orthographic information to discrimi-nate between them. Variants of this basic task havebeen used by a number of researchers to measureorthographic knowledge (Manis et al., 1993; Olsonet al., 1985; Stanovich & Siegel, 1994). Some re-searchers (Manis et al.; Vellutino et al., 1994) havetaken issue with this means of assessing ortho-graphic knowledge because it is possible for childrento make these decisions on the basis of their prior ex-perience reading the words, their spelling instruc-



tion, or the mere presence of more orthographic“neighbors” (i.e., rain, vain, stain, grain, brain) intheir mental lexicon for one option over the other.Still, it is probably the most widely used task to as-sess orthographic knowledge, so we included it inour study.

Second, we used a variant of the DoubletKnowledge Test, developed by Cassar and Treiman(1997), to assess metacognitive knowledge of dou-blets. This task assesses children’s understanding ofthe permissibility of consonants and vowels as dou-blets with regard to position (i.e., consonant dou-blets tend not to occur at the beginning of a wordwith a rare exceptions such as llama) and letter (i.e.,ii and uu cannot appear as doublets within a word;some consonants such as hh, jj, and vv are rarelydoubled). This task may also have some of the prob-lems of Experimental Spelling Task (i.e., neighbor-hood effects) but seems to relate less ambiguously totrue orthographic knowledge. In our variant of thistask, participants were asked to decide as quickly aspossible which of two items such as baff versus bbaf“looks more like a word should look.” Cassar andTreiman found that this type of knowledge emergesin kindergarten and is fully accurate by secondgrade. In our variant of this task, we presented chil-dren with the doublet task and asked them to makethe doublet decision as quickly as they could.

Text-reading fluencyTypically, assessments of text-reading fluency

entail asking children to read aloud from connectedtext while an examiner records, at minimum, thenumber of correctly read words per minute. At issueis exactly what additional benefit text-reading fluen-cy provides beyond those accrued from skilled, flu-ent word recognition.

There are a number of potential benefits thatthe fluent reading of connected text can provide overreading isolated words. Among them, fluent oralreaders verbally segment word strings into larger syn-tactic groupings that may support comprehension(Dowhower, 1987; Kuhn & Stahl, 2003; Rasinski,1990; Wolf & Katzir-Cohen, 2001; Young &Bowers, 1995). Text reading makes available context-facilitated word recognition as well (Stanovich,1980). To the extent that these processes are carriedout automatically, children can gain additional re-source benefits from text reading that they can usefor comprehension. The point of view representedby this model is supported by studies finding thattext-reading fluency accounts for additional variancein reading comprehension beyond that accounted

for by word-reading fluency alone (Jenkins et al.,2003). However, whether children in the early phas-es of learning to read are able to make use of the re-source benefits ascribed to text-level reading iscontroversial (Kuhn & Stahl, 2003; Schwanenflugelet al., 2004; Shinn et al., 1992). In the current study,to assess text-reading fluency, children read passagesfrom the Gray Oral Reading Test–Third Edition(GORT–3). This assessment is perhaps the mostwidely used standardized assessment of reading flu-ency (Marlow & Edwards, 1998).

The autonomy componentThe Stroop task directly taxes the autonomy

component in early reading by requiring participantsto name a primary stimulus (color or picture) whileignoring a printed word (e.g., naming the picture ofa desk with the word cat written on it, or naming theink color “blue” for the word red printed in blueink). In the picture–word version of the task, inter-ference results when the reader cannot help but readthe printed word while attempting to name the pic-ture. Interference is determined by comparing thetime it takes to complete the naming of the primarystimulus in a Stroop condition against the naming ofa control having random letter strings or no conflict-ing print at all.

Several studies have linked the development ofStroop interference to the development of readingskill. Studies by Stanovich, Cunningham, and West(1981) and Guttentag and Haith (1978) found anincrease in Stroop interference from the beginning tothe end of the first-grade year. Ehri (1976) foundthat early and proficient readers showed larger inter-ference, compared with late and less proficient read-ers. Similarly, Schadler and Thissen (1981) foundthat Stroop interference from words increased untilchildren’s reading skills reached a fourth-grade read-ing level, after which interference declined (see alsoGolinkoff & Rosinski, 1976). In later elementaryschool, as children become better able to inhibit dis-tractions, the level of interference is decreased(Pritchard & Neumann, 2004; Tipper, Bourque,Anderson, & Brehaut, 1989). Although the overallsize of the Stroop effect may change, the effect itselfdoes not disappear across the life span (Comalli,Wapner, & Werner, 1962).

Given that Stroop interference seems to developalong with reading skill, what exactly becomesprocessed autonomously as reading becomes automat-ic? Some early findings imply that not only are wordsprocessed autonomously but also letter units may be (Guttentag & Haith, 1978). Some researchers

hypothesize that children begin the process of learn-ing to read by using small units that map letters ontodistinct phonemes, called grapheme–phoneme corre-spondence rules (or GPC; Coltheart & Leahy, 1992;Perry, Ziegler, & Coltheart, 2002). A grapheme is asingle letter (e.g., t-) or letter cluster (sh-) that getsmapped on to a single phoneme or sound. Accordingto this view, children initially read using GPC units,making use of larger letter–sound mappings only laterwhen they have built up a large sight-word vocabu-lary. An alternative view is that children begin to readusing larger sublexical units that engage either wordanalogies or rime units for word decoding (Goswami,Ziegler, Dalton, & Schneider, 2003). Rime units arethe part of a single-syllable word used in a rhyme(e.g., cat, that; Treiman, Mullennix, Bijeljac-Babic, &Richmond-Welty, 1995). Stroop interference mightbe expected to be different depending on the size ofthe units used. In the current study, we presentedchildren with pictures containing pronounceablenonwords, words containing high-probability GPCunits, words containing high-probability rime units,and words containing both.

Reading comprehensionFinally, to determine whether there were bene-

fits from fluent text reading, we needed to have anassessment of reading comprehension. For this, weassessed children’s comprehension using the readingcomprehension subtest of the Wechsler IndividualAchievement Test (WIAT). We selected this test be-cause of its good psychometric properties and be-cause it seemed to cover a very basic ingredient ofreading comprehension, that is, that children shouldbe able to answer comprehension questions (bothsimple and more inferential) following the reading ofa text. Being able to answer questions regarding liter-al and inferential information following the readingof a text has been demonstrated to be a key discrimi-nator between skilled and less skilled comprehenders(Cain, Oakhill, & Bryant, 2004; Casteel, 1993;Long, Oppy, & Seely, 1997).

To sum, the purposes of the present study weretwo. The first was to develop an empirically basedmodel of the development of fluent and automaticreading in the early elementary school years. In par-ticular, we were interested in determining how lexicaland sublexical processes (word reading, nonwordreading, orthographic processing, and rapid objectnaming) and reading autonomy (Stroop) operate to-gether to produce the development of fluent textreading and good comprehension of text. The secondwas to determine whether fluent text reading provid-

ed additional benefits for reading comprehension be-yond those skills involved in word-reading fluencyalone. First-, second-, and third-grade children wereasked to complete a series of reading tasks targetingreading fluency, reading autonomy, and reading com-prehension. Performance on each task was examinedfor developmental change across grade levels. Then,structural equation modeling was carried out to eval-uate how these factors operate together to producefluent text reading and good comprehension.

MethodParticipants

Participants were 99 first-grade (mean age = 7years, 1 month; range = 5 years, 7 months to 8 years,2 months), 79 second-grade (mean age = 8 years, 1month; range = 7 years, 1 month to 9 years, 1month), and 71 third-grade (mean age = 9 years, 0months; range = 7 years, 11 months to 10 years, 4months) children attending four public schools lo-cated in communities in urban northeast Georgia orsuburban central New Jersey, USA. Approximately41% were African American, 27% were EuropeanAmerican, 23% were Hispanic American, 5% wereAsian American, and 4% were other or unknown;53% were female and 47% were male. All childrenin general education classrooms participated with theexception of those currently receiving English-as-a-second-language instruction or those in self-contained classrooms for special education. Childrenwere not excluded on the basis of reading disability.The children attended schools in which approxi-mately 76% received free and reduced-cost lunch.

Stimuli and proceduresChildren were given a mixture of standardized

and experimenter-constructed reading measures dur-ing the spring term (seventh and eighth month) ofthe school year. The order of the standardized mea-sures was counterbalanced with the experimentalreading measures, such that half the children re-ceived the standardized measures first and the otherhalf received the experimenter-constructed readingmeasures first.

Standardized reading measuresThe standardized reading measures were the

following:



(a) Test of Word Reading Efficiency (TOWRE)Sight Word Efficiency and Phonemic DecodingEfficiency subtests, Form A (1999). The Sight WordEfficiency subtest asks children to name all of thewords they can from a list of words in 45 seconds.The Phonemic Decoding Efficiency subtest requireschildren to pronounce all of the phonetically regularnonwords they can in a list in 45 seconds.Concurrent validity estimates reported in the testmanual have a median of .91 in grades 1 through 3.Alternate form reliabilities have a median of .97 ingrades 1 through 3.

(b) Wechsler Individual Achievement Test(WIAT; 1992) Reading Comprehension subtest. TheWIAT Reading Comprehension subtest consists of anumber of passages that increase in complexity.Moreover, passage questions move from assessing lit-eral to inferential comprehension, consistent with re-search on the development of readingcomprehension (Cain et al., 2004; Hannon &Daneman, 2001). In this assessment, a comprehen-sion question is asked when the child completes thereading of a passage, and the child answers the ques-tion aloud in his or her own words. The WIAT isdiscontinued once the child misses four sequentialquestions. Validity estimates of this subtest reportedin the test manual have a median of .74 and a relia-bility of .91 in grades 1 through 3.

(c) Gray Oral Reading Test–Third Edition(GORT–3, 1992). The GORT–3 is an assessment ofproficiency in the oral reading of text. The test is de-signed to evaluate text-reading fluency and consistsof a series of passages that increase in difficulty. Thetest has several subscales, but we focused on readingrate for the purpose of this study. Accuracy and rateare tabulated for each passage until the reading be-comes too slow and inaccurate and meets the discon-tinue rule established by standardization. Thereading of each passage received a rating for readingrate assigned by the test makers that is based on stan-dardized data regarding how long it took children toread each passage. These ratings are then summed toform a cumulative reading rate score. This scoreserved as our indicator of text-reading fluency.Validity estimates for the passage-reading rate scorespresented in the test manual range from .34 to .82,with a median validity of .65. The test has been in-dependently validated for a diverse population ofchildren (Craig, Thompson, Washington, & Potter,2004). Reliability estimates for the assessment ofpassage-reading rate range from .82 to .92, with amedian reliability of .90.

(d) Comprehensive Test of PhonologicalProcessing Rapid Object Naming subtest

(CTOPP–RON; 1999). The Rapid Object Namingsubtest of the CTOPP–RON measures the timechildren take to name six different objects alternatelypresented randomly over 72 objects. The test is saidto be a measure of the efficient retrieval of visual andphonological information from long-term memory.The manual reports reliability estimates for childrenin the current age range from .73 to .93 on this sub-test, with a median of .77, and validity estimatesagainst reading measures ranging from .30 to .54,with a median of .44.

Experimenter-constructed reading tasksThe experimenter-constructed reading tasks

were designed to provide a more fine-grained assess-ment on aspects of early reading not covered by thestandardized measures of reading, particularly thoserelated to the automaticity aspects of reading. Thesetasks were all presented on a Dell Inspiron 8000 lap-top computer using E-Prime experiment software(from Psychology Software Tools, Version 5.0;Schneider, Eschman, & Zuccolotto, 2001), whichwas connected to a serial response box (PsychologySoftware Tools, Model # 200A). Children were test-ed in a quiet location in their school. Each partici-pant was seated in front of the computer screen andasked to hold the microphone/response box whilethe experimenter read a set of instructions aloudfrom the computer screen for each task. For tasks us-ing the microphone, the accuracy of participants’ re-sponses was recorded directly by the experimenter asincorrect if the item was not appropriately named. Itwas scored as a mechanical error if the microphonewas not triggered by the child’s voice or if it was trig-gered by an extraneous noise unrelated to the experi-ment (e.g., background noise, cough, sneeze,pre-articulation response such as “tsk”).

(a) Stroop task. The purpose of the Stroop taskwas to provide an indicator of the degree to whichword reading was autonomous in the children. TheStroop task presented digitized line drawings of ob-jects provided in Cycowicz, Friedman, Rothstein,and Snodgrass (1997), each with a single conflictingword written in lowercase letters over the middle ofit (e.g., a picture of a cat with the word rock super-imposed on it). There were six conditions varying inthe grapheme–phoneme (GPC; Berndt, Reggia, &Mitchum, 1987) and rime-unit relationships repre-sented by the words superimposed on the pictures:control pictures without words, control pictures withrandom letter strings, pictures having words withhighly predictable GPC and highly consistent rimeunits, pictures having words with highly predictable

GPC units but with inconsistent rime units, pictureshaving words with low-predictability GPC units butwith highly consistent rime units, and pictures hav-ing nonwords that could be pronounced with highlyconsistent rime and predictable GPC units. Wordswere chosen so that they did not differ in normativeword frequency across conditions according to Zeno,Ivens, Millard, and Duvvuri (1995), p > .10. Itemswere presented in random order. There were 6 prac-tice and 60 experimental stimuli.

For each trial, a fixation point appeared whenthe experimenter pressed a button on the serial re-sponse box, followed a second later by the picturestimulus. Participants then named each picture stim-ulus aloud into the microphone. The accuracy of theparticipant’s response was recorded by the experi-menter. Spearman-Brown coefficient analysis indi-cated a split-half reliability on this task of .89 forreaction times and .79 for accuracy rates.

(b) Single-word–naming task. The purpose ofthe single-word–naming task was to provide an indi-cator of word-reading speed for a variety of wordtypes while controlling for part of speech. With theuse of normative ratings collected by Schwanenflugeland Noyes (1996) from third-grade children, thestimuli selected were a mixed set of nouns ranging inimageability (M = 4.5, SD = 1.7, range 1.4–6.8where 1 means “hard to think of a picture for theword” and 7 means “easy to think of a picture for theword”), rated context availability (M = 5.5, SD = .7,range 4.1–6.6 where 1 means “hard to think of asentence for the word” and 7 means “easy to think ofa sentence for the word”), frequency (M = 63, SD =44, range 1 to 171 in the third-grade corpus ofCarroll, Davies, & Richman, 1971), and ortho-graphic regularity based on Venezky (1999). Thus,this task included words ranging in semantic and or-thographic difficulty.

Participants were presented with words one ata time. Each child was asked to “say the word asquickly as you can when the word comes up on thescreen.” For each trial, a fixation point appeared forone second in the middle of the computer screen,and this was followed by the presentation of theword. When the word was named, it disappearedfrom the screen, reaction time (RT) was recordedautomatically, and the accuracy of the response waskeyed in by the experimenter. Pilot testing had indi-cated that the task was difficult for some children;therefore, we decided to include a ceiling rule so thatthe children with less reading skill would not be un-duly frustrated by the task. If the child made morethan six errors in a row, the task was discontinued.To ensure that there were observations from each

condition, items were presented in a fixed order ran-domized within blocks of four trials, randomly or-dering condition within each block. There were 8practice and 48 experimental items. Spearman-Brown coefficient analysis indicated a split-half relia-bility on this task of .80 for reaction times and .97for accuracy rates.

(c) Doublet knowledge task. The purpose of thedoublet knowledge task was to provide an indicatorof the degree to which knowledge of spelling pat-terns could be accessed automatically with increasingreading fluency. The task was modeled after theCassar and Treiman (1997) Doublet KnowledgeTask with the exception that the current task wastimed to better determine automaticity of thisknowledge. Children were given pairs of nonwordletter strings having consonant or vowel doublets(e.g., holl versus hhol, stee versus staa) and asked to“pick the item that looks more like a word shouldlook.” For each trial, a fixation point appeared forone second and was followed by the letter-stringpair. If the item on the left was chosen, then thechild pressed the left button on the response box; ifthe one on the right was chosen, then the childpressed the button on the right. The items were se-lected from ones used by Cassar and Treiman suchthat there were 8 practice and 22 experimental items.RT and accuracy were recorded automatically.Spearman-Brown coefficient analysis indicated asplit-half reliability on this task of .91 for reactiontimes and .73 for accuracy rates.

(d) Experimental spelling task. The purpose ofthe experimental spelling task was to provide an ad-ditional assessment of the automaticity of accessingorthographic knowledge. The experimental spellingtask was adopted from one developed by Olson et al.(1985; Olson et al., 1991). In this task children wereprovided with item pairs, one of which was a cor-rectly spelled version of a word (e.g., rain) and theother of which was a phonologically identical but in-correctly spelled version of the word (e.g., rane). Thechildren were told, “Two groups of letters will appearon the computer screen. Both sound like a real word,but only one is a real word. Pick the one that youthink is the real word by pushing the button that ison the same side as the real word.... Do this asquickly as you can without making mistakes.” Therewere 8 practice and 22 experimental trials. Half ofthe correct spellings appeared on the right and theother half on the left. RT and accuracy was recordedautomatically. Spearman-Brown coefficient analysisindicated a split-half reliability on this task of .87 forreaction times and .69 for accuracy rates.



ResultsThe findings were analyzed in two phases.

First, we carried out simple cross-sectional analysesto assess developmental change in the computerizedtasks and standardized tests. Then, with knowledgeof how children progressed on the individual tasks,we carried out tests of theory using structural equa-tion modeling of the relationships among the tasks.

Analyses of developmental change ontasks and tests

Analyses of standardized measuresWe began by carrying out basic analyses of

variance of the standardized test data(TOWRE–SWE and PDE, WIAT–RC,CTOPP–RON, and GORT–3) using Grade as a between-subjects factor, focusing on raw scores forthe tests. The raw score means and standard devia-tions for these subtests and their corresponding stan-dard scores can be found in Table 1.

For each assessment, there was improved per-formance throughout the early elementary schoolgrades. For the TOWRE, there was a significantmain effect of Grade on both the SWE subtest, F (2,246) = 83.45, and the PDE subtest, F (2, 246) =57.197, both p < .001. Bonferroni post-hoc follow-up tests indicated that third graders read significantlymore sight words and nonwords than secondgraders, who, in turn, read more words and non-words than first graders, all p < .001. For theWIAT–RC, there was also significant main effect ofGrade, F (2, 246) = 80.21, p < .001. Again, post-hoctests indicated that older children answered morepassage questions correctly than younger children, allp < .001. Finally, for the CTOPP–RON, there was asignificant main effect of Grade, F (2, 246) = 22.12,p < .001. Post-hoc tests indicated that first graderstook longer to name all the pictures than secondgraders, who took longer than third graders, all p <.05. For the GORT–3, there was a significant maineffect of grade on reading rate sum scores, F (2, 246)= 75.74, p < .001. Post-hoc tests indicated that firstgraders read the texts at a slower rate than second

Grade

Assessment 1 2 3

TOWRE–SWEWords/45 sec. M 31.88 49.05 60.44

SD 16.17 14.37 12.04Standard score M 105 103 105

SD 12 13 12

TOWRE–PDENonwords/45 sec. M 13.15 22.25 30.68


SD 11 13 13

WIAT–RCNo. passages correct M 8.91 14.19 19.75


SD 15 11 15

CTOPP–RONSeconds M 92 78 65

SD 34 22 15Standard score M 9.07 8.75 9.76

SD 3.36 2.69 2.86

GORT–3, Passage 1Rate sum score M 2.08 5.70 13.04

SD 3.96 5.53 7.77Standard score M 8.41 8.56 9.39

SD 2.07 2.36 2.70

TABLE 1MEAN OF THE STANDARDIZED ASSESSMENTS BY GRADE LEVEL

graders, who in turn read more slowly than thirdgraders, all p < .001.

Computerized reaction time (RT) tasksFor all computerized RT tasks, to identify RTs

and error rates that qualified for further analysis, thefollowing steps were conducted on the raw data.First, all RTs for incorrect responses were eliminatedfrom the data and their responses counted as errors.Then, all mechanical errors (as defined above) andtheir corresponding RTs were eliminated from thedata. Next, mean RT and standard deviations werecalculated for every participant in the study. Then,RT trials greater than 2 standard deviations aboveeach participant’s mean and less than 200 ms wereremoved from the data as outliers. Finally, for eachparticipant, means were recalculated for RT and er-ror rates for each condition. These means formed thebasic data for subsequent analyses (see Table 2).

(a) Stroop task. Because error rates approachedfloor for all conditions and grades, we focused onpicture-naming time means as the major dependentvariable of interest. A 3 Grade � 6 StroopCondition ANOVA was carried out with Grade as abetween-subjects factor and levels of Stroop (picturecontrol, letter-string control, nonword labels, Low-GPC + High-Rime Labels, High-GPC + High-RimeLabels, High-GPC + Low-Rime Labels) as a within-subjects factor. There was a significant main effect ofGrade, F(2, 246) = 6.89, suggesting that third-gradechildren named the pictures more quickly in generalthan first- and second-grade children. It is importantto note that there was a significant main effect of

Stroop, F(5, 1230) = 23.57, p < .001, and a non-significant interaction between Stroop and Grade,F(10, 1230) = 1.39, p = .178, suggesting similar pat-terns of interference across grade. Further, wedropped the picture control condition so that the in-fluence of sublexical units beyond the mere letterscould be examined directly. This analysis, too, founda significant main effect of Stroop Condition, F(4,984) = 15.01, p = .003, and Grade, F(2, 246) =5.85, p = .003, and no interaction between Gradeand Stroop Condition, F(8, 984) = 1.12, p = .348,suggesting similar patterns of interference acrossgrade. Follow-up Bonferroni tests, adjusting alpha bythe number of contrasts (.05/4 = .013), indicatedthat compared to the letter-string controls, non-words, High-GPC + Low-Rime words, and High-GPC + High-Rime words interfered significantlywith picture-naming time, but not Low-GPC +High-Rime words. This suggests that the operativesublexical reading unit in Stroop effects in early ele-mentary school is the presence of a highly pre-dictable GPC unit.

The presence of Stroop interference is generallyrecognized as the operation of autonomous process-ing of words, the degree of which varies across par-ticipants. Thus, for the remainder of the analyses, wecalculated Stroop interference directly for each par-ticipant by subtracting each one’s mean RT for letterstrings from their RT from nonwords, High-GPC +Low-Rime, and High-GPC + High-Rime words.These interference scores served as indexes of Stroopinterference for the assessment of the role of au-tonomous reading in subsequent analyses.


Grade

Task 1 2 3

StroopPicture control 1203 (.04) 1172 (.01) 982 (.01)Letter-string control 1192 (.04) 1274 (.03) 1029 (.03)Nonword 1339 (.05) 1341 (.04) 1171 (.03)Low GPC + High Rime 1231 (.07) 1266 (.05) 1148 (.04)High GPC + High Rime 1310 (.09) 1299 (.05) 1172 (.04)High GPC + Low Rime 1344 (.09) 1400 (.06) 1227 (.06)

Single-word naming 2231 (.73) 1314 (.42) 858 (.20)

Doublet Knowledge TaskVowel doublet 2261 (.29) 2421 (.12) 2122 (.07)Consonant doublet 2164 (.33) 2143 (.26) 1878 (.28)

Experimental spelling task 2323 (.42) 2209 (.25) 1801 (.12)

TABLE 2MEAN RTS IN MILLISECONDS (AND ERROR RATES) FOR EXPERIMENTER-CONSTRUCTEDREADING TASKS


(b) Single-word–naming task. The findings forthe single-word–naming task indicated that youngerchildren found decoding this set of words slow anddifficult. A perusal of error rate means indicated thaterrors were not at floor, so a separate analysis ofthem was conducted as well. A one-way ANOVA in-dicated a significant main effect of Grade for bothnaming time, F(2, 246) = 55.25, and error rates,F(2, 246) = 93.28, both p < .001, such that worddecoding became quicker and more accurate for thisset of words throughout the early elementary schoolyears. So that text-reading and word-reading skillcould be more directly compared, RTs were convert-ed into words/minute and error rates were convertedinto accuracy rates. Thus, for subsequent analyses,words per minute (WPM) and accuracy rates servedas indicators of single-word–naming skill.

(c) Doublet Knowledge Task. For this task, wewere interested in the speed and accuracy with whichchildren could make decisions regarding which dou-blet nonword “looked more like a word shouldlook.” ANOVA using Doublet Type (vowel or con-sonant) as a within-subjects factor and Grade as a between-subjects factor indicated a significant maineffect of Doublet Type, F(1, 246) = 36.61, such thatit took children longer to decide that vowel doubletslooked more like a word should look than consonantdoublets. There was a nonsignificant Doublet Type� Grade interaction, F(2, 246) = 2.91, p = .056, forRTs. There was a main effect of Grade for decisionsregarding consonants for both RT, Grade, F(2, 246)= 3.34, and error rates, Grade, F(2, 246) = 47.11,both p < .05, such that third graders took less timeand made these decisions more accurately than firstgraders. There was a main effect of Grade for deci-sions regarding vowels for accuracies, Grade, F(2,246) = 3.14, p = .045, but not for RT, Grade F(2,246) = 2.21, p = .112. Condition means on RTs andaccuracies served as indicators for the developmentof doublet knowledge in younger and older childrenin subsequent analyses.

Experimental spelling taskFor this task, we were interested in whether

children’s knowledge of correct spellings could be ac-cessed quickly with age. An analysis of variance com-paring mean RTs and error rates for this task as afunction of Grade indicated a significant main effectof Grade on RTs, F(2, 246) = 9.72, and error rates,F(2, 246) = 98.64, both p < .001. Third graders ac-cessed this knowledge more quickly than first gradersand more accurately than children in the earliergrades (p < .01). Second graders made these deci-

sions more accurately than first graders (p < .001).For subsequent analyses, mean RTs and error ratesserved as indicators of the development of spellingknowledge in these children.

In sum, between first and third grade in earlyelementary school, we found evidence of develop-ment on all tasks potentially related to fluent readingand automatic reading in young children. Comparedwith younger children, third-grade children pro-nounced sight words and nonwords at a faster rateon the TOWRE, read text passages at a faster rate onthe GORT–3, and read single words more quicklyand accurately. Older children made doublet deci-sions regarding consonants more quickly and accu-rately and regarding vowels more accurately. Theydetermined correct spellings of words more quicklyand accurately. They also named pictures morequickly on the CTOPP–RON. Moreover, thirdgraders could answer more comprehension questionscorrectly after reading passages than first graders onthe WIAT–RC. The sole exception to this was thepattern of developmental change on the Stroop task,which found similar patterns of interference withStroop interference as a function of grade, althoughthird graders showed less interference overall.

Structural equation modelingStructural equation modeling in LISREL was

conducted to examine the relations between the vari-ous measures potentially related to the developmentof fluency during the early elementary school years.The structural equation modeling was carried out intwo phases to evaluate particular models: (a) a mea-surement modeling phase to identify whether theobserved variables fit the latent factor structure pro-posed by theory, and (b) a structural modeling phaseto test the explanatory relationships indicated be-tween predictor and outcome variables by particularmodels (Jöreskog & Sörbom, 1996).

Measurement modeling phaseIn this phase, relations among variables thought

to be potentially relevant to reading fluency accordingto variants of automaticity theory were determined us-ing LISREL being as inclusive as possible with regardto the variables that might be considered to be rele-vant. In the measurement modeling phase, we includ-ed all tasks. We used raw scores from the standardizedassessments rather than age-based standard scores be-cause the correlations between the raw and age-basedstandard scores correlated > .90, but the raw scores al-lowed us to evaluate the grade effects directly in terms

of potential changing relationships between variablesas a function of age. In this initial phase, text-readingfluency and reading comprehension were consideredoutcome variables. We defined our interest in readingfluency to be those aspects of reading that relate tobetter and worse comprehension.

The correlation matrix for each grade level wasused separately as input to LISREL 8.30 (Jöreskog& Sörbom, 1996, 1999) and is the focus of theanalysis. We considered that autonomous readingwould consist of a latent factor comprising the meandifference in RT between letter controls and thethree conditions in which the interfering stimulicontained a highly predictable GPC unit (High-GPC + Low-Rime, High-GPC + High-Rime, andNonwords). We assumed a latent factor of word-reading fluency comprising the number of sightwords read for the TOWRE sight-word efficiency,the mean WPM calculated for the single-word–naming task, and the mean accuracies for that task.We also included in this latent word-fluency factorthe number of nonwords read in TOWRE phone-mic-decoding efficiency and the number of secondsto complete the CTOPP–RON task. As noted earli-er, nonword reading is generally considered a key in-dicator of a child’s ability to read unknown words.Further, rapid autonomous naming is thought toshare many of the skills underlying word-identifica-tion speed, such as quickly perceiving stimuli, re-trieving symbolic information from long-termmemory, and phonological production (Wolf et al.,2000). Thus, it seemed appropriate to have this vari-able serve as an indicator of word-reading fluency.Finally, we also included in this latent factor the RTand accuracy means for all orthographic knowledgetask conditions because orthographic knowledge isthought to be important to the fluent identificationof words (Cassar & Treiman, 1997; Olson et al.,1985). Both of these factors (word-reading fluencyand autonomous reading) served as predictors of theGORT–3 and the WIAT.

Measurement modeling was carried out foreach grade separately. We used several indicators forlack of model fit recommended by Kline (2005) andSchumacker and Lomax (1996). It is traditional topresent the model χ2 fit index, which tests “the dif-ference in fit between a given overidentified modeland a just-identified version of it” (Kline, 2005, p.136). A nonsignificant p > .05 is an acceptable fit.This is generally not the preferred fit index for vari-ous reasons and needs to be supplemented with oth-er indicators of fit. We included the root meansquare error of approximation (RMSEA) as an as-sessment of simple model misspecification where

RMSEA < .08 indicates an acceptable and < .06 agood fit. We also provide the goodness of fit index(GFI), an absolute fit index that estimates propor-tion of variability in the sample matrix explained bythe predicted covariance matrix, and theTucker–Lewis index (TLI), an incremental fit indexsensitive to misspecified factor loadings that is parsi-mony adjusted. For these, fits > .90 are consideredacceptable and > .95 a good fit.

Overall, measurement model results indicatedan unacceptable fit for first grade, χ2 (100) = 256.71,p < .001; RMSEA = .11; GFI = .77; TLI = .80. Thesame was true for second grade, χ2 (100) = 326.70, p< .001; RMSEA = .17; GFI = .66; TLI = .61; and forthird grade, χ2 (100) = 276.69, p < .001; RMSEA =.14; GFI = .70; TLI = .57. However, a perusal of fit-ted standardized residuals indicated the source of theill fit. We found that large residuals (> 2.58) wereconsistently associated with the orthographic tasks ateach grade level (5 of 5 for first grade, 7 of 10 forsecond grade, and 7 of 9 for third grade). Althoughwe made several other attempts to include these vari-ables in the model, every other attempt produced thesame result. Thus, the results of this measurementmodeling indicated that it was best simply to excludethese orthographic variables from further model test-ing. When these variables were excluded, the modelsobtained reached an acceptable fit to data for first, χ2

(33) = 40.44, p = .170; RMSEA = .034; GFI = .93;TLI = .99; and third grade, χ2 (33) = 51.30, p =.022; RMSE = .059; GFI = .90; TLI = .92, althoughthe fit for the second grade did not quite meet estab-lished levels, χ2 (33) = 62.18, p = .002; RMSEA =.10; GFI = .87; TLI = .91. However, in the interestof parsimony, we deemed it better to proceed bytesting the same models for all grades.

Tests of the hypothesized structural modelsStructural equation modeling was used to ex-

amine the role of word-reading fluency, autonomousword reading, and text-reading fluency in benefitingreading comprehension skills. We tested two theoret-ical versions of the way that these skills related to oneanother: (a) text reading as mediator model and (b)simple reading fluency model.

(a) Text reading as mediator model. In this mod-el, the autonomous reading interference and word-reading–fluency aspects of reading were viewed aspredictors of text-reading fluency and reading com-prehension. As noted above, the latent factor reflect-ing word-reading fluency was indicated by bothsubtests of the TOWRE, single-word–naming accura-cy and WPM from the single-word–naming comput-



erized task, and the CTOPP–RON. This latent factorserved both as a predictor of text-reading fluency andreading comprehension. However, in this model,text-reading fluency was considered a mediator of therelation between word-reading fluency and readingcomprehension. As noted earlier, the mediating quali-ty of text reading is said to emerge from the specialfeatures of text reading not available from fluent wordreading alone. A second latent factor consisting of thethree interference scores from the Stroop task servedas an indicator of the autonomy of lexical processing.To the extent that the presence of autonomous wordreading is a component of automaticity, it should berelated to word-reading fluency, text-reading fluency,and, perhaps, reading comprehension.

The proposed theoretical model was submittedto LISREL. The resulting analysis provided an ac-ceptable fit according to most of the fit indexes forthe first-grade data, χ2 (31) = 37.54, p = .190; RMSEA = .03; GFI = .93; TLI = .99; and third-grade data, χ2(31) = 50.10, p = .016; RMSEA =.067; GFI = .90; TLI = .91, but a less acceptable fitfor the second-grade data, χ2(31) = 61.99, p < .001;RMSEA = .11; GFI = .87; TLI = .91. However, giv-en the similarity between first- and third-grade data,it seemed more parsimonious to examine the samemodel for second grade. The model is presented inFigures 2a, 2b, and 2c, and exact t-levels for stan-dardized estimates for paths in Table 3; all factorloadings for each grade were significant at the p < .05level. Solid arrows represent significant paths anddashed lines represent nonsignificant paths.

An examination of the paths leading to text-reading fluency as an outcome measure indicated asignificant direct path between the latent word-reading–fluency factor and text-reading fluency forall grades (p < .05). In contrast, the direct path be-tween the autonomous reading interference latentfactor and text-reading fluency was not significant atany grade (all p > .10). Moreover, the correlation be-tween autonomous word reading and the word-reading–fluency latent factors was small and non-significant at first (r = .06), second, (r = .06) andthird grades (r = .06). Taken together, the structuralequations for this model accounted for 61% of thevariance in text-reading fluency in first grade, 74%of the variance in second grade, and 73% of the vari-ance in third grade.

Examination of the direct paths leading toreading comprehension as an outcome measure indi-cated the importance of word-reading fluency inreading comprehension. The direct path between theword-reading fluency factor and reading comprehen-sion was significant at all grades (p < .05). Moreover,

the path between autonomous reading interferenceand reading comprehension was significant for firstgraders (p < .05) but nonsignificant for the othergrades, suggesting this aspect of automaticity is notimportant once basic automaticity is established. It isimportant, however, that this theoretical model pre-dicts that the indirect path from the text-reading flu-ency to reading comprehension outcomes should besignificant. However, at all grades, this relationshipwas nonsignificant (all p > .10). Thus, the anticipat-ed additional variance from text-reading fluency forcomprehension beyond that predicted from word-reading fluency predicted by this model simply wasnot confirmed by this data. Thus, if there are impor-tant contributors of fluent text reading on compre-hension, it is unclear what they are for youngreaders. Taken together, the amount of variance ac-counted for in reading comprehension by the struc-tural equations in this model decreased from first(76%), to second (45%), to third grade (38%).

Simple reading fluency modelThis model would state that reading fluency is

a rather general trait of skilled readers and that, atleast for young readers building fluency, it is a gener-al trait that predicts comprehension. This modeldoes not distinguish text-reading from word-readingfluency skill and represents text-reading fluency asmerely another indicator of reading fluency. Thus, inthis model, children’s scores on both subtests of theTOWRE, mean single-word–reading time and accu-racy on the single-word–naming task,CTOPP–RON, and scores on the GORT–3 allserved as indicators of a reading fluency factor.

Similar to the previous model, it was predictedthat autonomous reading interference should be cor-related with the development of reading fluency, par-ticularly as reading automaticity emerges. This viewwould also seem to state that the emergence of all as-pects of automaticity, that is, fluent reading and au-tonomous reading, should be related to improvedcomprehension. Thus, as before, children’s interfer-ence scores for the Stroop task served as indicators ofthe autonomous reading factor.

We should note that this and the previousmodel are identical in most respects with the excep-tion that the parameters linking text-reading fluencyto reading comprehension and automatic reading totext-reading fluency are now eliminated. With twofewer parameters included, this model is simpler andmore parsimonious than the previous one.

The proposed theoretical model provided areasonable fit according to most of the fit indexes for


FIGURES 2A, B, AND CTEXT READING AS MEDIATOR MODEL

High-GPCHigh-Rime

CTOPP Rapid ObjectNaming


First grade

Autonomousreading

interference


Single-word–naming WPM TOWRE

SWE

.751.04

1.001.00

1.03

.06

.84

–.15

1.03

.20

–.02

1.00.84

–.36

TOWREPDE

GORT–3 Text-Reading Rate

WIAT ReadingComprehension

High-GPCLow-Rime

Nonword

High-GPCHigh-Rime



Second grade

Autonomousreading

interference

Single-wordnaming–accuracy


SWE

.781.05

.911.00

2.33

.06

.94

–.05

.78

.20

.04

1.00.91

–.48

TOWREPDE



High-GPCLow-Rime

Nonword

FIGURE 2B

FIGURE 2A


the first-grade data, χ2(33) = 40.44, p = .170; RM-SEA = .034; GFI = .93; TLI = .99; and third-gradedata, χ2(33) = 51.30, p = .022; RMSEA = .059; GFI= .90; TLI = .92, but a less acceptable fit for the second-grade data χ2(33) = 62.18, p = .002; RMSEA= .10; GFI = .87; TLI = .91. However, as before, giv-en the similarity between first- and third-grade data,it seemed more parsimonious to examine the samemodel for second grade. All factor loadings were sig-nificant at the p < .05 level for all relevant variablesat each grade level. Standardized path weights and t-values can be found in Table 4. All factor loadingsfor each grade were significant at the p < .05 level.

As noted in Table 4 and Figures 3a, 3b, and 3c,for all grades there was a significant path betweenreading fluency and reading comprehension (p < .05)indicating that, as children become fluent readers,they also comprehend what they read better. As be-fore, autonomous reading interference was uncorre-lated with reading fluency at first grade, r = .03, and

second grade r = .07, p > .10, but, by third grade,children who continued to show large amounts ofautonomous reading interference tended to displaysignificantly worse reading fluency (r = –.24, t(31) =2.30, p < .05). Autonomous reading interference wasa significant predictor of reading comprehension infirst grade, t(31) = 2.33, p < .05, but not in secondor third grades, both p > .10. Thus, what utility au-tonomous reading aspect of reading has for readingcomprehension disappears once basic reading fluen-cy is established. Moreover, continued difficulty re-covering from Stroop interference seems to beassociated with difficulties with establishing readingfluency.

It is important to note, however, that an exam-ination of the variance accounted for in readingcomprehension by this structural equation modelpointed to the declining influence of general readingfluency and autonomous reading on reading com-prehension with age. Taken together, the amount of

FIGURES 2A, B, AND C (CONTINUED)TEXT READING AS MEDIATOR MODEL

High-GPCHigh-Rime



Third grade

Autonomousreading

interference



SWE

.99.85

.971.00

1.08

.06

1.07

.07

.72

.18

.13

1.00.45

–.41

TOWREPDE



High-GPCLow-Rime

Nonword

Notes. Solid lines represent significant paths; dashed lines nonsignificant paths. TOWRE = Test of Word-Reading Efficiency (SWE: Sight-WordEfficiency subtest, PDE: Phonemic-Decoding Efficiency subtest); CTOPP–Rapid Object Naming = Comprehensive Test of Phonological Processing,Rapid Object Naming subtest; WIAT = Wechsler Individual Achievement Test; GORT–3 = Gray Oral Reading Test–Third Edition; Figure 2a: firstgrade; 2b: second grade; 2c: third grade.

FIGURE 2C


Path Weight SE t a

First grade Direct pathsWord-reading fluency to

Text-reading rate .84 .08 10.97Reading comprehension 1.03 .10 10.13

Autonomous reading interference toText-reading rate –.02 .10 .25Reading comprehension .20 .09 2.30

Indirect pathText-reading rate to

Reading comprehension –.15 .09 1.67

Second grade Direct pathsWord-reading fluency to

Text-reading rate .94 .08 11.16Reading comprehension .78 .21 3.69

Autonomous reading interference toText-reading rate .04 .12 .35Reading comprehension .10 .16 .62


Reading comprehension –.05 .18 .27

Third grade Direct pathsWord-reading fluency to

Text-reading rate 1.07 .14 7.89Reading comprehension .72 .32 2.23

Autonomous reading interference toText-reading rate .13 .13 1.00Reading comprehension .18 .18 .99


Reading comprehension .07 .23 .31

a tcrit (31) = 2.04, p < .05

TABLE 3STANDARDIZED PATH WEIGHTS, STANDARD ERRORS, AND T-VALUES FOR THE TEXTREADING AS MEDIATOR MODEL

Path Weight SE t a

First grade Direct pathsReading fluency to reading comprehension .90 .10 10.13Autonomous reading interference to reading comprehension .20 .09 2.33

Second grade Direct pathsReading fluency to reading comprehension .73 .10 6.96Autonomous reading interference to reading comprehension .09 .16 .57

Third grade Direct pathsReading fluency to reading comprehension .79 .15 5.27Autonomous reading interference to reading comprehension .17 .17 1.00

a tcrit (31) = 2.04, p < .05

TABLE 4STANDARDIZED PATH WEIGHTS, STANDARD ERRORS, AND T-VALUES FOR THE SIMPLEREADING FLUENCY MODEL


FIGURES 3A, B, AND CSIMPLE READING FLUENCY MODEL

High-GPCHigh-Rime


Reading fluency

First grade

Autonomousreading

interference



SWE

.75.84 1.00

1.00

1.03

.03

.83.90

.201.00.84

–.36

TOWREPDE



High-GPCLow-Rime

Nonword

High-GPCHigh-Rime


Reading fluency

Second grade

Autonomousreading

interference



SWE

.771.06 .91

1.00

2.26

.07

.95.73

.091.00.91

–.48

TOWREPDE



High-GPCLow-Rime

Nonword

FIGURE 3B

FIGURE 3A

variance accounted for in reading comprehension bythe structural equations in this model decreasedfrom 75% in first grade, to 45% in second grade,and 39% in third grade. This supports the commonobservation that, as children become older and theirtexts become more difficult, we need to look beyondreading fluency to other skills and knowledge (i.e.,vocabulary knowledge, world knowledge, inferenc-ing ability) to account for reading comprehensionability.

DiscussionDespite the popularity of automaticity views

regarding the development of reading fluency overthe past 30 years, there have been remarkably few in-vestigations examining the interrelationships betweenthe proposed features of automaticity and the devel-opment of reading. Most typically, investigations

have focused on two or three elements of this view si-multaneously (e.g., Fuchs et al., 2001; Jenkins et al.,2003; Stanovich et al., 1981; Young & Bowers,1995). Although it is impossible to capture in a sin-gle study all of the elements that might be relevant tothe developing role of reading fluency in readingcomprehension, the current study made a compre-hensive attempt to capture most of the importantfactors on which automaticity views have focused.This study provides some clarification regarding howthe elements of automaticity, word-reading fluency,text-reading fluency, and autonomous reading, oper-ate together to produce good comprehension in theearly elementary school years.

Our study examined the suitability of two ver-sions of the automaticity view of reading. For thetext reading as mediator model, fluent readers are saidto be better able to utilize text fluency-related skillssuch as sentence-parsing skills (Schreiber, 1980) andthe contextual activation of word meanings (Jenkins


FIGURES 3A, B, AND C (CONTINUED)SIMPLE READING FLUENCY MODEL

High-GPCHigh-Rime


Reading fluency

Third grade

Autonomousreading

interference



SWE

.98.85 .97

1.00

1.02

–.24

1.02.79

.171.00.44

–.41

TOWREPDE



High-GPCLow-Rime

Nonword

Notes. Solid lines represent significant paths; dashed lines nonsignificant paths. TOWRE = Test of Word-Reading Efficiency (SWE: Sight-WordEfficiency subtest, PDE: Phonemic-Decoding Efficiency subtest); CTOPP–Rapid Object Naming = Comprehensive Test of Phonological Processing,Rapid Object Naming subtest; WIAT = Wechsler Individual Achievement Test; GORT–3 = Gray Oral Reading Test–Third Edition; Figure 3a: firstgrade; 3b: second grade; 3c: third grade.

FIGURE 3C


et al., 2003) to aid in the comprehension of textonce their basic word reading is automatic enough tofree up cognitive resources. The current study foundno evidence for this mediating role of text-readingfluency. Instead, a simpler model treating text-read-ing fluency as merely another indicator of readingfluency skill provided a more appropriate account forour data. Thus, our data supported a simple readingfluency model for the early elementary school years.

This lack of a mediating effect between word-reading fluency and reading comprehension skill fortext-reading fluency is surprising given recent find-ings regarding the contribution of text fluency toreading comprehension (Cain et al., 2004; Jenkins etal., 2003). For example, Jenkins et al. found thattext-reading fluency was the primary factor in pre-dicting reading comprehension.

What can account for the differences betweenthese results and ours? We think this is where a de-velopmental account of reading skill must take place.In virtually all the studies finding additional benefitsfrom text reading, children have tended to be olderand more fundamentally fluent than the children inthe current study (Cain et al., 2004; Jenkins et al.,2003; Nation & Snowling, 1998). Even the skilledolder readers in our study were unable to use the ad-ditional cognitive resources gained by fluent textreading to comprehend text. We know that fluentearly elementary school children use syntactically ap-propriate prosody and expression when they read(Clay & Imlach, 1971; Zutell & Rasinski, 1991).However, some research suggests that the full under-standing of syntax is still under development duringthe age ranges we tested here (Traxler, 2002; Willows& Ryan, 1986). It may be that the imposition ofsyntactically appropriate phrasing during the readingof text is not yet automatic enough to free up re-sources that can be used for comprehension. Further,it is unclear whether children can orchestrate auto-matic word and text reading in the pursuit of betterreading comprehension at this stage of reading (Wolf& Katzir-Cohen, 2001).

Comprehension is always an interaction be-tween skills of the reader and characteristics of thetext. Consequently, it may be that children at thisage are, indeed, able to use their cognitive resourcesgained from fluent text reading to assist in compre-hension, but that early texts are simply not complexenough to demand these additional resources. Aschildren get older, the demands of the texts them-selves will become greater (Hiebert & Fisher, 2005).It may be that it is not until later that children canuse these additional benefits from text-level fluencyto improve their comprehension.

Thus, we conclude that, in the early stages ofthe development of reading fluency, children usetheir word-reading skills to comprehend the mean-ings of text, so the simple reading fluency model ap-plies. In fact, a study by Shinn et al. (1992) suggeststhat this may be the case. They found that, for thirdgraders, a unitary factor placing word- and text-levelskills and comprehension skills on the same factorprovided the best fit for third graders, but by fifthgrade, text-level skills had more closely alignedthemselves with reading comprehension and re-quired a multifaceted view of reading skill.

On a cautionary note, however, the samplesizes that we had available to us were somewhatsmaller than ideal. One rule of thumb for decidingsample sizes in SEM is approximately 100 cases persample. Another is 5 to 10 observations per parame-ter. Our sample sizes are clearly toward the lower endof these guidelines. Consequently, parameters thatwe did not find to be significantly significant or ill-fitting should not be dismissed completely out ofhand. It was the case, however, that the critical para-meter that distinguished the two models (i.e., thelink between text-reading rate and reading compre-hension) varied little across grade level, was veryclose to zero, and was in the opposite direction intwo of the three cases. We think it unlikely that theaddition of 20 or more participants at each gradelevel would have altered our findings dramatically.

One disappointment was our inability to dis-cern the role of orthographic knowledge in the de-velopment of reading fluency and readingcomprehension. We had supposed that includingtimed orthographic tasks might provide us with evi-dence that orthographic knowledge is related to oth-er reading fluency skills. In support of this, we didfind that older children were quicker and more accu-rate in making doublet decisions and in discerningthe correct spelling of a word. This is generally con-sistent with other orthographic research using timedtasks (Gayan & Olson, 2003; Olson et al., 1985).However, as a group, the variables from our ortho-graphic tasks were a source of ill-fit in the measure-ment modeling. Measurement issues are notuncommon in research on orthographic processing.For example, Cunningham, Perry, and Stanovich(2001) found low reliability on several commonlyused orthographic tasks, making interpreting theirrole in reading difficult. Findings on one ortho-graphic task may not always generalize to very simi-lar sorts of tasks (Stanovich & Siegel, 1994). Thedimensionality or covariance among orthographictasks may also change during the early elementaryschool years (Gayan & Olson; Notenboom &

Reitsma, 2003). Finally, the actual use of ortho-graphic information during reading may be partlyattributable to task features that encourage or dis-courage the use of orthographic information(Martensen, Dijkstra, & Maris, 2005). Thus, themere validity of orthographic tasks such as the oneswe used can be questioned (see also Vellutino et al.,1994). Consequently, we hesitate to say that ortho-graphic information is not useful for fluent readingand merely conclude that our current data do notprovide us with a theoretical basis for deciding howit might be used.

A second potential issue for our conclusions isthat our treatment of reading comprehension maynot have been as complex as it needed to be in con-trast with our treatment of reading fluency. It mayhave been that, if we had used a different measure ofreading comprehension, we might have found a dif-ferent role for text-reading fluency on comprehen-sion. While this is possible, we believe our measureof reading comprehension included key features im-portant in the early elementary school years. We se-lected the assessment that we did after evaluating anumber of other standardized assessments of earlyreading comprehension. The WIAT ReadingComprehension test was chosen for its consistencywith what many teachers consider a key indicator ofreading comprehension (i.e., the child can answerquestions about the text; Byers, 1998; Richardson,Anders, Tidwell, & Lloyd, 1991), while avoiding thepresence of false test floors that include items outsideof the domain of reading comprehension (e.g., sin-gle-word decoding only, letter recognition) that aresometimes found in standardized reading compre-hension tests for early elementary school. Other re-search on this test suggested it correlates well withother tests of reading comprehension and that teach-ers view it as valid (Byers; Foegen, Espin, Allinder, &Markell, 2001; Smith & Smith, 1998). Moreover,the test we used assessed both literal and inferentialcomprehension, all important to reading compre-hension (Cain et al., 2004). Still, it would be impor-tant to replicate these findings by using a broaderrange of comprehension measures than was presentin the current study. Moreover, future researchshould consider the issue of the complexity of the to-be-comprehended texts in considering the relation-ship between text-reading fluency andcomprehension.

A third potential issue for our conclusions isthat our measure of text-reading fluency may nothave been as complex as it needed to be because werelied heavily on a single text-reading measure, theGORT–3. However, reading fluency is typically de-

fined as comprising text reading that is quick, auto-matic, and expressive. The current study lacked ameasure of expressiveness during the reading of text.An earlier study by Schwanenflugel et al. (2004) hadfound that one measure of expressiveness, prosody,did not account for significant variance beyondword-reading fluency in predicting reading compre-hension in second- and third-grade children. On theother hand, it is possible that other measures of ex-pressiveness, such as those described by Cowie,Douglas-Cowie, and Wichmann (2002), may haveaccounted for significant variance above and beyondthat accounted for by the GORT–3. The GORT–3stresses rate and accuracy and not expressiveness.Expressiveness may be more relevant to the syntacticprocessing features of text fluency than rate and ac-curacy are. Future research should include a broaderrange of measures directly targeted at this expressive-ness feature of text-reading fluency.

With these limitations in mind, having foundsupport for a simple reading fluency model for earlyelementary school children, we outline our findingswith regard to the development of each of the com-ponents of automaticity. We describe how these ele-ments might fit together to produce goodcomprehension.

A number of descriptions of reading skill notethe foundational quality of quick and accurate wordand text reading to reading skill as a whole (Fuchs etal., 2001; Stanovich, 1980). In the current study, werepresented this skill through a combination of psy-chometric and computerized assessments at theword and text level, and we assessed both accuracyand speed of reading. We also included nonwordreading as an indicator of the ability to recognizeunknown words. We included rapid object naming,because of its linkage in prior research to fluentword and text reading (Wolf & Katzir-Cohen,2001). On all fluency-related tasks, children werefound to increase their speed and accuracy from firstto third grade. Further, this reading-fluency factorwas an important predictor of reading comprehen-sion. In each grade, children who had superior read-ing fluency had better reading comprehension.Thus, our data were consistent with automaticityviews that predict that as children become quickerand more accurate, there are freed-up resources thatcan be used for improved comprehension, but theseresources emanate mainly from superior word-read-ing skills and not from the special skills associatedwith fluent text reading per se.

Autonomy is a second aspect of automaticityaccounts of the development of reading fluency.Autonomous reading has been shown to develop ear-



ly in the development of reading skill. This compo-nent was indicated by interference scores obtainedfrom the Stroop interference task. We found an in-teresting pattern in the relationship between this au-tonomous reading component and othercomponents of our model suggesting that Stroopinterference may change in its meaning between firstand third grade. In first graders, although the effectwas small, Stroop interference was positively relatedto better reading comprehension skill. In secondgrade, Stroop interference was related to neitherreading fluency nor reading comprehension, presum-ably because most children were displaying readinginterference. By third grade, children who were stillshowing large degrees of interference from the con-flicting labels were likely to be less fluent readers.Further, because we identified that the interferencefound in the Stroop task occurred in words havinghigh-probability GPC units, we can state that theirdifficulty probably stemmed from difficulties in uni-tizing these sublexical features.

We think that our study helps better place theautonomy component within automaticity ap-proaches to the development of reading skill. Earlierdiscussions of this component viewed the uncertainrelationship between Stroop interference and wordfluency as problematic for automaticity views ofreading (Stanovich, 1990). Autonomy was originallyconsidered to be the by-product of a large amount ofpractice (Shiffrin & Schneider, 1977) and findinginterference from almost the onset of reading wasviewed as problematic. Recent reconceptualizationsof automaticity suggest that, instead, automatic re-trieval and its concomitant benefits for attentionalresources accrue very quickly in learning a new skill(Logan, 1988, 1997). Thus, the development of au-tonomous reading early in the process of learning toread is to be expected as sublexical patterns becomelearned and have begun to be practiced. This new-found autonomy contributes to the resource benefitsthat enable good reading comprehension. Thus, thecurrent study indicates that there is a place for thiscomponent early in the acquisition of reading andthat it is indicative of increasing skill.

This Stroop interference takes on a differentmeaning once children have achieved autonomousreading. Most children continue to develop increasedfluency in recognizing new words throughout thirdgrade (as our data indicate). By third grade, the re-duced resources that word reading in general nowtakes allow them to use these resources in the Strooptask to inhibit the task features that distract them.However, the persistence of a difficulty in overcom-ing interference from irrelevant print (as indicated by

large interference scores) is a sign of a basic fluencyproblem, which manifests as a significant negativecorrelation between reading fluency and au-tonomous reading interference in our data.

Finally, we have shown that reading fluencyand automatic reading account for considerable vari-ance in children’s reading comprehension through-out the early elementary school years. However,findings also point to the diminishing role that auto-maticity plays in reading comprehension as childrenget older. The variance accounted for by predictorsin the model declined steadily from first to thirdgrade. Presumably, once most children are fluentreaders, factors other than reading fluency becomeimportant for good comprehension. As noted earlier,this may be because other factors, such as generaloral language skills and the ability to draw appropri-ate inferences, become more important as childrenmove from learning to read to reading to learn(Chall, 1996; Nation & Snowling, 1998).

Despite the emphasis of the current study onfluency as a key skill for comprehension, we do notwish to be misunderstood as to what we believe theimplications of our study are for instruction duringthe early grades. No matter what the age of the stu-dents, comprehension is the ultimate goal of readingand should receive emphasis instructionallythroughout. At all grades, there was significant vari-ance in children’s reading comprehension not ac-counted for by fluency. Thus, if anything, ourfindings support the importance of carrying out fluency-oriented instruction alongside comprehen-sion instruction. Further, the key is to rememberthat fluency is an important bridge to comprehen-sion but not the ultimate destination. Once fluentreading has been achieved by children, fluency-oriented instruction can be phased out because, asour findings show, individual differences in compre-hension begin to be increasingly related to thingsother than reading fluency.

In sum, the current study has found supportfor a simple reading fluency view in automaticity ac-counts of the development of fluent and automaticreading during the early elementary school years.Future research should assess the continuing devel-opmental path of reading fluency and these otherlanguage factors in producing good comprehensionas children become older.

PAULA J. SCHWANENFLUGEL is a professor of educationalpsychology, psychology, linguistics, and cognitive science at theUniversity of Georgia, where she teaches courses on childdevelopment, cognition, and psycholinguistics as applied toeducational settings. She has recently been engaged in grants andresearch on reading fluency, preliteracy skills, and vocabulary, and

classroom practices related to these topics. She can be contacted atthe Department of Educational Psychology and InstructionalTechnology, 325R Aderhold Hall, University of Georgia, Athens, GA30602, USA, or by e-mail at [email protected].

ELIZABETH B. MEISINGER is a center manager for Youth & FamilyServices in the Dallas Independent School District. Her researchinterests include the normal development of reading skills with a focuson reading fluency, the assessment of children with readingdisabilities, and the integration of academic and psychologicalinterventions for children in the public school setting. She can becontacted at Youth & Family Services, Dallas Independent SchoolDistrict, 2909 N. Buckner Drive, Dallas, TX 75228, USA, or by e-mail at [email protected].

JOSEPH M. WISENBAKER recently retired as an associate professorof educational psychology at the University of Georgia, where hetaught courses in statistical methods and research design. In additionto working with colleagues across a wide range of disciplines, he hasworked extensively with schools in the evaluation of educationalprograms. He can be contacted at Department of EducationalPsychology & Instructional Technology, 325 Aderhold Hall, Universityof Georgia, Athens, GA 30602, USA, or by e-mail [email protected].

MELANIE R. KUHN is an assistant professor in the Department ofLearning and Teaching at the Rutgers Graduate School of Education,where she teaches courses on assessment and instruction forstruggling readers. She has been engaged in research grants onfluency, assistive technology, and, most recently, as part of the Mid-Atlantic Collaborative for Applied Research in Education. Her researchinterests also include literacy instruction for struggling readers,comprehension development, and vocabulary instruction. She can becontacted at the Graduate School of Education, Rutgers University, 10Seminary Place, New Brunswick, NJ 08901-1183, USA, or through e-mail at [email protected].

GREGORY P. STRAUSS is a doctoral student in clinical psychology atthe University of Nevada–Las Vegas. He conducts research in theareas of cognition, neuropsychology, psychopathology, and readingfluency. He can be reached at the Department of Psychology,University of Nevada, Las Vegas, 4505 Maryland Parkway, Box 5030,Las Vegas, NV, USA 89154-5030, USA, or through e-mail [email protected].

ROBIN D. MORRIS is the vice president for research and RegentsProfessor of Psychology at Georgia State University. He also holds ajoint appointment in the Department of Educational Psychology andSpecial Education in the College of Education. His current research isfocused on reading and language development, reading disabilitiesand dyslexia, bilingual language and reading development, and theneuroimaging of the developing brain. He can be contacted at theDepartment of Psychology, Georgia State University, PO Box 5010,Atlanta, GA 30302-5010, USA, or through e-mail [email protected].

R E F E R E N C E SADAMS, M.J. (1990). Beginning to read: Thinking and learning about

print. Cambridge, MA: MIT Press.ANDERSON, J.R. (1987). Skill acquisition: Compilation of weak-

method problem solutions. Psychological Review, 94, 192–210.BACKMAN, J., BRUCK, M., HEBERT, M., & SEIDENBERG,

M.S. (1984). Acquisition and use of spelling–sound correspondences in

reading. Journal of Experimental Child Psychology, 38, 114–133.BERGEN, L., GRIMES, T., & POTTER, D. (2005). How attention

partitions itself during simultaneous message presentations. HumanCommunication Research, 31, 311–336.

BERNDT, R.S., REGGIA, J.A., & MITCHUM, C.C. (1987).Empirically derived probabilities for grapheme-to-phoneme correspon-dences in English. Behavior Research Methods, Instrumentation, &Computers, 19, 1–9.

BOWERS, P.G. (1993). Text reading and rereading: Determinants offluency beyond word recognition. Journal of Reading Behavior, 25,133–153.

BOWERS, P.G., & SWANSON, L.B. (1991). Naming speed deficitsin reading disability: Multiple measures of a singular process. Journal ofExperimental Child Psychology, 51, 195–219.

BREZNITZ, Z., & BERMAN, L. (2003). The underlying factors ofword reading rate. Educational Psychology Review, 15, 247–265.

BYERS, J.A. (1998). Content and concurrent validity of the WIATand WJ–R reading subtests for second grade students. DissertationAbstracts International Section A: Humanities and Social Sciences, 58, 3100.

CAIN, K., OAKHILL, J., & BRYANT, P. (2004). Children’s readingcomprehension ability: Concurrent prediction by working memory, ver-bal ability, and component skills. Journal of Educational Psychology, 94,31–42.

CARROLL, J.B., DAVIES, P., & RICHMAN, B. (1971). TheAmerican Heritage word frequency book. New York: Houghton-Mifflin.

CASSAR, M., & TREIMAN, R. (1997). The beginnings of ortho-graphic knowledge: Children’s knowledge of double letters in words.Journal of Educational Psychology, 89, 631–644.

CASTEEL, M.A. (1993). Effects of inference necessity and readinggoal on children’s inferential generation. Developmental Psychology, 29,346–357.

CHALL, J.S. (1996). Stages of reading development (2nd ed.). FortWorth, TX: Harcourt Brace.

CLAY, M.M., & IMLACH, R.H. (1971). Juncture, pitch, and stressin reading. Journal of Verbal Learning and Verbal Behavior, 10, 133–139.

COLTHEART, V., & LEAHY, J. (1992). Children’s and adults’reading of nonwords: Effects of regularity and consistency. Journal ofExperimental Psychology: Learning, Memory, and Cognition, 18, 718–729.

COMALLI, P.E., JR., WAPNER, S., & WERNER, H. (1962).Interference effects of Stroop color-word test in childhood, adulthood,and aging. Journal of Genetic Psychology, 100, 47–53.

COWIE, R., DOUGLAS-COWIE, E., & WICHMANN, A. (2002).Prosodic characteristics of skilled reading: Fluency and expressiveness in8–10 year-old readers. Language and Speech, 45, 47-82.

CRAIG, H.K., THOMPSON, C.A., WASHINGTON, J.A., &POTTER, S.L. (2004). Performance of elementary grade African-American students on the Gray Oral Reading Tests. Language, Speech andHearing Services in Schools, 35, 141–154.

CUNNINGHAM, A.E., PERRY, K.E., & STANOVICH, K.E.(2001). Converging evidence for the concept of orthographic processing.Reading & Writing: An Interdisciplinary Journal, 14, 549–568.

CUNNINGHAM, A.E., & STANOVICH, K.E. (1990). Assessingprint exposure and orthographic processing skill in children: A quick mea-sure of reading experience. Journal of Educational Psychology, 82, 733–740.

CYCOWICZ, Y.M., FRIEDMAN, D., ROTHSTEIN, M., &SNODGRASS, J.G. (1997). Picture naming by young children: Normsfor name agreement, familiarity, and visual complexity. Journal ofExperimental Child Psychology, 65, 171–237.

DOWHOWER, S.L. (1987). Effects of repeated reading on second-grade transitional readers’ fluency and comprehension. Reading ResearchQuarterly, 22, 389–406.

EHRI, L.C. (1976). Do words really interfere in naming pictures?Child Development, 47, 502–505.

FOEGEN, A., ESPIN, C.A., ALLINDER, R.M., & MARKELL,M.A. (2001). Translating research into practice: Preservice teachers’ beliefsabout curriculum-based measurement. Journal of Special Education, 34226–236.

FOORMAN, B., FRANCIS, D., FLETCHER, D., & LYNN, A.(1996). Relation of phonological and orthographic processing to earlyreading: Comparing two approaches to regression-based reading-level-match designs. Journal of Educational Psychology, 88, 639–652.

FUCHS, L.S., FUCHS, D., HOSP, M.K., & JENKINS, J.R. (2001).Oral reading fluency as an indicator of reading competence: A theoreti-cal, empirical, and historical analysis. Scientific Studies of Reading, 5,239–256.

GAYAN, J., & OLSON, R.K. (2003). Genetic and environmental in-



fluences on individual differences in printed word recognition. Journal ofExperimental Child Psychology, 84, 97–123.

GOLINKOFF, R.M., & ROSINSKI, R.R. (1976). Decoding, se-mantic processing, and reading comprehension skill. Child Development,47, 252–258.

GOSWAMI, U., ZIEGLER, J.C., DALTON, L., & SCHNEIDER,W. (2003). Nonword reading across orthographies: How flexible is thechoice of reading units? Applied Psycholinguistics, 24, 235–247.

GOTTARDO, A., CHIAPPE, P., SIEGEL, L.S., & STANOVICH,K.E. (1999). Patterns of word and nonword processing in skilled andless-skilled readers. Reading and Writing: An Interdisciplinary Journal, 11,465–487.

GOUGH, P.B. (1996). How children learn to read and why theyfail. Annals of Dyslexia, 46, 3–20.

GUTTENTAG, R.E., & HAITH, M.M. (1978). Automatic process-ing as a function of age and reading ability. Child Development, 49,707–716.

HANNON, B., & DANEMAN, M. (2001). A new tool for measur-ing and understanding individual differences in the component processesof reading comprehension. Journal of Educational Psychology, 93, 103–128.

HARBER, J.R. (1979). Syntactic complexity: A necessary ingredientin predicting readability. Journal of Learning Disabilities, 12, 13–19.

HASBROUCK, J.E., & TINDAL, G. (1992). Curriculum-based oralreading fluency norms for students in grades 2 through 5. TeachingExceptional Children, 24, 41–44.

HIEBERT, E.H., & FISHER, C.W. (2005). A review of the NationalReading Panel’s studies on fluency: The role of text. The ElementarySchool Journal, 105, 443–460.

HORN, C.C., & MANIS, F.R. (1987). Development of automaticand speeded reading of printed words. Journal of Experimental ChildPsychology, 44, 92–108.

JENKINS, J.R., FUCHS, L.S., VAN DEN BROEK, P., ESPIN, C.,& DENO, S.L. (2003). Sources of individual differences in reading com-prehension and reading fluency. Journal of Educational Psychology, 95,719–729

JÖRESKOG, K., & SÖRBOM, D. (1996). LISREL 8: User’s referenceguide. Chicago: Scientific Software International.

JÖRESKOG, K., & SÖRBOM, D. (1999). LISREL 8.3 [Computersoftware]. Chicago: Scientific Software International.

JUEL, C. (1988). Learning to read and write: A longitudinal study of54 children from first to fourth grades. Journal of Educational Psychology,80, 437–447.

KATZIR, T., SHAUL, S., BREZNITZ, Z., & WOLF, M. (2004).The universal and the unique in dyslexia: A cross-linguistic investigationof reading and reading fluency in Hebrew- and English-speaking chil-dren with reading disorders. Reading & Writing, 17, 739–768.

KLINE, R.B. (2005). Principles and practice of structural equationmodeling. New York: Guilford.

KUHN, M.R., & STAHL, S.A. (2003). Fluency: A review of devel-opmental and remedial practices. Journal of Educational Psychology, 95,3–21.

LABERGE, D., & SAMUELS, S.J. (1974). Toward a theory of au-tomatic information processing in reading. Cognitive Psychology, 6,293–323.

LAING, E., & HULME, C. (1999). Phonological and semanticprocesses influence beginning readers’ ability to learn to read words.Journal of Experimental Child Psychology, 73, 183–207.

LOGAN, G.D. (1985). Skill and automaticity: Relations, implica-tions, and future directions. Canadian Journal of Psychology, 39, 367–386.

LOGAN, G.D. (1988). Toward an instance theory of automatization.Psychological Review, 95, 492–527.

LOGAN, G.D. (1997). Automaticity and reading: Perspectives fromthe instance theory of automatization. Reading and Writing Quarterly:Overcoming Learning Difficulties, 13, 123–146

LOGAN, G.D., & KLAPP, S.T. (1991). Automatizing alphabetarithmetic: I. Is extended practice necessary to produce automaticity?Journal of Experimental Psychology: Learning, Memory, and Cognition, 17,179–195

LONG, D.L., OPPY, B.J., & SEELY, M. (1997). Individual differ-ences in readers’ sentence and text-level representations. Journal of Memoryand Language, 36, 129–145.

LYON, G.R. (1995). Toward a definition of dyslexia. Annals ofDyslexia, 45, 3–27.

LYON, G.R., SHAYWITZ, S.E., & SHAYWITZ, B.A. (2003). A de-finition of dyslexia. Annals of Dyslexia, 53, 1–14.

MANIS, F.R., CUSTODIO, R., & SZESZULSKI, P.A. (1993).

Development of phonological and orthographic skill: A 2-year longitudi-nal study of dyslexic children. Journal of Experimental Child Psychology, 56,64–86.

MANIS, F.R., SEIDENBERG, M.S., & DOI, L.M. (1999). See DickRAN: Rapid naming and the longitudinal prediction of reading subskillsin first and second graders. Scientific Studies of Reading, 3, 129–157.

MARLOW, A., & EDWARDS, R.P. (1998). Test review: Gray OralReading Test, Third Edition (GORT–3). Journal of PsychoeducationalAssessment, 16, 90–94.

MARTENSEN, H., DIJKSTRA, T., & MARIS, E. (2005). A werd isnot quite a word: On the role of sublexical phonological information in vi-sual lexical decision. Language and Cognitive Processes, 20, 513–552.

MCFALLS, E.M., SCHWANENFLUGEL, P.J., & STAHL, S.(1996). Influence of word meaning on the acquisition of a reading vo-cabulary in second-grade children. Reading and Writing: AnInterdisciplinary Journal, 8, 235–250.

MORRIS, R.D., STUEBING, K.K., FLETCHER, J.M., SHAY-WITZ, S., LYON, R., SHANKWEILER, D., ET AL. (1998). Subtypesof reading disability: Variability around a phonological core. Journal ofEducational Psychology, 90, 347–373.

NATION, K., & SNOWLING, M.J. (1998). Semantic processingand the development of word-recognition skills: Evidence from childrenwith reading comprehension difficulties. Journal of Memory & Language,39, 85–101.

NATIONAL INSTITUTE OF CHILD HEALTH AND HUMANDEVELOPMENT. (2000). Report of the subgroups: National ReadingPanel. Washington, DC: U.S. Government Printing Office.

NICHOLSON, T. (1999). Reading comprehension processes. InG.B. Thompson & T. Nicholson (Eds.), Learning to read: Beyond phonicsand whole language (pp. 127–149). New York: Teachers College Press.

NOTENBOOM, A., & REITSMA, P. (2003). Investigating the di-mensions of spelling ability. Educational & Psychological Measurement,63, 1039–1059.

OLSON, R.K., GILLIS, J.J., RACK, J.P., DEFRIES, J.C., & FULK-ER, D.W. (1991). Confirmatory factor analysis of word recognition andprocess measures in the Colorado Reading Project. Reading & Writing, 3,235–248.

OLSON, R.K., KLIEGL, R., DAVIDSON, B.J., & FOLTZ, G.(1985). Individual and developmental differences in reading disability.In G.E. MacKinnon & T.G. Waller (Eds.), Reading research: Advances intheory and practice, Vol. 4 (pp. 1–64). New York: Academic Press.

PAAP, K.R., & OGDEN, W.C. (1981). Letter encoding is an oblig-atory but capacity-demanding operation. Journal of ExperimentalPsychology: Human Perception and Performance, 7, 518–527.

PENNINGTON, B.F., LEFLY, D.L., VAN ORDEN, G.C.,BOOKMAN, M.O., & SMITH, S.D. (1987). Is phonology bypassed innormal or dyslexic development? Annals of Dyslexia, 37, 62–89.

PERFETTI, C.A. (1985). Reading ability. London: Oxford UniversityPress.

PERFETTI, C.A., & HOGABOAM, T. (1975). Relationship be-tween single word decoding and reading comprehension skill. Journal ofEducational Psychology, 67, 461–469.

PERRY, C., ZIEGLER, J.C., & COLTHEART, M. (2002). Howpredictable is spelling? Developing and testing metrics of phoneme-grapheme contingency. The Quarterly Journal of Experimental Psychology,55A, 897–915.

PRITCHARD, V.E., & NEUMANN, E. (2004). Negative primingeffects in children engaged in nonspatial tasks: Evidence of early develop-ment of an intact inhibitory mechanism. Developmental Psychology, 40,191–203.

RACK, J.P., SNOWLING, M.J., & OLSON, R.K. (1992). The non-word reading deficit in developmental dyslexia: A review. Reading ResearchQuarterly, 27, 28–53.

RASINSKI, T.V. (1990). Investigating measures of fluency.Educational Research Quarterly, 14, 34–44.

RICHARDSON, V., ANDERS, P., TIDWELL, D., & LLOYD, C.(1991). The relationship between teachers’ beliefs and practices in readingcomprehension instruction. American Educational Research Journal, 28,559–586.

SCARBOROUGH, H.S. (1998). Predicting the future achievementof second graders with reading disabilities: Contributions of phonemicawareness, verbal memory, rapid naming, and IQ. Annals of Dyslexia, 48,115–136.

SCHADLER, M., & THISSEN, D.M. (1981). The development ofautomatic word recognition and reading skill. Memory & Cognition, 9,132–141.

SCHNEIDER, W., DUMAIS, S.T., & SHIFFRIN, R.M. (1984).Automatic and control processing and attention. In R. Parasuraman &R. Davies (Eds.), Varieties of attention (pp. 1–27). New York: AcademicPress.

SCHNEIDER, W., ESCHMAN, A., & ZUCCOLOTTO, A.(2001). E-prime user’s guide. Pittsburgh, PA: Psychology Software Tools.

SCHREIBER, P.A. (1980). On the acquisition of reading fluency.Journal of Reading Behavior, 12, 177–186.

SCHUMACKER, R.E., & LOMAX, R.G. (1996). A beginner’s guideto structural equation modeling. Mahwah, NJ: Erlbaum.

SCHWANENFLUGEL, P.J., & AKIN, C.E. (1994). Developmentaltrends in lexical decisions for abstract and concrete words. ReadingResearch Quarterly, 29, 251–263.

SCHWANENFLUGEL, P.J., HAMILTON, A.M., KUHN, M.R.,WISENBAKER, J., & STAHL, S.A. (2004). Becoming a fluent reader:Reading skill and prosodic features in the oral reading of young readers.Journal of Educational Psychology, 96, 119–129.

SCHWANENFLUGEL, P.J., & NOYES, C.R. (1996). Contextavailability and the development of word reading skill. Journal of LiteracyResearch, 28, 35–54.

SHIFFRIN, R.M., & SCHNEIDER, W. (1977). Controlled andautomatic human information processing: II. Perceptual learning,Automatic attending, and a general theory. Psychological Review, 84,127–190.

SHINN, M.R., GOOD, R.H., KNUTSON, N., TILLY, W.D., &COLLINS, V.C. (1992). Curriculum-based measurement of oral read-ing fluency: A confirmatory analysis of its relation to reading. SchoolPsychology Review, 21, 459–479.

SMITH, T.D., & SMITH, B.L. (1998). Relationship between theWide Range Achievement Test 3 and the Wechsler IndividualAchievement Test. Psychological Reports, 83, 963–967.

STAGE, S.A., SHEPPARD, J., DAVIDSON, M.M., & BROWN-ING, M.M. (2001). Prediction of first-graders’ growth in oral reading flu-ency using kindergarten letter fluency. Journal of School Psychology, 39,225–237.

STANOVICH, K.E. (1980). Toward an interactive-compensatorymodel of individual differences in the development of reading fluency.Reading Research Quarterly, 16, 32–71.

STANOVICH, K.E. (1985). Explaining the variance in readingAbility in terms of psychological processes: What have we learned? Annalsof Dyslexia, 35, 67–96.

STANOVICH, K.E. (1990). Concepts in developmental theories ofreading skill: Cognitive resources, automaticity, and modularity.Developmental Review, 10, 72–100.

STANOVICH, K.E., CUNNINGHAM, A.E., & WEST, R.F.(1981). A longitudinal study of the development of automatic recognitionskills in first graders. Journal of Reading Behavior, 13, 57–74.

STANOVICH, K.E., NATHAN, R.G., & VALA-ROSSI, M.(1986). Developmental changes in the cognitive correlates of reading abil-ity and the developmental lag hypothesis. Reading Research Quarterly, 21,267–283.

STANOVICH, K.E., NATHAN, R.G., & ZOLMAN, J.E. (1988).The developmental lag hypothesis in reading: Longitudinal and matchedreading-level comparisons. Child Development, 59, 71–86.

STANOVICH, K.E., & SIEGEL, L.S. (1994). Phenotypic perfor-mance profile of children with reading disabilities: A regression-basedtest of the phonological-core variable-difference model. Journal ofEducational Psychology, 86, 24–53.

STROOP, J.R. (1935). Studies of interference in serial verbal reac-tions. Journal of Experimental Psychology, 18, 643–662.

TIPPER, S.P., BOURQUE, T.A., ANDERSON, S.H., & BRE-HAUT, J.C. (1989). Mechanisms of attention: A developmental study.Journal of Experimental Child Psychology, 48, 353–378.

TORGESEN, J.K., WAGNER, R.K., & RASHOTTE, C.A. (1999).Test of Word Reading Efficiency. Austin, TX: Pro-Ed.

TRAXLER, M.J. (2002). Plausibility and subcategorization prefer-ence in children’s processing of temporarily ambiguous sentences:Evidence from self-paced reading. Quarterly Journal of ExperimentalPsychology, 55A, 75–96.

TREIMAN, R., GOSWAMI, U., & BRUCK, M. (1990). Not allnonwords are alike: Implications for reading development and theory.Memory & Cognition, 13, 357–364.

TREIMAN, R., MULLENNIX, J., BIJELJAC-BABIC, R. &RICHMOND-WELTY, E.D. (1995). The special role of rimes in the de-scription, use, and acquisition of English orthography. Journal ofExperimental Psychology: General, 124, 107–136.

VELLUTINO, F.R., SCANLON, D.M., & TANZMAN, M.S.(1994). Components of reading ability: Issues and problems in opera-tionalizing word identification, phonological coding, and orthographiccoding. In G.R. Lyon (Ed.), Frames of reference for the assessment of learn-ing disabilities: New views on measurement issues (pp. 279–332). Baltimore:Paul H. Brookes.

VENEZKY, R.L. (1999). The American way of spelling: The structureand origins of American English orthography. New York: Guilford.

WATERS, G.S., SEIDENBERG, M.S., & BRUCK, M. (1984).Children’s and adults’ use of spelling-sound information in three readingtasks. Memory & Cognition, 12, 293–305.

WILLOWS, D.M., & RYAN, E.B. (1986). The development ofgrammatical Sensitivity and its relationship to early reading achievement.Reading Research Quarterly, 21, 253–266.

WOLF, M., BALLY, H., & MORRIS, R. (1986). Automaticity, re-trieval processes, and reading: A longitudinal study in average and im-paired readers. Child Development, 57, 988–1000.

WOLF, M., & BOWERS, P. (1999). The double-deficit hypothesisfor the developmental dyslexias. Journal of Educational Psychology, 91,415–438.

WOLF, M., BOWERS, P.G., & BIDDLE, K. (2000). Naming-speedprocesses, timing, and reading: A conceptual review. Journal of LearningDisabilities, 33, 387–407.

WOLF, M., & KATZIR-COHEN, T. (2001). Reading fluency andits intervention. Scientific Studies of Reading, 5, 211–239.

YOUNG, A. & BOWERS, P.G. (1995). Individual difference andtext difficulty determinants of reading fluency and expressiveness. Journalof Experimental Child Psychology, 60, 428–454.

ZENO, S.M., IVENS, S.H., MILLARD, R.T., & DUVVURI, R.(1995). The educator’s word frequency guide. Brewster, NY: TouchstoneApplied Science Associates.

ZUTELL, J., & RASINSKI, T.V. (1991). Training teachers to attendto their students’ oral reading fluency. Theory Into Practice, 30, 211–217.

A U T H O R S ’ N O T EWe thank Samantha Johnson, Caroline Groff, Franklin Turner, Ann

Marie Hamilton, Barbara Bradley, Matthew Quirk, and Deborah Woofor assisting in the collection of data. The project was supported by theInteragency Education Research Initiative, a program of research managedjointly by the National Science Foundation, the Institute of EducationSciences in the U.S. Department of Education, and the National Instituteof Child Health and Human Development in the National Institutes ofHealth (NICHD/NIH). Funding for the project was provided byNICHD NIH Grant 7 R01 HD040746-06.

Received May 26, 2005Final revision received March 31, 2006

Accepted April 4, 2006


Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Documents

Becoming a fluent and automatic reader ... - web.macam.ac.ilweb.macam.ac.il/~offprint/79934.pdfautomática en los primeros años de escuela primaria y b) determinar si las habilidades