This article was downloaded by: [Washburn University]On: 02 November 2014, At: 22:39Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registeredoffice: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Journal of Research in ChildhoodEducationPublication details, including instructions for authors andsubscription information:http://www.tandfonline.com/loi/ujrc20

Phonemic Awareness and BeginningReading and WritingConstance Kamii a & Maryann Manning aa The University of Alabama at BirminghamPublished online: 03 Nov 2009.

To cite this article: Constance Kamii & Maryann Manning (2002) Phonemic Awareness andBeginning Reading and Writing, Journal of Research in Childhood Education, 17:1, 38-46, DOI:10.1080/02568540209594997

To link to this article: http://dx.doi.org/10.1080/02568540209594997

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the“Content”) contained in the publications on our platform. However, Taylor & Francis,our agents, and our licensors make no representations or warranties whatsoever as tothe accuracy, completeness, or suitability for any purpose of the Content. Any opinionsand views expressed in this publication are the opinions and views of the authors,and are not the views of or endorsed by Taylor & Francis. The accuracy of the Contentshould not be relied upon and should be independently verified with primary sourcesof information. Taylor and Francis shall not be liable for any losses, actions, claims,proceedings, demands, costs, expenses, damages, and other liabilities whatsoever orhowsoever caused arising directly or indirectly in connection with, in relation to or arisingout of the use of the Content.

This article may be used for research, teaching, and private study purposes. Anysubstantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Journal of Research in Chi ldhoo d Educa tion200 2. Vol. 17. No. I

Copyright 2002 by the Assoc iation forChildhood Education International

0256-8543/02

Phonemic Awareness and Beginning Reading and Writing

Constance KamiiMaryann ManningThe University of Alabama at Birmingham

Abstract. vernon and Ferreiro (1999) took Spanish-speaking kindergartners'levels of writing as an independent variable and their performance on oral­segmentation tasks as the dependent variable. They found a highly significantmultiple correlation R and concluded that: 1) children's ability to deal withoral-segmentation tasks seems to depend on their knowledge of the writing sys­tem and 2) the ability to segment words into phonemes is not a cause or prereq­uisite for learning to read and write. To find out ifa similar relationship can befound among English-speaking kindergartners, we gave 68 kindergartners awriting task and two oral-segmentation tasks similar to those used by vernonand Ferreiro. In the writing task, we asked each child to write four pa irs ofwords-"ham" and "ham ster," "butte r" and "butterfly," "key" and "m onkey," and"gum " and "bubblegum ," We, too, found a close relationship between children'slevels of writing and their levels oforal segmentation, and concluded that the irknowledge of our writing system enables them to write at a higher level and tosegment words phonemically.

Summarizing what is known about phone­mic awareness and beginning reading,Goswami (2000) said, "Agreement on theimportance of phoneme awareness for read­ing development is universal. It is prob­ably true to say that every study that hasmeasured the relationship between phone­mic awareness and progress in reading hasfound a positive connection" (p. 255). Someresearchers go further and state that pho­nemic awareness causes progress in earlyreading, as can be seen in the followingquote from Torgesen (1998):

Th e most common cause of difficulties acquir­ing early word reading sk ills is weakness in theability to process the phonological features oflanguage (Liberm an, Shankweiler, & Liberman,1989). This is perhaps the most important dis­covery about reading difficulties in the lasttwenty years . . . . Much of our new confidencein being able to identify children at risk for read­ing failure before reading instruction beginsdepends on the use of tests of phonemic aware­ness, since this ability has been shown to becausally related to the growth of early word read­ing skills (Lundberg, Frost, & Peterson, 1988;Wagner et al., 1997). (p. 33)

38

Many others, su ch as Adams, Foorman,Lundberg, and Beeler (1998a, 1998b) andStanovich (1993), have likewise stated thatthere is a causal relationship between pho­nemic awareness and progress in earlyreading.

On the basis of more than two decades ofresearch inspired by Piaget's theory, Vernonand Ferreiro (1999 ) reviewed the researchon phonemic awareness and focused theirattention on young children's writing. Theypointed out that reading is emphasized overwriting in English-speaking countries, butthat we can more clearly infer children'sknowledge about the nature of our writingsystem when we ask them to write a wordor a phrase. One ofthe hypotheses Vernonand Ferreiro advanced was that children'sknowledge of the writing system as ex­pressed in their writing could be consideredan independent variable to predict theirperformance on oral segmentation tasks.Their reason for advancing this hypothesiswas that they were familiar with youngchildren's progress in writing from scrib­bling to syllabic writing and then toalphabetico-syllabic writing. They per­formed a multiple regression analysis in

Dow

nloa

ded

by [

Was

hbur

n U

nive

rsity

] at

22:

39 0

2 N

ovem

ber

2014

PHONEMIC AWARENESS AND BEGINNING READING AND WRITING

which a child's score for level of writing wasan independent variable, and his or herscore for phonetic segmentation was thedependent variable; they found a multiplecorrelation R of .83817 (p = .0000 ). Theyconcluded that children's ability to deal withoral-segmentation tasks "seemed to dependon their overall knowledge of the writingsystem as expressed in their written pro­ductions" (p. 410 ). In other words, accord­ing to Vernon and Ferreiro's research,ability to segment words into phonemes isnot a cause or prerequisite for learning howto read and write.

Vernon and Ferreiro's research was con­ducted in Mexico with Spanish-speakingchildren, and we wondered whether a studywith English-speaking American childrenmight lead to the same conclusion. In thepast, we found that English-speakingAmerican children develop in ways that arevery similar to Spanish-speaking children,but we also found a few differences becauseof the differences between the two lan­guages (Kamii, Long, Manning, & Manning,1990). We therefore decided to replicateparts oftheir study to examine the relation­ship between children's development inwriting and their development in phonologi­cal analysis.

MethodThe participants were 68 kindergartnersattending a public school in a middle- toupper-middle-class neighborhood near Bir­mingham, Alabama. The children were insix kindergarten classes and had been het­erogeneously assigned to each group. Allthe children in the six classrooms whoseparents gave permission for them to be in­terviewed were included in the study.Eighty-five percent of the children werewhite, and the other 15% were AfricanAmerican. There were 36 boys and 32 girls,with the average age of 5 years, 5 months,and a range from 5 years, 0 months to 5years, 11 months at the time of the inter­views, which took place in October and earlyNovember.

All the students had received a few weeksof holistic reading and writing instruction

as part ofthe regular kindergarten curricu­lum, including daily journal writing, dem­onstration writing, and shared reading.Thus, the phonemic awareness and phon­ics instruction was not done in isolation.

Each child was individually interviewedtwice, on two consecutive days if at all pos­sible. A writing task was given on the firstday, and two phonemic-segmentation taskswere given on the second day.

The Writing TaskThe child was brought to a small, quietroom for the interview and, after a briefconversation, was given a pencil and ablank sheet of paper and asked to write hisor her name on it. The interviewer thenasked the child to listen carefully and ut­tered a pair of words-"ham" and ''ham­ster"-with the syllables pronounced in adeliberate manner so that "ham" wouldclearly be heard twice. The child was thenasked to repeat both words. After thispreparation, the words were uttered one ata time, and the child was asked to writethem and to read them back, pointing tothe letters he or she had written.

Four pairs of words were given in thisway-"ham" and "hamster," "butter" and"butterfly," "key" and "monkey," and "gum"and "bubblegum." These words were cho­sen because they were familiar and unlikelyto have been practiced, and because thelonger word of each pair contained the samesounds as the shorter word. We found in aprevious study (Kamii & Manning, 1999)that these kinds of words enabled us toidentify children who could not yet writewith "invent ed" spelling, but who wrotemore letters for longer spoken words,thereby revealing their knowledge thatwriting is related to speech. For example,one child wrote ATKO for "ham" andMSTGR for "hamster," eTPK for "butter"and PAWHQ for "butterfly," BjiW for "berry"and iNARS for "str awberry," and KRVe for"melon" and JNASQ for "watermelon." Thischild did not use letters conventionally, butconsistently wrote four letters for theshorter of each pair of words, and five let­ters for the longer word.

39

Dow

nloa

ded

by [

Was

hbur

n U

nive

rsity

] at

22:

39 0

2 N

ovem

ber

2014

Figure 1An Example ofSublevel IA

bubblegumFigure 2

An Example ofSublevel IB

KAMII AND MANNING

The criteria used to categorize the re­sponses. The criteria used to categorize theresponses were based on previous researchby Ferreiro and Teberosky (1979/1982),Vernon and Ferreiro (1999), Kamii, Long,Manning, and Manning (1990), Kamii andManning (1999), and Kamii, Long, andManning (2001). Itwas necessary to modifyVernon and Ferreiro's criteria considerablybecause their work was based on Spanish­speaking children, who generally begin torelate writing to speech by writing one let­ter (often a vowel) for each syllable (e.g.,writing "OlE" for "perico" [parrot], a three­syllable word). Syllables and vowel soundsare much clearer in Spanish than in En­glish. English is primarily a stress-timedlanguage, in which vowel sounds and syl­labic boundaries are clear only in stressedsyllables. English-speaking children thusbegin to relate writing to speech by writingconsonants (e.g., writing "PNMT" for "pun­ishment").

Together, we categorized the 68 children'sresponses into the following six levels. M­ter categorizing the responses, we made anoverall judgment of each child's level byusing the criterion of "more than half ofthe responses ." For example, when wewanted to decide whether or not a childwrote more letters for the longer of each pairof words, we asked whether the child wrotemore letters for the longer word in at leastthree of the four pairs of words.

od

ham

hamster

butter

butterfly

key

monkey

gum

ham

hamsterbutter

butterfly

key

monkey

gum

bubblegum

Levell. Children at this level have notconstructed the theory about the natureof the English writing system that writ­ing is related to speaking. Level I in­cludes the following three sublevels,which were considered distinct in ourprevious studies:

Sublevel IA Drawing a picture foreach word (see Figure 1)

Sublevel IE. Writing strings of ran­dom letters, with no relationship be­tween the length of a spoken wordand the number ofletterswritten (see Figure 2)

Sublevel Ie. Writing all the words witha minimum quantity (usu ally two )

40

Figure 3An Example ofSublevel Ie

ham

hamster

butter

butterfly

key

monkey

gum

bubblegum

Dow

nloa

ded

by [

Was

hbur

n U

nive

rsity

] at

22:

39 0

2 N

ovem

ber

2014

PHONEMIC AWARENESS AND BEGINNING READING AND WRITING

and approximately the same num­ber ofletters, usually two, three, four,or five letters. For example, the childwho produced Figure 3 had thetheory that all words are writtenwith about four letters. This theoryis truly a child's construction becausethere is no rule in external realitystating that all words must be writ­ten with about four letters.

Level II. Children at Level II are similarto those at Sublevel Ie, but they writethe first letter of each word convention­ally. For example, the child who pro­duced Figure 4 wrote three letters forall the words, but the first letters ofseven of the eight words are convention­ally correct. This is the first time chil­dren can be said to be revealing thetheory, or belief, that writing is relatedto speaking.

Level III. Children at this level are likeLevel-II children in that they write thecorrect consonant at the beginning ofeach word, but are more advanced in thatthey 1) write more letters for the longerof two spoken words, and 2) use morethan the "correct" initial letter to repre­sent sounds. In Figure 5, for example,"ham," ''butter," and "gum" are written

with two letters, but "hamster," "butter­fly," and "bubblegum" are written withthree letters. Examples ofmore than the"correct" initial letter are "Hr"for the be­ginning of"ham" and "hamster," and"Br"for the beginning of "butter" and "but­terfly." The first two letters of each pairare not conventionally correct, but thischild clearly noticed that "h am" and"hamster" begin with the same soundsand that "butter" and "butterfly" beginwith the same sounds . Other advancesin letter-sound correspondences can alsobe noted in the "E" at the end of ''key"and "monkey, " and the ''b'' in the middleof ''bubblegum."

Level IV. It is almost possible to read thewords in Figure 6, and Level IV can becalled "low consonantal" in contrast withthe next level , which can be considered"high consonantal." Many consonantsare still missing in Figure 6, but thereare many more at Level IV than at LevelIII.

Level V. This level can be called "h igh ­consonantal" because even more conso­nants appear at this level. In Figure 7,for example, "HAMSTR" and "BTR" caneasily be read. The striking characteris­tic of Level V, however, is the appear-

Figure 4 Figure 5An Example ofLevel II An Example ofLevel III

1--1 1 !VI ham H0 ham

hD/v hamsterHhT hamster

8TD B~rbutter

butter

tL 9 butterfly ~ butterfly

kHI key Kf- key

1\~ N+ monkey I'{\E monkey

SUE gum 6£\ gum

B bl)BUt bubblegum bubblegum

41

Dow

nloa

ded

by [

Was

hbur

n U

nive

rsity

] at

22:

39 0

2 N

ovem

ber

2014

KAMII AND MANNING

ance of many vowels. In Figure 7, six ofthe eight words are written with vowels.

Level VI. Figure 8, an example ofLevel VI,is almost conventional. Not a single con­sonant is missing except for the "N" in"monkey,"and the conventional spellingsof "er," "ey," and "urn" are striking in"HAMSter," "Butter," "key," and "guM."

The Phonemic Segmentation TasksWe gave two oral-segmentation tasks asVernon and Ferreiro did, and recordedchildren's responses on video- and audio­tapes. One task used six pictures; for theother task, we presented five cards (1" x 2"),each showing a written word. Like Vernonand Ferreiro, we gave the picture task firstto halfofthe children and the written-wordtask first to the other half to counterbal­ance the possibility that one task was morelikely than the other to facilitate phonemicsegmentation.

The picture task. The six pictures showeda sofa, money, coffee, a taco, a tepee, and alady, which were all two-syllable words.There were no printed words on the cards.With these cards, the children were askedto segment each word in a guessing gamedevised by Vernon and Ferreiro. In thisgame, the child was asked to choose a cardand to say the name of the object "in small

bits" so that the interviewer, covering hereyes, would not be able to guess it.

Before introducing the picture cards,many exercises were given to make sure thechild understood what was expected. Theinterviewer first demonstrated phonemicsegmentation of the names of objects in theroom , such as a desk, a pencil, and a table.The interviewer said, for instance, "I cansay /pencil/ or /pen-cil/ (syllables), but itwould be too easy to guess what I am think­ing about. If! say /pen-c-i-l/ (a syllable fol­lowed by three phonemes), it's not so easyto guess, but if! say /p-e-n-c-i-l/ (phonemes),it's really hard to guess the word. You saythe words in little bits so that I'll have areal hard time guessing."

All the cards were then examined to­gether to reach agreement on what to calleach object depicted. The child was thenasked to choose a card and say the name ofthe object in little bits so that the inter­viewer, covering her eyes, would have ahard time guessing it. For the first twocards chosen, the interviewer gave feedbackto the child such as , "That's too easy. I canguess that and win the card. Try making itharder. Say it in smaller bits, so I can'tguess it. " The game was played until allthe pictures had been used.

The written-word task. Each of the five

Figure 6 Figure 7An Example ofLevel N An Example ofLe vel V

~1\\,\\ ham H"~\ ham

\\~D~hamster H~NS tt, hamster

BI~ butter

BR butter

~TPfI butterfly

Rff butterfly

K t key

~1:key

N\ fC f- monkey

~\KEmonkey aM gum

1it\s~\gum

Boer--t/lbubblegumbubblegum

42

Dow

nloa

ded

by [

Was

hbur

n U

nive

rsity

] at

22:

39 0

2 N

ovem

ber

2014

PHONEMIC AWARENESS AND BEGINNING READING AND WRITING

cards had a word written on it in 28-point,lower-case, Times New Roman. (Vernonand Ferreiro used upper case, but we usedlower case because our children were morefamiliar with lower case.) The words usedwere "baby," "mama ," "pony," "soda," and"tuna ," which were familiar, two-syllable,consonantlvowellconsonantlvowel words.

Each child was shown a card, and the in­terviewer read the word aloud and askedthe child to point to each letter while say­ing the word in little bits. Naming of let­ters was not accepted. The interviewer firstgave three examples-"body," "ruby," and"city," showing strict correspondences be­tween letters and phonemes. The inter­viewer then read the first card aloud andasked the child to say it in little bits whilepointing to each letter-one bit for each let­ter. The same procedure was repeated un­til all the cards had been used.

The criteria used to categorize the re­sponses. We used essentially the same cri­teria as Vernon and Ferreiro's (1999) tocategorize the children's responses, butcombined their Levels 4 and 5. Vernon andFerreiro conceptualized a Level 4 for theirSpanish-speaking children , who seg­mentated only the second syllable, and aLevel 5 for those who segmented only thefirst syllable because "the first real segmen-

Figure 8An Example ofLev el VI

~~~ ham

HA(V)StZi hamster

8\J -I-'t-- r r butter

B\)1+:tI F1y butterfl y

t'(!'( key

!\00k: JJ '( monkey

g U0~ gum

BU'0\\3 \AM bubblegum

tation seems to take place for the last partof the word" (p. 407 ). We combined theselevels because our English-speaking chil­dren did not seem to segment the last partof the word first.

Levell:No segmentationThe child at this level said the com­plete word in spite of the feedbackfrom the interviewer. For example,he or she said "pony."

Level 2:Syllabic segmentationFor example, the child said "po-ny."

Level 3:Partial isolation ofaconsonant or vowelThe child produced a segmentationthat was basically syllabic, butmade an attempt to isolate one ofthe sounds. Examples are "p-en-ny"and "p-o-on-ny,"

Level 4:Phonological segmentation ofone of the two syllablesAn example is "po-n-y" or "p-o-ny."

Level 5:Complete phonologicalsegmentation"P-o-n-y" is an example of completephonological segmentation.

The two authors categorized each re­sponse together according to the precedingcriteria. Like Vernon and Ferreiro, whena child segmented a word at more than onelevel, we categorized the response at thehighest level exhibited for that word. Achild's final level of segmentation was de­termined by averaging the five or six num­bers thus obtained and using only thewhole number of the quotient. For ex­ample, if a child came out at Level 1.5 , heor she was categorized at Level L,

ResultsVernon and Ferreiro performed a multipleregression analysis , with level of writingas an independent variable and oral-seg­mentation level as the dependent variable.In our study, a simple examination of therelationship between children's level ofwriting and their level of oral segmenta­tion seemed sufficient. As can be seen inTable 1, most of those writing at Level I

43

Dow

nloa

ded

by [

Was

hbur

n U

nive

rsity

] at

22:

39 0

2 N

ovem

ber

2014

KAMII AND MANNING

segmented words at Levell or 2 ofthe oral­segmentation task, and only those writingat Level VI segmented words at the high­est levels. Those in the middle ofthe rangein writing segmented words at Levels 2, 3,and 4, in the middle of the range. The re­lationship between the two variables isstrong, with Somers' d BA = .63, which issignificant beyond the .001 level. Somers'd is a statistic used to assess the degree ofassociation between two ordered variables(Siegel & Castellan, 1988). Table 2 is simi­lar but deals with the written-word task.Somers' d BA was found to be .65 in thisanalysis , which is also significant beyondthe .001 level.

Vernon and Ferreiro found that childrensegmented words at a higher level in thewritten-word task than in the picture task.They stated, "At every writing level, chil­dren perform in a more analytic way whenthey are asked to take into account eachone of the letters on a printed card" (p .410 ). In our study, by contrast, only atLevel V of writing (th e high-consonantallevel) did children do slightly better in thewritten-word task. At every other level ofwriting, we found as many children whodid better on the picture task as those whodid better on the written-word task. Thisdifference was probably due to the socio­economic , environmental, and instructionaldifferences between the two groups. Ourkindergartners were probably more ad-

vanced in phonemic segmentation becausethey came from middle- to upper-middle­class homes, whereas the children in theMexican sample came from a lower-middle­class background. Vernon and Ferreiro alsocharacterized the school environment inMexico in the following ways, which are insharp contrast with the characteristics ofour American school: no previous instruc­tion in reading or phonics, very few read­ing materials available in the classroom,and infrequent instances of the teacherreading aloud to the children.

DiscussionWe replicated only parts of Vernon andFerreiro's study because they had alreadyfound a high multiple correlation R of.83817 (p=.0000). The only question thatremained unanswered was: Will we alsofind a strong relationship between a child'slevel of writing and his or her level of oralsegmentation? The answer is clearly in theaffirmative.

It is perhaps useful to distinguish be­tween the knowledge children constructabout the nature of our writing system andtheir skills, such as visual and auditory dis­crimination, short-term visual memory, andphonological segmentation. Children con­struct their own theories about the natureof our writing system, and our writing taskassesses this knowledge. At Sublevel IB,for example, children have constructed the

Table 1Relationship Between Level ofWriting and Level ofSegmentation for Picture Task

Level of segmentation1 2 3 4

Level ofwriting

I 16 6 1II 8 5 1III 1 2 4 3IV 3 1 2V 1 3 7VI 1 2

(Total) (25) (17) (11) (14)

Somers'dBA

= .63p < .001

44

5

1

(1)

(Total)

(23)(14)(10)( 6)(11)( 4 )

(68)

Dow

nloa

ded

by [

Was

hbur

n U

nive

rsity

] at

22:

39 0

2 N

ovem

ber

2014

PHONEMIC AWARENESS AND BEGINNING READING AND WRITING

belief that writing is different from draw­ing pictures, but they pay attention only towhat writing looks like. At Sublevel IC theyare still focusing only on what writing lookslike, but have constructed the theory thatall words must be written with two, three,four, or more letters. This is a surprisingtheory that we at first had a hard time be­lieving, but it turned out to be a robusttheory that young children seem to hold fora rather long time (Kamii, Long, & Man­ning,2001). When, at Level II , children fi­nally construct the theory that writing isrelated to speaking, they begin to makeprogress toward conventional writing.

For many years, researchers such asCarol Chomsky (1971, 1979) have empha­sized the importance of writing forchildren's learning to read. Vernon andFerreiro, too, have pointed out that whenteachers encourage children to write andto reflect on their writing, analysis of speechwill take place. This analysis of speech,which children engage in on their own ini­tiative, is reflected in the positive relation­ship we can see in Tables 1 and 2, showingthat as children advance toward conven­tional spelling, they also advance in oralsegmentation.

It is not difficult to get young children toengage in phonemic exercises out of con­text, and children who are already at a cer­tain level of development can occasionallybenefit from these exercises. However, we

agree with Vernon and Ferreiro and withGoodman (1993), who says,

My research shows that kids learn to use pho­nological , orthographic and phonic information,along with syntactic and semantic information,as they try to make sense of print. . . . Thereis no need to require kids to do unnatural ab­stract analysis as a prerequisite to recognizingwords as a prerequisite to comprehending! (p.85)

Assessment of phonemic awareness isusually done orally with a test. However, amore informative method is to ask childrento write as we did. When we watch themwrite and read what they have written,pointing to the letters, we can assess not onlytheir level of oral segmentation but also thegrapho-phonic information they are using.

We began this article by quotingGoswami's (2000) summary of what isknown about the relationship between pho­nemic awareness and beginning reading.Many researchers thus emphasize readingover writing, but it may be better to shiftthe focus of attention to children's writing.Their writing is highly informative whenwe want to assess their phonemic aware­ness. It is also important to encourage be­ginning readers to write because, whilewriting, children analyze their own speech,thereby becoming better able to differenti­ate phonemes.

Table 2Relationship Between Level of Writing and Level ofSegmentation for Written- Word Task

1Level of segmentation

2 3 4 5 (Total)Level ofwriting

IIIIIIIVVVI

(Total)

Somers' dBA

= .65p < .001

1461

(21)

752

(14)

1 133 3 15 13 6 21 1 2

(16) (12) ( 5)

(23)(14)(10)( 6)(11 )( 4)

(68)

45

Dow

nloa

ded

by [

Was

hbur

n U

nive

rsity

] at

22:

39 0

2 N

ovem

ber

2014

KAMII AND MANNING

ReferencesAdams, M.J ., Foorman, B. R , Lundberg, I., & Beeler,

T. (1998a). The elusive phoneme. American Edu­cator; 22(1 & 2), 18-29.

Adams, M. J.,Foorman, B. R.,Lundberg, I.,& Beeler,T. (1998b). Phonemic awareness in young chil­dren: A classroom curriculum Baltimore: PaulH.Brookes.

Chomsky, C. (1971). Write first, read later. Child­hood Education, 47, 296-299.

Chomsky, C. (1979). Approaching reading throughinvented spelling. In L. B. Resnick & P.A Weaver(Eds.), Theory and practice of early readers (pp.43-65 ). Hillsdale, NJ: Erlbaum.

Ferreiro, K, &Thberosky,A (1982). Literacy beforeschooling (K Goodman Castro, Trans.), Exeter,NH: Heinemann. (Original work published 1979)

Goodman, K (1993). Phonics phacs. Portsmouth,NH: Heinemann.

Goswami, U. (2000). Phonological and lexical pro­cesses. In M. L. Kamil, P. B. Mosenthal, P. D.Pearson, & R. Barr (Eds.), Handbook ofreadingresearch, Vol. III (pp. 251-267). Mahwah, NJ:Erlbaum.

Kamii, C., Long, R., & Manning, M. (2001). Kinder­gartners' development toward "invented" spellingand a glottographic theory. Linguistics and Edu-cation, 12, 195-210 . .

Kamii, C., & Manning, M. (1999 ). Before "invented"spelling: Kindergartners' awareness thatwritingis related to the sounds of speech. Journal ofRe­search in Childhood Education, 14, 16-25 .

Kamii, C., Long, R., Manning, M., & Manning, G.(1990). Spellingin kindergarten: Aconstructivistanalysis comparing Spanish-speaking and En­glish-speaking children. Journal ofResearch inChildhood Education, 4, 91-97.

Liberman, I. Y., Shankweiler, D., & Liberman, AM. (1989). The alphabetic principle and learningto read. In D. Shankweiler, & I. Y. Liberman,(Eds. ), Phonology and reading disability: Solvingthe reading puzzle (pp. 1-22). Ann Arbor, MI:University of Michigan Press.

Lundberg, I., Frost, J ., & Peterson, O. (1988). Ef­fects ofan extensive program for stimulating pho­nological awareness in pre-school children.Reading Research Quarterly, 23 , 263-284.

Siegel, S., & Castellan, N . J ., Jr. (1988). Nonpara­metric statistics for the behavioral sciences. NewYork: McGraw-Hill.

Stanovich, K K (1993). Romance and reality. TheReading Teacher; 47, 280-291.

'Ibrgesen, J . K (1998). Catch them before they fall.American Educator; 22(1 & 2), 32-39 .

Vernon, S. A, & Ferreiro, K (1999). Writing devel­opment: Aneglected variable in the considerationof phonological awareness. Harvard EducationalReview, 69,395-415.

Wagner, R.K, 'Ibrgesen,J.K ,Rashotte, C.A, Hecht,S.A , Barker, T.A , Burgess, S. R., Donahue, J ., &Garon, T. (1997). Changing causal relations be­tween phonological processing abilities and word­level reading as children develop from beginningto fluent readers: A five-year longitudinal study.Developmental Psychology,33, 468-479.

46

Dow

nloa

ded

by [

Was

hbur

n U

nive

rsity

] at

22:

39 0

2 N

ovem

ber

2014