See discussions, stats, and author profiles for this publication at: [Link]

net/publication/283091736

Prediction and Prevention of Reading Failure

Chapter · January 2000

CITATIONS READS
0 129

1 author:

Jane Flynn
Saint Mary's University of Minnesota
17 PUBLICATIONS 247 CITATIONS

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Prevention of Reading Failure View project

All content following this page was uploaded by Jane Flynn on 23 October 2015.

The user has requested enhancement of the downloaded file.

 Contents
PR~DICTIONAND Foreword vii

PREVENTION OF Introduction ix
1. Screening for Early Reading Failure 1

READ"ING FAILURE Yaleiie Muter

2. Do Preschool Orthographic Skills Contribute to Prediction of
Reading? 31
Nathlie A. Badian
--- 3. Systemic Screening and Intervention of
Reading Difficulty 57
Edited by Nathlie A. Badian Angela f. Fawcett and Roderick 1. Nicolson
4. Application of Frith's Developmental Phase Model to the
Process of Identifying At-Risk Beginning Readers 87
Diane f. Sawyer, Iwa K. Kim, and Sally Lipa- Wade
ht 2000 by York Press"Ine. S. Prevention and Prediction of Reading Problems 105
I
Bente E. Hagtvet
s reserved, including the right to reproduce this book or portions 6. From]cleiITification to Intervention: Improving Kindergarte
in any form except for the inclusion of brief quotations in a re- [Link]'Of ReacJing FaihTre 133
I inquiries should be addressed to York Press, Inc., P.O. Box 504, 1une~[Link]'ytp1
im, MarYland 21094.
7. Preschool Phonological Awareness and
Success in Reading 153
ik was manufactured in the United States of America.
Carol A. Chtistensen
8. Questioning the Role of Syllables and Rimes in Early
phy by Type Shoppe II Productions Ltd.
! Phonological Awareness 179
Myra Gipstein, Susan A. Brady, and Anne Fowler
, and binding by Data Reproductions Corporation
i
esign by joseph Dieter, Jr.

of Congress Cataloging-in Publication Data

[udith A. Bowey
217

10. What Do You Get if You Add /mmm/ to Ice? raining

F
9. A Case for Early Onset-Rime Sensitivity Training in At-Risk
Preschool and Kindergarten Children

Phoneme Awareness in Kindergarten: An Int rvention Study

rn and prevention of reading failure / edited by Nathlie A. Radian. with Children of Dyslexic Parents 247
[Link]. : DO/the Klint Petersen.
eludes bibliographical references and index.
BN 0-912752-57-2
11. Phonemes and Rhyme in the Development of Reading and
1. Readirig (Early childhood)--AbiIitytesting. I. Badian, Narhllo A, Metaphonology: The Dundee Longitudinal Study 275
Lynne G. Duncan and Philip H. K. Seymour
39.R4 P7~ 2000
~3--dc21 ()O.O'13,%,. Index 299
132 I Hagtvet

APPENDIX 1

OFF-TASK-BEHAVIOR-A CHECKLIST x)

2. Shakes
Chapter.6

From Identification to
Intervention
5. Bites nails
Improving Kindergarten
6. Hands shake Screening for Risk of
7. Fidgets on chair ReadinqFailure
8. Twists hair

9. Plays with hands

Jane M. Flynn
10. Bites lips

Many studies document the efficacy of early

12. Negative self oriented evaluation intervention in prevention of school failure (Ball and Blachman
1991; Berrieta et a1.1984; Byrne and Fielding-Bamsley 1991; Casto
and Mastropieri 1986; Felton and Brown 1990; Hurford et a1.1994;
Lundberg, Frost, and Peterson 1988; Strag 1972). As early as 1972,
14. Long pause bttore response Strag found that when students were diagnosed as reading disabled
in grades 1 or 2, 82% were eventually able to achieve at grade level,
15. Talks about tfrelevantmatters although only 10% to 15% of students identified in grades 5 through
7 reached grade level functioning. More recently, phonological
; throat; awareness programs in kindergarten and first grade have reduced
reading failure in later grades (Ball and Blachman 1988, 1991; Felton
Total: and Brown 1990; Hurford et a1.1994; Lundberg, Frost, and Peterson
1988). These studies have highlighted the importance of early identi-
x) Inspired by Wine, 1971, 1980 ficationof children at risk for school failure, leading to increased in-
terest in preschool and kindergarten screening programs.
Despite the consensus regarding the value of early identifica-
tion, however, experts disagree on how best to accomplish this
goal. Some favor teacher rating scales, while others maintain that
1 screening tests are more useful in early prediction. Adelman (1982)
 ":I~
734 I Flynn From Identification to Intervention I 135

and Algozzine and Ysseldyke (1986) argued for the cost effective- We turn now to the second question. Because screening tests
ness and predictive validity of asking teachers who are already fa- generally identify a reasonable number of children at risk, is it im-
miliar with the children to predict those at risk for school failure. portant to improve teacher prediction? We maintain that it is.
However, researchers generally have found that tests predict those Teachers who do not have a conceptual framework for observing
at risk for reading failure better than teachers. Reported teacher reading-related risk signs may have lowered expectations for some
prediction rates are disappointingly low, ranging from 15% to children. They are likely to espouse discriminatory entrance age
41% correct identification of children at risk (Feshbach, Adelman, policies that are not supported by later outcomes (Gredler 1992;
and Fuller 1974: Fletcher and Satz 1984; Flynn and Rahbar 1998a; Flynn and Rahbar, 1993a, 1993b; Shepard and Smith 1986, 1989).
LaTorre et a1.1982: Stevenson et a1.1976). In contrast, test identifi- For example, many kindergarten and primary teachers believe that
cation rates of .71 and .80 were reported by Fletcher and Satz younger children, boys in general, are more at risk for school fail-
(1984) and Flynn and Rahbar (1998a) respectively. ure. Therefore, many schools develop policies that adjust entrance
These findings suggest that when detection of risk is impor- ages upward, and parents of younger children (again, boys in par-
tant, screening tests are much better predictors than teacher rat- ticular) are often advised to give their children the gift of time" by
II

ings. Three questions arise from these data. First, why are teachers, delaying school entrance for another year (Ames 1967). Younger
who have spent most of a year interacting with their students, less children and boys are more likely to be retained in kindergarten
able to predict risk than brief screening tests? Second, if tests are (Gredler 1992; Shepard and Smith 1989).
able to identify reasonable numbers of at-risk children, is it impor- The lack of educationally relevant screening tasks and follow-
tant to improve teacher prediction rates? Finally, if teacher predic- up information may contribute to the prevalence of the matura-
tion of risk is important, how can screening programs be tional view that readiness resides in the child and unfolds naturally
improved to help teachers become better observers of their stu- with time. This view persists despite empirical and theoretical sup-
dents and, ultimately, better able to intervene to reduce the inci- port for the interaction between school conditions and child-back-
dence of reading failure? ground characteristics in determining reading success or failure
A number of factors may contribute to the low teacher identi- (Ausubel and Sullivan 1970; Bruner 1966; Gredler 1992; Iohannson
fication rate. One reason, reported by many of the kindergarten 1965; Vygotsky 1930-35/1978). Johansson (1965, p. 25) stated that
teachers in our screening program (Flynn and Rahbar 1998a), is "... if a child fails in one beginner's class it need not be on account
that teachers are reluctant to predict failure in young children at a of characteristics of the child; a child who's not ready for school in
time of rapid and unpredictable growth spurts. Also, many teacher one class may well succeed in another, when the teacher applies
training programs lack adequate instruction in the theoretical and different methods" .Iin Gredler 1992, p.17). By attributing the
scientific underpinnings of reading development (Brady 1997), source of risk to something within the child and his or her back-
e.g., how speech maps to print (Moats 1994), therefore many ground, teachers are ignoring the potentially powerful effect of
teachers are unprepared to observe skill development in areas cru- ,1,1, early intervention in reducing or eliminating school failure.
cial to reading success. In addition, many screening programs test Under-expectations for younger children and boys may occur
general developmental tasks rather than research-validated precur- also because of lack of information regarding later school achieve-
sors critical to reading success, e.g., linguistic skills (Adams 1990; ment. It is true that younger children, and boys in general, often
Brady 1997; Badian 1994; Blachman 1983; Stanovich and Siegel score lower on tests in kindergarten, first, and second grade (Flynn
1994; Wolf 1998). Such screening programs do not provide oppor- and Rahbar 1994; Shepard and Smith 1989; Davis, Trimble, and
tunities for teachers to observe their children engaged in reading- Vincent 1980). Research shows comparable achievement by third
related tasks, which would lead to development of a conceptual grade (Flynn and Rahbar 1994: Shaywitz et a1.1990; Shepard and
framework for assessing risk. Nor do the results of general screen- Smith 1989), but kindergarten teachers who lack long-term follow-
ing tests result in information that can be used to design specific up data on their students may remain skeptical of the research
interventions (Satz and Fletcher 1988; Majsterek and Ellenwood, evidence.
1995). Too often, these instruments are used to make inappropri- Given the accumulating evidence for the social constructivist
ate placement decisions, e.g., delayed school entrance, retention view that the child's development evolves as a result of the in-
in grade, or transition programs (Gredler 1992). teraction between personal-familial characteristics and school
 From Identification to Intervention 1 137
7361 Flynn

sized that reading is an interactive process composed of lower-

experiences, improving teacher prediction and intervention skills
level phonological, logo graphic, and orthographic processes oper-
is extremely important. Kindergarten teachers need to know how
ating simultaneously and synergistically with higher-level
to observe developmental levels in research-validated precursors to
semantic and syntactic processes. This view of reading has been
reading achievement. They also need to know how to match chil-
supported by a number of sources influential in current reading
dren's profiles with research-validated interventions in order to in-
theory (Adams 1990; Blachman 1983; LaBerge and Samuels 1974;
tervene effectively. In this way screening becomes a mediating
Liberman and Shankweiler 1985; Stanovich 1986), by empirical
experience that leads to appropriate intervention rather than de-
and clinical studies of reading disabilities (Flynn and Boder 1991;
layed school entrance or retention (Gredler 1992).
Lyon 1985; Stanovich, Siegel, and Gottardo 1997), and by our re-
Having considered the question of whether improving
search on neurophysiological characteristics of dyslexic subtypes
teacher prediction is an important goal and responding in the af-
(Flynn and Deering 1989; Flynn et a1. 1992; Ramaden 1997).
firmative, we turn now to the focus of this chapter: How to help
To capture the changing relationship of reading processes to
kindergarten teachers improve their ability to predict risk of read-
success from first through fifth grade, we 'specified developmental
ing failure. When kindergarten teachers were asked to predict
stages for acquisition of these processes. We postulated that early
reading achievement on a five-point scale, we, like previous re-
reading success depends on recognition of single words as 10-
searchers, found a low teacher-identification rate (.30) of children
gographs and on extraction of phonological principles. This
who later failed (Flynn and Rahbar 1998a). We hypothesized that
Accuracy Stage of reading development corresponds to the Glued
asking teachers to rate their children's current skill levels on pre-
to Print Stage described by Chall (1983). By second or third grade
cursors to reading would improve valid positive rates compared to
normally developing readers have reached the Automaticity Stage
our original rating scale which asked that they predict future read-
(Chall's Ungluing From Print stage), characterized by instanta-
ing success or failure. We also hypothesized that combining
neous access to orthographic features. This, in turn, results in in-
screening battery scores and teacher ratings on a skills-based in-
creased expressiveness and fluency as children are freed from the
strument would improve overall prediction rates. Therefore, we
mechanics of word recognition. With the attainment of auto-
developed a new rating scale that includes behavioral descriptions
maticity, the child's syntactic and semantic skills can be deployed
of emerging literacy skills. The Teacher Rating Scale (TRS) (Flynn
more fully as the child enters the Focus on Meaning Stage in third
1997a) was designed to mirror the skills tested in our kindergarten
grade and later.
screening test, the Literacy Screening Battery (LSB) (Flynn 1997b).
We hypothesized that disruption in one or more of these
We studied the utility of each instrument alone and in combina-
processes results in failure at one of these three crucial stages for
tion for predicting poor readers. Since the teacher prediction in-
reading development. Difficulty in acquiring phonological, ortho-
strument was based on our screening instrument, the Literacy
graphic, or automaticity processes leads to early reading failure
Screening Battery (LSB)is discussed first.
(first or second grade), although those children who have deficits
in language comprehension but intact phonological/orthographic
processing skills begin to fail at upper grades (third and up) when
DEVElOPMENT OF THE LITERACY SCREENING BATTERY (LSB)
the demands of understanding text increase. This developmental
sequence has been supported by empirical findings (Fletcher, Satz,
To meet the goal of early identification and effective intervention
and Scholes 1981; Jansky 1986; 1988; Satz et al. 1978; Shanahan
for children at risk for reading failure, we adopted three intermedi-
1984) and by theoretical writings (ChaIl1983; Frith 1985).
ate objectives: to develop a theory-based, psychometrically sound
After defining this developmental reading model, we devel-
kindergarten literacy screening battery that could be administered
oped screening tasks to measure precursors to phonological aware-
to a group; to follow a large sample from kindergarten through
ness, logographic/orthographic skills, semantics, and syntax. We
fifth grade; and to determine whether the screening battery would
were unable to identify a kindergarten-level precursor to auto-
add predictive utility over and above teacher ratings for identifica-
maticity that could be administered to a group. The Literacy
tion of children at risk of reading failure.
Screening Battery (LSB) consists of seven subtests: Vocabulary,
To accomplish these goals we used Rumelhart's (1977) read-
Syntax, Alphabet, Sounds, Phonemic Segmentation, Formcopy,
ing model to develop screening tasks. Rumelhart (1977) hypothe-
738 I Flynn
From Identification to Intervention I 139
and Visual Discrimination. A description of each subtest and ratio-
nale for inclusion in the screening battery illustrates how the LSB many kindergarten screening batteries have identified children
embodies our hypotheses of how and when reading develops. with visual processing deficits who later failed in reading (Badian
1988; Jansky 1986, 1988; Jansky and de Hirsch 1972; Sears and
Keogh 1993; Silver, Hagin, and Beecher 1978). Badian (1994) pre-
Phonological Processes sented strong evidence for the predictiveness of visual discrimina-
tion/orthographic tasks in identifying those at risk for reading
To measure phonological awareness at the kindergarten level, we failure. Letter naming, writing letters and sounds to dictation, and
constructed two subtests, Phonemic Segmentation and Sounds. The alphabet recitation have consistently been among the most predic-
Phonemic Segmentation subtest, a visual adaptation of the tive screening measures in early identification studies (Badian
Auditory Analysis Test (Rosner and Simon 1971), measures the abil- 1986; Jansky and de Hirsch 1972; Jansky 1986; Satz et a1.1978).
ity to hold a sequence of sounds in working memory and to recon-
struct an auditory gestalt. Children were asked to listen to a word, Semantic Processes
silently reconstruct that word with a syllable or sound deleted, and
to circle the picture of the reconstructed word. Twenty-six stimuli Knowledge of word meanings was represented in our battery by
were used, ranging from syllable deletion ("Cowboy," take away the Vocabulary subtest that consisted of 44 items, each with three
"boy" and circle the picture of what's left) to phoneme deletion pictorial choices. Items were chosen from our pilot studies and
("feet," take away /f/, circle the picture of what's left). We also hy- from reviews of existing tests of vocabulary and basic concepts
pothesized that knowledge of sound-symbol relationships con- (Bracken 1984; Boehm 1986; Dunn and Dunn 1981; Gardner
tributes to phonological processing skill. The Sounds subtest 1985). Research support for testing receptive word knowledge as a
required children to write the letters corresponding to ten sounds predictor of reading achievement was provided by Shanahan
dictated by the examiner. Phonological deficits, including difficulty (1984) who found that while early reading was predicted by
segmenting spoken language and identifying sounds or syllables, sound-symbol skill, vocabulary diversity made a greater contribu-
have been cited as a major cause of reading disabilities (Adams tion to advanced reading status. Screening studies have also re-
1990; Bradley and Bryant 1985; Ehri and Wilce 1985; Liberman and ported increasing correlation of kindergarten vocabulary scores
Shankweiler 1985; Liberman et a1. 1974; Wagner and Torgesen with upper grade reading scores (Jansky 1986, 1988; Satz et a1.
1987). Phonological awareness tasks have proved to be sensitive 1978).
screening tasks at the kindergarten level (Badian 1994; Ball and
Blachman 1988, 1991; Majsterek and Ellenwood 1995). Syntactic Processes

The Syntax subtest consisted of 26 spoken sentences with four pic-

Logographic/Orthographic Processes torial choices for each. Syntactical structures included noun and
In the Alphabet subtest, children were asked to write the 26 letters verb phrases, pronouns, indirect objects, negation, compound sen-
of the alphabet dictated in random order. In addition, because our tences, and embedded clauses. Fletcher, Satz, and Scholes (1981)
early studies indicated that by the end of kindergarten, letter writ- found that kindergarten syntactical tasks contributed to reading
ing alone was too easy for discrimination of achievement groups, achievement in upper grades" and measures of listening compre-
children were asked to write as many words as they knew from hension have consistently correlated with later reading achieve-
memory, with ten spaces provided for responses. The Visual ment (Curtis 1980; Jansky 1986, 1988; Stanovich, Cunningham,
and Freeman 1984).
Discrimination subtest consisted of matching 15 sequences of geo-
metric shapes, letters, numbers, or simple words. In the Formcopy The LSB met or exceeded recommended reliability and pre-
subtest, children were asked to copy 12 geometric shapes of in- dictive validity standards for screening instruments (Brown 1983;
creasing difficulty. Salvia and Ysseldyke 1991). Reliability studies revealed adequate
Although much of the current literature emphasizes linguis- internal consistency indices (» .80) for use of a Language Cluster
tic skill as the major contributor to reading success and failure, score (Vocabulary + Syntax + Phonemic Segmentation), an
Achievement Cluster score (Alphabet + Sounds), and the total
 740 I Flynn From Identification to Intervention I 147

score for screening decisions and for predictive follow-up studies. Sample
The LSB correctly predicted 73% to 81 % of poor readers in three Our sample consisted of all kindergarten children (n = 1,972) pre-
samples of children followed through third grade. In addition, five sent the day of screening in 26 Midwestern school districts. This
of the seven subtests met the .80 reliability criterion recom- cohort was part of our five-year study of reading development in
mended by Brown (1983) and Salvia and Ysseldyke (1991) for in- 5,317 children in 26 Minnesota and Wisconsin districts (Flynn
structional grouping decisions, the primary purpose of the LSB. and Rahbar 1993a, 1993b,1998a, 1998b). District K-12 enrollments
Formcopy and Visual Discrimination did not meet the .80 crite- for the first year of the study ranged from 418 to 7,455 students,
rion. Construct and criterion-related studies also supported the with a median enrollment of 950. Both rural and small town com-
theoretical base of the LSBand its relationship to established, indi- munities, ranging from 500 to 50,000 inhabitants, were repre-
vidually administered tests. For a complete discussion of the psy- sented, with median incomes from $8,600 to $21,500 (1988
chometric qualities of the LSB,see Flynn and Rahbar (1998b). Chamber of Commerce data by community). Ethnic background
of the communities served by these districts was primarily north-
ern European, predominantly Norwegian, Irish, Polish, and
DEVELOPMENT OF THE TEACHER RATING SCALE
German. Two districts reported minority populations of Hmong
children.
The Teacher Rating Scale (TRS) reflected the same precursors to
reading represented in the LSB. First, we specified 10 skill cate- Procedures and Materials
gories: Letter Names, Sight Words, Visual Discrimination, Visual-
Motor Skill, Letter Sounds, Rhyming, Decoding, Receptive Kindergarten children in 221 classrooms were screened in March or
Vocabulary, Oral Language, and Expressive Vocabulary. Next, we April by the principal investigator and three research assistants.
wrote behavioral descriptions for rating each characteristic on a Because screening is not diagnosis, but rather a system for identify-
scale of 1 to 10. For example, under "Letter Names," 1 was de- . ing children who need further assessment, the issue of personnel
scribed as "recognizes less than 10 letters" and 10 as "recognizes costs favors a group-administered screening test, few of which ex-
and names all letters." SCOlesin each category were summed to isted at the time this study was implemented. To ensure reliable and
produce a total reading prediction score with a minimum score of valid individual scores in a group setting, we devised developmen-
10 and a maximum of 100. Table I depicts the relationship of the tally appropriate screening procedures. Private "offices" in the form
TRS categories to the companion LSB subtests and to the reading of tabletop carrels were used to reduce error variance associated with
components identified by Rumelhart. copying. Teachers and paraprofessionals assisted the screeners so
that the adult-child ratio was no more than 1:6. Each subtest was
contained in a separate booklet with no more than four items per
page. Children were encouraged to mark an item only if they were
sure of the answer to reduce error variance associated with guessing.
The examiner moved to the next item after monitors signaled that
Table I. Relationship of Teacher Rating Scale and Literacy Screening Battery
all children had completed the current item. Because phonological
to Rumelhart's Reading Model
deletion tasks were not a part of kindergarten curricula at that time,
Rumelhart Reading Process LSB Subtest TRS Category
Phonological process Phonemic Segmentation Rhyming
a training session using large pictures and a one-page pretest with
Sounds Letter Sounds feedback regarding correct responses was completed before adminis-
Decoding tration of the Phonemic Segmentation subtest. The battery was ad-
Logographic process Visual Discrimination Visual Discrimination ministered on two separate days, with an activity break between
Formcopy Visual Motor each subtest. Each subtest required 10 to 20 minutes to complete.
Orthographic process Alphabet Letter Names
Sight Words
Total administration time, including breaks, averaged two hours. A
Syntactic process Syntax Oral Language very small number of children who became anxious or had diffi-
Expressive Vocabulary . culty staying with the tasks were taken to an alternative play setting
Semantic process Vocabulary Receptive Vocabulary and tested individually by the project screener.
 From Identification to Intervention I 743
7421 Flynn

Kindergarten teachers completed the Teacher Rating Scale for

'I}
.~t.
;:(.
components analysis of the LSB subtests suggested a three-factor
each of their students prior to the screening days. Some teachers solution consisting of language, orthographic/achievement, and
elected to use the previous five-point prediction instrument, re- visual processing factors, partially confirming our hypothesized
sulting in combined TRS and LSB data on 1,634 of the 1,972 chil- model of reading development. This solution was stable across
dren screened. It is important to note that district personnel were three different cohorts and two revisions of the LSB (Flynn and
not informed of screening results to avoid biasing outcome. Our Rahbar 1998b). For the jointTRS-LSB analysis, we submitted scores
goal, at this point in our research program, was to identify valid from both instruments to principal components analysis with
precursors to reading achievement. Subsequent research was varimax rotation. In both a three-factor and four-factor solution,
planned to investigate the interaction between risk profile and in- eight of the ten teacher ratings sorted in a largely uninterpretable
terventionprograms, with inservice training to teachers on how to factor, with some ratings contributing roughly equivalent loadings
use the LSBto identify risk signs and match children with effective on two factors. The four-factor solution, reported in table II, ac-
interventions. counted for a larger proportion of the variance explained and con-
tained three interpretable factors: Facto!"II-an achievement factor
consisting of four teacher statements plus LSB Alphabet; Factor
COMPARISON OF THE TEACHER RATING SCALE AND III-LSB language subtests; and Factor IV-LSB Visual Processing
THE LITERACY SCREENING BATTERY subtests. Thus, it appears that, for the most part, teacher ratings re-
fleeted a more global and different view of children's pre-reading
Content and criterion-related validity studies used data from 1,634 abilities compared to their performance on the LSB.
kindergarten children for whom both TRS and LSB scores were To investigate the predictive validity of the TRS separately
available. First, we computed intercorrelations of TRS categories to and in conjunction with the LSB, we studied a subset of children
determine whether the ratings contributed unique information for whom first and third grade reading achievement scores were
suggestive of different skill domains. We expected relatively low available. These scores were provided by participating districts
correlations between items representing different domains, e.g., and because testing schedules varied, the follow-up sample for
Receptive Vocabulary and Visual Motor Skills, however, all teacher
ratings were moderately to highly correlated (.55 to .81, P = .0001), Table". Principal Components Analysis of LSB Subtests and TRS
suggesting a "halo" effect rather than differential rating of specific Statements--4 Factor Solution
precursors to reading. In contrast, the LSBintercorrelations were Variable Factors
all in the low to moderate range (.16 to .51) with the exception of I. II. III. IV.
Alphabet and Sounds subtests, which correlated .67. TRS Decoding Ability .715 .501
Next, correlations of TRS ratings with LSBsubtest scores were TRS Expressive Vocabulary .879
computed to examine the degree of relationship between teacher TRS Oral Language .874
ratings and subtest scores hypothesized to test common processes. TRS Receptive Vocabulary .805
Teacher Rating Scale early literacy ratings (Letter Names, Sight TRS Rhyming Ability .764
TRS Sight Words .598 .571
Words, Letter Sounds, and Decoding) correlated moderately to TRS Visual Discrimination .779
highly (.48 to .72) with LSB subtests Alphabet and Sounds, indicat- TRS Visual Motor Skill .613
ing that teachers' observations matched children's performance on LSBAlphabet .775
these tasks fairly well. However, other TRS-LSBcorrelations were TRS Letter Naming .776
relatively low, ranging from .31 to .43. This suggested that teacher TRS Letter Sounds .555 .680
LSB Sounds .770
ratings of linguistic and visual processing skills for the most part LSB Phonemic Segmentation .634
were not highly related to children's performance on screening LSB Syntax .760
tasks measuring those skills. LSBVocabulary .781
We next used principal components analysis with varimax LSB Visual Discrimination .613
rotation to determine whether teacher ratings and LSB subtests LSB Formcopy .828
clustered in theoretically meaningful ways. Previous principai Note: 74% of variance accounted for. Factor loadings $.50 not reported.
 From Identification to Intervention I 145
144 I Flynn

predictive studies consisted of 210 children who had been tested percent of the children predicted to fail by this criterion were good
in both first and third grade. This sample did not differ signifi- readers at the beginning of third grade. Predictive validity results
cantly from the screening cohort in distribution of age, socioeco- for the TRS,the LSB,and the TRS-LSBare summarized in table HI.
nomic status, place of residence, or sex.
Based on a 35% base rate of reading failure by third grade in
our first cohort, we defined a score at or below the 35th percentile DISCUSSION
on the TRS as at risk for reading failure. Teacher ratings on the TRS
resulted in a mean of 74 and a standard deviation of 19. Therefore, This study, using a skills-based rating instrument rather than
we used the raw score of 66 (corresponding to the 35th percentile) teacher prediction of future achievement, resulted in a 34% im-
as the cut-off score for teacher prediction of children at risk. Risk provement in teacher identification of children at risk. However,
status on the LSB was defined as a score at or below the 35th per- 36% of the children who later failed in reading were not identified
centile on either the Language or Achievement Cluster. by the Teacher Rating Scale. Overall, this study suggests that addi-
Outcome measures provided by the participating districts tional procedures are needed to help teachers identify and inter-
consisted of identification for special reading services and reading vene early. First, kindergarten teachers would benefit from
percentile scores (composite or total reading percentile) on stan- training in differential diagnosis and observation, particularly of
dardized tests, e.g., Iowa Test of Basic Skills (ITBS) ( Hieronymus, precursors to literacy that are not part of a traditional kindergarten
Hoover, and Lindquist 1986), the Metropolitan Achievement Test curriculum. We also believe that multiple screenings, including
(MAT6) (Prescott et a1. 1985), or the Stanford Achievement Test one at the beginning of kindergarten and another at the begin-
(Gardner et a1. 1982). Reading failure was defined as total reading ning of first grade, as well as the current end-of-kindergarten
percentile at or below the 40th percentile or enrollment in Title 1 screening, would ensure more accurate identification of children
or LD reading services in first, second, and/or third grade. The at risk. Finally, studies of how child risk factors interact with dif-
40th percentile was chosen as the achievement test cutoff because ferent instructional environments could facilitate the ultimate
most districts use this criterion fOI Title 1 eligibility. Although the goal of screening programs-to match child risk profiles with in-
40th percentile is within the average range for reading achieve- terventions proven effective in reducing reading failure.
ment in the national norming samples for these tests, this cutoff The TRSand other rating scales that identify children's initial
identified only the lowest 23% of our follow-up sample. levels in reading precursors can serve as inservice tools to train
After specifying predictive and outcome cutoff scores, we teachers on observation of skills outside of traditional reading
constructed 2 X 2 classification tables to examine the utility of the readiness areas. It would be important to provide this inservice
TRS and LSBin predicting children who failed in reading. Teachers training early in the kindergarten year. In our study, the rating
correctly predicted 64% of poor readers and missed 36% of those scale was completed late in the year and teachers were not trained
who failed. Although the TRS represented a substantial improve- in differential diagnosis. The fact that their ratings on letter names
ment over our original teacher rating scale (30% correct identifica- and sounds generally matched screening results in these domains
tion of children who failed), identification rates were still suggests that training on phonological, orthographic, and visual
substantially below that obtained by the group-administered LSB,
which correctly predicted 80% of poor readers in this sample.
Despite the high identification rate of the LSBcompared with
most sCIeening instruments (GredleI 1992; Keogh 1977; Mercer, Table III. Predictive Validity of the Teacher Rating Instrument (TRS), the
Algozzine, and Triffiletti 1988) we were concerned that 20% of the Screening Test (LSB), and Teacher Rating and Screening Test
poor readers were not identified by the test. We wondered Combined Criteria (TRS/LSB)
whether a combination of the TRS and LSB cut scores might im- Variable TRS LSB TRS/LSB
prove identification. Using a risk classification rule of either test Valid positives .64 .80 .88
(LSB)or teacher (TRS)cut score at or below the 35th percentile re- Valid negatives .86 .72 .57
False positives .23 .31 .39
sulted in a positive identification rate of 88% of poor readers.
False negatives .37 .20 .12
However, the rate of over-identification also increased: Thirty-nine
1461 Flynn From Identification to Intervention 1 747
l
discrimination precursors would improve their ability to rate their it" agnostic assessments were characteristic of the most successful
,I.:
children in these important emerging literacy skills. l'~~••
classroom, in which 77% of the children with relatively low "men-
We beJieve that screening children early in the kindergarten tal ages" (4-5 to 5-11at the beginning of the year) were successful
year would not only facilitate observation of important literacy readers. This percentage of successful readers contrasted sharply
precursors but also provide information for interventions during with findings from another classroom, where only 15% with simi-
the kindergarten year. To achieve the goal of identifying children lar readiness levels met the criterion for reading success.
early in their kindergarten year, we recently developed the Early In another study of the interaction between first-grade readi-
Literacy Screening Battery (ELSB), a downward extension of the ness, Gesell developmental age scores, and instructional tech-
LSB. The overlap of tasks and items between the ELSB and LSB niques, Zenka and Keatley (1985, in Gredler 1992) found that
holds promise for documenting the effects of instruction as well as when the Writing to Read Program (Martin and Friedberg 1986)
for early identification. For example, in a pilot study of seven was added to the regular kindergarten curriculum, the incidence of
kindergarten classrooms screened in October (n = 154), there were "immature" children at the end of kindergarten was 29% less than
no significant differences on ELSBsubtests measuring vocabulary, in the non-enriched classrooms. It should be noted, however, that
syntax, alphabet recognition, sound recognition, or visual discrim- this study used a score at or above the 40th percentile on the
ination. However, at the end of the kindergarten year, the LSB Metropolitan Readiness Test as the criterion for readiness for first
results revealed significant differences between groups on vocabu- grade. Whether this criterion and intervention program fully cap-
lary (p = .0001). Children in the classroom that focused on pre- tures the probability of children's success in reading is debatable.
academic skills with minimal attention to enriching oral language Still fewer studies have investigated the effects of instruc-
made on average only two points gain on the Vocabulary subtest, tional programs on children with speciftc risk profiles, with the ex-
in contrast to gains of five to seven points for the other classrooms ception of the accumulating data concerning the effects of
that were primarily literature-based. phonological awareness programs on children with deficits in pro-
Prediction of reading failure and effective intervention may cessing speech sounds (Felton and Brown 1990; Hurford et al.
also be enhanced if screening takes place at the beginning of first 1994; Lundberg, Frost, and Peterson 1988; see, however, Olson
grade. Satz and Fletcher (1988) recommended re-screening to in- and Wise 1997 and Torgesen 1997 for cautions regarding the
crease the confidence with which we identify children at risk. transfer of phonological training to reading fluency). Our early
Equally important, re-screening will provide first grade teachers kindergarten pilot study suggests that similar studies of vocabulary
with instructionally relevant information. Multiple screenings are enrichment will yield important information for prevention of
feasible if testing batteries are administered to groups in order to later reading failure. We know that the teacher whose children
reduce time and personnel cost. Although classroom teachers can achieved the highest average gains in vocabulary (for both high
conduct their own screenings with the ELSB and LSB, it may be and low scorers on the ESLB)used a literature-rich approach with
more valuable to allow teachers time to make observations of the deliberately planned redundancy in new concepts and vocabulary.
children without the constraints of having to attend to the whole Another teacher was especially effective in raising vocabulary
group as the main screener. In this way screening becomes an op- scores of children who had been identified most at risk (lowest
portunity for teachers to make personal structured observations 33% at the beginning of the year). Further observation of and in-
that can be used in planning interventions (Iansky 1988). terviews with this teacher may yield specific techniques for inter-
The potentially powerful interaction between child character- vention with low-scoring children. Given the demographics of
istics and instructional interventions must also be considered in this rural district where many children score low on the vocabu-
assessment of risk. Although many studies have looked at child lary and syntax subtests, these data suggest the importance of
risk variables, few have studied the effects of instructional envi- studying classroom-based interventions for reduction of reading
ronments. Gates (1937) investigated the influence of instructional failure in upper grades.
conditions on the reading success of children in first grade who These studies highlight the importance of documenting
had been tested previously to establish "mental ages" (Ames teacher and instructional variables as well as child characteristics.
1967). He found that a high quality of teacher instruction, a wide Adequately assessing risk depends on assessment of the quality of
range of easy and interesting reading materials, and the use of di- the subsequent teaching instruction that a child will receive, one
748 I Flynn
From Identification to Intervention I 749
of the primary reasons early screening programs seldom identify
more than 75% of the children who later fail (Gredler 1992; Badian. 1994, Preschool prediction: Orthographic and phonological skills,
and reading. Annals o(Dyslexia 44:3-25,
Jansky 1986, 1988; Keogh 1977). Matching children with specific
f'h Ball, E., and Blachman, B. 1988, Phoneme segmentation training: Effect
risk profiles to instructional interventions proven to reduce on reading readiness. Annals of Dyslexia 38:208-25.
the risk of failure would be a major improvement in the a11-too- Ball, E., and Blachman, B. 1991. Does phoneme segmentation training in
prevalent practice of screening children, deeming them not ready kindergarten make a difference in early word recognition and develop-
mental spelling? Reading Research Quarterly 26:49-66.
for reading instruction, and recommending delayed school en-
Berrieta, c., Schweinhart, L., Barnett, W" Epstein, A., and Weikart, D. 1984.
trance, transition room placement, or retention in kindergarten. Changed Lives: The Effects of the Perry Preschool Program on Youths through
By examining the patterns of improvements in children at risk Age 19, Ypsilanti, MI: High/Scope Educational Research Foundation,
who had been screened with instructionally relevant tasks at the Blachman, B. 1983, Are we assessing the linguistic factors critical in early
beginning and end of kindergarten and given well-documented in- reading? Annals o(Dyslexia 33:91-109.
structional programs, effective interventions could be identified. Boehm, A, 1986. Boehm Test of Basic Concepts Pre-School Version. San
Antonio, TX: The Psychological Corporation.
Teachers could make immediate use of screening information in Bracken, B. 1984. Bracken Basic Concept Scale. San Antonio, TX: The
matching child characteristics with instructional programs. Psychological Corporation,
Our conclusion from these studies is that it is important to im- Bradley, L., and Bryant, P. 1985. Rhyme and Reason in Reading and Spelling,
prove teacher prediction of children at risk of failure, not only to Ann Arbor, MI: The Psychological Corporation.
identify the greatest number of those at risk but also to provide in- Brady, S. 1997. Informed L'1struction for Reading Success: Foundations for
Teacher Preparation. Position paper of The Orton Dyslexia Society. MD:
service training for teachers on effective interventions. To move Baltimore.
screening from identification to intervention, it is imperative that Brown, F. 1983. Principles o( Educational and Psychological Testing (3rd ed.).
screening programs use theory-based, research-validated instru- New York: Holt, Rinehart, and Winston.
ments that provide instructional guidance. These screenings should Bruner,], 1966, Toward a Theory o( Instruction. Cambridge, MA: Harvard
University Press.
occur at crucial points in kindergarten and first grade in order to
Byrne, B" and Fielding-Barnsley, R. 1991. Evaluation of a program to
identify potential problems in acquisition of emerging literacy skills. teach phonemic awareness to young children, Journal of Educational
Repeated screening should also provide information on the interac- PsychOlogy 83(4):451-55,
tion between classroom instruction and child characteristics in pre- Casto,· G" and Mastropieri, M, 1986, The efficacy of early intervention
venting or lessening the impact of reading failure. It is our hope that programs: A meta-analysis. Exceptional Children 52(5):417-25.
Chall,]. 1983. Stages of Reading Development. New York: McGraw-Hill.
kindergarten screenings will ultimately result in curriculum being Curtis, M. 1980. Development of components of reading skill. Journal of
adapted to meet the needs of the child, rather than the child being Educational Psychology 72:656-69.
asked to wait until he or she is "ready" for the curriculum. Davis, B., Trimble, c., and Vincent, D, 1980, Does age of entrance affect
school achievement? Elementmy School Iournal 80: l33-43,
Dunn, L., and Dunn, L. 1981. Peabody Picture Vocabulary Test Revised.
REFERENCES Circle Pines, MN: American Guidance Service.
Ehri, L, and Wilce, L. 1985. Movement into reading: Is the first stage of
Adams, M. 1990. Beginning to Read: Thinking and Learning About Print. printed word learning visual or phonetic? Reading Research Quarterly
Cambridge, MA: MIT Press. 20:163-79.
Adelman, H. 1982. Identifying learning problems at an early age: A criti- Felton, R., and Brown, 1. 1990, Phonological processes as predictors of spe-
cal appraisal. Journal of Clinical Child Psychology 11:255-61. cific reading skills in children at risk for reading failure. Reading and
Algozzine, B.,and Yssledyke,]. 1986. The future of the learning disabilities Writing: An Interdisciplinmy JoumaI2:39-59.
field: Screening and diagnosis. Journal of Learning Disabilities 19:394-38', Feshbach, S., Adelman, H., and Fuller, W. 1974. Early identification of chil-
Ames, L. 1967, Is Your Child In the Wrong Grade? New York: Harper and Row, dren with risk of reading failure, Journal o(Leaming Disabilities 7:639-44.
Ausubel , D" and Sullivan, E, 1970. Theory and Problems of Child Fletcher, J" and Satz, P. 1984. Test-based versus teacher-based predictions
Development. New York: Grune and Stratton, of academic achievement: A three-year longitudinal follow-up. Journal
Badian, N, 1986, Improving the prediction of reading for the individual of Pediatric Psychology 9(2): 193-201.
child: A four-year follow-up, Journal of Learning Disabilities 19:262-70, Fletcher, ]., Satz, P·i and Scholes, R. 1981. Developmental changes in the
Badian, N, 1988. The prediction of good and poor reading before kinder- linguistic performance correlates of reading achievement: Differences
garten entry: A nine-year follow-up. Journal of Learning Disabilities in what predicts reading at different ages. Brain and Language l3:78-90.
21(2):98-103. Flynn.}, 1997a, Teacher Rating Scale, LaCrosse, WI: LaCrosse Area Dyslexia
Research Institute, Ine.
1501 Flynn
From Identification to Intervention 1151
.,,'

Flynn. 1997b. Literacy Screening Battery. LaCrosse, WI: LaCrosse Area

Dyslexia Research Institute, Ine.
J.:t~ LaTorre, R., Hawkhead, F., Kawahira, R., and Bilow, L.1982. Kindergarten
screening pilot project in Vancouver schools 1979-1980: A two-year
Flynn, j., and Boder, E. 1991. Clinical and electrophysiological correlates following of the McCarthy Screening Test. British [ournal of Special
of dysphonetic and dyseidetic dyslexia. In Vision and Visual Dyslexia. 1.' Education 6:23-41.
Vol. 13, ed. J. Stein. Basingstoke, England: MacMillan Press. Liberman, I., Shankweiler, D., Fischer, F., and Carter, B.1974. Explicit syl-
Flynn, J., and Deering, W. 1989. Subtypes of dyslexia: Investigation of lable and phoneme segmentation in the young child. Journal of
Boder's system using quantitative neurophysiology. Developmental Experimental Child Psychology 18:201-12.
Medicine and Child Neurology 31:215-31. Liberman, I., and Shankweiler, D.198S. Phonology and problems of learn-
Plynn.j., Deering, W., Goldstein, M., and Rahbar, M. 1992. Electrophysio- ing to read and write. RASE: Remedial and Special Education 6:8-17,
logical correlates of dyslexic subtypes. Journal of Learning Disabilities Lundberg, I., Frost,]., and Peterson, O. 1988. Effects of an extensive pro-
25:133-41. gram for stimulating phonological awareness in preschool children.
Flynn, J., and Rahbar, M. 1993a. The effect of age, sex, socioeconomic sta- Reading Research Quarterly 23:263-84.
tus, and kindergarten screening scores on reading achievement. Pro-· Lyon, G. R. 1985. Identification and remediation of learning disability
ceedings of the 1993 American Statistical Association Conference 78S-90. subtypes: Preliminary findings. Learning Disabilities Focus 1:21-35.
Flynn]., and Rahbar, M.1993b. Effects of age of school entrance and sex Majsterek, D., and Ellenwood, A. 1995. Phonological awareness and be-
on achievement: Implications for pediatric counseling. Developmental ginning reading: Evaluation of a school-based screening procedure.
and Behavioral Pediatrics 14(S):304-7. Journal of Learning Disabilities 28:449-56.
Flynn, J., and Rahbar, M. 1994. Prevalence of reading failure in boys com- Martin,]., and Friedberg, A. 1986. Writing to Read. New York: Warner.
pared with girls; Psychology in the Schools 11:66-71. Mercer, c., Algozinne, B., and Triffiletti, J. 1988. Early identification: An
Flynn, J., and Rahbar, M. 1998a. Improving teacher prediction of children analysis of the research. Learning Disability Quarterly 11: 176-88.
at risk for reading failure. Psychology in the Schools 35(2):163-72. Moats, L. 1994. The missing foundation in teacher education: Knowledge
Flynn, J., and Rahbar, M. 1998b. Kindergarten screening for risk of read- of the structure of spoken and written language. Annals of Dyslexia
ing failure. [oumal of Psychoeducational Assessment 16:1S-3S. 44:81-104.
Frith, U. 1985. Beneath the surface of developmental dyslexia. In Surface Olson, R., and Wise, B. 1997. Problems in transfer from improved phono-
Dyslexia: Neuropsychological and Cognitive Studies of Phonological Reading. logical skills to reading growth. Paper read at the 48th annual confer-
eds. J. Marshall and M. Coltheart. London: Erlbaum. ence of The Orton Dyslexia Society, Minneapolis.
Gardner, M. 1985. Receptive One-Word Picture Vocabulary Test. Novato: CA: Prescott, G., Balow, I., Hogan, T., and Farr, R. 1985. Metropolitan Achievement
Academic Therapy. Tests: MA T6. San Antonio, TX: The Psychological Corporation.
Gardner, E., Rudman, H., Karlsen, B., and Merwin, J. 1982. Stanford Ramaden, Z. 1997. Artificial intelligence: A data analysis system to sub-
Achievement Test. San Antonio, TX: The Psychological Corporation. type dyslexic children using electorencephalograms. Ph.D. dissertation.
Gates, A. 1937. The necessary mental age for beginning reading. Elemen- Marquette University, Milwaukee, WI.
tary School [ournal 8:497-S08. Rosner, J., and Simon., D. 1971. The Auditory Analysis Test: An initial re-
Gredler, G. 1992. School Readiness: Assessment and Educational Issues. port. Journal of Learning Disabilities 4:384-92.
Brandon, Vermont: Clinical Psychology Publishing Company. Rumelhart, D. 1977. Toward an interactive model of reading. In Attention
Hieronymus, A., Hoover, H., and Lindquist, E. 1986. Iowa Tests of Basic and Perfonnance IV. eds. S. Dornic and P. Rabbitt. Hillsdale, NJ: Erlbaum.
Skills. Chicago: Riverside. Salvia, J., and Ysseldyke, J. 1991. Assessment: 5th Edition. Boston:
Hurford, D., Johnston, M., Nepote, P. Hampton, S., Moore, S., Neal, J., Houghton-Mifflin.
Mueller, A., McGeorge, K., Huff, L., Awad, A., Tatro, c., Juliano, c., and Satz, P., and Fletcher, J. 1988. Early identification of learning disabled
Huffman, D. 1994. Early identification and remediation of phonological- children: An old problem revisited. Journal of Consulting and Clinical
processing deficits in first-grade children at risk for reading disabilities. Psychology 56:824-29.
[oumal of Learning Disabilities 27(10):647-59. Satz, P., Taylor, G., Friel, J. and Fletcher, J. 1978. Some developmental
Jansky, ]. 1986. The case for prediction. Paper read at the 37th annual and predictive precursors of reading disabilities: A six year follow-up.
conference of The Orton Dyslexia Society, Philadelphia. In Dyslexia: An Appraisal of Current Knowledge. eds. A. Benton and
Jansky, J. 1988. Prediction: A six-year follow up. Paper read at the 39th D. Pearl. New York: Oxford Press.
, annual conference of The Orton Dyslexia Society, Tampa. Sears, S., and Keogh, B. 1993. Predicting reading performance using the
Jansky, J., and de Hirsch, K. 1972. Preventing Reading Failure: Prediction, Slingerland procedures. Annals of Dyslexia 43:78-89.
Diagnosis, Intervention. New York: Harper and Row. Shanahan. 1984. Nature of the reading-writing relation: An exploratory
Johansson, B. 1965. Criteria for School Readiness. Stockholm: Almquist and multivariate analysis. [oumal of Educational Psychology 76:466-77_
Wiksell. Shaywitz, S., Shaywitz, B., Fletcher, J., and Escobar, M. 1990. Prevalence of
Keogh, B.1977. Early ID: Selective perception or perceptive selection? reading disability in boys and girls. Journal of the American Medical
Academic Therapy 12(3):268-73. Association 264:998-1002.
LaBerge, D., and Samuels, S. 1974. Toward a theory of automatic informa- Shepard, L., and Smith, M. 1986. Synthesis of research on school readi-
tion processing in reading. Cognitive Psychology 6:293-323. ness and kindergarten retention. Educational Leadership 78-86.
 752 I Flynn

Shepard, L., and Smith, M. 1989. Flunking Grades: Research and Policies on
Retention. London: Falmer Press.
Silver, A., Hagin, R., and Beecher, R. 1978. Scanning, diagnosis, and inter-
vention in the prevention of reading disabilities: 1. SEARCH:The scan-
ning measure. [oumal of Learning Disabilities 11(7):439-49.
Stanovich, K. 1986. The Matthews effects in reading: Some consequences
of individual differences in the acquisition of reading. Reading Research
Quarterly 21:360-407.
Stanovich, K., Cunningham, A., and Freeman, D. 1984. Intelligence, cog- ·1, ..
Chapter- 7
nitive skills, and early reading progress. Reading Research Quarterly
19:278-303.
Stanovich, K., and Siegel, 1.1994. The phenotypic performance profile of
reading-disabled children: A regression-based test of the phonological preschoo~OnOIOgiCal
core variable-difference model. Journal of Educational Psychology
86:24-53.
Stanovich, K., Siegel, L., and Gottardo, A. 1997. Converging evidence for
Awar ness and
phonological and surface subtypes of reading disability. Journal of
Educational Psychology 89(1):114-27.
Succe in Reade
Stevenson, H., Parker, T., Wilkinson, A. Hegion, A., and Fish, E. 1976.
Predictive value of teachers' ratings of young children. Journal of
Educational Psychology 68:506-17.
Strag, G.1972. Comparative behavioral ratings of parents with severely
mentally retarded, special learning disability and normal children.
Journal of Learning Disabilities 5:52-6.
Torgesen,]. 1997. Issues and conclusions about interventions: A research-
based perspective. Paper read at the 48th annual conference of The
Orton Dyslexia Society, Minneapolis.
Vygotsky, L. 1930-35/1978. Mind in Society: The Development of Higher . e en remarkably produc-
Psychological Processes. Cambridge, MA: Harvard University Press. tive in terms of research on readini Wit a consistency that is
Wagner, R., and Torgesen,]. 1987. The nature of phonological processing , the fac ors that can explain
and its causal role in the acquisition of reading skills. Psychological reading success and failure hav ecome clea ly documented. In
Bulletin 101:192-212. one of the most comprehensive nalyses of this search, Stanovich
Wolf, M. 1998. What time may tell: Towards a reconceptualization of de- (1986) wrote, "Evidence is ounting that the rimary specific
velopmental dyslexia. The Norman Geschwind memorial lecture pre-
sented at the 49th annual International Dyslexia Association meeting, mechanism that enables rly reading success phonological
San Francisco. awareness: conscious acc s to the phonemic leve of the speech
Zenka, L., and Keatley, M. 1985. Progress toward excellence: Tulsa's stream and some ability 0 cognitively manipulate r resentations
kindergarten program. ERS Spectrum 3:3-8. at this level" (p. 362). n the decade following Stan ich's paper,
research has continu d to be productive and to confirm the pivotal
role phonological areness plays in learning to read.
Historically, some general measures of cognitive pability
such as general ntelligence and language processing ski shave
had a significan positive relationship with reading. Howe er, re-
search has shown that phonological awareness is the most p tent
predictor of who will learn to read successfully and who is a risk
of failure (Bradley and Bryant 1978; Share et al. 1984; Tunme and
Nesdale 1985). In fact, Muter et al. (1998) found that IQ does not
have an independent relationship with reading other than its asso-
ciation with phonological awareness. In other words, the relation-
ship between reading and IQ is fully accounted for by the
relationship between phonological awareness and reading.

View publication stats