Speech And Language Characteristics Of Children With Developmental Dyslexia

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content 71- 12,420
VENT, Faythe, 1926-
SPEECH AND LANGUAGE CHARACTERISTICS OF CHILDREN
WITH DEVELOPMENTAL DYSLEXIA.
University of Southern California, Ph.D.,
1970
Speech Pathology
University Microfilms. A X E R O X Company. Ann Arbor, Michigan
;
*»va«Twmi*BVAM war* b w u u T A D A v rr u r n p v a o » pt V AO D m P T l f r n
SPEECH AND LANGUAGE CHARACTERISTICS OF
CHILDREN WITH DEVELOPMENTAL DYSLEXIA
by
Faythe Vent
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(Communicative Disorders)
August 1970
UNIVERSITY O F SO UTHERN CALIFORNIA
TH E GRADUATE SCHOOL
UNIVERSITY PARK
LOS ANGELES. CALIFORNIA 8 0 0 0 7
This dissertation, written by
............................. Faythe..Vent.................................
under the direction of h&z.... Dissertation C om
mittee, and approved by all its members, has
been presented to and accepted by The G radu
ate School, in partial fulfillment of require
ments of the degree of
D O C T O R OF P H I L O S O P H Y
TABLE OF CONTENTS
Page
LIST OF TABLES
iv
LIST OF ILLUSTRATIONS
v
Chapter
I. INTRODUCTION 1
Introduction
Purposes of the Study
II. REVIEW OF THE LITERATURE 6
Speech and Language Development
Incidence and Type of Speech Disorders
Incidence and Type of Language Deficits
Possible Approaches to Language
Measurement
Variability of Children with
Developmental Dyslexia
Description of the Sample
Standardized Speech and Language
Measures
Non-standardized Speech and Language
Measures
Testing Procedures
Scoring Procedures
Statistical Procedures
III. METHODS AND PROCEDURES 29
ii
lhapter Page
IV. THE RESULTS................................... 44
Question 1
Question 2
Question 3
Question 4
V. SUMMARY AND CONCLUSIONS.................. 86
Summary
Conclusions
Limitations of the Study
Suggestions for Further Research
APPENDIXES
Appendix A: Scoring Procedures for the Modified
: Picture Story Language Test Produc-
! tivity and Syntax Scales ............... 93
I
j Appendix B: Normative Extrapolation Curves for
Subtests of the Illinois Test of
Psycholinguistic Abilities ............. 103
Appendix C: Modified Picture Story Language Test
Data for 30 Dyslexic Children Com
pared with Wilson's (1968) Normative
Group ..............................116
Appendix D: ITPA Profiles of 30 Dyslexic Children
Shown in Quotients and Based in Part
on Extrapolated Norms ................. 119
(BIBLIOGRAPHY.............................................122
LIST OF TABLES
Table Page
1. Speech and Language Development History
as Reported by Parents....................... 47
2. Range of Quotients on Five Tests of
Language Comprehension and Usage ............. 54
3. Inter-test Correlations: Three ITPA
Subtests, WISC Verbal Scale IQ and
WRAT Quotient............................. 55
4. Productivity Measures on the Modified
Picture Story Language Test for 30
Dyslexic Children and 22 of Wilson's
(1968) Preliminary Normative Group ........... 58
5. Clinical Diagnoses of Speech and Language
Characteristics of 30 Dyslexic Children . . . 69
6. Summary of Scores on All Subtests of the
ITPA by 30 Dyslexic Children.............. 72
7. Correlation Matrix: ITPA Subtest and WRAT
Scores for 30 Dyslexic Children ............. 74
8. ITPA Subgroups Based on Sequencing Skills
for 30 Dyslexic Children.................. 77
9. Subgroups of Dyslexic Children Based on
Clinical Impressions of Language Adequacy . . 83
iv
LIST OF ILLUSTRATIONS
Figure
1.
I
I
! 2.
Page
Median Total Words Corrected Produced by
30 Dyslexic Children on Modified Picture
Story Language Test, Compared with Wilson's
Preliminary Norms........................... 59
Median Total Sentences Produced by 30 Dys
lexic Children on Modified Picture Story
Language Test, Compared with Wilson's
Preliminary Norms ........................... 61
Median Total Words Per Sentence Produced by
30 Dyslexic Children on Modified Picture
Story Language Test, Compared with Wilson's
Preliminary Norms ........................... 62
v
CHAPTER I
INTRODUCTION
Introduction
It has been estimated that as many as 10 per cent of
jail school children read significantly below the expected
'level for their mental ages (Critchley, 1964; Thompson,
1966; Rabinovitch, 1968). Many causes of reading failure
have been postulated, and poor readers have been scrutinized
from many points of view. An increasing amount of attention
has been focused in recent years on developmental dyslexia
as one cause of reading failure. In 1964 the Council for
Exceptional Children of the National Education Association
adopted the term "specific language disability" (SLD) to
designate this problem.
From the neurologist's point of view, developmental
dyslexia or specific language disability is a constitutional
defect in the ability to interpret written symbols (Herman,
1959; Critchley, 1964) . Orton (1925) estimated its
1
2
incidence at 2 per cent of the school population. Rabino
vitch (1968) is in close agreement, estimating that one-
fourth to one-third of school children with reading retarda
tion have some degree of developmental dyslexia. Critchley
(1964) has estimated that developmental dyslexia is present
in 1 to 2 per cent of all males. The problem is thought to
be hereditary and occurs three to four times more frequently
in boys than girls (Orton, 1928; Hallgren, 1950; Eisenberg,
11966) .
j According to Critchley (1964), developmental dys
lexia is distinguishable from other causes of poor reading
by the peculiar and specific errors made in reading and
writing, the extreme persistence of the disability, a strong
family history of reading and communication problems, and
frequent association with other symbol defects. The marked
variability among children with developmental dyslexia has
hindered better understanding of the problem, however.
Establishment of meaningful subgroups or types of develop
mental dyslexics has proven difficult. Rabinovitch (1968)
has commented that regardless of what diagnostic correla
tions are attempted, the behavior or defect questioned is
rarely found in more than 15 to 20 per cent of the dyslexic
children studied.
3
i
Its frequent association with other symbol defects
lhas led some workers to regard developmental dyslexia as a
part of a more generalized language disorder affecting con
cept formation and the acquisition of communication skills
(Bender, 1958). Some writers (Orton, 1928; Saunders, 1962;
DeHirsch, 1967) have described "characteristic speech and
language deficits" associated with developmental dyslexia.
Hard data on speech and language characteristics of children
with developmental dyslexia are lacking, however.
Most research concerning the relationships of speech,
language, and reading deficits has been based on a multiple
causation theory of reading failure. There has been no
attempt to correlate the incidence and type of communication
deficit with the etiology of the reading disability. Diag
nosis of speech and language deficits has been made, in many
cases, by persons without specialized training in speech and
language pathology.
There appears to be a need for careful description
and documentation of speech and language characteristics of
children with developmental dyslexia. If, as has been pos
tulated, the disability is a part of a more pervasive lan
guage disorder, then detailed study of speech and oral lan
guage patterns might lead to better understanding of it.
4
It seems possible that speech and oral language character
istics might be an effective basis for determining whether
subgroups or subtypes of dyslexic children exist. Knowledge
of "characteristic deficits" could aid in earlier detection
of potential reading failures and point the way toward more
effective remedial techniques.
Purposes of the Study
The purposes of the present study were (1) to docu-
I
ment and describe the speech and language characteristics of
a group of children with developmental dyslexia and (2) to
jdetermine whether recognizable subgroups could be differen
tiated among them on the basis of such characteristics. The
following specific questions were asked:
1. How many of the children in this study present
histories of slow speech and language development or other
communication problems?
2. How many of the children deviate from the norm
on some measurable dimension of speech or oral language at
this time? To what extent do they deviate?
3. Does analysis of profiles on the Illinois Test
of Psycholinguistic Abilities show consistent areas of
strength or weakness?
5
4. Can recognizable subgroups be differentiated on
the basis of speech and language characteristics? On the
basis of these plus WISC findings?
CHAPTER II
REVIEW OF THE LITERATURE
It has been estimated that since 1896 a total of
20,000 professional books and articles have been written on
the subject of reading failure (Ellinson, 1967). Extensive
reviews of this literature are available in Hallgren (1950),
Critchley (1964), and Thompson (1966). For the purpose of
this study, review will be limited to literature pertaining
to speech and language characteristics of children with
reading problems and to bases for subgrouping children with
developmental dyslexia.
Speech and Language Development
Several writers (Critchley, 1964; Ingram, 1963;
DeHirsch, 1966) have commented on the disproportionately
large number of children with slow speech and language de
velopment who later experience reading difficulty. Others
(Orton, 1937; Money, 1966; Cole and Walker, 1966) have noted
6
7
the high frequency with which dyslexic children present
histories of delayed speech and language development.
Critchley (1968) has estimated that 50 per cent of all chil
dren with developmental dyslexia were somewhat slow in talk
ing.
Hallgren (1950) reported that 10 per cent of the 180
dyslexic children he studied were described by their parents
as having had delayed speech and language development.
Ingram (1963), reporting on a group of 78 dyslexic children,
stated that 58 per cent had said their first words after age
18 months, 35 per cent after age two years, and 8 per cent
after age three years; about a third of the group did not
speak in phrases until after three and one-half years of
age. Clark (1966) found that all of the 24 children she
evaluated because of "general language disability" were at
least five months slower in talking than in walking; the
median age for walking was 12 months compared to 27 months
for talking.
Discussion
Data are so limited that it is not possible to draw
conclusions about the relationship between slow speech and
language development and reading disabilities. The
8
difficulties of obtaining such data are immediately appar
ent, because documentation of early speech and language
history is dependent upon parental recall and definition of
onset of speech. The relative unreliability of data in this
area has been documented by Darley and Winitz (1961).
Incidence and Type of Speech Disorders
The high incidence of speech disorders among chil
dren whose reading disabilities are attributable to develop
mental dyslexia was first commented on by Orton (1925), who
noted that severe dyslexia was often associated with severe
stuttering. He later (1931) reported that articulation
defects were often present as residuals of infantile speech.
The consonants /r/, /l/, /s/, and /0/ were commonly defec
tive, although he also noted lack of clear-cut differentia
tion of the short vowels. Critchley (1964) has listed the
/r/, /9/, and /v/ sounds as those most frequently causing
difficulty, but notes that these problems are usually over
come by the late teens.
The need to listen to the dyslexic child's sponta
neous, propositional speech has been stressed by several
writers (Orton, 1931j Saunders, 1962; DeHirsch, 1968).
Subtle, irregular problems of articulation, unusually soft
9
or monotonous voice, and choppiness or non-fluency of speech
are frequently mentioned characteristics.
"Spoonerisms" or "wherryisms" often characterize
dyslexic speech (Thompson, 1966; Arnold, 1967; Shephard,
1967). Both terms refer to inadvertent transpositions of
sounds or syllables in words. (The Rev. W. A. Spooner of
New College, Oxford and the late Senator Kenneth S. Wherry
of Nebraska were both well known for such slips. Often
quoted examples are "our queer old dean" for "our dear old
queen" attributed to Spooner, and "opple amportunity" at
tributed to Wherry [Arnold, 1966].) "Slurvian" is another
term sometimes applied to dyslexic speech because of a ten
dency to slur over sounds and syllables so that, for exam
ple, "I am an animal lover" becomes "I am enamel lover"
(Shephard, 1967).
Several European studies have documented the inci
dence of speech disorders in children diagnosed as having
developmental dyslexia. KSgen (1943), in a study cited by
Hallgren (1950), reported that 30 per cent of his 100 dys
lexic children had "retarded speech development." Hallgren
(1950) compared 38 dyslexic children with 173 normals, find
ing speech defects in 26 per cent of the dyslexic children
and 7 per cent of the normals (34 per cent of the dyslexic
10
children were felt by their parents to have speech prob
lems). He also compared 180 dyslexic children, many of them
siblings, with 103 non-dyslexic siblings. Twenty-three per
cent of the dyslexic group were found to have speech prob
lems as compared to 11 per cent of their siblings. (Forty-
nine per cent of the dyslexics and 18 per cent of the sib
lings were reported by their parents to have speech prob
lems.) Hallgren provided breakdowns for the frequency of
stuttering, lisping, and cluttering but most of the speech
problems fell into the category designated as "other,"
Kaiser (1956), cited by Arnold (1966), studied 20 dyslexic
children and reported that all were clutterers.
A number of studies done in this country have docu
mented the frequency of speech problems among children
designated only as poor readers. Arthur (1927) compared
school records of 14 non-readers aged eight to 14 years old
with those of 14 normal readers aged six to 10 years. He
found that eight of the non-readers were described as having
"infantile speech, oral inactivity, and poor enunciation,"
compared with only two of the normal readers, who were
younger and had superior IQ's.
Anderson and Kelly (1931) matched 100 reading dis
ability cases with 100 normal controls on the basis of age,
11
sex, and IQ. They found that 27 per cent of the poor read
ers had speech defects compared with 10 per cent of the good
readers.
Monroe (1932) compared 415 defective readers with
101 normal children. Her group included 315 children (mean
age 10 years nine months) who had been referred to a child
guidance clinic for various behavior problems and had been
found to have coincidental reading retardation, 155 children
I
(mean age 11 years one month) referred specifically because
of reading problems, and 45 children (mean age 12 years
eight months) with borderline to defective intelligence who
were reading below the expected level for their mental ages.
Two per cent of the children had significant hearing losses.
A younger control group (mean age eight years four months)
was chosen to match "stage of learning." Monroe used two
classifications of speech problems: stuttering or stammer
ing and articulation defects (including infantile speech,
lisping, and mispronunciations). Nine per cent of her de
fective readers had stuttering problems compared to 1 per
cent of the normal readers; 18 per cent of the poor readers
had articulation defects compared to 7 per cent of the nor
mal reading group.
Gaines (1941) and Wong (1944) reviewed the literature
12
on the relationship of speech and reading disorders and
concluded that the amount of research done was too small
either to disprove or substantiate any assumptions that
might be made. Both noted that the diagnosis of speech dis
orders had been made by variously qualified persons and that
lack of standardized definitions for either speech or read
ing disabilities as well as lack of normative data precluded
comparisons.
Jackson (1944) compared 214 advanced readers with
234 retarded readers in the fourth through sixth grades and
found that 23 per cent of the poor readers had speech de
fects compared to 10 per cent of the advanced readers.
Artley (1948) again reviewed the literature and con
cluded that a relationship existed between speech and read
ing disabilities but that there was a lack of agreement as
to the extent of that relationship.
Yedinack (1949) surveyed 2,147 second-grade children
and found that 3 per cent had functional articulation prob
lems and 3 per cent had reading disabilities. One per cent
were deficient in both articulation and reading. Forty per
cent of the poor readers had articulation defects and 38 per
cent of the children with articulation defects had reading
deficits. Four groups of children were chosen for further
13
study: 72 normals with neither speech nor reading deficits,
42 with articulation deficits only, 35 with reading dis
abilities only, and 27 with both speech and reading prob
lems. Mean IQ for all groups was 97.8. Children with ar
ticulation deficits only did not differ in IQ from the nor
mal group. Children with reading deficits or reading and
articulation deficits averaged somewhat lower in IQ.
yedinack concluded that children with articulation defects
are significantly inferior to normals in both oral and si
lent reading and that children with reading disabilities
often have concomitant articulation problems.
Several speech pathologists have investigated the
frequency of reading problems among children with speech
deficits. Gibbons (1934) and Moss (1938) found that second-
grade children with speech problems did significantly more
poorly than normal speakers on selected reading tests.
Moore (1947) checked school records of 233 speech handi
capped high school students and found that their median per
formance on the Iowa Silent Reading Test was the same as
that of the rest of the high school population.
Weaver et al. (1960) investigated the relationship
between articulatory competence and reading readiness in
638 first-grade children. They concluded that the two were
14
to some extent related but that the proportion of variance
common to both measures was small. Others (Jones, 1951;
Zedler, 1956; Sommers et al.. 1961; Irwin, 1962) have in
vestigated the effect of speech therapy or speech improve
ment on reading with inconclusive results.
Discussion
Despite the lack of valid bases for comparison of
studies, it seems safe to conclude that poor readers as a
group have more speech deficits than normal readers. When
considered separately, there may be a higher incidence of
speech deficits in children with developmental dyslexia than
in a heterogenous population of poor readers.
Kaiser's (1958) findings are especially pertinent,
since many of the deviations described by Orton (1931),
Saunders (1962), Arnold (1967), Thompson (1966), and others
fall into the general category of cluttering. This term is
used more widely in Europe than in the United States; it is
used more frequently by physicians than by speech patholo
gists .
Bakwin and Bakwin (1961) define cluttering as a
pathological exaggeration of normal, physiologic errors of
speech. Arnold (1963) lists two elements of cluttering:
15
tachylalia. primarily a disorder of speaking rate, and
paraphrasia, a disorder of language formulation that in
cludes repetitions of syllables and words, inversions of
words in the sentence, and dysgrammatic distortions of syn
tax. Both Arnold (1963) and Weiss (1964) report that clut-
terers may have little difficulty with individual sounds and
words. Their speech improves markedly when they concentrate
on speaking correctly. Thus, cluttering might not be iden
tified in a formal, highly structured test situation.
Incidence and Type of Language Deficits
Characteristic language deficits of dyslexic chil
dren have been noted by many writers. Orton (1927),
DeHirsch (1968), Rabinovitch (1968), and others have re
marked that grammar is often poor and that difficulties in
word finding and syntax are evident. Thompson (1966) and
Saunders (1962) have commented on the frequency of mala-
propisms and scrambling of cultural cliches. Conceptual
deficits (inability to form normal judgments in relation to
size, time, and distance) and confusion of concepts (up for
down, first for last, etc.) are also noted (DeHirsch, 1968;
Johnson and Myklebust, 1967; Rabinovitch, 1962; Saunders,
1962). Rabinovitch (1968) has hypothesized that a specific
16
conceptual deficit in orientation is the basis of these
language inadequacies. The dyslexic child has difficulty
translating perceptions into symbols. While able, for ex
ample, to appreciate which of two people is taller, he may
be unable to define their height in feet and inches.
DeHirsch (1968) agrees that dyslexic children are weak in
conceptualization, but suggests that their generally poor
organizational ability is the basis of the grammatical and
syntactic deficits.
Critchley (1968) doubts the presence of linguistic
deficits in developmental dyslexia. He suggests that the
presence of a language disorder in a dyslexic child may in
dicate that the problem is not one of developmental dyslexia
but of symptomatic or secondary dyslexia subsequent to brain
lesion. Joos (1966), viewing the problem as a linguist,
states that unless the non-reader is actually in or above
the normal range of language ability and skill, he is not
dyslexic.
DeHirsch (1968) studied the records of 16 dyslexic
boys aged 11 to 15 years. IQ's ranged from 115 to 142. All
of the boys showed problems in auditory discrimination, word
finding, and formulation of thoughts. Oral and written out
put was disorganized. None of the boys was able to tell a
17
complex story coherently and clearly. Myklebust (1965) has
documented language deviations in written work of dyslexic
children, but his findings were not subjected to statistical
analysis.
Several studies have explored language development
in children described only as poor readers. Monroe (1932)
and Yedinack (1949) in their previously cited studies noted
that their poor readers tended to have poorer vocabularies
than normal readers. Monroe also noted that some of the
poor readers had difficulty with sentence structure. Yedi
nack, however, found no significant differences between good
and poor readers in length of response or complexity of
grammatical structure.
Shire (1945) gave reading tests to 300 first graders,
then selected the 30 best and the 30 poorest readers for
further study of language development. She found signifi
cant differences in mean length of response, number of com
plete grammatical sentences, and number of elaborated sen
tences . Mean IQ for the good readers was 118 compared to
99 for the poor readers.
DeHirsch et al. (1964) compared 51 prematurely born
and 55 maturely born children on an extensive battery of
tests at kindergarten age. In a follow-up study (1966)
18
standardized reading, writing, and spelling tests were given
when the children were completing second grade. Eleven of
the children were having significant reading difficulties;
10 of these children were clearly identifiable on the basis
of kindergarten tests. Language measures that proved to be
reliable predictors included number of words used in telling
a story, organization of a story (ability to integrate de
tails into a meaningful whole), and categorizing (capacity
verbally to group objects or events in terms of their common
denominator).
Discussion
On the basis of the limited number of reported
studies, it is not possible to draw conclusions about the
pervasiveness or extent of language deviations among chil
dren with reading problems.
Possible Approaches to Language
Measurement
From the literature it is apparent that children
with developmental dyslexia are most likely to demonstrate
speech and language deficits during spontaneous, proposi-
tional verbalizations. Various measures of spontaneous
speech and language have been devised. A comprehensive
19
discussion of conventional measures is available in Johnson,
Darley, and Spriestersbach (1963). The use of rating scales
is generally accepted for evaluation of speech character
istics. The evaluation of spoken language, however, pre
sents a more difficult problem.
Normative data on language development, based on
Templin's studies (1957), extend only through eight years of
age. The reliability of some of her measures as adapted for
I
clinical use has been questioned by Minifie, Darley, and
Sherman (1963). Also, according to Shriner (1969), recent
developments in psycholinguistics have pointed up many in
adequacies in such measures. Work by Menyuk (1964) and Lee
(1966), for example, suggests that children with retarded
language development are not merely following the normal
developmental pattern at a slower rate but are failing to
make linguistic generalizations on which syntactic develop
ment depends.
Shriner (1967) found that mean length of response,
one of the most commonly used measures, was not a signifi
cant indicator of expressive language development in chil
dren over five years of age. For older children a measure
of structural complexity appeared to be of greater value.
VanDemark and Mann (1965) had reached similar conclusions
20
after comparing the language skills of 50 speech defective
children and 50 normal children aged eight to 13 1/2 years.
Of the seven conventional measures used, only the sentence
complexity scores differentiated the two groups. Unfortu
nately the standard sentence complexity measure (Structural
Complexity Score) has been found by Templin (1958) and
Darley and Moll (1960) to be relatively unreliable.
Marge (1964) studied oral communication skills in
143 pre-adolescent youngsters by means of factor analysis.
Teachers' and speech specialists' ratings of grammatical
usage, pronunciation, complexity of sentence structure, and
vocabulary emerged as highly significant factors in the
over-all judgment of language maturity.
Wilson (1968), whose major concern was developing a
standardized method of obtaining language samples, adapted
the Picture Story Language Test (PSLT) (Myklebust, 1965) for
use with spoken language. The PSLT was designed as an in
strument to study written language from both a developmental
o
and diagnostic point of view. It consists of a stimulus
picture about which the subject is asked to write a story.
Stories are scored on the basis of Productivity (Total
Words, Total Sentences, Total Words Per Sentence), Syntax
Quotient (a total score that includes word usage, word
21
order, and grammatical correctness), and an Abstract-
Concrete Scale that attempts to measure the quality of ideas
expressed. The test was standardized on 747 children be
tween the ages of seven and 17 years and found to be a re
liable measure of written language. Although not intended
to be a measure of spoken language, Myklebust (1965) felt
that comparison of verbal and written performance on the
test was of value in the diagnosis of communication dis
orders .
Wilson (1968) taped responses elicited with the
stimulus picture for a group of 40 normal subjects aged
three to 17 years, then scored typed transcripts of the
stories using a modification of Myklebust's scoring tech
nique. She found the stimulus picture to be an adequate
means of eliciting language samples. All of the measures
except Total Sentences and Syntax Quotient demonstrated an
orderly progression of increasing language facility with
increasing age. The Total Sentences discrepancy was thought
to be an artifact of the small sample size, as the means
were affected by extreme scores at some age levels. The
problem of the Syntax Quotient was more complex.
Myklebust (1965) had found that inter-scorer relia
bility on this measure was closely related to the training
22
level of the scorers. With trained scorers (doctoral level
clinicians who were highly skilled in the use of objective
tests) inter-scorer reliability was .85 compared to .79 for
untrained scorers (graduate students with limited training
in the use of educational and psychological tests). Wilson,
using untrained scorers, obtained an inter-scorer relia
bility coefficient of -.28. She suggested the possibility
that her modifications of scoring technique had affected the
weighting of scores used in obtaining the Syntax Quotient.
She attributed the greater part of the difficulty, however,
to lack of clearly defined criteria for determining the
correctness of language usage.
The syntax scale as modified by Wilson provided for
errors of omission, addition, or substitution of whole words
or word endings, and also for errors in word order. She
exempted the words "and" and "and so" from penalty as word
additions because of their common overuse by the subjects
in her study.
McCarthy (1930) noted that much of adult spoken
language is composed of phrases and other groups of words
that are not strictly sentences. Templin (1957) reported
that 9.5 per cent of all utterances made by eight-year-olds
in her study were incomplete sentences. Only 76.1 per cent
23
of the remarks made by her eight-year-olds were grammati
cally correct. This suggests that a Syntax Quotient for
spoken language might be somewhat lower than for written
language. Myklebust (1965) found that the average Syntax
Quotient for written language was 89 per cent for eight-
year-old males and 92 per cent for eight-year-old females.
Wilson's eight-year-olds achieved Syntax Quotients of 98.5
per cent.
i
| Templin (1957), obtaining her language sample in
part in a non-story-telling situation, found that beginning
sentences with "and" occurred only two and one-half times
per thousand words. Occurrence was more frequent among the
younger children than in older age groups. This suggests
that over-use of "and" may be a dimension of language that
has developmental significance. It would seem that, if a
measure is to reflect the language changes that occur with
maturation or pathology, rigorous adherence to a carefully
defined Stringency Level is mandatory regardless of the
frequency with which certain errors occur.
Variability of Children with
Developmental Dyslexia
Children who experience reading difficulty may do so
for a variety of reasons. Developmental dyslexia is thought
24
to account for only a small portion (25 to 30 per cent) of
reading disability cases (Rabinovitch, 1968). There is evi
dence to suggest that even this small group may be hetero
geneous .
Orton (1928) saw dyslexic children as a relatively
homogeneous group who differed in degree but not in type of
impairment. He suggested that the basis of the problem was
a defect in building associations between the auditory and
visual modalities. He viewed speech deficits and non-verbal
apraxia, often seen in association with developmental dys
lexia, as separate but related disorders (1930, 1931). In
1946 he pointed out that the only general process common to
reading difficulties, speech disabilities, and non-verbal
apraxia was sequence building.
Johnson and Myklebust (1967) recognize a visual dys
lexia and an auditory dyslexia. They report that visual
dyslexia is characterized by deficits in visual discrimina
tion or perception, reversal or inversion tendencies in
reading and writing, and difficulties in re-visualizing,
visual sequencing, visual analysis, and visual synthesis.
The auditory dyslexic, on the other hand, reportedly has
deficits in auditory perception and discrimination, may not
perceive similarities of sound, and has difficulty with
25
rhyming; auditory analysis and synthesis are poor; he cannot
dissect words and put them together again. Difficulties in
re-auditorization and auditory sequencing are also present.
Other writers (DeHirsch, 1966; Rabinovitch, 1968) doubt this
clear-cut division. They point out that both auditory and
visual memory are usually impaired in children with develop
mental dyslexia.
Bannatayne (1966) identifies two main groups of dys-
i
lexic subjects: pure dyslexics with an inherited inability
to use language fluently, and persons who, along with the
dyslexia, show evidence of neurologic abnormalities. He
sees the inability to sequence associated sounds in phase
with visual symbols as the basis of dyslexia. He regards
the "pure dyslexics" as a relatively homogenous group repre
senting the tail end of the normal ability continuum.
Wolfe (1967) compared 32 dyslexic children with 23
normal readers on 198 variables and found the dyslexic chil
dren as a group to be significantly poorer than the normals
in auditory discrimination, auditory blending, and auditory
memory. Three subgroups of dyslexic children emerged on the
basis of Wechsler Intelligence Scale for Children (WISC)
findings: those who did more poorly on the verbal tests
than on performance, those who did better on verbal than on
26
performance subtests, and those vho did about the same on
both tests. No attempt was made to correlate other findings
with these subgroupings.
Kinsbourne and Warrington (1966) studied 13 dyslexic
subjects who readily fell into two subgroups. Group I con
sisted of six children and adults whose performance scores
on the WISC or Wechsler Adult Intelligence Scale (WAIS) ex
ceeded the verbal by at least 20 points. Performance on
reading and spelling tests showed about the same degree of
retardation in both areas. Histories of slow speech and
language development were common, and all showed persisting
evidence of inadequate verbal expression and reception. All
showed impairments on tests of verbal learning. The writers
suggested that this group might represent a developmental
Gerstmann's Syndrome and postulated that the reading re
tardation was secondary to the over-all language deficits.
Herman (1959) has also suggested that a developmental Gerst
mann's Syndrome might be the basis of dyslexia.
Group II consisted of seven children whose verbal
score on the WISC exceeded the performance score by at least
20 points. Spelling age was about a year below reading age,
and reading retardation was more severe in this group than
in Group I. These subjects, like Bannatayne's second
27
subgroup, showed more evidence of neurologic impairment than
Group I. Reading disability was attributed to deficits in
sequential ordering.
Discussion
No definite conclusions can be drawn from these
limited data. The question arises, however, whether some of
Bannatayne's "pure dyslexics" and Kinsbourne and Warring
ton's Group I may not be identical to those cases described
by Weiss (1964) and Arnold (1967) in their discussions of
cluttering and tachyphemia. Weiss notes that reading prob
lems are almost always associated with cluttering. However,
he differentiates between these reading problems and iso
lated dyslexia, which represents a defect in the ability to
handle symbols. Weiss states that the clutterer usually has
the ability to learn to read and that his reading achieve
ment improves when his problems in attention are brought
under control. Although a true dyslexia and cluttering may
occur together, Weiss feels this association is relatively
rare. It has also been noted by Rabinovitch (1968) that the
dyslexic child's spelling is usually well below his reading
level. Such was not the case with Kinsbourne and Warring
ton's Group I.
28
Rabinovitch (1968) has pointed out that regardless
of what diagnostic correlations are attempted, the behavior
or. defect questioned is rarely found in more than 15 to 20
per cent of the dyslexic children studied. This suggests
that classification into two subgroups may not provide an
adequate framework for accurate description of children with
developmental dyslexia.
CHAPTER III
METHODS AND PROCEDURES
Description of the Sample
The sample consisted of 30 children (29 boys and one
girl) who were selected from a larger group of children
evaluated in the Mayo Clinic Section of Pediatric Neurology
and diagnosed as having some degree of developmental dys
lexia. Selection was made on the basis of the following
criteria:
1. A quotient of 90 or above on either the Verbal
or Performance Scale of the WISC. Johnson and Myklebust
(1967) have stated that a quotient of 100 or above on either
scale is essential for a diagnosis of a specific learning
disability. It was necessary to lower this criterion to 90,
however, in order to obtain an adequate sample. Verbal IQ's
ranged from 80 to 124. Mean Verbal IQ was 98, and the me
dian verbal IQ was 96. Performance IQ's ranged from 95 to
129, with a mean of 108 and a median of 106. Full Scale
29
30
IQ's ranged from 88 to 125. Mean Full Scale IQ was 103,
median was 101. Thus, all of the subjects fell within the
generally accepted range of normal intelligence.
2. A Reading Quotient on the Wide Range Achievement
Test (WRAT) that was at least 10 points below the WISC Full
Scale IQ. Jastak and Jastak (1965) have indicated that this
discrepancy represents a degree of reading retardation that
is severe enough to suggest reading disability. WRAT quo
tients ranged from 61 to 107; mean and median WRAT quotients
were 79. WISC-WRAT discrepancies ranged from 12 to 55. The
mean discrepancy was 24 points, while the median was 21
points.
The children ranged in age from six years 11 months
to 14 years seven months. Mean age was 10 years five
months; median age was 10 years four months. All had normal
or corrected vision, and normal hearing in one or both ears.
The subjects came from the upper midwestern states,
primarily from rural or small city areas. All were white,
native English speakers; none came from bilingual homes.
Socioeconomic backgrounds varied. About a third of the
youngsters' parents were in the professions or held mana
gerial positions. A number of the parents were farmers;
some were blue collar workers in factories. Two of the
31
children were adoptive siblings; there were no natural sib
lings in the study. In all cases there was a history of
close relatives having had difficulty in learning to read.
About a fourth of the children had a parent or sibling who
was a non-reader.
Standardized Speech and Language
Measures
The test battery was selected to provide measures of
articulation ability, fluency, verbal comprehension, oral
language usage, and psycholinguistic abilities. The follow
ing standardized tests were used:
1. Templin-Darley Tests of Articulation, Revised
Edition (Templin, Darley, 1969). These tests are an out
growth of Templin*s earlier study (1957) of speech and lan
guage development in children. Test items and norms are
based on a sample of 480 children between the ages of three
and eight years, the age at which most children are able to
produce most sounds correctly in all positions. A complete
description of the tests is to be found in the manual. For
the purpose of this study the Screening Test (50 items
closely associated with significant progress in the develop
ment of articulation) and the Consonant Singles (28 addi
tional items) were used.
32
2. Peabody Picture Vocabulary Test (PPVT) (Dunn,
1959). This widely-used test of vocabulary recognition was
standardized on 4,012 cases ranging in age from two years
six months to 18 years. Test reliability was established
during the standardization studies using the test-retest
method with alternate forms. Description of the test and
reliability data are to be found in the manual. For the
purpose of this study, Form B was used because of its some
what higher correlation with other tests (Dunn, 1959).
3. Illinois test of Psycholinguistic Abilities
(ITPA), Revised Edition (Kirk, McCarthy, Kirk, 1968). This
is a revision of the Experimental Edition first introduced
in 1961 as a diagnostic test of communication abilities.
Revisions were based on validation studies summarized by
Bateman (1965). The Revised Edition was standardized on
approximately 1,000 children between the ages of two and 10
years. Psychometric characteristics have been described by
Paraskevopoulos and Kirk (1969) .
The test is based on an adaptation of Osgood's
theoretical model of communication and postulates three
dimensions of cognitive abilities: Channels of Communica
tion (modalities through which impressions are received and
means by which response is made); Psycholinguistic Processes
33
(receptive, expressive, organizing); and Levels of Organiza
tion (automatic and representational). In its present form
the ITPA consists of 10 basic and two supplementary sub-
tes ts :
Functions Tested at the Representational Level
1. The Receptive Process: Auditory and Visual Recep
tion
2. The Organizing Process: Auditory and Visual
Association
3. The Expressive Process: Verbal and Manual Expres
sion
Functions Tested at the Automatic Level
1. Closure: Grammatic, Visual, and Auditory; Sound
Blending
2. Sequential Memory: Auditory and Visual Sequencing
Descriptions of the various subtests are to be found in the
manual.
Although normative data for the ITPA are available
only through age 10, the test has been found clinically
useful with older children who present language problems.
Children whose performances on various subtests fall below
the norms can be assigned a psycholinguistic age score for
those subtests. Areas of difficulty can be pinpointed in
this way, although it is not possible fully to evaluate the
extent of the difficulty._____________________________________
34
Several investigators have used the Experimental
Edition of the ITPA in the study of reading ability in var
ious groups. Kass (1962) studied 21 children with normal
IQ's and severe reading disabilities. Age range was seven
to 10 years. In addition to the ITPA, five other tests at
the automatic-sequential level were used. As a group, the
children scored significantly below the norms on the ITPA
auditory association and sequencing subtests. On the com-
I
bined test battery (14 subtests) they showed deficiencies in
seven out of eight tests at the automatic-sequential level
and one out of six at the representational or meaningful
level.
Bateman (1963) studied 131 partially seeing children
with normal IQ's enrolled in special school programs. As a
group the children read less than a half-month below age
level and about six months below grade level. Their scores
on the auditory-vocal subtests of the ITPA agreed closely
with the norms. As a group they scored below normal on
visual-motor subtests, although children with milder visual
deficits showed no defect in visual association and only
mild defects in visual reception. Reading achievement was
positively correlated with achievement on the automatic-
sequential subtests.
35
Sutton (1963) used the ITPA visual-motor sequential
subtest as a part of a larger battery of tests with educable
mentally retarded (EMR) children. She found that children
who were achieving well in reading did somewhat better on
the test than did poor readers. Ragland (1964) compared
ITPA performances of EMR children who were high achievers in
reading with those of EMR children who were reading below
the expected level for their mental ages. The poor readers
scored less well than the good readers on all subtests ex
cept visual and auditory reception and did significantly
more poorly on both auditory and visual sequential tests.
McLeod (1968) studied factors involved in reading at the
second grade level and concluded that the relevant basic
skills for reading are likely to be in terms of the ITPA
model at the automatic sequential level of organization and
in the auditory-vocal modality.
Only one related study has been published to date
using the Revised Edition of the ITPA. Bannatayne and
Wichiarajote (1969) used the automatic sequencing and clo
sure subtests as a part of a test battery for predicting
spelling achievement in 50 normal third graders. They found
that the Sound Blending subtest was significantly correlated
with spelling ability (r =.40). Auditory Closure was
36
significantly correlated with Sound Blending but not with
spelling. There was no significant correlation between the
auditory and visual sequencing tests and spelling achieve
ment.
Non-standardized Speech and Language
Measures
Analysis of spontaneous
speech and language samples
The taped speech and language samples used in this
study consisted of stories told by the children in response
to an action picture. The picture, showing Paul Revere
warning the colonists of the British approach, was not the
same as that used by Myklebust (1965) and Wilson (1968).
The Abstract-Concrete Scale, based on normal subjects' re
sponses to a standardized picture, was not applicable for
this study. The productivity measures and syntax scale were
used with slight modification, however, as they are not
dependent upon story content.
Productivity measures.— The Myklebust/wilson pro
ductivity measures were used without further modification.
Total Words Corrected (TWC) is the number of words used by
the child in telling a story, corrected for the dysfluencies
37
that normally occur in speech. Rules used for handling
repetitions were those outlined by Johnson, Darley, and
Spriestersbach (1963). Total Sentences (TS) is the number
of sentences as defined by McCarthy (1930) and further de
lineated by Myklebust (1965). Words Per Sentence (WPS) is
computed by dividing the number of words (TWC) by the number
of sentences (TS).
I Syntax quotient (SQ).— For the purposes of this
i
j
study, the Myklebust/wilson error categories pertaining to
whole word usage, word endings, and word order were re
tained. An additional error category, "Revision," was also
used because of the frequency with which the subjects in
this study changed or corrected entire phrases to clarify
meaning or resolve grammatical dilemmas. The words "and"
and "and so" were penalized as additions vdien incorrectly
used. Two arbitrary decisions were made regarding Strin
gency Level: the overuse of qualifying expressions (e.g.,
"probably," "I guess") was considered a matter of style so
long as they were grammatically correct within the sentencej
intended verb tense was determined on the basis of the story
as a whole, and inconsistencies were penalized as errors of
substitution even though their use within the sentence might
38
be correct.
A complete description of scoring procedures, along
with sample stories illustrating their use, is to be found
in Appendix A.
Performance ratings
Taped speech and language samples were rated on
three dimensions by a panel of three senior staff members of
the Mayo Clinic Speech Pathology Section. Two of the panel
members held the Ph.D. in Speech Pathology; all held the
Certificate of Clinical Competence in Speech Pathology and
had previous experience with similar rating scales.
A seven-point equal-appearing-interval scale was
used on which a rating of "one" indicated normal adult per
formance and a rating of "seven" indicated severe impairment
as described below.
Articulation Performance Rating (APR).— Speech is
characterized by frequent distortions, substitutions, or
omissions of sounds. Consonant sounds lack precision or
adequate sharpness.
Fluency Performance Rating (FPR).— Speech is char
acterized by interjections, hesitations, inappropriate
39
pauses, prolongations, repetitions, revisions, incomplete
sentences, and broken words.
Language Performance Rating (LPR).— Mid-point and
extreme values were defined as follows:
1— Normal adult language
4— Immature language. Sentences are short and
simply constructed. Simple grammatical forms are
used. Grammatical errors are typical of young
children. Vocabulary is limited but reasonably
accurate.
7— Deviant language; severe impairment. Sentence
structure is poor and grammatical errors are
bizarre. Vocabulary is imprecise and/or inaccu
rate. Language usage distracts listener from
content.
Raters worked independently and were instructed to
rate each dimension separately; thus, each rater listened to
each sample a total of three times. Ratings of the judges
were then averaged to provide a more stable measure.
Testing Procedures
The speech and language evaluation was done as a
part of a comprehensive clinical work-up which included
40
examinations by a pediatric neurologist, audiologist,
psychologist, reading specialist, and other consultants as
needed. The children were seen by the investigator for two
one-hour sessions, usually on the same day.
Tests were administered in the same order to all of
the children. The ITPA was routinely given during the first
session; the Templin-Darley Tests of Articulation and the
PPVT were given during the second session. The taped speech
and language sample was obtained near the close of the sec
ond session. Scheduling of parent interviews varied, de
pending upon the amount of time used for testing and the
child's appointments with other specialists.
Scoring Procedures
Standardized tests were scored by the investigator
in accordance with procedures outlined in the test manuals.
The language samples were scored jointly by the investigator
and the supervising committee member, utilizing both the
typed transcripts and original tapes. The transcripts fur
nished reliable records of the subject's words, but it was
often necessary to listen to pauses and intonation patterns
to determine intended meaning. The scorers worked together
and the resulting scores represent their combined judgment.
41
Statistical Procedures
ITPA extended norms
The chronological ages of more than half the chil
dren in this study were beyond the standardized norms. Al
though all subjects scored below published norms on one or
more subtests, careful study of the test profiles suggested
that analysis based only on these norms would be grossly
incomplete or misleading. Profiles of the younger children
revealed diffuse deficits or varying severity. Profiles of
the over-age children reflected only their most severe defi
cits. Age scores for the older children did not take into
account the age at which performance might be expected to
level off on the subtest in question. At the same time, the
test data themselves provided information that appeared to
warrant detailed analysis. For this reason, it was decided
to extend the published norms by means of extrapolation so
that the test data would be available for further study.
Normative scores for each of the subtests were plotted on
graphs and the resulting curves were extended with the aid
of a French Curve. The Psycholinguistic Age (PLA) score was
not extrapolated, as it was anticipated that this global
score would compound any extrapolation errors occurring on
42
individual subtest scores. The fact that expected perform
ance on individual subtests leveled off at varying ages also
suggested the possibility that the weighting of scores would
be destroyed in extrapolation. Extrapolation curves for all
subtests appear in Appendix B.
Analysis of data
It had been recognized from the outset of the study
that its design would present problems in the statistical
handling of data. The extreme heterogeneity of the sample,
which had not been anticipated, created further difficul
ties. Factor analysis might have appeared to be the ap
proach of choice. The validity of its use was questionable,
however, because of the relatively small sample size and
large number of variables.
Data were examined first by means of scattergrams,
which failed to show expected relationships among measures
that were thought to be logical choices as predictor vari
ables. Sorting techniques failed to disclose subgroups that
were mutually exclusive on the basis of more than a single
variable. Preliminary steps in pattern analysis indicated
that this technique was not adaptable to the data at hand.
Accordingly it was decided to present the data
43
exclusively in descriptive terms. Inter-relationships of
test scores and measures were explored by means of a 26 x 26
correlation matrix provided through the computer facilities
of the Mayo Clinic Section of Medical Statistics, Epidemi
ology, and Population Genetics.
CHAPTER IV
THE RESULTS
The results of the study will be presented in terms
of the specific questions asked at the outset of the study.
; Question 1
i
I
Question 1: How many of the children present his
tories of slow speech and language development or other com
munication problems?
Information on speech and language development was
obtained from parents during an informal interview. It was
possible to verify information on only five subjects by
means of previous clinic records.
Onset of speech
Seven (23 per cent) of the children were reported
to have been slow in talking. Two children were said to
have had little or no speech until three years of age.
Others were described only as having been slower to talk
44
45
than their siblings.
Twenty-three (77 per cent) were reported to have
started talking at what the parents considered to be the
"normal" age. Only a few parents were able to give an
approximate age for the onset of speech. For most parents,
the standard of normalcy was the age at which non-dyslexic
siblings had started to talk.
History of communication
problems
Twenty (67 per cent) of the children were recognized
by their parents as having some difficulties in verbal ex
pression. Thirteen (43 per cent) were reported to have had
specific speech or language deficits, and 12 (40 per cent)
had received one or more years of speech and/or language
therapy.
Seven (23 per cent) of the children were described
as poor communicators although their parents considered
speech and language development to have been normal. All
had started talking at the expected age, but they had never
expressed themselves verbally as well as their siblings.
Many of these children were described as being generally
immature; four were said to have difficulty understanding
and remembering verbal instructions. Communication
46
difficulties were ascribed variously to shyness, daydream-
ling, emotional reaction to academic failure, and basic per
sonality characteristics.
Types of specific problems
reported
Twelve of the 13 children with histories of specific
speech and language problems were described as having had
articulation deficits. Six had concomitant language defi
cits, and three presented histories of stuttering in addi
tion to articulation difficulties. The thirteenth child's
problems had not been diagnosed. He had been slow in talk
ing and, according to his mother, speech had never been
clear. She attributed the difficulty to his tendency to
speak very rapidly. Data from parent interviews are sum
marized in Table 1.
Comparison with previous
studies
Hallgren (1950) reported that only 10 per cent of
the dyslexic children he studied were described as slow in
talking. Ingram (1963), on the other hand, found that 58
per cent of his group had been somewhat delayed in speech
and language development. At least a part of the variation
in findings may reflect differences in interview techniques
47
TABLE 1
SPEECH AND LANGUAGE DEVELOPMENT HISTORY
AS REPORTED BY PARENTS
(N = 30)
Number
of Cases
Percentage
of Total
Onset of Speech
Slow in talking 7 23
Talked at "normal" age 23 77
Historv of Communication Problems
Specific speech and/or language
deficits 13 43
Non-specific communication
deficits 7 23
Types of Specific Problems Reported
Articulation defect only 3 10
Articulation and language deficits 6 20
Articulation defect and stuttering 3 10
Undiagnosed 1 3
Number of Children Who Had Received
Speech Therapy 12 40
48
rather than differences in the samples themselves. The
present investigator was unable to obtain as specific devel
opmental information as that reported by Ingram. A number
of parents in the present study reported other problems,
however, that are often associated with slow speech and
language development. This fact suggests that delayed ac
quisition of speech and language may have been more common
than was reported.
Forty-three per cent of the children in this study
were described as having had specific speech and/or language
problems . This finding is in close agreement with Hallgren's
(1950) report that 49 per cent of the dyslexic children in
his study were believed by their parents to have speech
deficits.
Question 2
Question 2: How many children deviate from the
norm on some measurable dimension of speech or language at
this time? To what extent do they deviate?
The dimensions of speech and language measured in
this study included articulation, fluency, comprehension,
and language usage. Data were collected during the course
of a two-hour clinical evaluation using standardized and
49
non-standardized measures as well as informal observation.
Results will be presented first in terms of group perform
ance on each of the various measures, then summarized in
terms of clinical diagnoses.
Articulation ability
Articulation was evaluated by means of the Templin-
Darley Tests of Articulation (TDTA) and informal observa
tions of spontaneous connected speech. Taped samples of
spontaneous speech were also rated by a panel of judges to
obtain an Articulation Performance Rating (APR).
Tempiin-Parlev Tests of Articulation.— The tests
used comprised 78 items. The expected score is 100 per cent
by age eight years. On the basis of chronological age, 28
of the children would have been expected to score 100 per
cent on the tests. Six of the children fell below the norm.
Number of errors ranged from four to 16; most frequent er
rors were distortions of /r/, /s/, and /s/ clusters.
Only one of the children scoring below the norm was
under eight years. This youngster demonstrated a moderate
articulation defect vdiich was not attributable to age.
The TDTA scores were positively correlated with
chronological age (r = .407). This correlation is
50
significant at the .05 level of confidence. There was no
significant correlation between the TDTA and other tests or
measures used in the study.
Informal evaluation of articulation in connected
speech.— Twelve children made consistent articulation errors
in connected speech. The most frequent errors were distor
tion or omission of /a/, substitution of /f/ for /8/, and
isubstitution of /v/ for /3/.
Articulation performance rating (APR).— The APR was
based on a seven-point equal-appearing-interval scale on
which a rating of "one" represented normal adult speech and
"seven" indicated severe articulation impairment. On the
basis of chronological age, 28 of the children would be ex
pected to receive a "one" or "two" rating. A rating of
"three" might be considered normal for the two youngest
children in the study. Obtained ratings were as follows:
Rating Frequency
51
Obtained APR's were in agreement with the informal
observation that 12 children made consistent articulation
errors in connected speech. The APR was negatively corre
lated with chronological age (r = -.430) at the .05 level of
confidence. Correlation with the Templin-Darley Tests of
Articulation was also negative (r = -.355) but this was not
statistically significant. The APR was positively corre
lated with several other test measures at the .05 level:
WISC Verbal IQ (r = .422); WISC Full Scale IQ (r = .442);
PPVT (r = .422); Verbal Expression subtest of the ITPA (r =
.463); and the Language Performance Rating (r = .422).
Fluency
seven-point equal-appearing-interval scale on which a rating
of "one" represented normal adult fluency and "seven" indi
cated severe impairment. Fluency norms are not available.
Agreement among panel members was poorer for this scale than
for the APR, suggesting differences in opinion regarding
normal fluency. Obtained ratings were as follows:
The Fluency Performance Rating (FPR) was based on a
Rating
1
2
3
4
Frequency
1
11
15
3
52
Correlations between the FPR and other measures were
not obtained.
Comprehens ion
Two tests were used to measure verbal comprehension:
the Peabody Picture Vocabulary Test (PPVT) and the Auditory
Reception (AR) subtest of the ITPA. The PPVT is essentially
a test of verbal recognition, utilizing single words as
stimuli. The AR subtest measures comprehension of contex
tual language. In this study scores on the two tests were
significantly correlated (r = .672) at the .01 level of con
fidence. To facilitate comparison, standard scores (quo
tients) are shown for the PPVT. AR scores have been con
verted to quotients (obtained score divided by expected
score and multiplied by 100). ARQ's for children over 10
years two months of age are based on extrapolated norms.
Peabody Picture Vocabulary Test.— Standard scores
ranged from 76 to 138, with a median score of 99.5. Four
teen of the children attained scores of 100 or better; only
three scored below 85. Correlations between the PPVT and
three other measures were significant at the .01 level of
confidence: WISC Verbal IQ (r = .657); WISC Full Scale IQ
(r = .515); WRAT (r = .544). Range of scores and other
53
information are shown in Table 2.
I
Auditory Reception Quotient (ARQ).— ARQ's ranged
from 64 to 132, with a median of 95.5. Ten subjects ob
tained ARQ's of 100 or above; eight scored below 85. Like
the PPVT, the ARQ was positively correlated with the WISC
Verbal IQ (r = .529) and WRAT (r = .545) at the .01 level of
confidence. Correlation with the WISC Full Scale IQ (r =
1.397) was significant at the .05 level of confidence. Range
of scores and other information are shown in Table 2.
Language usage
Oral language usage was evaluated on the basis of
three ITPA subtests (Auditory Association, Verbal Expres
sion, and Grammatic Closure), analysis of taped spontaneous
language samples, and informal observations of spontaneous
conversational language. Scores on the ITPA subtests have
been converted to quotients to facilitate comparison.
Scores for the older children are based on extrapolated
norms. Inter-test correlations for the ITPA subtests, WISC,
and WRAT are shown in Table 3.
Auditory Association Quotient (AAQ).— The Auditory
Association subtest is a test of verbal analogies. AAQ's
TABLE 2
RANGE OF QUOTIENTS ON FIVE TESTS OF LANGUAGE
(N = 30)
COMPREHENSION AND USAGE
Range of
Quotients3
Median
Quotient3
Percentage
Scoring
Below 85
Percentage
Scoring at
or Above 100
Lanquaqe Comprehension
Peabody Picture Vocabulary Test 76-138 99.5 10 47
Auditory Reception— ITPA Subtest 64-132 95.5 23 33
Lanquaqe Usaqe
Verbal Expression— ITPA Subtest 49-111 73.0 77 10
Auditory Association— ITPA Subtest 67-133 97.0 37 33
Grammatic Closure— ITPA Subtest 67-131 87.5 43 10
aBased in part on extrapolated norms.
U1
TABLE 3
INTER-TEST CORRELATIONS: THREE ITPA SUBTESTS,
WISC VERBAL SCALE IQ AND WRAT QUOTIENT
AA VE GC WISC-V WRAT Q
Auditory Association . 389a .580b .676b .442
Verbal Expression . 733b ,489b .364
Granunatic Closure . 510b .475
WISC Verbal Scale IQ .656
WRAT Quotient
a.05 level of significance.
k.01 level of significance.
ranged from 67 to 133, with a median AAQ of 97.0. Ten of
the children received AAQ's of 100 or better; seven scored
below 85. Range of AAQ's is shown in Table 2; inter-test
correlations are shown in Table 3.
Verbal Expression Quotient (VEQ).— Scores on the
Verbal Expression subtest are based on the number of dis
crete, relevant, and approximately factual concepts the
jchild is able to verbalize about a common object. VEQ's
I
ranged from 49 to 111, with a median of 73. Only three
children made scores of 100 or better; 23 (77 per cent)
scored below 85. Range of VEQ's is shown in Table 2; inter
test correlations are shown in Table 3.
Grammatic Closure Quotient (GCQ).— This subtest
utilizes a sentence completion approach to measuring syn
tactic development. GCQ's ranged from 67 to 131; median
GCQ was 87.5. Thirteen of the children had quotients below
85. Range of GCQ's is shown in Table 2; inter-test correla
tions are shown in Table 3.
Evaluation of taped language
samples
Taped samples of spontaneous language were evaluated
by means of modified Picture Story Language Test (PSLT)
57
measures and the Language Performance Rating (LPR) . Lan
guage was evaluated in terms of productivity, syntax, and
over-all performance.
Productivity.— Productivity measures were the same
as those used by Wilson (1968) in her preliminary normative
study. Twenty-two of her subjects fell within the age group
covered by this study. Number of cases in each age group,
range of scores, and group means for both the normal and
dyslexic groups are to be found in Appendix C.
Total Words Corrected (TWC).— The dyslexic children
showed a narrower range of variation in the number of words
used to tell a story than did Wilson's normals. In addi
tion, the expected trend was reversed in that Total Words
Corrected decreased steadily with increasing age. Compara
tive Median Total Words for both groups are presented in
Table 4 and graphically displayed in Figure 1.
Total Sentences (TS).— Total sentences for the dys
lexic children ranged from two to 13, compared with a range
of two to 61 for the normal subjects in Wilson's study.
Again, the expected trend was reversed, with the dyslexic
children using fewer sentences with increasing age. As with
58
TABLE 4
PRODUCTIVITY MEASURES ON THE MODIFIED PICTURE STORY
LANGUAGE TEST FOR 30 DYSLEXIC CHILDREN AND 22 OF
WILSON'S (1968) PRELIMINARY NORMATIVE GROUP
Dyslexic
Children
Wilson's
Normals
Median Total Words Corrected
6-1 to 8-0 years 72.5 55.0
8-1 to 10-0 years 70.0 62.5
10-1 to 12-0 years 66.5 182.0
12-1 to 14-0 years 64.0 204.0
14-1 to 16-0 years 55.0 115.5
Median Total Sentences
6-1 to 8-0 years 7.5 4.5
8-1 to 10-0 years 8.0 5.5
10-1 to 12-0 years 4.5 16.0
12-1 to 14-0 years 5.0 13.0
14-1 to 16-0 years 3.0 6.5
Median Total Words Per Sentence
6-1 to 8-0 years 9.7 11.9
8-1 to 10-0 years 9.3 12.3
10-1 to 12-0 years 13.0 11.5
12-1 to 14-0 years 14.2 19.3
14-1 to 16-0 years 18.3 17.9
200
150
□ Q Present study
-O Wilson study
100
50
14-1
16-0
12-1
14-0
10-1
12-0
8-1
10-0
6-1
8-0
Years
Fig. 1.— Median total words corrected produced by 30 dyslexic children on modi
fied Picture Story Language Test, compared with Wilson's preliminary norms.
U 1
60
Total Words Corrected, the six- to 10-year-old dyslexic
children scored somewhat higher than the normals. This is
difficult to interpret in view of the extremely small sam
ples in both studies. Within this general age span, Wil
son's subjects were somewhat younger than the dyslexic
children but it is doubtful that this age discrepancy could
account for the marked differences in scores. Comparative
Median Total Sentences for both groups are presented in
Table 4 and graphically displayed in Figure 2.
I
Total Words Per Sentence (TWS).— The TWS measure
showed a wider range of variation for the dyslexic children
than for the normals. The dyslexic group followed the ex
pected trend of increasing sentence length with increasing
age but used shorter sentences at three of five age levels
than did the normals. Comparative Median Total Words Per
Sentence for both groups appear in Table 4 and are graphi
cally displayed in Figure 3.
Syntax Quotients.— Syntax Quotients for the children
in this study are not comparable to those of Wilson's group
because of the modifications made of her scale. SQ's for
the dyslexic group ranged from 77 to 96. Using the age
breakdowns shown for other measures, median SQ's were as
20
10
o -
14-1
16-0
12-1
14-0
10-1
12-0
6-1
8-0
8-1
10-0
Years
Present study
O— — Wilson study
Fig. 2.— Median total sentences produced by 30 dyslexic children on modified
Picture Story Language Test, compared with Wilson's preliminary norms.
20
10
12-1 14-1 10-1 8-1 6-1
8-0 10-0 12-0
Years
14-0 16-0
Present study
— -OWilson study
Fig. 3.— Median total words per sentence produced by 30 dyslexic children on
Modified Picture Story Language Test, compared with Wilson's preliminary norms.
ro
follows:
Age Range Median SQ
6-1 to 8-0 years 89.0
8-1 to 10-0 years 90.0
10-1 to 12-0 years 85.6
12-1 to 14-0 years 86.0
14-1 to 16-0 years 93.0
The most frequent syntactic errors (45 per cent)
were errors of whole word addition. Sixty-six per cent of
these errors involved the over-use of the word "and" and
were made by 26 of the 30 children. Twenty-one per cent of
the errors involved whole word omissions, made by 18 of the
children. Nineteen per cent of the errors involved whole
word substitutions. Twenty of the children made this type
of error.
The Syntax Quotient was not significantly correlated
with other measures of language usage.
Language Performance Rating (LPR).— Over-a11 lan
guage performance was rated from taped samples of sponta
neous language on a seven-point equal-appearing-interval
scale. A rating of "one" indicated normal adult language
64
performance while a rating of "seven" indicated severe lan
guage impairment. This scale differed from the other per
formance rating scales in that a midpoint (four) was defined
with a description of language that might have been consid
ered normal for the two youngest children in the group.
Distribution of ratings was as follows:
Rating Frequency
1 0
2 0
3 3
4 21
5 6
The frequency with which a rating of "four" was
given suggests that the judges may have been unduly influ
enced by the midpoint definition. At the same time, it is
evident that the judges considered the language performance
of most of the children to be below the expected level. The
LPR was not significantly correlated with any of the other
language measures.
Clinical diagnoses
Clinical diagnoses were based on history, test per-
formance, and informal observations of the children during
65
the two-hour test sessions.
Articulation defects.— Twelve (40 per cent) of the
children demonstrated defective articulation in connected
speech. Of these, six were identified on the Templin-Darley
Tests of Articulation. Articulation Performance Ratings on
the seven-point APR Scale ranged from three to five for the
twe lve ch i ldre n.
I The habitual articulation patterns of four addi-
j
tional children were considered sub-standard, although spe
cific errors were inconsistent. Only 14 (47 per cent) of
the subjects demonstrated essentially normal articulation at
all times. Three of the children with normal articulation
had received extensive speech therapy in the past. Nine of
the children with articulation defects were receiving speech
therapy at the time they were seen by the investigator.
Therapy was recommended for two other children following
evaluation.
Fluency problems.— Three children demonstrated
residual stuttering problems. All of these presented his
tories of stuttering as well as articulation deficits. All
received a Fluency Performance Rating of "four" on the
seven-point FPR Scale.
66
Fifteen children received FPR ratings of "three,"
suggesting that the judges considered them somewhat less
fluent than average. Dysfluencies were not marked enough,
however, to be considered pathologic.
Cluttering.— More than half of the children demon
strated exaggerations of normal, physiologic errors of
speech which, by Bakwin and Bakwin's (1951) criterion, would
be considered cluttering. Such errors were consistent in
only two of the children, however. Both demonstrated tachy
lalia and paraphrasia as described by Arnold (1967). They
were the only two subjects given the clinical diagnosis of
cluttering.
Language deficits.— Clinical impressions of language
deficiencies were based on test performances, spontaneous
language samples, and informal conversation during the two-
hour test sessions. As a group, the children performed near
the average on tests of language comprehension but did
poorly on measures of language usage. This was especially
true of those language measures based on productivity: the
Verbal Expression subtest of the ITPA and the productivity
measures of the modified Picture Story Language Test. In
most cases the children were more talkative during relaxed,
67
give-and-take conversation periods when language demands
were less specific.
Three of the children were felt to have essentially
normal language. All of these had essentially normal
speech. Two of the children with normal language scored at
or above age level on the Verbal Expression subtest of the
ITPA. The third scored well below average on this subtest
but was able to express himself adequately on topics that
held specific interest for him.
Eight children used language that was immature or
substandard but not really deviant. Most of these young
sters came from lower socioeconomic backgrounds. They dis
played less social maturity than some of the other children.
Eleven of the children appeared to have unusually
limited language facility. They were the least verbal of
the children, often communicating in monosyllables. Their
sentences were short and simply constructed, but grammatical
errors were rare. Vocabularies were somewhat limited, al
though obvious word-finding difficulties were absent. One
of the children in this group presented a history of severe
language disability which was verified by means of previous
clinical records. His language was no longer characterized
by the bizarre syntactic errors documented in earlier
68
records. When seen as a part of this study, the problem
appeared as one of limited language facility rather than
specific disability.
Eight children demonstrated severe language diffi
culties . Their communication was marked by word-finding
difficulties, confusion of words expressing similar con
cepts, and atypical errors of grammar and syntax.
Clinical diagnoses are summarized in Table 5.
Comparison with previous studies.— Comparison with
previous studies of the incidence of speech defects among
dyslexic children is complicated by variation in terminology
and lack of information regarding diagnostic criteria.
Fourteen (47 per cent) of the children in the present study
demonstrated specific speech pathology, while three other
children had speech patterns that were considered substand
ard. Kagen (1943) and Hallgren (1950) reported a somewhat
lower incidence of speech deficits among their groups. On
the other hand, Kaiser (1956) considered all of his subjects
to be clutterers.
The children's generally poor performance on meas
ures of productivity in language usage is in close agreement
with previous findings about poor readers. Shire (1945)
69
TABLE 5
CLINICAL DIAGNOSES OF SPEECH AND LANGUAGE CHARACTERISTICS
OF 30 DYSLEXIC CHILDREN
Clinical Diagnosis
Number
of Cases
Percentage
of Total
Essentially normal speech and
language 3 10
Essentially normal speech with
immature or substandard language 3 10
Essentially normal speech with
I limited language facility 5 17
Essentially normal speech with
evidence of language disability 2 7
Substandard Articulation and
immature or substandard language 2 7
Substandard Articulation and
limited language facility 1 3
Articulation defect with immature
or substandard language 3 10
Articulation defect and limited
language facility 4 12
Articulation defect and evidence
of language disability 2 7
Articulation defect, stuttering, and
limited language facility 1 3
Stuttering and evidence of language
disability 2 7
Cluttering and evidence of language
disability 2 7
70
found her poor readers to be well below the normals in mean
length of response and total number of complete sentences.
DeHirsch et al. (1964, 1968) found that the total number of
words used in telling a story at kindergarten age was a good
predictor of future reading achievement. Kass (1961), how
ever, found that her reading disability cases performed in
a way that was "similar to normals" on the Vocal Encoding
(Verbal Expression) subtest of the ITPA. The children in
i
the present study achieved well below the expected level on
this subtest, with a median quotient of 73.0. More than
three-fourths of the present group scored below 85 per cent
of the expected level of performance.
Question 3
Question 3: Does analysis of ITPA profiles show
any consistent areas of strength or weakness?
For the purpose of this analysis, a quotient of 85
was considered the lower end of the normal range of per
formance. This procedure is in keeping with conventional
handling of IQ scores and provides an effective basis for
comparison.
Individual profiles of subtest quotients showed
diffuse areas of difficulty (as indicated by quotients below
71
85) in many cases. All but two of the children scored below
85 on at least one subtest; two children scored below 85 on
nine subtests. The median number of deficits (scores below
85) was four, with the majority (21) of the children showing
deficits in three to six areas. Range of subtest quotients,
median quotients, and number of children scoring below 85 or
above 100 are shown in Table 6.
Consistent areas of weakness
I
j There were five subtests on which a substantial
number of children did poorly;
Verbal Expression.— As previously reported, 23
children scored below 85 on this test. Median quotient was
73.0 and only three children scored at or above 100. Verbal
Expression appeared as the lowest quotient in the profiles
of 11 children.
Auditory Closure.— The median score on this test was
79, with 18 children scoring below 85. Only four children
scored at or above 100. Auditory Closure appeared as the
lowest quotient in the profiles of four children.
Auditory Sequencing.— Fourteen children scored below
85 in Auditory Sequencing, although the median score was
SUMMARY OF SCORES ON
TABLE
ALL SUBTESTS OF
6
THE ITPA BY 30 DYSLEXIC CHILDREN
ITPA Subtest
Range of
Scores3
Median
Score3
Percentage
Scoring
Below 85
Percentage
Scoring at
or Above 100
Auditory Reception 64-132 94.5 27 33
Visual Reception 61-148 100.0 20 53
Auditory Association 67-133 97.0 23 37
Visual Association 61-115 96.5 37 40
Verbal Expression 49-111 73.0 77 10
Manual Expression 55-131 104.5 13 73
Grammatic Closure 67-131 87.5 43 10
Visual Closure 60-126 100.5 17 63
Auditory Sequencing 44-118 91.0 47 40
Visual Sequencing 63-126 100.0 20 57
Auditory Closure 54-118 79.0 60 13
Sound Blending 60-145 102.0 27 63
aShown in quotients and based in part on extrapolated norms.
-j
ro
73
91.0 and 12 children scored at or above 100. Generally
speaking, children who obtained low scores on this subtest
did equally poorly on one or more additional subtests. The
ASQ was "tied" with some other subtest for lowest score in
the profiles of seven children.
Grammatic Closure.— Thirteen subjects scored below
85 on this subtest. Median score was 87.5. Although only
three children scored at or above 100, the Grammatic Closure
Quotient appeared as the lowest score in only one profile.
Visual Association.— Twelve subjects scored below 85
on this test. Median score was 96.5 and 12 children scored
at or above 100. Visual Association was the lowest score on
two profiles.
Individual ITPA profiles for all of the subjects are
to be found in Appendix D. A correlation matrix showing the
relationship of the various subtests to each other and to
the Wide Range Achievement Test in Reading (WRAT) appears in
Table 7.
Performance on ITPA sequencing
tasks
The subjects fell into four subgroups on the basis
of their performance on the ITPA sequencing tests.
TABLE 7
CORRELATION MATRIX: ITPA SUBTEST AND WRAT SCORES FOR 30 DYSLEXIC CHIIDREN
AR VR AA VA VE ME GC VC AS VS AC SB WRAT-Q
Auditory Reception -.108 .488b .476“ .406* .565b .464“ .314 .432* .253 .300 .002 .545b
Visual Reception .016 .264 -.056 -.026 -.095 -.180 -.154 .243 -.325 .132 -.159
Auditory Association .288 .389“ .498b .580b .195 .181 .129 .339 -.100 .442*
Visual Association .303 .313 .169 .146 .152 .051 .137 .137 .308
Verbal Expression .572b .733b .370 .139 -.087 .357 .245 .364
Manual Expression • 651b .241 .337 -.001 .309 .134 .360
Grammatic Closure .376* .342 .082 .325 .358 .475*
Visual Closure .084 -.001 .289 .111 .236
Auditory Sequencing .291 .063 .251 .354
Visual Sequencing .209 .073 .067
Auditory Closure -.036 .261
Sound Blending .153
WRAT Quotient -.037
“significant at the .05 level.
^Significant at the .01 level.
75
Group I.— Group I consisted of 14 children who
showed no evidence of significant difficulties in either
auditory or visual sequencing. Scores on the Auditory Se
quencing subtest ranged from 91 to 118 with a median score
of 105.5. Ten of the children in this subgroup had ASQ's of
100 or better. Visual Sequencing scores ranged from 99 to
126, with a median quotient of 104.5.
As a group, these children did somewhat better on
i
Jail ITPA subtests than did the other children. Two of the
children scored 85 or better on all subtests. The number of
subtest scores below 85 in individual profiles ranged from
zero to six; median number of deficits (subtest scores below
85) was three, compared to four for the entire group. Pat
terns of deficits varied widely among the individual pro
files .
Group II.— Group II consisted of 10 children who
scored below 85 on the Auditory Sequencing task but showed
no marked difficulties in Visual Sequencing. ASQ's ranged
from 44 to 77, with a median score of 66.0. Visual Sequenc
ing Quotients ranged from 91 to 107 with a median VSQ of
100. 0.
The number of subtest scores below 85 in individual
76
profiles ranged from three to nine. Median number of sub
test scores below 85 was five, compared to four for the en
tire sample and three for Group I. No consistent pattern of
deficits was shown in individual profiles, although nine of
the 10 children scored poorly in Auditory Closure.
Group III.— Group III consisted of two children who
scored below 85 on the Visual Sequencing subtest while doing
( well in Auditory Sequencing. Their VSQ's were 69 and 80,
I
I
compared to ASQ's of 109 and 105. One subject scored below
85 in nine areas, the other in six areas. Both children did
poorly in Auditory Reception, Auditory Association, Visual
Association, and Auditory Closure.
Group IV.— Group IV consisted of four children who
scored below 85 on both the Auditory and Visual Sequencing
subtests. ASQ's ranged from 61 to 70, with a median of 68.
VSQ’s ranged from 63 to 84, with a median of 78.5. All of
the children scored below 85 on four or five subtests.
Data on the ITPA subgroups are summarized in Table
8.
Comparison with previous
studies
Kass (1962) reported ITPA (Experimental Edition)
TABLE 8
ITPA SUBGROUPS BASED ON SEQUENCING SKILLS FOR 30 DYSLEXIC CHILDREN
Subgroup
Number
of Cases
Median
ASQ
Median
VSQ
Median Number of
Subtest Scores
Below 85
I. No sequencing problems 14 105.5 104.5 3.0
II. Auditory sequencing problems only 10 66.0 100.0 5.0
III. Visual sequencing problems only 2 107.0 74.5 7.5
IV. Visual and auditory sequencing problems 4 68.0 78.5 4.5
78
findings in terms of group means. Her reading disability
cases, as a group, did poorly on the Auditory-Vocal Associa
tion (Auditory Association), Auditory-Vocal Automatic (Gram
matic Closure), and Visual-Motor Sequencing (Visual Sequenc
ing) subtests, but were near or above the norm on other sub
tests .
Unlike Hass's group, the subjects in the present
study did most poorly on the Verbal Expression Subtest,
while performing at the average level on Visual Sequencing.
Like Hass's group, performance was poor on the Grammatic
Closure task.
Bateman (1963) reported that reading achievement
among partially seeing children was positively correlated
with performance on the automatic-sequential tasks of the
ITPA. In the present study, correlation was positive with
WRAT Quotients for both Auditory Sequencing (r = .354) and
Visual Sequencing (r = .067) but the correlations were not
statistically significant.
Question 4
Question 4: Can recognizable subgroups be differ
entiated on the basis of speech and language characteris
tics? On the basis of these plus WISC findings?
79
It proved possible to subgroup the children on the
basis of any one of several parameters. When other dimen
sions were added, however, subgroups proved not to be mutu
ally exclusive. The most promising approach to subgrouping
appeared to be that of clinical diagnoses of language ade
quacy. Four subgroups emerged when children were grouped by
these classifications.
Subgroups based on language
adequacy
Group A .--Group A consisted of three children who
were considered to have essentially normal language. All
had normal speech when seen; histories of communication
problems were negative. WISC Full Scale IQ's were all above
100; WRAT quotients ranged from 24 to 29 points below the
Full Scale IQ’s.
Two of the youngsters scored above 85 on all sub
tests of the ITPA. The third did poorly on three subtests.
This third child was observed by both the psychologist and
speech pathologist to demonstrate rather marked symptoms of
emotional disturbance vfcich, by history, antedated the read
ing difficulty.
Group B.--Group B consisted of eight children whose
80
language was immature or substandard but not deviant. Three
of the children had specific speech problems; two had sub
standard speech patterns. One was reported to have been
slow in talking; five were considered poor communicators by
their parents.
WISC Full Scale IQ's ranged from 88 to 113 with a
median IQ of 103.5. WRAT Quotients ranged from 15-34 points
below IQ; median WISC-WRAT discrepancy was 21.
i
All of the children in Group B scored somewhat below
average on the ITPA Verbal Expression subtest and well above
average on the test of Manual Expression. One child scored
above 85 on all subtests of the ITPA, but most of the chil
dren showed diffuse patterns of low scores.
Group B children seemed generally immature for their
chronological ages and came largely from lower socioeconomic
backgrounds. One child presented a history of juvenile de
linquency and was reported by the psychologist to show
sociopathic tendencies on projective tests.
Group C.— Group C consisted of 11 children who were
classified as having limited language facility. Six of the
children had specific speech problems vrtien seen. One child,
whose speech was normal at the time of evaluation, had a
81
history of a severe speech and language disability. Seven
members of this group were recognized by their parents as
being poor communicators.
WISC Full Scale IQ's ranged from 91 to 116, with a
median IQ of 94. Verbal IQ's ranged from five to 27 points
below Performance IQ's. WRAT Quotients ranged from 13 to 55
points below Full Scale IQ's. Median WISC-WRAT discrepancy
was 19.
All of the children in this group scored below aver
age on both the Grammatic and Auditory Closure subtests of
the ITPA.
Many of the children in Group C gave the impression
of being "slow learners" despite normal intelligence. They
needed repeated explanations during testing; their work,
while fairly adequate, was often painfully slow.
Group D.— Group D consisted of eight children with
deviant language, characterized by word-finding difficul
ties, word confusion, and atypical errors of grammar and
syntax. Seven of this group had specific speech problems;
five presented histories of slow speech and language devel
opment .
WISC Full Scale IQ's ranged from 96 to 119; median
82
IQ was 101. WRAT Quotients ranged from 12 to 35 points be
low IQ; median WISC-WRAT discrepancy was 22.5. All scored
below average on the Verbal Expression and Grammatic Closure
subtests of the ITPA.
As a group, these youngsters were much more outgoing
and verbal than those in Group C. One boy, like his adop
tive brother in Group B, had a history of juvenile delin
quency and, according to the psychologist's report, showed
sociopathic tendencies on projective tests.
Data on Language Subgroups are summarized in Table
9.
Neurologic findings
The children as a group were described by the ex
amining pediatric neurologist as being somewhat clumsy and
poorly coordinated. Specific positive neurologic findings
were few, however. Five of the children presented problems
of confused laterality. One child was described as having
generalized hypotonia. One child (Language Group C) showed
mild spasticity of the legs and arms, hyperactive reflexes,
and upgoing toe signs; his EEG showed a bi-parietal grade 2
dysrhythmia. This youngster had been followed at the Mayo
Clinic for a number of years because of a severe speech and
TABLE 9
SUBGROUPS OF DYSLEXIC CHILDREN BASED ON CLINICAL IMPRESSIONS OF LANGUAGE ADEQUACY
(N = 30)
Subgroup
Number
of Cases
Median Full
Scale NISC
Median
WISC-WRAT
Discrepancy
Other Characteristics
A. Essentially normal
language
3 122.0 27.0 Normal speech and language development history
Good performance on most ITPA subtests
a. Immature or substandard
language -
8 103.5 21.0 Variable speech and language development history
Specific speech problems common
Above average scores on Manual Expression (ITPA)
Below average scores on Verbal Expression (ITPA)
Social immaturity
Lower socioeconomic backgrounds
c. Limited language facility 11 94.0 19.0 Speech problems common
WISC Verbal IQ lower than Performance IQ
Below average scores on Grammatic Closure and
Auditory Closure (ITPA)
D. Deviant language . 8 101.0 22.5 All but one had specific speech problems
Slow speech and language development common
Below average scores on Verbal Expression and
Grammatic Closure (ITPA)
84
language disability which had responded well to therapy.
Eight of the children were reported to demonstrate dyscal-
culia as well as dyslexia.
There was no apparent relation between neurologic
findings and other parameters used in this study.
WISC findings
On the basis of WISC findings, the subjects fell
into three subgroups: (1) fourteen of the children obtained
Verbal and Performance Scale IQ's that were within 10 points
of each other, (2) one child had a verbal IQ 15 points
higher than Performance Scale IQ, and (3) Verbal scores for
15 children were 11 or more points below the Performance
score. Of these, six showed a discrepancy of 20 or more
points.
The WISC Verbal/Performance IQ discrepancy was
positively correlated with three other measures at the .01
level of confidence: PPVT (r = .591); Auditory Reception
(r = .489); Auditory Association (r = .600). Correlations
significant at the .05 level of confidence were: Manual
Expression (.422) and Grammatic Closure (r = .395).
All of the children in Language Subgroup C (Limited
Language Facility) had Verbal Scale IQ's that were somewhat
85
below Performance Scale IQ's. Other than this, the WISC
Verbal/Performance IQ discrepancy showed little relation to
parameters on which subgrouping could be based.
Comparison with previous
studies
In general, results of the present study are in
agreement with Rabinovitch's (1968) comment that, regardless
of what diagnostic correlations are attempted, the behavior
jor defect questioned is rarely found in more than 15 to 20
per cent of the dyslexic children studied.
WISC profiles for the present group followed the
patterns described by Wolfe (1967) in that some children did
about the same on both scales, some did better on the Per
formance Scale, and others did better on the Verbal Scale.
Although the Verbal/Performance discrepancies were
not so great as those reported by Kinsbourne and Warrington
(1966), the children in Language Group C (Limited Language
Facility) appear to be similar to those in Kinsbourne and
Warrington's Group I. They might also fit Bannatayne's de
scription of "pure dyslexics," representing the lower end of
the normal language facility continuum. The other children
in the present study do not fit so readily into the other
subgroups described by these writers.
CHAPTER V
SUMMARY AND CONCLUSIONS
Summary
The present study was designed to document and de
scribe speech and language characteristics of a group of
children with developmental dyslexia and to determine
whether meaningful subgroups could be defined on the basis
of these characteristics.
Thirty children ranging in age from six years 11
months to 14 years seven months who had been diagnosed as
developmental dyslexics by a pediatric neurologist were in
cluded in the study. WISC Full Scale IQ's ranged from 88
to 125. All of the children had a score of at least 90 on
either the Verbal or Performance Scale of the WISC. Reading
disabilities were classified as severe, with WISC-WRAT Quo
tient discrepancies ranging from 12 to 55 points.
Data were obtained by means of parent interviews and
standardized tests of articulation, language comprehension,
86
87
and psycholinguistic abilities. Non-standardized measures
of over-all articulation performance, fluency, and language
usage were also used.
Twenty-three per cent of the children were reported
to have been slow in talking, while 43 per cent had histo
ries of communication disorders. Upon evaluation, 47 per
cent of the children demonstrated specific speech problems
and an additional 10 per cent had speech patterns that were
considered substandard. Only 10 per cent of the children
were found to have essentially normal language. Twenty-
seven per cent of the group had immature or substandard lan
guage patterns. Thirty-six per cent demonstrated extremely
limited language facility, and 27 per cent had markedly
deviant language.
ITPA profiles showed diffuse patterns of deficits
for most of the children. As a group, they scored most
poorly on the Verbal Expression, Auditory Closure, and
Grammatic Closure subtests. In the case of each subtest,
however, a few children scored poorly while others achieved
above the expected level.
Clinical diagnoses of over-all language adequacy
appeared to be the most promising basis for the definition
of meaningful subgroups. Such subgroups were not mutually
88
exclusive, however, when other parameters were taken into
consideration.
Conclusions
On the basis of this study, it seems safe to con
clude that children with developmental dyslexia demonstrate
an unusually high incidence of speech defects and that most
have diffuse language deficits. This finding lends support
to the thesis that developmental dyslexia is a part of a
I
more pervasive language disorder affecting the acquisition
of communication skills. No single pattern of communication
deficits appears to characterize all dyslexic children, how
ever .
From the literature, it seems probable that common
patterns of disabilities are sensed intuitively by experi
enced clinicians. Such commonalities are yet to be opera
tionally defined, however, so that they can be surveyed and
measured objectively.
Limitations of the Study
The present study was based on a small sample of
heterogeneous children. It utilized some measures that are
frankly experimental in order to explore dimensions of
speech and language on which normative data are lacking.
89
ITPA data for more than half of the subjects are based on
extrapolated norms which are yet to be validated.
It must also be remembered that, although some
i
positive findings contribute to a diagnosis of developmental
I
dyslexia, such diagnosis is based primarily on negative
findings. The diagnosis of developmental dyslexia pre
supposes adequate motivation, adequate exposure to con
ventional teaching methods, and the absence of gross in
tellectual, neurologic, sensory, emotional, or environ
mental deficits. Few clinics have the facilities for sys
tematically assessing all of these variables. Thus, the
diagnosis may depend more on the orientation of the special
ists doing the evaluation than upon a specific pattern of
presenting problems. Not all of the children included in
this study would be considered developmental dyslexics by
all clinicians. Many of the children eliminated from the
study because of other diagnoses might have been considered
developmental dyslexics elsewhere.
In an attempt to define the sample so that it might
be duplicated more readily in another clinical setting, two
criteria were imposed in addition to a neurologic diagnosis
of developmental dyslexia: a score of at least 90 on either
the Verbal or Performance Scales of the WISC, and a WRAT
Quotient that was at least 10 points below the Full Scale
WISC IQ. Findings, however, must be viewed as purely de
scriptive. They may not be applicable to other groups of
dyslexic children who do not meet these additional criteria.
Suggestions for Further Research
The areas of speech and language appear to offer
fruitful avenues of approach to better understanding of
developmental dyslexia. There is considerable need for
basic, descriptive data, however, before hypotheses can be
formulated and tested.
1. Hard data on early speech and language develop
ment of children with developmental dyslexia are almost
totally lacking. It is doubtful that such information will
become available until extensive longitudinal studies of
speech and language development in presumably normal chil
dren are completed. Follow-up studies of children on whom
Well-Baby Clinic records exist might provide important pre
liminary data, however.
2. Evaluation of language deficits in dyslexic
children is hindered by lack of normative data. The gather
ing of normative data is delayed by a lack of adequate
91
measures. On the other hand, many of the language deficits
described by clinicians familiar with dyslexics are far too
subtle to be picked up on measures designed for normal chil
dren. More extensive study and description of such deficits
could lead to the development of appropriate measuring in
struments .
3. Most research studies of dyslexic children have
been directed toward the discovery of common factors. It
seems possible that many valuable data have been lost in
group means. Further study of the extent and type of varia
tion among dyslexic children might prove to be a shorter
route to the discovery of commonalities.
4. The question arises as to whether the marked
heterogeneity among dyslexic children is indigenous to the
disorder or an artifact of ambiguous criteria for diagnosis.
The incidence of developmental dyslexia is estimated at
roughly 2 per cent of the population. This means that re
search samples will be almost inevitably small in size.
There is a need for careful definition and description of
such samples in terms of objective criteria, so that result
ing data will be more readily available for comparison and
samples may be duplicated more readily in other research
settings.
92
APPENDIX A
SCORING PROCEDURES FOR THE MODIFIED PICTURE
STORY LANGUAGE TEST PRODUCTIVITY
AND SYNTAX SCALES
93
APPENDIX A
| SCORING PROCEDURES FOR THE MODIFIED PICTURE
STORY LANGUAGE TEST PRODUCTIVITY
AND SYNTAX SCALES
Productivity Scale
Total Words Corrected (TWC)
Wilson (1968) modified Myklebust's Total Words (TW)
to correct for dysfluencies that occur in speech. Rules for
handling repetitions are those outlined by Johnson, Darley,
and Spriestersbach (1963).
1. Any sound or combination of sounds that can
stand alone as an utterance is counted as a word if it is an
integral part of the sentence (e.g., "The lion is going to
grrr the monkey"). Interjections not considered dictionary
items and functioning solely to connect words or phrases
e.g., "er," "um," "uh,") should not be counted.
2. Hyphenated words are counted as one word.
3. All names and initials are counted as separate
words. .
4. Story titles and phrases such as "The end" and
94
95
"That's all" are not counted.
5. Contractions of the verb and negative (e.g.,
can't, won't, didn't) are counted as one word.
6. Contractions of the subject and verb are counted
as two words (e.g., he's, they're, you're).
7. Lists are defined as four or more words in a
series occurring within a sentence. No more than three
units should be counted, the first two words and the last
I
word in the series (e.g., "The boy is playing with blocks,
toys, [chairs, dolls, a mother, books] and some other
things.") The words within the brackets are not counted.
8. When the same word is repeated several times
consecutively (e.g., "He [he, he] went home") it should be
counted only once.
9. When a phrase is repeated (e.g., "he was [he
was] hurt") it should be counted only once. However, if at
least one different word is added when the phrase is re
peated (e.g., "And he hit— and he spanked him") all the
words in the repetition should be included in the count.
10. Repetitions used for enumerative purposes or for
starting a new thought unit should not be excluded. When it
cannot be determined, however, whether a repetition is used
for emphasis or constitutes a dysfluency, the repetition
96
should be excluded.
Total Sentences (TS)
The following criteria are used for determining sen-
I
jtences:
1. A response set off by pauses from the preceding
or succeeding remarks is considered a separate unit even if
it is not a grammatically complete sentence (McCarthy,
1930) .
2. When there is undue use of connectives, each
grammatically independent unit is counted as a separate sen
tence, with the division made before the unnecessary con
nective. If it is not possible to judge whether the utter
ance is intended as one sentence or two, score as one.
3. A diagonal bar is placed at the end of each
response-sentence and the bars are counted to determine the
number of sentences: e.g., "the boy has a dog / and he's
playing with dolls / and there are toys on the shelf."/
Count as three sentences. The following compound sentence,
in which the relationship between the parts is logical and
evident, would be scored as one sentence: "The boy has a
chair in his hand and he will give it to the girl."/ (Mykle-
bust, 1965) .
I 97
jwords Per Sentence (WPS)
This is computed by dividing the number of words
|(TWC) by the. number of sentences (TS) (Myklebust, 1965).
Syntax Scale
The Syntax Scale was designed as a measure of accu
racy within the standards of Informal English. It is con
cerned with conformity to correct, accepted usage. Common
colloquial expressions are acceptable within the stringency
levels of Informal English.
Myklebust's Syntax Scale is not directly concerned
with meaning or style. Intended meaning must often be de
duced, however, in order to determine correctness. Style,
if it deviates to the extent that it causes incorrect usage
of words or word endings, is indirectly scored. The child
is not penalized, however, for inferior style so long as he
is grammatically correct.
The Syntax Scale as modified by Wilson (1968) pro
vides for errors in whole word usage (penalized one point)
and word endings (penalized 1/3 point). Error categories
include addition, omission, or substitution. For the pur
pose of this study, an additional category, "Revision," is
also used. Rather than count all of the words of revised
98
phrases as "additions," a penalty (one point) is given for
the entire phrase.
Three arbitrary definitions of Stringency Level are
also used in this study.
1. The use of "and" or "and so" at the beginning of
a sentence is considered an error of word addition.
2. Over-use of qualifying expressions such as
"probably," "I guess," and "maybe" is considered a matter of
style so long as the words are used correctly within the
sentence and is not penalized.
3. Proper verb tense is determined on the basis of
the story as a whole. Following the rule of "least pen
alty, " the correct tense is that which is inferred most fre
quently. Deviations are scored as errors of substitution
even though their use within the sentence is grammatically
correct.
Scoring of the Syntax Scale
Errors in Word Usage— 1 error point
Type of Error Notation
Addition S '
Omission
Substitution — ■
99
Word Order
Revision C 3
Errors in Word Endings— 1/3 error point
Type of Error Notation
Addition - —
Omission
t
Substitution
Computing the Syntax
Quotient (SQ)
1. Count the number of words in the sentences (TWC)
2. Add Total Omissions (TO) to TWC to get Total
Unit (TU) .
3. Subtract Total Errors (TE) to get Total Correct
(TC). Total Omissions are included here as well as in TU.
4. Divide TC by TU.
5. Multiply by 100.
/ TWC + TO - TE .
6. Summary: ( ------------- ) 100 = SQ
TU
Representative Stories Told by
Dyslexic Children
Example 1
Well it's about^long time ago .j. . There is a man
100
shouting to three more men .J. . on a horse, [the man's on a
horse ^ The horse is . . . (is) rurfnlng . . ( * . mmmm . .
starting to run.j The man on the horse is holding a stick
. . . holding the reins too.I There's a man running with a
I \aa
shotgun. The houses are made out of wood./'* No leaves on the
trees but the moon's up. There is a couple of houses
That's a 11.^
Scoring
TWC: 76
TS: 9
TWS: 8.4
ERRORS:
•I
Omission
Addition
Substitution
Word Order
Whole Word
3 ___
Word Ending
Revision
Total
SYNTAX QUOTIENT: 92 per cent
Example 2
George Washington.1 He's going to ride .1 . . and
b i
there's a man coming out in . . . with his pajamas on/and
101
there's a man running to-^uh^--George Washington's horse^and
George Washington ^uhj . . . is taIking.j It is nighttime
. . . about^middle of the night. | One guy has a lantern.J I
think it's winter or fall.j
Scoring
TWC: 54
TS: 8
TWS: 6.8
ERRORS:
Whole Word Word Ending
Omission 3
Addition 4
Substitution 1
Word Order 0
Revision 0
Total ____8
SYNTAX QUOTIENT: 86 per cent
APPENDIX B
NORMATIVE EXTRAPOLATION CURVES FOR
SUBTESTS OF TOE ILLINOIS TEST OF
PSYCHOLINGUISTIC ABILITIES
102
ITPA— AUDITORY RECEPTION (AR)
50
40
25-122 months Norms
< D 20
Possible score: 50
10
48 24 36 84 72 96 108 120
Age In months
144 168 60 132 156 180
ITPA— VISUAL RECEPTION (VR)
50r
40
30
25-130 months Norms
20
Possible score: 40
10
24 36 48 60 72 84 96 108
Age in months
120 132 144 168 180 156
Numerical score
ITPA— AUDITORY ASSOCIATION (AA)
50
40
30
30-131 months Norms
20
Possible score: 42
10
24 48 36 84 60 72 96
Age in months
108 120 132 144 156 168 180
ITPA— VISUAL ASSOCIATION (VA)
50
40 “
• H
30-123 months Norms
Possible score: 42
10
24 36 48 60 84 72 96 108 120
Age in months
132 144 156 168 180
ITPA— VERBAL EXPRESSION (VE)
50
40
u 30
•rl
26-131 months Norms
Possible score 50
10
24 48 84 96 108
Age in months
132 144 168 36 60 72 120 156 180
Numerical score
ITPA— MANUAL EXPRESSION (ME)
50r
40
30
24-124 months Norms
20
Possible score: 42
10
24 144 168 180 48 84 96 108
Age in months
120 132 156 23 60 72
ITPA— GRAMMATIC CLOSURE (GC)
50
40
a 30
• H
20 Norms: 24-124 months
Possible score 33
10
24 36 48 84 60 72 96 108 120
Age in months
132 144 156 168 180
Numerical score
VISUAL CLOSURE (VC)
60
50
40
30
26-126 months Norms
Possible score 58
20
10
144 156 168 180 24 48 84 108 120 132 36 60 72 96
Age in months
AUDITORY SEQUENCING (AS)
60
50
40
24-123. months Norms
Possible score 56
20
10
144 168 180 84 132 156 24 48 96
Age in months
108 120 36 60 72
VISUAL SEQUENCING (VS)
50
40
I -I
• H
25-125 months Norms
20
Possible score 50
10
168 180 48 96 108 120 132
Age in months
144 156 24 84 26 60 72
to
AUDITORY CLOSURE (AC)
50
40
•H
28-120 months Norms 20
Possible score 30
10
168 180 108 144 156 84 120 132 24 48 60 72 96
Age in months
36
Numerical score
50
40
ITPA— SOUND BLENDING (SB)
28-103 months Norms
Possible score: 32
30
20
10
0
24 36 48 60 72 84 96 108 120
Age in months
132 144 156 168 180
114
APPENDIX C
MODIFIED PICTURE STORY LANGUAGE TEST DATA
FOR 30 DYSLEXIC CHILDREN COMPARED WITH
WILSON'S (1968) NORMATIVE GROUP
115
116
COMPARATIVE SAMPLES
Age Range
(years)
Dyslexic Children Normative Group
No.
Cases
Mean
Age
Median
Age
No.
Cases
Mean
Age
Median
Age
6-1 - 8-0 2 7-5 7-5 6 6- 9 6-11
8-1 - 10-0 12 9-2 9-5 4 9- 0 8-11
10-1 - 12-0 12 11-0 11-1 4 10-10 10-10
12-1 - 14-0 3 13-2 13-7 4 13- 4 13- 5
14-1 - 16-0 1 14-7 14-7 4 15- 1 15- 0
TOTAL WORDS CORRECTED
Age Range
(years)
Range
of TWC
Mean Median
Range
of TWC
Mean Median
6-1 - 8-0 66- 79 72.5 72.5 28- 83 56.3 55.0
8-1 - 10-0 54-119 76.4 70.0 38- 75 59.5 62.5
10-1 - 12-0 26-152 74.3 66.5 24-895 320.5 182.0
12-1 - 14-0 48- 71 61.0 64.0 99-375 220.5 204.0
14-1 - 16-0 55- 55.0 55.0 74-210 128.8 115.5
117
TOTAL SENTENCES
Age Range
(years)
Mean
Of TS
Median R*n?e Mean Median
of TS
6-1 - 8-0 7-8 7.5 7.5 3-7 4.8 4.5
8-1 - 10-0 5-13 8.0 8.0 3- 6 5.0 5.5
10-1 - 12-0 3-10 5.5 4.5 2-61 24.0 16.0
12-1 - 15-0 2- 5 4.0 5.0 4-24 13.5 13.0
14-1 - 16-0 3- 3.0 3.0 4-10 7.0 6.5
TOTAL WORDS PER SENTENCE
Age Range
(years)
Dyslexic Children Normative Group
Range
_ Mean
of TWS
Med. ^*1*1 Mean
of TWS
Med.
6-1 - 8-0 9.4- 9.9 9.7 9.7 9.3-13.2 11.4 11.9
8-1 - 10-0 5.2-18.8 10.1 9.3 10.6-12.7 12.0 12.3
10-1 - 12-0 8.3-25.3 13.9 13.0 10.8-14.7 12.1 11.5
12-1 - 14-0 12.8-24.0 17.0 14.2 12.7-24.7 19.0 19.3
14-1 - 16-0 18.3 18.3 18.3 12.5-26.2 18.6 17.9
APPENDIX D
ITPA PROFILES OF 30 DYSLEXIC CHILDREN SHOWN
IN QUOTIENTS AND BASED IN PART ON
EXTRAPOLATED NORMS
118
+»
o
®
•r>
A
AR VR AA VA VE ME GC VC AS VS AC SB
P
CO
1 86 95 111 80 70 127 101 116 115 105 98 60
2 128 85 133 106 105 131 131 116 110 99 92 108
3 94 80 88 97 76 125 88 85 67 83 88 104
4 118 130 126 81 92 124 116 126 91 125 118 103
5 64 97 67 88 66 101 75 96 61 91 92 81
6 114 85 98 114 80 119 81 114 118 113 78 122
7 103 117 103 115 83 97 95 118 73 91 65 145
8 109 116 99 110 83 111 88 121 112 105 107 123
9 73 110 84 61 61 72 74 80 109 69 54 138
10 107 103 93 76 73 109 90 104 109 110 104 136
11 94 111 97 105 111 106 94 108 94 100 103 133
12 82 83 69 78 106 106 92 101 105 80 81 101
13 103 61 111 91 84 105 90 107 70 74 74 96
14 92 89 89 81 86 104 89 106 77 105 68 76
15 91 107 87 102 82 103 84 104 65 103 79 114 119
AR VR AA VA VE ME GC VC AS VS AC SB
16 100 74 96 98 99 114 98 100 72 93 80 124
17 76 103 104 110 63 104 98 84 114 111 89 124
18 97 94 104 74 65 104 84 75 114 99 79 91
19 84 148 100 81 85 73 75 71 44 107 73 75
20 132 94 99 106 59 105 80 89 91 126 72 69
21 97 102 106 102 79 108 93 106 69 63 85 63
22 100 100 84 91 73 110 90 60 93 104 75 116
23 96 102 106 100 64 102 70 87 70 92 70 65
24 96 100 81 96 69 106 79 101 61 84 86 94
25 87 71 97 98 55 67 67 101 102 100 72 65
26 90 94 89 101 49 92 87 97 101 100 85 111
27 95 126 67 112 55 91 70 100 100 100 74 103
28 76 119 103 72 56 100 85 85 64 100 58 100
29 80 80 70 67 56 55 80 100 62 100 62 100
30 76 105 85 81 62 89 70 100 67 100 63 100
120
121
BIBLIOGRAPHY
Anderson, M., and Kelly, M. "An Inquiry into Traits Asso
ciated with Reading Disability." Smith College
Studies in Social Work. II (1931), 46-63.
Arnold, Godfrey E. "Signs and Symptoms." Studies in
Tachvphemia. New York: Speech Rehabilitation In
stitute, 1967.
Arthur, G. "An Attempt to Sort Children with Specific Read
ing Disability from Other Non-readers." Journal of
Applied Psychology. XI (1927), 251-263.
Artley, A. S. "A Study of Certain Factors Presumed to Be
Associated with Reading and Speech Difficulties."
Journal of Speech and Hearing Disorders. XIII
(1948), 351-360.
Bakwin, R. M., and Bakwin, H. "Cluttering." Journal of
Pediatrics. XL (1952), 393-395.
Bannatayne, Alex. "The Color Phonics System." The Disabled
Reader. Edited by J. Money. Baltimore: The Johns
Hopkins Press, 1966.
________ , and Wichiarajote, Penny. "Relationships between
Written Spelling, Motor Functioning and Sequencing
Skills." Journal of Learning Disabilities. II
(1969), 4-16.
Bateman, Barbara D. The Illinois Test of Psycholinquistic
Abilities in Current Research: Summaries of Stud
ies . Urbana, 111.: Institute for Research on Ex
ceptional Children, University of Illinois, 1965.
122
123
Bateman, Barbara. Reading and Psvcholinguistic Processes of
Partially Seeing Children. CEC Research Monograph
A5, 1963.
Bender, Lauretta. "Problems in Conceptualization and Com
munication in Children with Developmental Dyslexia."
Psychopathology of Communication. Edited by P. H.
Hoch and J. Zubin. New York: Grune and Stratton,
1958.
Clark, Ruth M. "General Language Disability: Use of Psy
chological Tests in Diagnosis." Studies in Tachv-
phemia. New York: Speech Rehabilitation Institute,
1967.
Cole, E. M., and Walker, Louise. "Reading and Speech Prob-
| lems as Expressions of Specific Language Disability."
j Bulletin of the Orton Society. XVI (1966), 55-73.
Critchley, Macdonald. Developmental Dyslexia. Springfield,
111.: Charles C. Thomas, 1964.
________ . "Topics Worthy of Research." Dyslexia: Diagno
sis and Treatment of Reading Disorders. Edited by
A. Keeney and V. Keeney. St. Louis, Mo.: C. V.
Mosby Co., 1968.
Darley, F. L., and Moll, K. I. "Reliability of Language
Measures and Size of Language Sample." Journal of
Speech and Hearing Research. Ill (1960), 166-173.
parley, F. L., and Winitz, Harris. "Age of First Word:
! Review of Research." Journal of Speech and Hearing
Disorders. XXVI (1961), 272-290.
DeHirsch, Katrina. "Diagnoses of Developmental Language
Disorders." Studies in Tachvphemia. New York:
Speech Rehabilitation Institute, 1967.
________ . "Discussion." Dyslexia: Diagnosis and Treatment
of Reading Disorders. Edited by A. Keeney and V.
Keeney. St. Louis, Mo.: C. V. Mosby Co., 1968.
124
beHirsch, K., and Jansky, J. "Kindergarten Protocols of
I High Achievers, Slow Starters and Failing Readers."
1 Dyslexia: Diagnosis and Treatment of Reading Dis
orders . Edited by A. Keeney and V. Keeney. St.
Louis, Mo.: C. V. Mosby Co., 1968.
________; Jansky, J.; and Langford, W. S. "Early Prediction
of Reading Failure." Bulletin of the Orton Society.
XVI (1966).
________. "The Oral Language Performance of Premature Chil
dren and Controls." Journal of Speech and Hearing
Disorders. XXIX (1964), 60-69.
________ . Predicting Reading Failure: A Preliminary Study.
New York: Harper and Row, 1966.
Dunn, L. M. Peabody Picture Vocabulary Test. Minneapolis,
Minn.: American Guidance Service, 1959.
Eisenberg, Leon. "The Epidemiology of Reading Retardation
and a Program for Preventive Intervention." The
Disabled Reader. Edited by John Money. Baltimore:
The Johns Hopkins Press, 1966.
Ellinson, Careth. The Shadow Children. Chicago: Topaz
Books, 1967.
Foster, Suzanne. "Language Skills for Children with Per
sistent Articulatory Disorders." Unpublished doc
toral dissertation. Texas Women's University, 1963.
Gaines, F. P. "Interrelationship of Speech and Reading
Disabilities." Quarterly Journal of Speech. XXVII
(1941), 104-110.
Gibbons, H. "The Relation of Reading Ability to Skill in
Articulation." Proceedings of the American Speech
Correction Association. IV (1934), 7-12.
Hallgren, Bertil. "Specific Dyslexia (Congenital Word
Blindness." Acta Psvchiatrica et Neurologica.
Suppl. 65 (1950).
125
(Herman, Knud. Reading Disability: A Medical Study of Word
| Blindness and Related Handicaps. Springfield, 111.:
! Charles C. Thomas, 1959.
i
Ingram, T. T. "The Association of Speech Retardation and
Educational Difficulties." Proceedings of the Royal
Society of Medicine, LVI (1963), 199-203.
Irwin, Ruth Becky. "Speech Therapy and Children's Linguis
tic Skills." Journal of Speech and Hearing Re
search, V (1962), 377-381.
Jackson, J. A. "A Survey of Psychological, Social and En
vironmental Differences between Advanced and Re
tarded Readers." Journal of Genetic Psychology. IXV
(1944), 113-131.
I
jjastak, J. F., and Jastak, S. R. The Wide Range Achievement
Test. Wilmington, Del.: Guidance Associates, 1965.
Johnson, D. J., and Myklebust, H. R. Learning Disabilities.
New York: Grune and Stratton, 1967.
Johnson, W.; Darley, F. L.; and Spriestersbach, D. C. Diag
nostic Methods in Speech Pathology. New York: Har
per and Row, 1963.
Jones, M. V. "The Effect of Speech Training on Silent Read
ing Achievement." Journal of Speech and Hearing
Disorders. XVI (1951), 258-263.
Joos, L. W. "Linguistics for the Dyslexic." The Disabled
Reader. Edited by John Money. Baltimore: The
Johns Hopkins Press, 1966.
Kagen, B. "Om ordblindhet." Pedagog. Skrifter. LX (1943),
179-180. Cited by Hallgren, B., "Specific Dyslexia
(Congenital Word Blindness)," Acta Psvchiatrica et
Neurologica. Suppl. 65 (1950).
Kaiser, L. "Dyslectische Kinderen." Ned. Tydschr. Geneesk. ,
C (1956), 756. Cited by Arnold, G. E., Studies in
Tachvphemia. 1967.
126
Kass, Corrine E. "Some Psychological Correlates of Severe
Reading Disability." Unpublished doctoral disserta
tion. University of Illinois, 1962.
i
Kinsbourne, Marcel, and Warrington, Elizabeth K. "Develop
mental Factors in Reading and Writing Backwardness."
The Disabled Reader. Edited by John Money. Balti
more: The Johns Hopkins Press, 1966.
Kirk, S. A.; McCarthy, J. J.; and Kirk, W. Examiner's Man
ual: Illinois Test of Psycholinguistic Abilities.
Rev. ed. Urbana, 111.: University of Illinois
i Press, 1968.
i
Lee, Laura. "Developmental Sentence Types: A Method for
Comparing Normal and Deviant Syntactic Development."
Journal of Speech and Hearing Disorders. XXXI
(1966), 311-330.
i
^McCarthy, Dorothea. Language Development of the Pre-school
Child. Child Welfare Monographs, No. 4. Minneapo
lis, Minn.: University of Minnesota Press, 1930.
McCarthy, J. J., and Olson, J. L. Validity Studies on the
Illinois Test of Psycholinguistic Abilities. Madi
son, Wis.: Photo Press, Inc., Xerlith Service,
1964.
McLeod, John. "Reading Expectancy for Disabled Learners."
Journal of Learning Disorders. I (1968), 97-105.
Marge, M. "A Factor Analysis of Oral Communication Skills
in Older Children." Journal of Speech and Hearing
Research, VII (1964), 31-46.
Menyuk, P. "Comparison of Grammar of Children with Func
tionally Deviant and Normal Speech." Journal of
Speech and Hearing Disorders. VII (1964), 109-120.
Minifie, F. D.; Darley, F. L.; and Sherman, D. "Temporal
Reliability of Seven Language Measures." Journal of
Speech and Hearing Disorders. VI (1963), 139-149.
127
Money, John. "On Learning and Not Learning to Read." The
I Disabled Reader. Edited by John Money. Baltimore:
The Johns Hopkins Press, 1966.
I
iMonroe, Marion. Children Who Cannot Read. Chicago: Uni
versity of Chicago Press, 1932.
(Moore, C. E. "Reading and Arithmetic Abilities Associated
with Speech Defects." Journal of Speech Disorders.
XII (1947), 85-86.
Moss, M. A. "The Effect of Speech Defects on Second Grade
Reading Achievement." Quarterly Journal of Speech.
| XXIV (1938), 642-654.
I
Myklebust, Helmer R. Development and Disorders of Written
' Language. VI: Picture Story Language Test. New
York: Grune and Stratton, 1965.
I
Olson, J. L. "A Comparison of Receptive Aphasic, Expressive
Aphasic, and Deaf Children on the Illinois Test of
Abilities." Unpublished doctoral dissertation.
University of Illinois, 1960.
Orton, Samuel Tory. "Familial Occurrence of Disorders in
Acquisition of Language." Eugenics. Ill, No. 4
(1930), 139-146.
________ . "The Neurologic Basis of Elementary Education."
Arch. Neurol, and Psvchiat.. XXI (1929), 641-646.
I
, ________. "A Physiological Theory of Reading Disabilities
and Stuttering in Children." New England Journal of
I Medicine. CXCIX (1928), 1046-1052.
________ . Reading. Writing and Speech Problems in Children.
New York: W. W. Norton and Co., 1937.
________ . "Special Disability in Spelling." Bulletin of
the Neurological Institute of New York. I, No. 2
(1931), 167-200.
________ . "Specific Reading Disability— Strephosymbolia."
Journal of the American Medical Association. XC
(1928), 1095-1099.
128
Orton, Samuel Tory. "Word Blindness in School Children."
Arch. Neurol. and Psychiat.. IV (1925), 581-615.
i
|Paraskevopoulos, J. N., and Kirk, S. A. The Development and
Psychometric Characteristics of the Revised Illinois
Test of Psycholinguistic Abilities. Urbana, 111.:
University of Illinois Press, 1969.
Rabinovitch, Ralph D. "Dyslexia: Psychiatric Considera
tions ." Progress and Research Needs in Dyslexia.
Edited by John Money. Baltimore: The Johns Hopkins
Press, 1962.
________ . "Reading Problems in Children: Definitions and
Classifications." Dyslexia: Diagnosis and Treat
ment of Reading Disorders. Edited by A. Keeney and
V. Keeney. St. Louis, Mo.: C. V. Mosby Co., 1968.
Ragland, G. G. "The Performance of Educable Mentally Handi
capped Students of Different Reading Ability on the
ITPA." Unpublished doctoral.dissertation. Univer
sity of Virginia, 1964.
Saunders, R. E. "Dyslexia: Its Phenomenology." Progress
and Research Needs in Dyslexia. Edited by John
Money. Baltimore: The Johns Hopkins Press, 1962.
Shepherd, George. "Phonetic Description of Cluttered
Speech." Studies in Tachvphemia. II. New York:
Speech Rehabilitation Institute, 1967.
Shire, M. L. Sr. "The Relation of Certain Linguistic Fac-
! tors to Reading Achievement in First Grade Chil-
| dren." Unpublished doctoral dissertation. Fordham
| University, 1945.
Shriner, Thomas. "A Comparison of Selected Measures with
Psychological Scale Values of Language Development."
Journal of Speech and Hearing Research. X (1967),
828-835.
________ . "A Review of Mean Length of Response as a Measure
of Expressive Language Development in Children."
Journal of Speech and Hearing Disorders. XXXIV
(1969), 61-67.
129
Shriner, Thomas, and Sherman, Dorothy. "An Equation for
Assessing Language Development." Journal of Speech
! and Hearing Research. X (1967), 41-48.
Sommers, R. K.; Cockerville, C. E.; Paul, C. D.; Bowser,
D. C.; Fichter, G. E.; Fenton, A. K.; and Copetas,
F. G. "Effects of Speech Therapy and Speech Im
provement upon Articulation and Reading." Journal
of Speech and Hearing Disorders. XXVI (1961), 27-38.
Sutton, Peggy. "The Relationship of Visualizing Ability to
Reading." Unpublished master's thesis. University
of Illinois, 1963.
iTemplin, Mildred C. Certain Language Skills in Children.
! Minneapolis, Minn.: University of Minnesota Press,
; 1957.
Thompson, Lloyd J. Reading Disability; Developmental Dys
lexia . Springfield, 111.: Charles C. Thomas, 1966.
Van Demark, A. A., and Mann, M. B. "Oral Language Skills of
Children with Defective Articulation." Journal of
Speech and Hearing Research, VI (1965), 409-413.
Weaver, C. H.j Furbee, C.; and Everhart, R. W. "Articula
tion Competency and Reading Readiness." Journal of
Speech and Hearing Research. Ill (1960), 174-180.
Weiss, Deso A. Cluttering. Englewood Cliffs, N. J.:
Prentice-Ha11, Inc., 1964.
Wilson, M. E. H. "A Standardized Method of Obtaining a
Spoken Language Sample." Unpublished master's the
sis. University of Nebraska, 1968.
Wolf, Clifton W. "An Experimental Investigation of Specific
Language Disability." Bulletin of the Orton Society.
XVII (1967), 32-39.
Wong, H. "An Appraisal of Investigations Dealing with the
Relationship of Speech Defects to Reading Disabili
ties." Unpublished master's thesis. Stanford Uni
versity, 1944.
130
jtfedinack, Jeanette S. "A Study of the Linguistic Function
ing of Children with Articulation and Reading Disa
bilities." Journal of Genetic Psychology. LXXIV i
(1949), 23-59.
I
Zedler, Empress Young. "Effect of Phonic Training on Speech
Sound Discrimination and Spelling Performance."
Journal of Speech and Hearing Disorders. XXI (1956),
245-250.

Asset Metadata

Creator Vent, Faythe (author)

Core Title Speech And Language Characteristics Of Children With Developmental Dyslexia

Contributor Digitized by ProQuest (provenance)

Degree Doctor of Philosophy

Degree Program Communicative Disorders

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag health sciences, speech pathology,OAI-PMH Harvest

Format dissertations (aat)

Language English

Advisor Perkins, William H. (committee chair), [Hedernick], Robert P. (committee member), Darley, Frederic L. (committee member), Haney, Russell (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-462504

Unique identifier UC11362277

Identifier 7112420.pdf (filename),usctheses-c18-462504 (legacy record id)

Legacy Identifier 7112420.pdf

Dmrecord 462504

Document Type Dissertation

Format dissertations (aat)

Rights Vent, Faythe

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA