• TABLE OF CONTENTS
HIDE
 Main
 Acknowledgement
 Table of Contents
 Abstract
 Introduction to the study
 Review of the related literatu...
 The method and design of the...
 Results
 Summary
 Broad occupational groups utilized...
 Letter to parents, instruction...
 Sentences constructed for use with...
 Bibliography
 Biographical sketch






Title: Psycholinguistic abilities and academic achievement of hard of hearing students
CITATION PDF VIEWER THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00099535/00001
 Material Information
Title: Psycholinguistic abilities and academic achievement of hard of hearing students
Physical Description: viii, 140 leaves. : illus. ; 28 cm.
Language: English
Creator: Anderson, Martha Williamson, 1936-
Publication Date: 1974
Copyright Date: 1974
 Subjects
Subject: Intelligence levels -- Deaf   ( lcsh )
Deaf children   ( lcsh )
Speech thesis Ph. D   ( lcsh )
Dissertations, Academic -- Speech -- UF   ( lcsh )
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
 Notes
Abstract: The purposes of this study were; (1) to describe selected characteristics of a group of hard of hearing school-age subjects and (2) to determine possible relationships between their academic performance and the following variables: sex, socioeconomic status, characteristics of hearing aid use, special training, severity of the hearing loss, performance intelligence, language comprehension and expression, and speechreading skill. The study involved 26 subjects with bilateral, sensorineural hearing impairments, greater than 30 dB Re: ANSI 1969, who attended regular school classes, grades 4 through 12, in Duval County, Florida, during the 1973-74 academic year. Only subjects with no discernible additional handicaps were included in the study. Data were collected from questionnaires returned by parents of the subjects, from cumulative school records, from files maintained in the office of the school audiologist, and from tests administered by the investigator. The seven tests administered by the investigator were: 1. Performance sections of the Wechsler Intelligence Scale for Children and the Wechsler Adult Intelligence Scale 2. Peabody Picture Vocabulary Test 3. Test for Comprehension of Indirect/Direct Object Constituents 4. Test of Auditory Language 5. Screening Deep Test of Articulation 6. Picture Story Language Test 7. Utley Lipreading Test, Sentence Form The mean age of subjects in the study was 14.5 years. Eighteen of the subjects consistently wore hearing aids. The mean hearing loss of the group of subjects was 61 dB with a range of losses from 30 dB to 88 dB. The mean performance IQ of the subjects was 100.3. Academic performance of hard of hearing subjects was measured by the average of the national percentile scores for all subtests of the Stanford Achievement Test. The mean national percentile score for the hard of hearing subjects was 16.15. The hard of hearing subjects performed at a level substantially below expectation for their age level on all language measures. The ability to write grammatically, measured by Syntax Quotient of the Picture Story Language Test, was the variable most highly correlated with grade point average of the subjects. No significant differences were demonstrated in the mean academic performance of subjects categorized as to sex, socioeconomic status, or characteristics of hearing aid use. The severity of the hearing loss was not significantly correlated with academic performance or with any of the language skills tested. Speechreading was not significantly correlated with academic performance. Receptive vocabulary (Peabody Picture Vocabulary Test), Auditory Language (Test for Auditory Language), Written Language (Picture Story Language Test), and Syntactic Comprehension (Test for Comprehension of Indirect/Direct Object Constituents) were significantly correlated with academic performance. A three-variable regression model was developed for prediction of the criterion variable, academic performance. This model, including receptive vocabulary, syntactic comprehension, and performance intelligence, could account for 52 percent of the variance of the criterion variable, academic performance. The regression equation was significant at the 0.001 level of confidence.
Thesis: Thesis -- University of Florida.
Bibliography: Bibliography: leaves 134-139.
General Note: Typescript.
General Note: Vita.
General Note: A dissertation presented to the graduate council of the University of Florida in partial fulfillment of the requirements for the degree of Doctor Philosophy
 Record Information
Bibliographic ID: UF00099535
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: alephbibnum - 000580530
oclc - 14046259
notis - ADA8635

Downloads

This item has the following downloads:

PDF ( 5 MBs ) ( PDF )


Table of Contents
    Main
        Page i
        Page ii
    Acknowledgement
        Page iii
    Table of Contents
        Page iv
        Page v
    Abstract
        Page vi
        Page vii
        Page viii
    Introduction to the study
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
    Review of the related literature
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
    The method and design of the study
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
    Results
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
        Page 100
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
        Page 111
    Summary
        Page 112
        Page 113
        Page 114
        Page 115
        Page 116
        Page 117
        Page 118
        Page 119
        Page 120
        Page 121
        Page 122
        Page 123
        Page 124
        Page 125
    Broad occupational groups utilized by the United States Bureau of the Census
        Page 126
    Letter to parents, instruction sheet, permission slip and questionnaire
        Page 127
        Page 128
        Page 129
        Page 130
    Sentences constructed for use with the screening deep test of articulation
        Page 131
        Page 132
        Page 133
    Bibliography
        Page 134
        Page 135
        Page 136
        Page 137
        Page 138
        Page 139
    Biographical sketch
        Page 140
        Page 141
        Page 142
        Page 143
Full Text














PSYCHOLINGUISTIC ABILITIES AND ACADEMIC ACHIEVEMENT
OF HARD OF HEARING STUDENTS
















By
MARTHA WILLIAMSON ANDERSON























A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL
OF THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY





UNIVERSITY OF FLORIDA


1974


































DEDICATED TO THE MEMORY OF

ANN WILLIAMSON















ACKNOWLEDGEMENTS


The writer wishes to express appreciation to the Supervisory

Committee members, Dr. E.C. Hutchinson, Dr. Norman Markel, Dr. R.J.

Scholes, and Dr. Dan Sellers, and especially to the Chairman, Dr.

Thomas B. Abbott, for support and guidance during the preparation of

this dissertation.

Special thanks go to Mrs. Julia Wickersham, Director of Excep-

tional Child Education; Mrs. Eunice Carter, Coordinator, Language,

Speech, and Hearing Services; and Mrs. Sandra Village, Audiologist,

all of the Duval County Schools, Jacksonville, Florida, for their

assistance and encouragement in this research. The writer is especially

grateful to the hard of hearing students in Duval County, Florida,

who cooperated willingly in this endeavor.

Mrs. Gene DeLoach, graduate student in Speech Pathology at the

University of Florida, was of inestimable aid in helping to organize

the study. Mrs. Elizabeth Etheridge gave able assistance in proofread-

ing the manuscript.

Perhaps the most significant contributions toward the completion

of this project were made by the family of the writer. Her children,

Donny and Andrea, remained cheerful under sometimes trying circumstances,

and her husband, Gary, understood and remained supportive.

















TABLE OF CONTENTS


ACKNOWLEDGEMENTS . . . . . . . . ... ...... iii

ABSTRACT . . . . . . . . ... . . . . . vi

CHAPTER PAGE

I. INTRODUCTION TO THE STUDY. . . . . . . . .. 1

Statement of the Problem . . . . . . . . . 1

Purposes of the Study . . . . . . . .. . 2

Justification for the Study . . . . . . . .. 2

Scope and Limitations of the Study . . . . . . 3

Definition of Terms . . . . . . . .. ... 3

Theoretical Development . . . . . . . . . 8

Summary . . . . . . . . . . . . . 11

II. REVIEW OF THE RELATED LITERATURE . . . . . . .. .12

Introduction . . . . . . . . ... . . 12

Characteristics of Hard of Hearing School Children . .. 12

Verbal Ability of Hard of Hearing School Children . . 17

Academic Performance of Hard of Hearing Children ... . 22

Factors That May Influence the Academic Performance

of Hard of Hearing Children . . . . . ... 28

Summary . . . . . . . . . . . . . 31











CHAPTER PAGE

III. THE METHOD AND DESIGN OF THE STUDY . . . . .. 32

Introduction . . . . . . . . ... . . 32

Procedures Used in Data Collection . . . . .. 32

Experimental Design of the Research . . . . .. 35

Data Collection Instruments . . . . . . .. 40

Summary . . . . . . . . ... . . . 53

IV. RESULTS . . . . . . . . ... . . . 55

Introduction . . . . . . . . . . . 55

Findings Related to the First Purpose of the Study . 56

Findings Related to the Second Purpose of the Study . 96

Summary . . . . . . . . ... . . . 111

V. SUMMARY AND DISCUSSION . . . . . . . .. 112

Summary . . . . . . . . ... . . . 112

Discussion . . . . . . . . ... . . 120

APPENDIX A BROAD OCCUPATIONAL GROUPS UTILIZED BY THE

UNITED STATES BUREAU OF THE CENSUS . . . . .. 126

APPENDIX B LETTER TO PARENTS, INSTRUCTION SHEET, PERMISSION

SLIP, AND QUESTIONNAIRE . . . . . . ... 127

APPENDIX C SENTENCES CONSTRUCTED FOR USE WITH THE

SCREENING DEEP TEST OF ARTICULATION . . . . .. 131

APPENDIX D INTERCORRELATION MATRIX OF NUMERIC VARIABLES . 132

BIBLIOGRAPHY . . . . . . . . .. . . . . 134

BIOGRAPHICAL SKETCH . . . . . . . .... ... 140











Abstract of Dissertation Presented to the
Graduate Council of the University of Florida in
Partial Fulfillment of the Requirements
for the Degree of Doctor of Philosophy


PSYCHOLINGUISTIC ABILITIES AND ACADEMIC ACHIEVEMENT
OF HARD OF HEARING STUDENTS

By

Martha Williamson Anderson

December, 1974

Chairman: T.B. Abbott
Major Department: Speech

The purposes of this study were; (1) to describe selected

characteristics of a group of hard of hearing school-age subjects and

(2) to determine possible relationships between their academic per-

formance and the following variables: sex, socioeconomic status,

characteristics of hearing aid use, special training, severity of

the hearing loss, performance intelligence, language comprehension and

expression, and speechreading skill.

The study involved 26 subjects with bilateral, sensorineural

hearing impairments, greater than 30 dB Re: ANSI 1969, who attended

regular school classes, grades 4 through 12, in Duval County, Florida,

during the 1973-74 academic year. Only subjects with no discernible

additional handicaps were included in the study.

Data were collected from questionnaires returned by parents of

the subjects, from cumulative school records, from files maintained in

the office of the school audiologist, and from tests administered by

the investigator. The seven tests administered by the investigator

were:










1. Performance sections of the Wechsler Intelligence
Scale for Children and the Wechsler Adult
Intelligence Scale

2. Peabody Picture Vocabulary Test

3. Test for Comprehension of Indirect/Direct Object
Constituents

4. Test of Auditory Language

5. Screening Deep Test of Articulation

6. Picture Story Language Test

7. Utley Lipreading Test, Sentence Form


The mean age of subjects in the study was 14.5 years. Eighteen

of the subjects consistently wore hearing aids. The mean hearing loss

of the group of subjects was 61 dB with a range of losses from 30 dB

to 88 dB. The mean performance IQ of the subjects was 100.3.

Academic performance of hard of hearing subjects was measured by

the average of the national percentile scores for all subtests of the

Stanford Achievement Test. The mean national percentile score for the

hard of hearing subjects was 16.15.

The hard of hearing subjects performed at a level substantially

below expectation for their age level on all language measures. The

ability to write grammatically, measured by Syntax Quotient of the

Picture Story Language Test, was the variable most highly correlated

with grade point average of the subjects. No significant differences

were demonstrated in the mean academic performance of subjects cate-

gorized as to sex, socioeconomic status, or characteristics of hearing

aid use. The severity of the hearing loss was not significantly corre-

lated with academic performance or with any of the language skills tested.

Speechreading was not significantly correlated with academic performance.










Receptive vocabulary (Peabody Picture Vocabulary Test), Auditory

Language (Test for Auditory Language), Written Language (Picture Story

Language Test), and Syntactic Comprehension (Test for Comprehension of

Indirect/Direct Object Constituents) were significantly correlated

with academic performance.

A three-variable regression model was developed for prediction

of the criterion variable, academic performance. This model, including

receptive vocabulary, syntactic comprehension, and performance intel-

ligence, could account for 52 percent of the variance of the criterion

variable, academic performance. The regression equation was signifi-

cant at the 0.001 level of confidence.
















CHAPTER I
INTRODUCTION TO THE STUDY

Statement of the Problem


The acquisition of a native language is dependent upon lin-

guistic experience, and this experience is initially acquired through

audition (Hebb et al., 1973). The effect of profound, congenital

auditory deprivation on language development is obvious to those who

have been associated with deaf children. A number of studies are con-

cerned with linguistic analysis of oral and written communication of

the deaf and profoundly hard of hearing (Walter, 1955; Myklebust, 1964,

1965; Brannon and Murry, 1966; Power and Quigley, 1973; Pressnell, 1973).

The effect of a mild or a moderate congenital hearing impairment

on the language behavior and academic performance of students is less

well documented in published research. Moores (1972) noted that "research

on this type child and his special problems is almost non-existent"

(pp. 162,163). The Subcommittee on Human Communication and Its Disorders,

National Advisory Neurological Diseases and Stroke Council (1969)

stated that:


In summary, we find ourselves in the situation
of having some types of research knowledge on deaf
children but of lacking not only other types of knowledge
on the deaf, but also almost any research information
on either hard of hearing children or acoustically
impaired adults. (p. 67)










Purposes of the Study


This study investigated pertinent background information,

audiologic data, language characteristics, intellectual ability, speech-

reading skill, and academic performance of hard of hearing students

attending regular school classes.

Two specific purposes of the study were: (1) to describe

selected characteristics of this group of hard of hearing school-age

subjects and (2) to determine possible relationships between their aca-

demic performance and the following variables: sex, socioeconomic

status, characteristics of hearing aid use, special training, severity

of the hearing loss, performance intelligence, language comprehension

and expression,and speechreading skill.


Justification for the Study


The justification for this study was derived from the need to

document more fully the nature of deficits that may be imposed by a

significant, though not profound, sensorineural hearing impairment.

Determination of the relative importance of potentially significant

variables in predicting academic performance of these students should

be useful in planning appropriate habilitative programs for hard of hear-

ing students.

Although programs to assist hard of hearing school children have

been in existence for many years, the Subcommittee on Human Communication

and Its Disorders (1969) suggested that the educational management of

these children has evolved primarily from tradition. This Subcommittee











stated that:


We are in a research desert the moment we leave
the realm either of finding hard of hearing children
or of quantifying their hearing deficits.

Strong but conflicting traditions as to how to
manage hard of hearing children educationally and socially
have evolved 1) from "common sense," 2) from extrapo-
lation of the experiences of hard of hearing adults
and 3) from implications of research on psychological and
linguistic performance of normal hearing children. . .
On the whole, these traditions lack either confirmation or
refutation through research. (p. 72)


Scope and Limitations of the Study


This study was limited to those students enrolled in the schools

of Duval County, Florida, during 1973-74 who: (1) had bilateral,

sensorineural hearing impairments greater than 30 dB in the better ear

in the speech frequencies 500 Hz to 2000 Hz; (2) were enrolled in the

4th through the 12th grades in school; (3) attended regular school

classes; and (4) had no discernible additional handicap other than a loss

of hearing.


Definition of Terms


Definitions of terms and variables used in the study are

listed below.


Average Loss of Hearing

Unless otherwise specified, hearing levels quoted in this

study represent the average pure-tone audiometric threshold levels

for 500 Hz, 1000 Hz, and 2000 Hz for the better ear. Averages are

given as decibel (dB) levels Re: American National Standard Specification










for Audiometers (ANSI) 1969.


Normal Hearing

The zone of normal hearing is considered to extend from 0 to

25 dB, as Davis (1970) proposed.


Hard of Hearing

The handicap that is the result of auditory deprivation cannot

be judged on the basis of audiometric tests alone (Johnson, 1973).

However, it is necessary to specify as carefully as possible the severity

of the hearing impairment of the subjects under study. According to

Davis' (1970) classification, "the condition known as hard of hearing

begins at 27 dB" (p. 85). Davis further defined a zone of "uncertainty"

from 70 to 90 dB within which some people may function as socially deaf

but most are merely very hard of hearing (p. 84).

The Subcommittee on Human Communication and Its Disorders (1969)

described the hard of hearing child in this way:


The hard of hearing child is the youngster
who has sufficient hearing impairment to suffer
educational and social handicap yet whose hearing
is good enough so that it may be used as a major
communication channel. Most such children have
hearing losses that do not average more than about
65-70 dB when classified in terms of degree of
threshold deficit. (p. 71)


Deaf

Davis (1970) defined deafness as beginning at 93 dB (p. 85).

Berg and Fletcher (1970) described the deaf child in the following ways:


The deaf child is a hearing impaired person
who can identify through hearing at best only a few
of the prosodic and phonetic features of speech and then










not enough to permit auditory recognition of sound or
word combinations. He relies mainly or entirely on
speechreading or some other form of visual receptive
communication for perception of the spoken or manual
form of language. Provided the communicative context is
within his linguistic code, he understands language
in many instances. His linguistic code typically is less
well developed than that of a hard of hearing child. (p.7)


Type of Hearing Loss: Conductive

A conductive hearing loss is an impairment of hearing due to

damage or obstruction of the ear canal, drum membrane, or the ossicu-

lar chain in the middle ear. It involves a failure of air vibrations

to be adequately conducted to the cochlea (Wood, 1971, p. 10).


Type of Hearing Loss: Sensorineural

A sensorineural hearing loss is a loss that is due to pathology

in the inner ear, the VIIIth Cranial Nerve, or both (Wood, 1971, p.20).


Speech Reception Threshold

The Speech Reception Threshold (SRT) is measured in dB from the

level at which the average normal ear's SRT has been established (Newby,

1964). It represents the level of amplification that will enable

the subject to repeat correctly approximately half of the spondaic

words that are presented to him (Wood, 1971).


Speech Discrimination

Speech discrimination tests attempt to determine the maximum

percentage of words that are intelligible to the subject at the most

favorable intensity (Davis, 1970, p. 211). In the present study speech

discrimination refers to the percent of words correctly repeated from








6

the C.I.D. Auditory Test W-22 Phonetically Balanced Word Lists (Newby,

1964, pp. 369-377).


Regular Classroom

An educational placement in the schools other than in a full-

time special class for handicapped students is designated as a regular

classroom in this study. Some students in the study received special

assistance from resource teachers or clinicians, but the major part

of each student's day was spent in a classroom with normal hearing children.


Socioeconomic Status

The four broad occupational groups utilized by the United States

Bureau of the Census (1973) were used as the basis for determining the

socioeconomic status of each subject. A description of these occupa-

tional groups may be found in Appendix A. The occupation of the parent

whose employment represented the higher status was classified according

to: (1) white collar worker, (2) blue collar worker, (3) farm worker,

or (4) service worker.


Special Training

Special training refers to attendance in a preschool class for

the hearing impaired; speech, hearing, and language therapy services;

placement in a class for handicapped students; or special assistance

by a resource teacher while attending regular school classes.


Performance Intelligence

Performance intelligence refers to the IQ score obtained by

administration of the performance sections only of the Wechsler Intelli-







7

gence Scale for Children (WISC) (Wechsler, 1949) or the Wechsler Adult

Intelligence Scale (WAIS) (Wechsler, 1955).


Receptive Vocabulary

Receptive vocabulary refers to the standard score derived for

each subject from the Peabody Picture Vocabulary Test (PPVT) Form B

(Dunn, 1959).


Syntactic Comprehension

Syntactic comprehension refers to the raw score obtained on the

Test for Comprehension of Indirect and Direct Objects (Scholes et al.,

19731.


Auditory Language

Auditory language refers to the subject's combined score on the

Auditory Comprehension and Auditory to Visual Sections of the Test of

Auditory Language (Myklebust, 1973, pp. 61,62).


Reading Ability

Reading ability refers to the reading subtest score, stated in

national percentile score, on the Stanford Achievement Test.


Speech Articulation Proficiency

Speech articulation proficiency refers to the score, stated in

number correct, on the Screening Deep Test of Articulation (McDonald,

1968).


Written Language

Written language refers to the subject's Syntax Quotient,

stated in percentile form, on the Picture Story Language Test (Myklebust,

1965).









Speechreading Skill

Speechreading skill refers to the subject's score, stated in

percent correct, on the Utley Lipreading Test, Sentence Form, Test

Form A (Utley, 1947).

Academic Performance

In this study academic performance refers to the average of the

national percentile scores on all subtests of the Stanford Achievement

Test (SAT).


Grade Point Average

Grade point averages (GPA) were calculated for academic subjects

in which the hard of hearing students were enrolled during the school

year 1972-73. Grade points were assigned in the following manner:

4 points for each A, 3 points for each B, 2 points for each C, 1 point

for each D, and zero points for each F.


Theoretical Development


Questions to be Examined

In order to describe the students in the study and the deficits

that may have been imposed by their long-standing partial loss of hearing,

the following questions were examined in this study: (1) What are

the characteristics of hard of hearing subjects attending regular classes

with regard to age, sex, socioeconomic status, etiology of the hearing

impairments, use of amplification, special training, severity of their

hearing loss, performance intelligence, and speechreading skill?









(2) What are the language characteristics of these hard of hearing

children with regard to receptive vocabulary, syntactic comprehension,

auditory language, reading ability, speech articulation proficiency, and

written language? (3) How do the hard of hearing children in this study

rank academically in comparison to national norms for students at their

grade placement level? (4) What significant correlations exist between

the following variables: average loss of hearing, SRT, speech discrimina-

tion score, receptive vocabulary, auditory language, syntactic compre-

hension, reading ability, speech articulation proficiency, written

language, performance intelligence, age, speechreading skill, and

grade point average?


Assumptions


For the purpose of this study it was assumed that:

1. A primary goal of education is that each child
should perform academically at a level com-
mensurate with his or her abilities.

2. Achievement test scores represent an index of the
performance of hard of hearing students
in relation to normal hearing students at the
same grade levels.

3. The factors affecting the level of academic per-
formance of students are complex and interrelated.

4. Some of the factors that are associated with
academic performance of hard of hearing
students can be identified and manipulated.

5. Ancillary services for hard of hearing children
and youth attending regular school classes
can be improved.

6. Improvements in ancillary services for hard of
hearing students that may result in enhanced
educational performance are desirable.









Hypotheses


On the basis of the preceding assumptions and the review of the

related literature, eight hypotheses concerning the academic performance

of hard of hearing subjects in the study were developed. These hypo-

theses, stated in the null form, are listed below.


Null hypothesis one

There is no significant correlation between the severity of

the hearing impairment, stated as an average loss of hearing, and the

academic performance of hard of hearing subjects.


Null hypothesis two

There is no significant correlation between the performance

intelligence of hard of hearing subjects and their academic performance.


Null hypothesis three

There is no significant correlation between the receptive

vocabulary of hard of hearing subjects and their academic performance.


Null hypothesis four

There is no significant correlation between the syntactic

comprehension of hard of hearing subjects and their academic performance.


Null hypothesis five

There is no significant correlation between the auditory

language of hard of hearing subjects and their academic performance.

Null hypothesis six

There is no significant correlation between the written

language of hard of hearing subjects and their academic performance.










Null hypothesis seven

There is no significant correlation between the speechreading

skill of hard of hearing subjects and their academic performance.


Null hypothesis eight

Regression of selected predictor variables on the criterion

variable, academic performance, will result in no significant Beta

coefficients.


Summary


The effect of a significant, though not profound, sensorineural

hearing impairment on the language behavior and academic performance

of school-age children has not been well documented in published

research. The purposes of this study were: (1) to describe some of

the pertinent characteristics of a group of hard of hearing school-

age subjects and (2) to determine possible relationships between their

academic performance and selected variables.

The present study was limited to hard of hearing students in

the 4th through the 12th grades in Duval County, Florida, in 1973-74.

Four questions related to the first purpose of the study were posed

in this chapter, and eight null hypotheses related to the second pur- _

pose were stated.
















CHAPTER II
REVIEW OF THE RELATED LITERATURE

Introduction


The literature and research studies that are relevant to this

investigation are treated in four sections. The first section con-

tains a review of studies concerning pertinent characteristics of hard

of hearing school children. The second contains a review of studies

relating to the verbal abilities of hard of hearing school children.

The third section contains a review of studies concerning academic

performance of hard of hearing school children. The fourth contains

a review of studies regarding factors that may influence the academic

performance of hard of hearing school children.


Characteristics of Hard of Hearing
School Children


Prevalence

The present investigation concerned children with significant,

though not profound, bilateral, sensorineural hearing impairments.

The prevalence of this type of handicap in children attending regular

school classes is difficult, if not impossible, to estimate from cur-

rently available information.

From figures obtained in a study conducted in the public schools

of Pittsburgh, Pennsylvania, Eagles (1964) stated that "the percentage

of children with any significant handicap . is 1.7 percent" (p. 14).

Because of the design of the Pittsburgh study it is impossible to determine

12






13

what portion of these hearing losses were sensorineural and what portion

were conductive.

Johnson (1962) studied the problems of the partially hearing

child in "ordinary" schools in England and suggested that there might

be 2 per 1,000 school children, or 0.2 percent, with a marked bilateral

sensorineural or mixed hearing impairment sufficiently severe to require

some form of special help but not requiring special schooling (p. 22).

In reference to the prevalence of hearing handicaps among school-

age children Silverman stated:


Our best estimate supported by an extensive
detailed study of Pittsburgh school children is that
five percent of school-age children have hearing
levels in one ear at least outside the range of
normal and that from one to two of every 10 in this group
require special education attention (Eagles et al., 1963).
(These figures do not include children in special schools
for the deaf.) (Silverman, 1971a, p. 403)


Northcott (1973) cited the Bureau of Education for the Handi-

capped, U.S.Office of Education as estimating a 0.5 percent incidence

of children 5 to 19 years old who are hard of hearing. This prevalence

figure also appeared in a 1963 Health, Education, and Welfare publication

(Mackie et al., 1963). How the figure was derived and what types of hearing

impairments were included were not specified.

In a personal communication, Dr. Raymond J. Trybus, Deputy Direc-

tor of the Office of Demographic Studies, Gallaudet College stated that,

"To the best of my knowledge no recent prevalence figures on hard of

hearing students in public schools exists. The closest that is possible

to come at present involves a procedure with a number of estimates and

a good bit of guessing in between."









Etiology

The Annual Survey of Hearing Impaired Children and Youth,

Office of Demographic Studies, Gallaudet College (1973a), collects

information on the characteristics of hearing impaired students enrolled

in special educational programs in the United States. The most recent

data published were for the 1970-71 school year and concerned 41,000

hearing impaired students. In that year, information on the probable

cause of hearing loss was reported by respondents for only about 75

percent of the students involved. Another 25 percent indicated that

there was "no known cause" for the hearing loss (p. 10). The three

probable causes that were listed most frequently for the group of

students with hearing losses less than 85 dB (ISO) were: maternal ru-

bella, prematurity, and hereditary causes (p. 20).

Johnson (1962) determined that familial deafness and meningitis

accounted for more than one-third of the sensorineural hearing impair-

ments of students attending ordinary schools in Cheshire, England.

Forty-one percent of the total group had hearing losses of unknown

etiology.

In Hine's (1970) study of hard of hearing children, also in

England, 27 percent of the children had hearing losses of unknown cause.

He described the etiology of the remaining as: 36 percent prenatal, --

25 percent perinatal, and 12 percent postnatal. By a different classi-

fication, 25 percent had acquired some form of genetically determined

hearing impairment, while as many as 37 percent were judged to have

some form of brain damage leading to impaired hearing.


Severity of the Hearing Handicap

Complete data on the average hearing deficit of hard of hearing








15

children attending regular schools are not available. Research studies

concerning hard of hearing children attending regular classes have

reported average hearing losses ranging from 14 dB to 78 dB (Quigley,

1968; Young and McConnell, 1957; Lewis, 1972).

In a detailed study of a group of students with sensorineural

impairments, attending ordinary schools in Cheshire, England, Johnson

(1962) found that the mean hearing loss for the middle range of fre-

quencies was 54 dB.


Intelligence

The literature indicates agreement that mean intellectual

levels of hard of hearing school children are within the average range

when performance criteria, such as the performance sections of the

Wechsler Intelligence Scales (Wechsler, 1949, 1955), the Nebraska

Test of Learning Aptitude (Hiskey, 1955), or the Leiter International

Scale (Leiter, 1940), are used as measures (Reynolds, 1955; Young and

McConnell, 1957; Hardy et al., 1958; Kodman, 1963; Hine, 1970; Lewis,

1972).

When verbal measures are used to assess abilities of hard of

hearing children, most published research indicates that these children

perform substantially below the average for their age level (Madden,

1931; Young and McConnell, 1957; Hine, 1970; Lewis, 1972). In a study

conducted at Johns Hopkins University, however, Hardy et al. (1958)

reported that hard of hearing subjects obtained a mean IQ of 101.25 on

the Stanford Binet Scale, which is considered to be heavily weighted in

the verbal areas (Vernon and Brown, 1964).







16

Steer et al. (1961), in a study conducted at Purdue University,

chose a vocabulary test as an instrument upon which to base estimates

of the intelligence of hard of hearing public school subjects and

controls. These authors stated that:


The Ammons Full Range Picture Vocabulary Test
is individually administered and designed to provide
an estimate of mental ability. Since this is a per-
formance-type test, it was felt that the use of this
instrument would minimize the language handicap possibly
resulting from a hearing impairment and would thus
provide more valid comparisons between experimental and
control subjects. (pp. 48,49)


There may be disagreement with the premise that a vocabulary

test would minimize the language handicap of hard of hearing children

in any way. In view of other research (Young and McConnell, 1957;

Lewis, 1972) it is not surprising that on this verbal test Steer

found that "the control subjects consistently exhibited higher mean

intelligence scores than the experimental subjects" (p. 85). It is

surprising, however, that the scores reported for the experimental

subjects were within the range of average (97.93) for 15 students with

binaural hearing losses of 60 to 74 dB and only slightly below the

average range (91.56) for the nine students with binaural hearing losses

of 75 to 100 dB (p. 163).


Socioeconomic Background

Socioeconomic status of hard of hearing subjects is seldom

discussed in published research studies. Hine (1970) noted that the

class distribution of hard of hearing subjects in his study was "not

significantly different from that of the area from which the children

came" (p. 173).










Steer (1961) analyzed parents' occupations and showed insig-

nificant differences between hard of hearing subjects and controls.

Although Hardy (1958) reported no detailed information on socioeconomic

status, he stated that the experimental and control subjects were matched

according to socioeconomic status and that controls were drawn from

children of the medical faculty and residency staff of Johns Hopkins

University. Experimental subjects were drawn from the patient load in

the Hearing and Speech Center of Johns Hopkins Hospital.


Verbal Ability of Hard of Hearing
School Children


Verbal ability is recognized as a highly reliable predictor of

academic achievement in normal hearing children (Stroud, 1957).

Each of the studies reviewed in this section concerns the assessment of

verbal abilities of hard of hearing children.

With regard to hard of hearing children, O'Neill commented

that:


Language has been unduly neglected by
researchers. Some clinicians feel that hard of
hearing children show some retardation in language
usage and comprehension. Here again, we may have only
a generalization from random experience. (1964, p. 109)


Receptive Vocabulary

There is general agreement in the research literature that hard

of hearing children demonstrate a vocabulary deficit when compared

to normal hearing students. Young and McConnell (1957) administered the

Ammons and Ammons Full Range Vocabulary Test to a group of 20 hard

of hearing children with a mean hearing loss of 51 dB and to a










matched control group of normal hearing children. Mean age of both

groups was 11 years. The authors found a highly significant difference

in the vocabulary level favoring the control group as indicated by a

t score of 6.72, which well exceeded that required for the 0.01 level

of confidence.

Hardy's (1958) subjects, although within the range of normal

on the vocabulary section of the Metropolitan Series of Achievement

Tests, differed significantly from controls, who performed well above

average.

Lewis (1972) reported that students at the New York League

for the Hard of Hearing achieved a mean difference of 23.8 months be-

tween vocabulary age and chronological age on the Peabody Picture Voca-

bulary Test. Similar results were reported by Jones and Byers (1971).


Auditory Comprehension

No studies directly related to auditory comprehension of

language were found in the published research on hard of hearing

children. Since it appears that many hard of hearing children have

difficulty with single word, receptive vocabulary, it may follow that

their understanding of syntactic clues in the language might also lag

behind that of the normal hearing child.

Wilcox and Tobin (1974) employed a repetition task to investi-

gate syntactic patterns of 10 hard of hearing children. They determined

that the hard of hearing group achieved significantly lower means

in each grammatical form than the normal hearing controls. The hard

of hearing children showed a much wider range of performance than did

the normal hearing children and apparently found some constructions










relatively easy and others extremely difficult. The normal hearing

children had little difficulty with any of the constructions.

Power and Quigley (1973) studied deaf children's comprehension

of the passive voice by requiring them to move toys to demonstrate

the action of a sentence or to select a picture showing the action

of the sentence such as: "the car was pushed by the tractor" (p. 6).

They stated that virtually all hearing children have mastered both

comprehension and production of the passive voice by the age of

eight years; however, many deaf children had not achieved this mastery

by age 17 or 18 years.

Scholes (1973) investigated the ability of normal hearing

children to utilize syntactic clues in order to comprehend sentences.

His study involved the comprehension of indirect and direct object

constructions by children ages 5 through 13 years. The results of

this study indicated that the accuracy of response to disambiguating

clues of article placement and disjuncture increased gradually over

the age groups. Five-year-olds responded at a level of 61 percent

accuracy, 7-year-olds at 71 percent accuracy, 11-year-olds at 85 percent

accuracy, and 13-year-olds at 91 percent accuracy. The acquisition of

complex syntactic structures develops slowly in normal children. It

is possible that such subtle linguistic clues may be difficult for

congenitally hard of hearing children to assimilate.


Spoken Language

Calvert et al. (1968) studied the expressive language of a group

of partially hearing students in France. They concluded that the

children exhibited a deficit in expressing notions of relationship,










analogy, or opposition and that there was a syntactical "awkwardness

and a poorness in narrative" in their speech (p. 203).

Hardy et al. (1958) performed a type-token analysis of words

in a spoken language sample elicited from 20 hard of hearing children

and 20 normal hearing controls. The control group was not significantly

superior to the experimental group in number of different words employed

or in total number of words employed. In addition, the authors compared

the groups on the ratios of four major syntactical categories (actor,

action, connective, modifiers) to the total number of different words.

They found no basic differences in the means or standard deviations

between the groups. These findings led Hardy to the conclusion that

the hard of hearing children in that study:


. are by and large functioning well within
normal limits, as demonstrated by the controls in terms
of grasp of range and extent of their native language
and of their capacity to employ the language within
the scope of the experimental communicative situation.
(p. 11)


In commenting on the techniques used by Hardy (1958) and by

Brannon and Murry (1966), Wilcox and Tobin (1974) asserted that the

measures such as type-token ratios and mean length of response are

inadequate for an understanding and assessment of language ability

because they are limited to an observation of surface performance

(p. 286).

It is widely recognized that speech articulation problems often

exist in the presence of congenital, sensorineural hearing impairments

(Van Riper, 1963; Silverman, 1971b). Goetzinger et al. (1964) adminis-

tered the Templin-Darley Sentence Test of Articulation to a group of










children with mild sensorineural hearing losses and to matched con-

trols. Chronological age range for both the experimental and control

groups was 9 to 16 years. No child in the control group made arti-

culatory errors, while a total of 28 errors were made by the experi-

mental group.

Nielson (1967) found that 40 percent of his subjects with high

frequency sensorineural hearing losses exhibited "mild to moderate" arti-

culation disorders. Most commonly misarticulated phonemes were: Es3,

Cz3, Esh], and Ech].


Written Language

A review of the published research literature failed to reveal

studies of the written language of hard of hearing children. Watson

(1967) observed that:


The written work of children with partial
hearing loss of long standing has often been observed
to be exceptionally poor in comparison with their
spoken language. This is understandable in view of the
very incomplete patterns which they receive, but it also
seems to indicate that not enough stress has been laid
upon the development of reading skills. (p. 184)


Myklebust (1964, 1965, 1973) has extensively studied the

written language of deaf and hearing children. He developed the

Picture Story Language Test (Myklebust, 1965) and established develop-

mental norms for normal hearing children as well as for the deaf.

Using this instrument he was able to establish that hearing children

wrote longer stories than did deaf children at every age level except at

7 years. Sentences written by deaf children were short and simple

when compared to those written by normal children. Syntactic ability







22


was measured by patterns of formation and structure of sentences. Deaf

children were found to be inferior to hearing children in syntactic

ability at age 7 years and remained "substantially so at each of the

age levels studied" (Myklebust, 1964, p. 292).

With regard to the graphic medium as a method of teaching

language to the deaf, Lenneberg (1968) contended that:



Congenital deafness has a devastating effect
on the vocal facilitation for speech, yet presentation
of written material enables the child to acquire
language through a graphic medium without undue
difficulty. (p. 33)


Academic Performance of
Hard of Hearing Children


In some respects the research concerning the academic performance

of hard of hearing children seems equivocal. Results of some studies

indicate that hard of hearing children perform academically within

normal range (Madden, 1931; Sprunt and Finger, 1949; Reynolds, 1955;

Hardy, 1958). Other studies suggest that, as a group, hard of hear-

ing children perform below expectation academically (Steer et al., 1961;

Kodman, 1963; Nielson, 1968; Quigley, 1968; Hine, 1970). O'Neill

cautiously stated that:


There are some indications that the child with
a hearing loss may experience some retardation in academic
progress. There may be a slight retardation in language
learning but this retardation does not appear severe
enough to incapacitate the individual. (p. 114)


Interpretation of the research literature on the academic per-

formance of hard of hearing children is facilitated by the realization










that children designated as "hard of hearing" are not a homogeneous

group. In order to interpret the results of research concerning hard

of hearing children or to make generalizations from research results

to any population of hard of hearing children, it is necessary to be

cognizant of the handicap with which the particular study deals. At

the outset it is important to ascertain: the mean hearing level of the

hard of hearing children in the study; whether the experimental group

included children whose hearing losses were conductive, sensorineural,

or both; unilateral, bilateral, or both; acquired, congenital, or

both; and whether children with multiple handicapping conditions were

included in the study. Research studies on the academic performance

of hard of hearing students are examined below in the light of these

factors.

Madden (1931) studied the school performance of a group of

hard of hearing students in two schools in New York City. In 1931,

pure tone audiometric testing was relatively new, wearable hearing aids

were cumbersome, and antibiotic drugs were not in wide use. Madden

did not indicate whether his experimental group had been differentiated

as to type of their hearing losses. It would be interesting to know

what percentage of the group had conductive impairments as opposed to

sensorineural impairments. The criterion set for impaired hearing was

"15 Sensation Units for a 2-A Audiometer" (p. 23). The hard of hearing

subjects represented the 5 percent of the schools' population with

the "poorest hearing." Madden found no demonstrable differences between

the school achievement of the experimental group and that of the controls.







24

For reasons previously cited, it may be inappropriate to make

generalizations about the academic performance of hard of hearing

children from Madden's study except as the results may apply to the

5 percent of students in the school population with the poorest hearing.

Sprunt and Finger (1949) compared the Stanford Achievement Test

performance of 28 hard of hearing subjects in the 4th through the 7th

grades with a matched number of hearing students of equivalent nonverbal

IQ. Defective hearing was defined as an average loss in the better ear

of 10 dB or greater. Statistically significant differences in subtest

scores could not be demonstrated.

Reynolds (1955) compared the educational achievement, as

measured by the California Achievement Tests, of 36 children with "mini-

mal" hearing loss (mean hearing loss in the better ear of 21.26 dB) to

matched normal hearing controls. Assuming that this hearing level

reflects 1951 American Standards Association (ASA) audiometric stan-

dards, this level may be translated to approximately 11.26 dB Re:ISO 1964

or ANSI 1969 for purposes of comparison to more recent studies (Newby,

1971, p. 355). Reynolds found no statistically significant differences

between scores made by the two groups on the California Achievement Test.

Both the Reynolds (1955) and the Sprunt and Finger (1949) studies

concerned children whose hearing levels were within the range of

normal (Davis, 1970).

Hardy (1958) compared a group of 20 hard of hearing children

ranging in age from 6 to 15 years with a matched group of normal hear-

ing children. All experimental subjects in the study had sensorineural

impairments involving only the peripheral auditory mechanism. This

group of hard of hearing children had not undergone extensive adjunctive









training, had a "fair start in language development," and attended

regular school classes (p. 5). It was reported that 18 of the 20 were

regular hearing aid wearers. The mean average loss of hearing in the

better ear was 41 dB (approximately 31 dB Re: ANSI 1969).

Each child in the Hardy study was given a vocabulary subtest and

a reading subtest from the Metropolitan Series appropriate for his

age and grade level in school. There were no significant differences

between the reading quotients of the two groups, but they were signi-

ficantly different groups in terms of vocabulary quotient. However,

the vocabulary scores of the hard of hearing subjects were within normal

range.

In order to interpret studies of the academic performance of

hard of hearing children it is necessary to consider factors unrelated

to the actual hearing deficit of the subjects. It has been demonstrated,

for example, that academic performance of normal hearing students is

directly related to socioeconomic status (Lavin, 1965). The higher

one's social status, the higher one's academic performance is likely

to be. It is possible that both the control and experimental subjects

in the Hardy study represented a rather select socioeconomic group.

As stated previously, the experimental subjects were drawn from the

patient load in the Hearing and Speech Center of Johns Hopkins Hospital,-

and the matched control subjects were drawn from children of the medical

faculty and residency staff of Johns Hopkins University and from friends

and classmates of the children in the experimental group (Hardy, 1958).

If this premise is correct, it may be that the results of the Hardy

study should be generalized only to those children with rather high

socioeconomic status who have mild sensorineural hearing impairments

and who are consistent hearing aid users.









Kodman (1963) studied 100 Kentucky school children with SRT

scores ranging from 20 to 65 dB, with a mean SRT of 40 dB (30 dB Re:

ANSI 1969). Mean achievement test results indicated that these children

were achieving at a level one year below their expected grade level

according to chronological age. Kodman noted that "based on the degree

of organic hearing loss (30 dB or greater in the better ear), 65 per-

cent of the children were possible candidates for a hearing aid .

however, only 35 percent of the pupils were fitted with aids" (p. 298).

In a study at Purdue University, Steer et al. (1961) compared

hard of hearing subjects enrolled in regular classes to control sub-

jects on three subtests of the Stanford and Metropolitan Achievement

Tests and demonstrated a significant deficit in the reading subtest

scores for the experimental group but no significant differences in

spelling or in arithmetic. The total hard of hearing group included

students with monaural, binaural, conductive, and sensorineural hearing

losses from marginal to profound severity.

The Purdue study also included a group of hearing impaired stu-

dents attending special classes. Results of achievement testing with

these students led to the puzzling conclusion that academic retardation

was inversely related to degree of hearing loss. The worse the hear-

ing, the better the academic achievement (p. 76). As a possible expla-

nation, the authors offer the fact that a greater number of mentally

retarded subjects were within the mild and moderate hearing loss classi-

fication. Why the Purdue research design included a group of subjects

whose major handicap was mental retardation is not clear.

O'Neill (1964) apparently misread these data and erroneously









cited the Steer (1961) study to indicate that:


The hard of hearing children in special
classes showed academic achievement directly related
to the degree of hearing ability. The better the
hearing, the better the academic achievement. (p. 111)


Nielson (1968) studied the effects of high frequency hearing

impairments on academic performance of school-age children. He com-

pared the total achievement efficiency scores of students with bilateral,

high frequency sensorineural hearing losses at 2000 Hz and above to

national norms of the Metropolitan Achievement Test. Seventy-seven

3rd through 8th grade students in the Salt Lake City schools with

hearing losses ranging from 40.5 dB to 60.5 dB were subjects in this

study. A significant difference in achievement efficiency scores for

the hearing impaired children and the national norms was demonstrated,

suggesting that there may be educational retardation associated with

high frequency, sensorineural hearing loss in the population investigated.

Hine (1970) measured the intelligence, academic attainments,

and social adjustments of 100 children, aged 8 to 15 years, at a school

for the partially hearing in England. The mean hearing loss of the

children was 66.1 dB. Hine administered the Schonell Silent Reading

Tests A and B and Essential Mechanical and Problem Arithmetic Tests.

The research results were interpreted to indicate that on reading and

arithmetic the children were retarded relative to normal children and

that they fall proportionately further behind as they grow older.

Quigley (1968) administered the Word Meaning, Paragraph

Meaning, and Language subtests of the Stanford Achievement Test, Form W,

to a group of 116 students in the public schools of Elgin, Illinois,

who had been identified as having a hearing impairment but for whom









no special educational provisions had been made. The mean hearing

threshold level in the better ear for the group was 16.84 dB, while

the mean hearing threshold for the worse ear was 37.75 dB. The study

included students with conductive as well as sensorineural impairments

and unilateral as well as bilateral impairments.

Although no statistical procedures were utilized in the Quigley

study, the author reported the differences between expected performance

and actual performance of the subjects on the subtests of the Stanford

Achievement Test. For each subtest in every hearing level category,

actual performance was lower than expected performance. A steady

progression was noted in retardation on each subtest through the

various hearing threshold categories with the exception of the last

category, where only the Language subtest maintained the progression.

The expected grade placement of the subjects (based on individual birth-

date) was 6.90, and the actual grade placement was 5.78.

As noted previously, the mean hearing loss of the children in

the Quigley (1968) study was less than the level currently considered

to be handicapping (Davis, 1970).


Factors That May Influence the Academic Performance
of Hard of Hearing Children


Most research studies concerning the academic performance of

hard of hearing children suggest that, as a group, children with sig-

nificant, though not profound, sensorineural hearing impairments show

academic retardation of varying degrees (Kodman, 1963; Nielson, 1968;

Quigley, 1968; Hine, 1970). At least one study, Hardy et al. (1958),

seemed to indicate that hard of hearing children are not significantly

retarded in academic or verbal achievement.









It is reasonable to expect that as wide a range of motivation

and learning aptitudes exists among hard of hearing children as among

normal children. However, it is desirable to understand the factors

that may contribute to the academic performance of hard of hearing

children so that more effective measures may be taken to promote a

higher level of academic performance in this group of students.


Severity of the Hearing Impairment

Many authorities stress that the use a child makes of his

residual hearing is more important than the amount of his residual

hearing (O'Neill, 1964; Silverman, 1971b; Johnson, 1973). However,

it has been shown that hard of hearing children demonstrate a higher

level of language performance than the deaf (Brannon and Murry,

1966) and that their speechreading skills may be more advanced than

that of the deaf (Costello, 1957). Because of these differences,

which have been documented, it seems reasonable to expect that the degree

of hearing deficit suffered by a child will affect his performance in

the school situation. In this regard, however, Northcott (1972)

stated that:


The prognosis for success in an integrated
classroom setting is dependent not upon the degree
of hearing loss but on the "listening age" which dates
from the day a hearing aid is prescribed and worn and
auditory training is begun. (p. 8)

Silverman (1971b) has suggested that the use children are able to

make of their residual hearing:

. is not simply a matter of hearing level for
speech but also such different factors as age of onset,
of the severity and exact type of hearing loss, the intelligence








of the child, the amount of training that the child
has had, the age at which the training was begun, and
particularly the auditory and language environment
of the child. (p. 431)


Speechreading Skill

Frick (1973) noted that the ability to read lips is more

essential to the hard of hearing child in the public school than it

is in the school for the deaf. Costello (1957) was able to demonstrate

that reading showed a significant relationship with speechreading in

a group of deaf and hard of hearing children. However, other researchers

have been unable to achieve correlations high enough to permit pre-

diction between speechreading and school achievement (Jeffers and

Barley, 1971, p. 115).


Socioeconomic Status

Lavin (1965) reviewed 13 studies, which reported that socio-

economic status is directly related to academic performance in normal

hearing students. The higher one's social status, the higher one's

academic performance is likely to be. The relationship has been var-

iously attributed to motivational level, the "achievement syndrome,"

and a variety of cultural values (p. 124). If socioeconomic status

is correlated with academic performance in the general population, it

should also be correlated in the hard of hearing population.


Intelligence

The relationship between ability and academic performance is

well documented (Lavin, 1965). Travers (1949) found that correlation

between intelligence and grades ran between 0.50 and 0.75 on the 8th

to 10th grade level.










Vernon and Brown (1964) cautioned that "to be valid as a

measure of the intelligence of a deaf youngster, an IQ test must be a

nonverbal performance-type instrument" (p. 415). Stroud (1957) found

that nonverbal tests predicted academic achievement as well as the

verbal tests in normal children.


Verbal Ability

Verbal ability is recognized as a highly reliable predictor of

academic performance in normal children (Stroud, 1957). Verbal skills

present some of the most significant difficulties for hearing handi-

capped children (Young and McConnell, 1957; Myklebust, 1953, 1964, 1973;

Goetzinger et al., 1964; Brannon and Murry, 1966; Calvert et al.,1968;

Lewis, 1972; Power and Quigley, 1973; Wilcox and Tobin, 1974). It may

follow that a hard of hearing child's academic performance is bound

rather directly to his verbal skill.


Summary


The literature and research studies reviewed in this chapter were

treated in four sections. The first contained a review of studies per-

taining to such characteristics as prevalence, etiology and severity

of the hearing handicap, intelligence,and socioeconomic background of

hard of hearing school children. The second concerned a review of

studies relating to the verbal abilities of hard of hearing school children.

The third section contained a review of studies concerning academic per-

formance of hard of hearing school children. The fourth contained a

review of studies regarding factors that may influence the academic per-

formance of hard of hearing school children.
















CHAPTER III
THE METHOD AND DESIGN OF THE STUDY

Introduction


The purposes of this study were: (1) to describe selected

characteristics of a group of hard of hearing school-age subjects and

(2) to determine possible relationships between their academic per-

formance and the following variables: sex, socioeconomic status,

characterisitcs of hearing aid use, special training, severity of the

hearing loss, performance intelligence, language comprehension and

expression, and speechreading skill.

This chapter describes the procedures used in data collection,

the experimental design of the research, and the instruments used in

data collection.


Procedures Used in Data Collection


Setting of the Investigation

This research was conducted in the Duval County School District,

which encompasses the city of Jacksonville, Florida. The general popu-

lation of Jacksonville is 560,000, and the total school population of

the school district is 110,000. The present study concerns students

in grades 4 through 12. The total school population in those grades

is 78,559.

Duval County has well-established programs in hearing conser-

vation; speech, hearing, and language therapy; education of the deaf;









and education of the hard of hearing. A full-time certified audiologist

administers pure tone threshold tests to all students who fail hear-

ing screening tests in the county schools. Records are maintained for

those children for whom medical or other referrals are made or whose

audiograms indicate the presence of a loss of hearing. Those county

school students with known hearing handicaps are retested at least once

each school year.

Preschool training is provided for hearing handicapped students

beginning at age 3 years. Self-contained classes for the deaf and for

the hard of hearing are provided for those students who require full-

time special educational placement. Resource teachers for the hearing

handicapped are provided for students in grades 7 through 12 who attend

regular classes but require special academic assistance.


Selection of Subjects

An attempt was made to locate each child in the Duval County

Schools during the academic year 1973-74 who had a bilateral sensori-

neural hearing impairment of 30 dB or greater, who attended regular

school classes in grades 4 through 12, and who had no discernible handi-

cap other than the loss of hearing. The Coordinator of Speech, Hearing,

and Language Services supplied the original list of 37 students who

apparently met these criteria. In an attempt to locate more students, the

audiograms of all students for whom records had been kept within the

last 10 years were examined, and those suggesting the presence of a

sensorineural hearing loss greater than 30 dB in the better ear were

selected as possible subjects. This procedure yielded seven students

who apparently met the criteria and were not on the original list, bring-

ing to 44 the total number of potential subjects.








34

Individualized form letters explaining the research project were

mailed to parents of these 44 students. Each mailing contained the

letter, a questionnaire, a statement for the parent's signature grant-

ing permission for the student to take part in the project, and a

stamped, addressed return envelope. The letter, permission slip, and

questionnaire may be found in Appendix B.

Questionnaires and signed permission slips were returned for 34

students. This represented 77 percent of the letters mailed. Of the

34 subjects whose parents gave permission for their participation in

the study, eight were not included in the study for the following reasons:

three had multiple handicaps, four had hearing losses less severe than

that specified in the criteria for the study, and one subject who met

the criteria left the school district before the testing was completed.

Assuming that the 10 students whose parents did not return

questionnaires met the criteria for this study, a total of 37 students

with bilateral sensorineural hearing losses greater than 30 dB were

known to be attending regular school classes in Duval County in 1973-74.

This number represented 0.047 percent of the enrollment for grades 4

through 12.


Testing Procedures

Each subject who had not had an audiological evaluation within

the previous calendar year was requested to make an appointment for

such an evaluation. All testing procedures were individualized and

took approximately 3 hours for each subject. Frequent rest periods

were allowed. In order to negate any effect that fatigue might play

in test performance, the order of test administration was systematically

varied from subject to subject.









Achievement test data were taken from each child's permanent

record folder or from information provided by the Director of Testing

Services for the Duval County Schools. Students in the present study

took achievement tests along with their hearing classmates in the school

they attended. Apparently no special arrangements or procedures were

utilized because of their handicap.


Experimental Design of the Research


The first purpose of the study was to describe selected character-

istics of the hard of hearing subjects including deficits that may

have been imposed by their long-standing, partial loss of hearing.

In Chapter I the following questions concerning the characteristics

of hard of hearing subjects were posed:


1. What are the characteristics of hard of hearing subjects
attending regular classes with regard to age, sex,
socioeconomic status, etiology of the hearing loss,
use of amplification, extent of special training,
severity of the hearing impairments, performance intel-
ligence, and speechreading skill?

2. What are the language characteristics of these hard
of hearing children with regard to receptive vocabulary,
syntactic comprehension, auditory language, reading
ability, speech articulation proficiency, and written
language?

3. How do the hard of hearing children in this study rank
academically in comparison to national norms for students
at their grade placement level?

4. What significant correlations exist between the following
variables: average loss of hearing, SRT, speech discrimi-
nation, receptive vocabulary, auditory language, syntactic
comprehension, reading ability, speech articulation pro-
ficiency, written language, performance intelligence,
age, speechreading skill, and grade point average?







36

In order to examine these questions, 22 variables were selected for

use in describing the subjects in this study. The nature of the variables

selected required their separation into two types: classification

variables and numeric variables. The classification variables are

categorical in nature (their variability is limited to discrete

categories). The numeric variables are more nearly continuous and

numeric values can be assigned to their variates.

The eight classification variables in this study were:


1. Sex of the subject

(a) Female
(b) Male

2. Socioeconomic status of the subject's family

(a) Parent is employed as a white collar worker
(b) Parent is employed as a blue collar worker
(c) Parent is a farm worker
(d) Parent is a service worker or is unemployed

3. Age at which the subject first wore a hearing aid

(a) Before the age of 2 years
(b) Between the ages of 2 and 4 years
(c) Between the ages of 4 and 6 years
(d) Between the ages of 6 and 10 years
(e) Between the ages of 10 and 15 years
(f) The subject has never worn a hearing aid

4. The consistency of the subject's present hearing aid use

(a) Never wears a hearing aid
(b) Wears a hearing aid less than 3 times a week
(c) Wears a hearing aid from 3 to 6 hours every day
(d) Wears a hearing aid from 6 to 12 hours every day
(e) Wears a hearing aid for over 12 hours every day

5. Special training: preschool for the hearing impaired

(a) Attended a preschool class for the hearing impaired
(b) Did not attend a preschool class for the hearing
impaired









6. Special training: speech, hearing, or language therapy

(a) Has been enrolled in a speech, hearing, or
language therapy program
(b) Has never been enrolled in a speech, hearing,
or language therapy program

7. Special training: full-time class for the school-aged
hearing impaired

(a) Has been enrolled in a full-time class for
school-age hearing impaired students
(b) Has never been enrolled in a full-time class
for school-age hearing impaired students

8. Special training: academic assistance from a resource
teacher while the subject was enrolled in regular school
classes

(a) Has had assistance by a resource teacher
(b) Has never had assistance by a resource teacher


The following statistical treatments were used with the classi-

fication variables. The means and standard deviations of selected variables

in the classification categories were reported. The means of selected

numeric variables in certain classification categories were compared sta-

tistically by use of a one-way analysis of variance technique (Ferguson,

1966, p. 281). A chi square test (Ferguson, 1966, p. 191) was performed

to determine the significance of any difference between the proportions of

the subjects' parents who were employed in each of four broad occupational

categories and the proportions of the population in the United States who

were employed in each of the same occupational categories.

The 14 numeric variables in the study were:


1. Age
2. Average loss of hearing
3. Speech reception threshold (SRT)
4. Speech discrimination score
5. Performance intelligence
6. Receptive vocabulary
7. Syntactic comprehension









8. Auditory language
9. Reading ability
10. Speech articulation proficiency
11. Written language
12. Speechreading skill
13. Academic performance
14. Grade point average (GPA)


In order to further describe the subjects the means, standard

deviations, and intercorrelations with other variables are reported

for the numeric variables in the study.

The second purpose of this study was to determine possible

relationships between the academic performance of the subjects and the

following variables: sex, socioeconomic status, characteristics

of hearing aid use, special training, severity of the hearing loss,

performance intelligence, language comprehension and expression,

and speechreading skill. The relationships between the classification

variables listed and the dependent variable, academic performance,

were studied by comparing the mean academic performance of subjects

in the various classification categories by means of the one-way

analysis of variance technique.

In Chapter I, eight null hypotheses concerning the relationship

between academic performance of the subjects and selected numeric

variables were stated.


Null hypothesis one. There is no significant correlation
between the severity of the hearing impairment,
stated as an average loss of hearing, and the
academic performance of hard of hearing subjects.

Null hypothesis two. There is no significant correlation
between the performance intelligence of hard of hearing
subjects and their academic performance.

Null hypothesis three. There is no significant correlation
between the receptive vocabulary of hard of hearing
subjects and their academic performance.









Null hypothesis four. There is no significant correlation
between the syntactic comprehension of hard of hearing
subjects and their academic performance.

Null hypothesis five. There is no significant correlation
between the auditory language of hard of hearing
subjects and their academic performance.

Null hypothesis six. There is no significant correlation
between the written language of hard of hearing
subjects and their academic performance.

Null hypothesis seven. There is no significant correlation
between the performance intelligence of hard of hearing
subjects and their academic performance.

Null hypothesis eight. Regression of selected predictor
variables on the criterion variable, academic per-
formance, will result in no significant Beta coefficients.


Two methods of data analysis were utilized to investigate

these relationships: (1) the coefficients of correlation between the

academic performance of the subjects and the numeric variables included

in the null hypotheses were calculated and (2) a regression model

was developed to predict the criterion variable, academic performance.

Only numeric variables were utilized as potential predictor variables

in developing this model.

The IBM 370-165 Computer at the Northeast Regional Data Center

of the State University System of Florida was used to obtain means,

standard deviations, intercorrelations, multiple correlations, Beta

coefficients, t statistics, F ratios, and probability levels of the

results. The Statistical Analysis System (SAS) Forward Selection

STEPWISE Procedure (Service, 1972) was used to develop a regression

model designed to predict the criterion variable, academic performance.

The Forward Selection Procedure was programmed to include those predictor

variables which were significant at the 0.50 level.






40

The SAS printed the probability of each correlation coefficient

or regression coefficient indicated. Null hypotheses one through seven

were rejected when it was found that the correlation between the given

variables had reached the 0.05 level of significance. Rejection of null

hypothesis eight was indicated when it was found that the t value for the

Beta coefficient of a predictor variable was significant at the 0.05 level

of confidence. The 0.05 level was also accepted as the significance

level for F ratios in the analysis of variance procedure.


Data Collection Instruments


The data collected in this study that related to age, socio-

economic status, probable cause of the hearing loss, hearing aid use,

and special training were obtained from questionnaires returned by

parents of the subjects. The data related to academic performance,

reading ability, academic grades, and audiologic testing were taken

from existing school records. The data related to performance intel-

ligence, speechreading skill, receptive vocabulary, auditory language,

syntactic comprehension, articulation proficiency, and written language

were obtained through tests administered by the investigator. The

various instruments used in data collection are described in the para-

graphs below.


Parent Questionnaire

A questionnaire was developed by the investigator and consisted

of spaces for identifying information and six questions related to

the child's hearing handicap. It was designed to assist in the des-

cription of the subjects and for gathering pertinent background infor-

mation. A copy of the Parent Questionnaire may be found in Appendix B.










Audiometric Tests (Variable: Severity of the Hearing Loss)

Information concerning average loss of hearing, speech reception

threshold (SRT), and speech discrimination score for each of the hard

of hearing students in the study was taken from records in the Audiology

Section of the Exceptional Child Program in the Duval County Schools.

The audiometer used was a Maico-24 calibrated to ANSI Re: 1969 standards.

The subjects were tested in an IAC chamber. All testing was done by

the Duval County School District audiologist, who held the American

Speech and Hearing Association Certificate of Clinical Competence in

Audiology.


Wechsler Intelligence Scale for Children (Variable: Performance

Intelligence)

The Wechsler Intelligence Scale for Children (WISC) is composed

of 12 subtests, which are divided into two subgroups identified as

Verbal and Performance. The Performance Section of the WISC was

administered to subjects in this study who were 16 years old or younger.

The subtests in the Performance Section of the WISC include:

Picture Arrangement, Picture Completion, Block Design, Object Assembly,

and Coding or Mazes. The Coding subtest rather than the Mazes was

administered to each subject.

Vernon and Brown (1964) cautioned that to be a valid measure

of the intelligence of a hearing impaired child, a test must be of the

performance type as opposed to a verbal scale. These authors further

stated that "The Wechsler Performance Scale is at present the best test

for deaf children ages 9-16" (p. 416).


Wechsler Adult Intelligence Scale (Variable: Performance Intelligence)

The Wechsler Adult Intelligence Scale (WAIS) consists of 11










subtests. Six of these are grouped in the Verbal Scale; the remaining

five compose the Performance Scale. The WAIS Performance Scale was

administered to subjects 17 years old and older. Vernon and Brown

(1964) also suggested that the WAIS Performance Scale was the best test

presently available for deaf individuals over 16 years old.

The performance subtests of the WAIS include: Digit Symbol,

Picture Completion, Block Design, Picture Arrangement, and Object

Assembly. On both the WISC and the WAIS,


. the method used in the computation
of the deviation IQ was to set the average sum of
scaled scores on the Verbal, Performance, or Full
Scale equal to an IQ of 100 for each age group.
The standard deviations of the distributions of
sums of scaled scores were set at 15 IQ points for
each age group. Thus, for any age group, the dis-
tribution of IQ's has a mean of 100 and a standard
deviation of 15. The middle 50 percent of each
group will have IQ's between 90 and 110. (Wechsler,
1955, p. 3)


Peabody Picture Vocabulary Test (Variable: Receptive Vocabulary)

Receptive vocabulary was determined by administration of the

Peabody Picture Vocabulary Test (PPVT). The PPVT is an untimed, indivi-

dual intelligence test consisting of a booklet with 3 practice and 150

test plates, each with 4 numbered pictures. The examiner reads the

stimulus word and the subject responds by pointing to or giving the

number of the picture best illustrating the word.

For purposes of the present research, the test procedures were

modified to include graphic presentation of the stimulus word as it was

spoken by the examiner. For this purpose, each stimulus word was type-

written on a 3"x5" white card. Although this modification was contrary










to the instructions presented in the PPVT manual (Dunn, 1959) it

was considered necessary to insure that the hearing handicapped subjects

correctly perceived the stimulus word.

In the present study, standard scores were reported for each

subject. Dunn (1959) described the PPVT standard score equivalents

in the following manner:


The standard score norms were derived
by preparing separate raw score distributions
for each age level in the standardization sample.
The mean and standard deviation of raw scores
were then found for each distribution . an
IQ of 100 was arbitrarily assigned to the mean
raw score for each age level and the standard
deviation set at 15 IQ points. (pp. 28-29)


Test for Comprehension of Indirect and Direct Object Constituents

(Variable: Syntactic Comprehension)

Syntactic comprehension was assessed by administration of the

Test for Comprehension of Indirect and Direct Object Constituents

(Scholes et al., 1973). The purpose of this test is to determine the

ability of an individual to utilize syntactic clues to comprehend sen-

tences. Use of two types of clues are evaluated: placement of the

definite article to distinguish between such sentences as:

he showed her the baby pictures
and
he showed her baby the pictures


and auditory clues of disjuncture to distinguish between such sentences

as:
he showed her baby pictures
and
he showed her baby pictures.










Twenty-five different sentences are used in the test. Twenty

of the sentences are unambiguous and are presented once each. Five

of the sentences are ambiguous in meaning and these are presented twice

each. A total of 30 tape-recorded stimuli thus are used.

The types of sentences used are illustrated in the example set

below:

Clue
he showed her bird the seed
article placement
he showed her the bird seed


he showed her the bird seed
disjuncture
he showed her bird the seed


ambiguous he showed her bird seed


Five such sets were constructed. Scholes explained the

sentence construction in the following way:


For each set one sentence is ambiguous
as to which words of the predicate comprise the
indirect object and which words comprise the direct
object. In recording these sentences, the ambiguity
was preserved.
Of the four unambiguous sentences in each
set, two are made clear as to the indirect/direct
object constituents by means of a definite article
preceding the direct object, and, in the other
two sentences, this constituency is made clear by
disjuncture (a complexity of pause, intonation, and
stress cues) preceding the direct object (denoted
by "-" in the examples above). Within these pairs of
sentences disambiguated by article or disjuncture clues,
one member has only the last noun as the direct object
and the other member has the last two nouns as the
direct object. (p. 6)










The stimuli for the test consisted of tape-recorded sentences

and line-drawing pictures. For the purposes of the present research,

test procedures were modified to allow for the presentation of each

stimulus sentence in graphic as well as auditory form. Each stimulus

sentence was typewritten, without punctuation, on a 3"x5" white card

and was presented to the subject simultaneously with the tape-recorded

presentation.

Accompanying each sentence presented to the subject was an 8"xll"

paper containing four line drawings. One of the drawings corresponded

to the sentence presented, one of the drawings corresponded to the

alternate reading of the sentence, and two of the pictures corresponded

to the readings of a different set of sentences. Thirty different

sheets of four drawings each were used. Pictures were arranged randomly

on each page.

The subject heard a tape-recorded sentence at an individually

selected listening level and was simultaneously shown the same sentence

in written form. The subject selected from among the four pictures

presented the one that corresponded to the stimulus sentence.

The raw score obtained represented the number of correct choices

involving the 20 unambiguous test sentences. The responses to the 10

ambiguous test sentences were used to determine whether a possible bias

existed for one interpretation or the other.

As reported in Chapter II, Scholes (1973) used this instrument

to study the developmental comprehension of English syntax in normal

hearing children from age 5 to 13 years. His findings indicated that

the ability to use the syntactic clues of article placement and disjuncture










increased gradually over the age groups. Five-year-olds were

able to select the appropriate interpretation with 61 percent accu-

racy, seven-year-olds with 71 percent accuracy, eleven-year-olds with

85 percent accuracy, and thirteen-year-olds with 91 percent accuracy.

In a pilot study utilizing similar materials, college students main-

tained a 95 percent accuracy rate (Scholes, 1973).

Other findings were that ambiguous sentences such as:

He showed her baby pictures

were most often interpreted by normal hearing children to mean:

He showed her the baby pictures

rather than:

He showed her baby the pictures.

College-age students in a pilot study selected each of the two inter-

pretations with almost equal frequency. Scholes referred to the first

interpretation as the "B" interpretation and stated that:


Since such an interpretation cannot
be accounted for by syntactic, semantic, or phonologic
information contained in the sentence, the procedures
employed by subjects to make this interpretation is
termed a strategy. This particular strategy says
that a predicate sequence of nominal + nominal +
nominal is to be interpreted as:


Predicate



Indirect Object Direct Object


Nominal Nominal Nominal
(her) (baby) (pictures)


(Scholes, 1973, p. 8)










Auditory Language Test (Variable: Auditory Language)

The Test of Auditory Language (Myklebust, 1973, pp. 61-62) was

chosen as an instrument with which to assess the subjects' ability to

follow oral directions. Although the instrument was not designed for

hearing impaired children, it was chosen because it simulated in many

ways the types of instructions that students are required to follow

in the classroom.

Test One: Auditory Comprehension is composed of a group of 10

instructions of increasing difficulty. The subject was given a form to

use for his or her responses. Sentences were spoken slowly and dis-

tinctly but the direction was given only once. An example of one of

the directions is: "Write the day it was yesterday."

Test Two: Auditory to Visual was administered in the same manner.

The subject's response form contained five rows of letters, and the

directions involved instructions such as: "Draw a cross on the a's and

circle the t's."

Because the test author did not provide complete directions for

scoring the tests, the results obtained in this study were not compared

to the normative data provided by Myklebust (1973). The scores reported

in the present study represented the combined total correct score on

Test One: Auditory Comprehension and Test Two: Auditory to Visual.

An error was recorded in each instance that the direction given was not

explicitly followed.


Screening Deep Test of Articulation (Variable: Speech Articulation

Proficiency)

Speech Articulation Proficiency was quantified in this study as










the number correct on a modified version of the Screening Deep Test

of Articulation (McDonald, 1968). The Screening Deep Test elicits

10 productions each of the following consonants: Es], C13, Cr], Ech],

Cthl, Esh], Ek], Cf3, and It].

This instrument was chosen for use in this study because it

permitted the evaluation of commonly misarticulated consonants in a

variety of phonetic contexts. The test utilizes pairs of pictures to

elicit the subject's production of bisyllables. Because all the sub-

jects in the present study were readers, it seemed desirable to use

a sentence test in order to simulate more nearly the conditions of

connected speech. Therefore, for each pair of stimulus pictures on the

Screening Deep Test, a sentence was constructed containing the same

words in the same relationship as they appeared on the test. The stimu-

lus sentences that were used may be found in Appendix C.

McDonald (1968) cautioned against counting all of a child's cor-

rect productions and expressing these as a total articulation score.

He emphasized that a total score would not reveal the pattern of a child's

articulatory development. It was with some reservation, therefore, that

this method was used to express the articulation proficiency of the sub-

jects in this study. It is clear that this method of quantifying arti-

culatory proficiency leaves much to be desired. However, since all

subjects in the study were past the age at which considerable spontaneous

gains in articulation skill are expected (Powers, 1971) and because the

primary purpose for assessing articulation in this study was to describe

the general articulatory behavior of this group, it was determined that

this method of expressing the score was defensible.









Picture Story Language Test (Variable: Written Language)

Written language of the hard of hearing children in the present

study was surveyed by use of the Picture Story Language Test (Myklebust,

1965). The purpose of this test is to "serve as an instrument for

the study of language developmentally and diagnostically" (Myklebust,

1965, p. 70).

Administration of the test involves a stimulus picture which the

subject is able to view throughout the examination period. The subject

is provided with a pencil and paper and asked to write a story about

the picture. There is no time limit. The written story is scored

according to directions provided by the test author.

The score reported as "Written Language" in the present study

was actually the percentile equivalent (based on age) of the Syntax

Quotient. Myklebust (1965) described the Syntax Scale in the follow-

ing way:


In summary, the Syntax Scale incorporates a
measure of syntax, certain morphological aspects,
correctness of word choice, and punctuation. Therefore,
it cannot be compared with measures using only sentence
structure, complexity or parts of speech. Our objective
was to devise a scale which comprised all aspects pertinent
to the accuracy with which the written word is used.
Though referred to as the Syntax Scale because of the
lack of a more appropriate term, not only syntax was
included. We justified this approach theoretically in that
our interest was in studying the growth of language usage
in general, not a single component or facet such as syntax,
punctuation, or morphology. For this reason we developed a
generalized scale of correctness. (pp. 110-111)


Utley Lipreading Test (Variable: Speechreading Skill)

A live presentation of Part I, the Sentence Test, Form A, of

the Utley Lipreading Test was used to assess the subjects' speechreading







50

ability. Utley (1947) designed what is ". . probably the best known

and most widely used test of speechreading ability" (Jeffers and Barley,

1971, p. 336). This filmed version includes three parts. Part I is

a Sentence Test and consists of two forms, A and B; Part II is a Word

Test; and Part III is a Story Test.

Jeffers and Barley (1971) demonstrated that college students and

hard of hearing adults scored substantially higher when the Sentence

Test was presented live rather than in the original filmed version.

The presentation method described by Jeffers and Barley (p. 339) was

utilized in the present study. Each sentence was presented only once.

The subject wrote down what he perceived. Five practice sentences were

given before the test. No voice was used. The examiner used a fairly

slow, though normal, rate of speech and ample, but not exaggerated, lip

and jaw movement. Facial expression appropriate to the content was

employed.

The scoring method used in the present study was developed by

the Audiology Clinic at Northwestern University (Jeffers and Barley, 1971).

In this method, a sentence is scored correct if the content is perceived

with reasonable accuracy. There are 31 sentences in Test Form A, and

the score was reported as the percent of correct sentences.


Stanford Achievement Tests (Variables: Academic Performance and Reading)

The Stanford Achievement Tests are comprehensive achievement

tests designed to measure student progress in subject areas, skills,

and understandings generally accepted as desirable outcomes of elemen-

tary and secondary education (Buros, 1972).









There are six test battery levels in the Stanford Series:

Primary I and II, Intermediate I and II, Advanced, and High School

Basic Battery. The latter was added to the series in 1965 (Buros, 1972).

The Intermediate I Battery is used with students in grade 4

through the middle of grade 5. It measures reading, arithmetic, lan-

guage, spelling, social studies, science, and word study skills. The

items in this test battery are of a multiple choice variety. No

questions are dictated.

The Intermediate II Battery is designed for use with students

in the middle of grade 5 to the end of grade 6. Essentially the same

areas are measured in this battery as in the Intermediate I Battery.

The Advanced Battery is designed for use with students in the

7th, 8th, and 9th grades. Reading is measured by a single test, Para-

graph Meaning. Arithmetic is measured by three tests; and language,

spelling, social studies, and science are measured by one test each.

The High School Basic Battery is for use with students in grades

9 through 12. Tests include English, Numerical Competence, Mathematics,

Reading, Science, Social Studies, and Spelling.

A nationwide achievement testing program, conducted by the Office

of Demographic Studies, Gallaudet College, compiled substantial data

concerning achievement test results of hearing impaired students attending

special education programs (Annual Survey, 1972). Some of the major

findings in this extensive testing program should be noted. A survey

conducted in 1968 indicated that the Stanford Achievement Test Series

was the most widely used measure of academic achievement of hearing

impaired students in the United States.









In 1969, more than 12,000 Stanford Achievement Tests were admin-

istered through the Office of Demographic Studies to hearing impaired

students attending special educational programs. The results indicated

that the hearing impaired students scored substantially below the

average levels of attainment of their hearing agemates (Annual Survey,

1973b, p. 51).

In discussing the limitations of data from these tests and

tests subsequently administered by the same group in 1971, DiFrancesca

and Carey (1972) noted that:


The directions for taking these tests are
designed for oral dictation. . The language
level and structure of the test questions may be
biased against language handicapped students. A
student may know the proper answer to a question
but fail it because of inability to grasp the lan-
guage complexity. (p. 2)


Experience gained in 1969 led to the conclusion that measuring

instruments must be revised before an adequate measure of the academic

achievement of hearing impaired students could be obtained. Of central

importance was the finding that large numbers of students received test

batteries too advanced for their achievement level; and, therefore, those

tests failed to reveal true differences between the good and the poor

students. By assigning tests on the basis of age or grade placement,

many student scores were "at or below the level where guessing or

random responses become a major determinant of obtained score" (Annual

Survey, 1973b, p. 51). To alleviate this problem, a screening test

procedure was inaugurated to determine the appropriate level at which

the student should be tested.










In the present study, because of the age range and span of grade

placement of the subjects, it was determined that the most efficient

way to utilize the test results was in the form of national percentile

ranks. In this way, hard of hearing students could be compared according

to their respective rank among hearing students at their own grade level.

No assumptions were made concerning equivalence of percentile ranks

from grade to grade. The overall achievement test score was obtained

by averaging the national percentile scores on all the subtests of the

battery that was taken by each student. This score was used to represent

academic performance in the present study. The reading ability of the

subjects was measured by the Paragraph Meaning Subtest of the SAT Inter-

mediate and Advanced Batteries and the Reading Subtest of the High School

Basic Battery.

Current Stanford Achievement Tests were available for 21 of the

26 subjects. Seventeen subjects took the Intermediate or Advanced Bat-

teries and four took the High School Basic Battery. The five subjects

for whom current SAT scores were not available had taken either the

Metropolitan Achievement Test, the Science Research Associates Achieve-

ment Test, or the Wide Range Achievement Test. The averaged national

percentile rank for all subtests was reported for each of these five

subjects and was used in determining correlations with other variables.

Reading ability was measured by the appropriate subtest of each of these

batteries.


Summary


The purposes of this study were (1) to describe selected charac-

teristics of a group of hard of hearing children and youth and (2) to







54

determine possible relationships between their academic performance

and the following variables: sex, socioeconomic status, charac-

teristics of hearing aid use, special training, severity of the hearing

loss, performance intelligence, language comprehension and expression,

and speechreading skill.

The procedures used in data collection, the experimental design

of the research, and the instruments used in data collection were

included in this chapter.
















CHAPTER IV
RESULTS

Introduction


The purposes of this study were: (1) to describe selected

characteristics of a group of hard of hearing school-age subjects and

(2) to determine possible relationships between their academic perfor-

mance and the following variables: sex, socioeconomic status, charac-

teristics of hearing aid use, special training, severity of the hearing

loss, performance intelligence, language comprehension and expression,

and speechreading skill.

The study involved 26 children with sensorineural hearing impair-

ments greater than 30 dB in the better ear who attended regular school

classes in grades 4 through 12 in Duval County, Florida.

Presentation of the findings in this chapter is divided into

two sections. The first section contains data that are descriptive of

the subjects in the study and related to the first purpose of the study.

The findings are presented as answers to four questions posed in Chapter I.

These data concern the age, sex, socioeconomic status, etiology of the

hearing loss, the use of amplification, special training, severity of

the hearing deficits, performance intelligence, certain language skills,

and academic performance of the subjects in the study. The second section

concerns the relationships between the academic performance of the sub-

jects and selected variables. Tables illustrating the mean academic

performance of subjects according to various classification variables








56

and tables illustrating correlations between academic performance and

selected numeric variables appear in this section. A complete inter-

correlation matrix of numeric variables in the study is found in

Appendix D. Results concerning the hypotheses stated in Chapter I are

presented in the second section of this chapter.


Findings Related to the First Purpose of the Study


The first purpose of the study was to describe selected charac-

teristics of this group of hard of hearing school-age subjects. The

following questions relate to that purpose.


Question One

What are the characterisitcs of the hard of hearing subjects

in this study with regard to age, sex, socioeconomic status, etiology

of the hearing loss, use of amplification, special training, severity

of the hearing loss, performance intelligence, and speechreading skill?


Age of the hard of hearing subjects

The mean age of subjects in the study was 14.53 with a range in

age from 9 years to 19 years.


Sex of the hard of hearing subjects

Sixteen of the 26 hard of hearing subjects in the study were

males and 10 were females. Data in Table I demonstrate that the mean

loss of hearing for females was 69.40 dB, and the mean hearing loss

for males was 56.68 dB. The difference between means for males and

females on average loss of hearing did not reach a significant level.










TABLE I
AVERAGE LOSS OF HEARING OF
FEMALE AND MALE SUBJECTS


SEX AVERAGE LOSS OF HEARING
(dB)

Females
(N=10) 69.40

Males
(N=16) 56.68


F Ratio 3.82


*Critical F Ratio for significance at the .05 level = 4.26
(Ferguson, 1966, p. 409)


Socioeconomic status

Because it had been reported that socioeconomic status was

directly related to academic performance in normal hearing students

(Lavin, 1965) it was desirable to determine whether the subjects in this

study were typical of the United States population in terms of socio-

economic status. The occupation of one of the parents of each subject

was used to determine the socioeconomic status of that subject. The parent

whose occupation represented the higher status was used in the tabulation.

A description of the broad occupational groups used by the United States-

Bureau of the Census is found in Appendix A. Table II demonstrates the

comparisons of percentages of parents in the study who were employed in

four broad occupational groups to those of United States residents. No

parent in the sample was a farm worker, although approximately 3 percent

of the United States population was employed in that setting in 1970.

Forty-two percent of the parents in this study were white collar workers,











39 percent were blue collar workers, and 19 percent were service workers.

Forty-eight percent of the employed population in the United States in

1970 were white collar workers, 36 percent were blue collar workers,

3 percent were farm workers, and 13 percent were service workers.


TABLE II
PERCENT OF EMPLOYED PERSONS IN BROAD OCCUPATIONAL CATEGORIES
IN THE UNITED STATES AND AMONG PARENTS OF
HARD OF HEARING SUBJECTS




WHITE COLLAR BLUE COLLAR FARM SERVICE
WORKERS WORKERS WORKERS WORKERS


United States 48.2 35.9 3.1 12.8

Parents of
Subjects 42.3 38.5 0.0 19.2



Source: United States Bureau of the Census (1973)


The chi square test (Ferguson, 1966, p. 191) was used to deter-

mine whether the proportion of parents in this study who were employed

in each of the occupational groups was different from that of United

States residents. Data in Table III demonstrate that no significant

differences existed between the proportion of subjects' parents in each

occupational category and the proportion of United States residents who

were employed in each occupational category.









TABLE III
COMPARISON OF THE OBSERVED FREQUENCY OF PARENTAL EMPLOYMENT IN FOUR OCCUPATIONAL
CATEGORIES AND THE EXPECTED FREQUENCY OF SUCH EMPLOYMENT BASED ON PROPORTION
OF UNITED STATES RESIDENTS EMPLOYED IN EACH CATEGORY


OCCUPATIONAL OBSERVED EXPECTED O-E (O-E)2 (O-E)2
CATEGORY FREQUENCY FREQUENCY E
(Parents)

White Collar Workers 11 13 2 4 0.31


Blue Collar Workers 10 9 1 1 0.11


Farm Workers 0 1 1 1 1.00


Service Workers 5 3 1 1 1.33
x*
Chi Square = 2.75




Based on proportions of U.S. residents employed in the four broad occupational categories (United
States Bureau of the Census, 1973)
Not significant at the .05 level of confidence (Ferguson, 1966, p. 407)
Not significant at the .05 level of confidence (Ferguson, 1966, p. 407)







60


Table IV demonstrates that children whose parents were employed

as white collar workers demonstrated consistently higher scores on four

language tests than those whose parents were employed as blue collar

workers and that children whose parents were employed as blue collar

workers made higher scores on these tests than those whose parents were

service workers. The differences among the occupational groups reached

the 0.01 level of significance on receptive vocabulary and written lan-

guage. The variation among groups on syntactic comprehension was signi-

ficant at the 0.05 level.


TABLE IV
MEANS OF AVERAGE LOSS OF HEARING, AUDITORY LANGUAGE,
RECEPTIVE VOCABULARY, SYNTACTIC COMPREHENSION,
AND WRITTEN LANGUAGE OF HARD OF HEARING
SUBJECTS BY PARENTAL OCCUPATION CATEGORY




PARENTAL AVERAGE LOSS AUDITORY RECEPTIVE SYNTACTIC WRITTEN
OCCUPATION OF HEARING LANGUAGE VOCABULARY COMPREHENSION LANGUAGE
(Raw (Standard
(dB) Score) Score) (Percent) (Percentile)



White Collar
Worker
(N=11) 60.45 28.63 83.00 13.72 43.54

Blue Collar
Worker
(N=10) 68.50 23.20 72.30 13.40 12.90

Service
Worker
(N=5) 50.20 17.60 55.20 9.80 2.60


F Ratio* 2.14 3.01 9.36 4.23 5.96


Critical F Ratio:
For significance


at the .01 level = 5.61


For significance at the .05 level = 3.40










Etiology of the hearing loss

The probable cause of the hearing impairment of 16 children,

61 percent of the subjects in this study, was unknown to their parents.

Birth defects or prematurity were cited as the cause in four cases;

maternal rubella and spinal meningitis accounted for two cases each.

Measles was cited once, and kidney disease was named as the probable

cause of the hearing loss of one subject. These data are presented

in Table V.


TABLE V
NUMBERS AND PERCENTAGES OF SUBJECTS
BY PROBABLE CAUSE OF HEARING LOSS





PROBABLE CAUSE OF THE
HEARING LOSS AS STATED
BY THE PARENT NUMBER PERCENT




Unknown* 16 61.53

Birth Defect 3 11.53

Maternal Rubella 2 7.69

Meningitis 2 7.69

Prematurity 1 3.84

Measles 1 3.84

Kidney Disease 1 3.84




*This number includes 11 subjects whose parents indicated that
the cause of the hearing loss was unknown, four subjects whose parents
listed "nerve deafness" as the cause of the hearing loss, and one sub-
ject whose parent stated that the "ear grew in two parts because she
put a string in it."









Use of amplification

Twenty-three of the 26 subjects had previously been fitted with

hearing aids; however, parents of five subjects who had hearing aids

stated that their children never wore their aids. No child in the

study had been fitted with an aid before the age of two years. Eleven

subjects had been fitted before the age of 6 years,and 12 were fitted

after the age of 6 years. The variance in the severity of the hearing

losses of subjects fitted at different ages did not reach significance.


TABLE VI
PERCENT OF SUBJECTS AND AVERAGE HEARING LOSS OF SUBJECTS
BY AGE OF FIRST HEARING AID FITTING




AGE AT WHICH A
HEARING AID WAS PERCENT OF AVERAGE
FIRST FITTED SUBJECTS HEARING LOSS
(dB)

Before age 2
(N=0) 0.0

Between ages 2 and 4 15.38 67.50
(N=4)

Between ages 4 and 6
(N=7) 26.92 62.28

Between ages 6 and 10
(N=11) 42.30 65.63

Between ages 10 and 15
(N=l) 3.84 55.00

Never
(N=3) 11.53 39.33


F Ratio* 1.80


level = 2.84 (Fer-


Critical F Ratio for significance at the .05
guson, 1966, p. 409)







63


Parents of nine subjects stated that their children wore their

hearing aids "over 12 hours a day." Another five were said to wear

hearing aids from "6 to 12 hours a day," while four wore their aids "3

to 6 hours a day." There was a trend for more consistent hearing aid

use among those subjects with greater hearing losses. There were sig-

nificant differences between the mean hearing losses of subjects in the

various categories concerned with consistency of hearing aid use. Data

in Table VII indicate that the average loss of hearing of children

wearing their hearing aids "3 to 6 hours a day" was 51.75 dB while

those who wore aids "over 12 hours a day" had a mean loss of 74.11 dB.


TABLE VII
PERCENT OF SUBJECTS AND AVERAGE HEARING LOSS OF SUBJECTS
BY CONSISTENCY OF HEARING AID USE




NUMBER OF HOURS A DAY AVERAGE
THE HEARING AID IS PERCENT OF HEARING LOSS
PRESENTLY WORN SUBJECTS (dB)


Never
(N=8) 30.76 47.87

3 to 6 hours a day
(N=4) 15.38 51.75

6 to 12 hours a day
(N=5) 19.23 68.80

Over 12 hours a day 34.61 74.11
(N=9)


F Ratio 7.10




Critical F Ratio:
For significance at the .05 level = 3.05
For significance at the .01 level = 4.76
(Ferguson, 1966, p. 409)







64

Table VIII compares the consistency of present hearing aid use

with the age at which a hearing aid was first fitted. There is some

indication that the earlier the hearing aid was fitted, the greater the

likelihood that it was consistently worn at the time of this study.

Three out of four children fitted before the age of 4 years wore their

hearing aids over 12 hours a day while only three out of seven children

fitted between 4 and 6 years of age wore their aids over 12 hours a

day. Only three out of 11 who were fitted between the ages of 6 and

10 wore their aids over 12 hours a day.

TABLE VIII
CONSISTENCY OF PRESENT HEARING AID USE BY AGE OF
FIRST FITTING OF A HEARING AID


AGE OF FIRST
FITTING OF A
HEARING AID


CONSISTENCY OF PRESENT HEARING AID USE

Never 3-6 Hours 6-12 Hours Over 12 Hours
a Day a Day a Day
(N=5) (N=4) (N=5) (N=9)


Between ages 2-4
(N=4) 1 3

Between ages 4-6
(N=7) 1 3 3

Between ages 6-10 3 3 2 3
(N=11)

Between ages 10-15
(N=1) 1









Special training

Four types of special training were considered in the study:

attendance at a preschool for the hearing impaired; speech, hearing,

or language therapy; full-time enrollment in a class for school-age

hearing impaired children; and assistance provided by a resource teacher.

Parents were asked to indicate which of these services, if any, their

child had received and the length of time the service had been provided.

No attempt was made to define or control the "quality" of special train-

ing provided.

Six of the 26 subjects had attended a preschool for the hearing

impaired. The mean hearing loss of students attending a preschool for

the hearing impaired was 79 dB whereas the mean loss of hearing of those

not attending a preschool was 56 dB. Length of attendance varied from

2 months to two years. Twenty-three of the 26 subjects had been enrolled

in speech therapy programs, and 13 had at one time been enrolled in a

full-time class for school-age hearing impaired children. Twenty-one

of the 26 subjects had received special help from a resource teacher

while attending classes with normal hearing students.

Table IX illustrates the fact that the mean hearing loss of

those children receiving special help was greater than that of the

children not receiving special help. This was true for each type of

special assistance provided. Because the distribution of the numbers

of subjects in most of these special training categories was skewed,

tests for significance of the difference between means would not produce

meaningful data.













TABLE IX
NUMBERS AND MEAN HEARING LOSS OF SUBJECTS WHO RECEIVED
SPECIAL TRAINING AND SUBJECTS WHO
DID NOT RECEIVE SPECIAL TRAINING


SPECIAL TRAINING MEAN LOSS OF
NUMBER HEARING (dB)


Attended preschool 6 79.33

Did not attend preschool 20 56.25


Enrolled in speech
therapy 23 64.04

Never enrolled in
speech therapy 3 42.66


Enrolled in a full-
time class for the
hearing impaired 13 66.53

Never enrolled in a
full-time class for
the hearing impaired 13 56.61


Received special assistance
from a resource
teacher 20 65.45

Never received special
assistance from a
resource teacher 6 48.66










Severity of the hearing impairment

The three measures relating to the severity of the hearing

impairment of subjects in the study were: average loss of hearing,

speech reception threshold, and speech discrimination score. Table X

illustrates that the mean loss of hearing in the better ear for the

subjects in this study was 61.57 dB,with a range from 30 dB to 88 dB.

The average speech reception threshold was 51.61 dB,and the mean dis-

crimination score was 74.23 percent. Seven subjects in the present

study had losses greater than 70 dB. Davis (1970) stated that within

the range of 70 dB to 90 dB, some people are socially deaf but most are

merely very hard of hearing. The mean hearing loss of subjects in this

study was more severe than that of the subjects in 9 of the 11 studies

of hard of hearing children cited in Chapter II.


TABLE X
MEANS, RANGE, AND STANDARD DEVIATIONS FOR AVERAGE LOSS OF HEARING,
SPEECH RECEPTION THRESHOLD, AND SPEECH DISCRIMINATION SCORE
OF HARD OF HEARING SUBJECTS




SEVERITY OF THE MINIMUM MAXIMUM STANDARD
HEARING LOSS MEAN VALUE VALUE DEVIATION


Average Hearing
Loss (dB) 61.57 30.00 88.00 17.02

Speech Reception
Threshold (dB) 51.61 10.00 99.00 21.41

Speech Discrimina-
tion Score
(percent) 74.23 20.00 100.00 21.21










Performance intelligence

The performance section of the WISC was administered to subjects

who were 16 years old and younger, and the performance section of the

WAIS was given to those subjects who were 17 years old and older. Data

presented in Table XI indicate that the mean performance IQ of this group

of subjects was 100.30, with a range from 74.00 to 136.00. It should

be noted that criteria set for selection of subjects specified that

no children with discernible multiple handicapping conditions would be

included for study. Therefore, no child known to be mentally retarded

was selected as a possible subject.


TABLE XI
MEAN, RANGE, AND STANDARD DEVIATION OF PERFORMANCE INTELLIGENCE
OF HARD OF HEARING SUBJECTS



MINIMUM MAXIMUM STANDARD
MEAN VALUE VALUE DEVIATION


Performance IQ
(WISC, WAIS) 100.30 74.00 136.00 15.39





Speechreading

Speechreading skill was assessed by a live presentation of the -

Utley Lipreading Test, Part I, Sentence Test, Form A. The score reported

was the percentage of the test sentences perceived with reasonable accuracy.

Table XII demonstrates that the mean score for subjects in the study was

41.34 on this test of speechreading ability. The scores ranged from 0-0

percent to 78 percent.











TABLE XII
MEAN, RANGE, AND STANDARD DEVIATION OF SPEECHREADING SCORES
OF HARD OF HEARING SUBJECTS




MINIMUM MAXIMUM STANDARD
MEAN VALUE VALUE DEVIATION


Speechreading Skill
(Utley Lipreading
Test, Part I, Sen-
tence Test, Form A)
Percent correct 41.34 0.0 78.00 21.46




Table XIII compares the speechreading scores of hard of hearing subjects

in this study to a value scale suggested by Jeffers and Barley (1971)

based on their experience with lipreading ability of college students

and hard of hearing adults. According to this interpretation, 15 of the

subjects in this study could be considered poor lipreaders, 10 subjects

might be described as average, and one would be considered a good lip-

reader.

The speechreading scores of older subjects in this study were

not significantly better than those of younger students. The inter-

correlation matrix in Appendix D illustrates that the correlation between

age and speechreading ability was insignificant.












TABLE XIII
INTERPRETATION OF UTLEY LIPREADING TEST SCORES ACCORDING TO
VALUE JUDGMENTS PROPOSED BY JEFFERS AND BARLEY




PERCENT CORRECT ON
UTLEY LIPREADING NUMBER OF
TEST, SENTENCE FORM SUBJECTS INTERPRETATION



90 to 100 0 Excellent

78 to 87 1 Good

53 to 74 10 Average

Below 49 15 Poor



*
Source: Jeffers, Janet, and Barley, Margaret, Speechreading.
Springfield, Illinois: Charles Thomas (1971, p. 341).








Question Two

What are the language characteristics of hard of hearing

children in the study with regard to receptive vocabulary, syntactic

comprehension, auditory language, reading ability, speech articulation

proficiency, and written language?


Receptive vocabulary

The Peabody Picture Vocabulary Test (Dunn, 1959) was admin-

istered to each subject in the study. Test procedures were modified

to allow for presentation of the stimulus word in written as well as in

vocal form. Results were reported as standard scores. Standard scores

of 100 on the PPVT represent the mean raw score of the standardization

sample with the standard deviation set at 15 (Dunn, 1959).

Only one of the 26 hard of hearing subjects scored at or above

the mean for normal hearing children in the standardization sample.

Table XIV illustrates that the mean standard score for hard of hearing

subjects was 73.53, almost two standard deviations below the mean for

normal hearing students of comparable age. It was previously demonstrated

that the hard of hearing subjects performed as well as their normal

hearing agemates on the nonverbal Performance Section of the Wechsler

Intelligence Scales.

Receptive vocabulary (PPVT) scores of subjects in this study were

shown to be significantly related to the socioeconomic status of the

subjects. Similar relationships existed between socioeconomic status

and three other measures of verbal ability.










TABLE XIV
MEAN, RANGE, AND STANDARD DEVIATION OF RECEPTIVE VOCABULARY
SCORES OF HARD OF HEARING SUBJECTS




MINIMUM MAXIMUM STANDARD
MEAN VALUE VALUE DEVIATION



Receptive Vocabulary
(PPVT Standard
Score) 73.53 42.00 109.00 15.45






Syntactic comprehension

A test for comprehension of indirect and direct objects (Scholes,

1973) was used to examine the ability of hard of hearing subjects to

utilize syntactic clues to interpret the meaning of sentences having

double object predicate constructions.

The subjects' interpretation of the unambiguous sentences with

regard to type of disambiguating clue (article or disjuncture) formed

the basis of the analysis of the results. Comprehension of the major

lexical items in the sentences and interpretation of the ambiguous sen-

tences were also noted. Table XV indicates that the mean score obtained

by hard of hearing subjects on this test was 12.84 out of a possible 20

points.

TABLE XV
MEAN, RANGE, AND STANDARD DEVIATION OF SYNTACTIC COMPREHENSION SCORES
OF HARD OF HEARING SUBJECTS



MINIMUM MAXIMUM STANDARD
MEAN VALUE VALUE DEVIATION

Syntactic Comprehension 12.84 6.00 19.00 2.93







73

On this test the subjects heard a sentence, saw the same sentence

written without punctuation, were shown a page with four pictures, and

were asked to point to the picture that corresponded to the sentence.

Two of the pictures were totally inappropriate, referring to completely

different sentences than the stimulus. Two pictures corresponded to

different interpretations of the stimulus sentence. Subjects in this

study pointed to totally inappropriate pictures only 1.4 percent of the

time, indicating that they had no difficulty in comprehending the major

lexical items in the sentences. These types of inappropriate responses

were therefore ignored in the analysis of the results. The subjects

had substantial difficulty, however, in selecting between the two pic-

tures representing different interpretations of each stimulus sentence.

Their mean performance of 12.84 out of a possible 20 suggests that some

of the subjects could have made choices on a purely random basis.

Other analyses of these results include the subjects' inter-

pretation of ambiguous sentences. Table XVI demonstrates that 70 percent

of the ambiguous sentences such as:

he showed her baby pictures

were interpreted by the hard of hearing students to mean:

he showed her the baby pictures.

This is the interpretation which Scholes (1973) referred to as a "B"

interpretation. An "A" interpretation would translate this sentence

to mean:

he showed her baby the pictures.

The hard of hearing subjects in this sample demonstrated a bias toward

a "B" interpretation of ambiguous sentences. The normal hearing subjects

in Scholes' study who were 13 years old and younger demonstrated a similar

bias.







74

As would be expected, the hard of hearing students interpreted

the "B" readings of unambiguous sentences more accurately than the "A"

readings. In fact, when only "A" readings were considered, the hard of

hearing subjects performed at a "chance" level (49 percent) in choosing

the appropriate interpretation.

The hard of hearing students correctly interpreted 63 percent

of the 20 unambiguous sentences. This performance is comparable to

that of the 5-to 7-year-old normal hearing children in the Scholes study.

The type of clue, disjuncture or article placement, was apparently not

important to the correct interpretation of the sentences. Table XVI

illustrates that hard of hearing subjects correctly interpreted article

placement clues 63.84 percent of the time and correctly interpreted dis-

juncture clues 63.07 percent of the time. Since the disjuncture clue

was strictly auditory and the article placement clue was visual and audi-

tory, this finding is surprising.

In the sample of normal hearing subjects (Scholes, 1973) the

ability to use the syntactic clues of article placement and disjuncture

increased gradually over the age groups. No such increase in this ability

was observed as a function of age in the hard of hearing subjects in

this study. The intercorrelation matrix in Appendix D illustrates an

insignificant negative correlation between age and score on the Test

for Comprehension of Indirect/Direct Object Constituents (syntactic com-

prehension).









TABLE XVI
INTERPRETATION OF AMBIGUOUS AND UNAMBIGUOUS SENTENCES
BY HARD OF HEARING AND NORMAL HEARING SUBJECTS


HARD OF NORMAL NORMAL NORMAL NORMAL NORMAL
HEARING HEARING HEARING HEARING HEARING HEARING
N=26 N=20 N=20 N=20 N=20 N=20
MEAN AGE AGE AGE AGE AGE AGE
14.4 5 7 9 11 13

Percent Percent Percent Percent Percent Percent


AMBIGUOUS SENTENCES

B Interpretation 70.0 54.0 64.5 80.0 76.0 84.5

A Interpretation 28.5 43.0 35.0 16.5 18.5 15.5


UNAMBIGUOUS SENTENCES

Correct B
Interpretations 77.3 59.0 78.0 91.0 88.0 96.0

Correct A
Interpretations 49.6 63.5 64.0 71.5 71.0 85.5

Correct Interpretation
of Article Placement 63.8 58.5 74.5 86.0 83.5 94.5

Correct Interpretation
of Disjuncture Clue 63.0 64.0 67.5 76.5 75.5 86.0

Total Correct
Interpretations 63.4 61.25 71.0 81.25 80.5 90.5

*Source Scholes (1973)










Auditory language

Because the scoring procedures used with the subjects in this

study may have differed from those used with the standardization sample,

present scores on auditory language were not compared to test norms

(Myklebust, 1973). The most important use of the auditory language scores

in this study involved their correlations with other variables in the

study. The comparison and contrast between the ability to follow oral

directions (auditory language) and speechreading skill was interesting.

These data are presented in a later section of this chapter.

Under the scoring procedure used in this study, Test One and

Test Two of the Auditory Language Test were combined to yield a possible

total raw score of 46. Table XVII illustrates that the mean score of

subjects in this study was 24.


TABLE XVII
MEAN, RANGE, AND STANDARD DEVIATION OF AUDITORY LANGUAGE SCORES
OF HARD OF HEARING SUBJECTS



MINIMUM MAXIMUM STANDARD
MEAN VALUE VALUE DEVIATION


Auditory Language
(Raw Score) 24.42 5.00 43.00 9.23





Reading ability

Reading ability was measured by the Paragraph Meaning Subtest

of the SAT Intermediate and Advanced Achievement Test Batteries and the

Reading Subtest of the High School Basic Battery. On these subtests,

which measured reading skill, the mean national percentile rank of hard







77

of hearing subjects in the study was 17.30, with a range from the Ist

percentile to the 94th percentile. Stated differently, the mean reading

score of hard of hearing subjects represented a score which was exceeded

by 83 of 100 students in the standardization sample. These data are

presented in Table XVIII.


TABLE XVIII
MEAN PERCENTILE RANK, RANGE, AND STANDARD DEVIATION
OF SAT READING SUBTEST SCORES
OF HARD OF HEARING SUBJECTS



MEAN
NATIONAL MINIMUM MAXIMUM STANDARD
PERCENTILE RANK VALUE VALUE DEVIATION



Reading Subtest
Score 17.30 1.00 94.00 22.51





Speech articulation proficiency

A modified version of the Screening Deep Test of Articulation

(McDonald, 1965) was administered to the hard of hearing subjects in the

study. In this test, nine frequently misarticulated consonant phonemes

are tested in 10 phonetic contexts. Results are reported as total number

of correctly articulated consonant phonemes. Table XIX illustrates that

the mean score for subjects was 75.88 out of a possible 90, with a range

from 36.00 to 90.00. The speech of eight subjects was judged by the

investigator to be within the range of normal.

The most frequently misarticulated consonants were; Es3, Ech]

Esh], Eth], and Cr3. The voiced cognates of the voiceless phonemes










listed are not tested in the Screening Deep Test. The mean number of

correct productions of each of the phonemes tested are listed below:


Is] = 6.88
C13 = 9.46
[r] = 8.92
Cch] = 7.34
Cth3 = 7.92


[sh] = 7.84
Ek] = 8.23
Ef3 = 9.46
Et] = 9.38


The same information is illustrated graphically in Figure 1.


TABLE XIX
MEAN, RANGE, AND STANDARD DEVIATION OF SPEECH ARTICULATION
PROFICIENCY SCORES OF HARD OF HEARING SUBJECTS




MINIMUM MAXIMUM STANDARD
MEAN VALUE VALUE DEVIATION


Speech Articulation
Proficiency
(Screening Deep Test
Total Correct) 75.88 36.00 90.00 12.69
















Es] EC13 Er [ch] Eth] Esh] Ek) Cf] Et]


Phonemes Tested


FIGURE 1
MEAN CORRECT PRODUCTIONS OF EACH CONSONANT PHONEME
BY HARD OF HEARING SUBJECTS ON THE
SCREENING DEEP TEST OF ARTICULATION


Correct
Productions
10

9

8

7

6

5

4

3

2

1

0









Written language

The Picture Story Language Test (Myklebust, 1965) was admin-

istered to the hard of hearing subjects in the study as a measure of

the accuracy of their written language. In order to measure the corre-

lations between written language ability and other variables in the

study, the syntax quotient for each subject was converted to a percentile

rank for age and sex using the tables provided by the test author (Mykle-

bust, 1965). The mean of the percentile ranks of hard of hearing subjects

for syntax quotient was 23.88, with the minimum value at the 2nd per-

centile and the maximum at the 98th percentile. These data are presented

in Table XX.


TABLE XX
MEAN, RANGE, AND STANDARD DEVIATION OF WRITTEN LANGUAGE
SCORES OF HARD OF HEARING SUBJECTS




MINIMUM MAXIMUM STANDARD
MEAN VALUE VALUE DEVIATION


Written Language (Syntax
Quotient, Picture Story
Language Test)
Percentile Rank 23.88 2.00 98.00 30.07





Table XXI illustrates the means of the subjects' raw scores on

the productivity scale, the syntax scale, and the abstract-concrete

scale of the Picture Story Language Test. Comparisons of these means

to normative data provided by Myklebust (1965) demonstrate that the

written language performance of these hard of hearing subjects with a

mean chronological age of 14.5 was comparable to that of 7-to 9-year-old

normal hearing children.









TABLE XXI
MEANS AND AGE EQUIVALENTS ON THE PICTURE STORY LANGUAGE TEST
FOR PRODUCTIVITY, SYNTAX QUOTIENT, AND ABSTRACT-CONCRETE
SCORES BY HARD OF HEARING SUBJECTS




RAW SCORE MEAN AGE EQUIVALENT*


PRODUCTIVITY

Total Words 75.08 8 to 9
Total Sentences 7.52 8
Words Per Sentence 9.52 9




SYNTAX QUOTIENT 85.96 7


ABSTRACT CONCRETE 10.88 8


*Source: Myklebust (1965)


The types of grammatical errors made by the hard of hearing

subjects are summarized in Table XXII. Omissions and substitutions

were the most common errors involving the use of words. Omissions

were also the most common errors involving word endings and punctuation.










TABLE XXII
CATEGORIES AND TYPES OF ERRORS MADE BY HARD OF HEARING SUBJECTS
ON THE PICTURE STORY LANGUAGE TEST


PICTURE STORY LANGUAGE TEST


Error Type:




Additions


Omissions


Substitutions


Word Order


WORD USAGE

Total Mean No.
Errors Errors


25.0 1.0



56.0 2.2


46.0 1.84


5.0 0.19


CATEGORIES

WORD ENDINGS

Total Mean No.
Errors Errors


4.0 0.16


43.0 1.72


11.0 0.44


PUNCTUATION

Total Mean No.
Errors Errors


9.0 0.36


109.0 4.36


0.0 0.0










Question Three

How do hard of hearing children rank academically in comparison

to national norms for students at their grade level?


Academic performance

The principal source of data concerning the academic performance

of subjects in this study was results of the Stanford Achievement Tests.

For the purpose of determining correlations between academic performance

and other variables in the study, the average of the percentile scores

for all subtests of the most current achievement test available was

reported as "academic performance" for each subject.

Table XXIII illustrates that the mean percentile rank for hard

of hearing subjects in this sample was 16.15. Stated differently, 83

out of every 100 students in the national standardization sample scored

higher on overall academic achievement than did the hard of hearing

subjects in this study even though the performance intelligence of the

hard of hearing subjects was within the range of average.


TABLE XXIII
MEAN, RANGE, AND STANDARD DEVIATION OF ACADEMIC PERFORMANCE
OF HARD OF HEARING SUBJECTS



MINIMUM MAXIMUM STANDARD
MEAN VALUE VALUE DEVIATION


Academic Performance
(Average of the National
Percentile Scores on
All Subtests of the
SAT) 16.15 3.00 78.00 16.29







84

Table XXIV illustrates the national percentile rank for each

subject on each subtest of the SAT. Five subjects for whom current

SAT scores were not available were not included in this tabulation.

The highest mean percentile rank obtained by hard of hearing

students taking the Intermediate and Advanced Batteries of the SAT

was the 22nd percentile on the spelling subtest. The highest mean per-

centile rank obtained by the four students taking the High School

Basic Battery was the 30th percentile on the math subtest. The lowest

scores obtained on the tests were in the areas of science and social

studies.
















HHO OI I I I
0 ) (N c |





N000) 0 I)H0)H0(N0
(NOOHNOHH 00000000
. . . ~~. . .


H C0)0 0Hl(-lNl







0l 0000l^r 00
r iO N0 0
0 0 0CO000 0


.0 0 (N ON HN
0000N HN


. ON m H o 0 V 0 H o N -0 0 (N a H








S H(NOOOO0)0H0 ) 00
OHOO0)OmO(NHOOHH H
o ~ mI


D0 0 0) 00 Hi 0 0 'I N 0D 0


0 -H


CO O N cO N MONO 00 0 DID
O O ON O N IOO O









0 NOON00N 0 IN 0 0 00
OH O NOHH OOOOH









Hi (N nl 0 m0 D 0 0 0 H (N mm r) C n 0 D 0
i-l ml -d il ^O i- i- i-


0
O
N 1













(0

-i









0






H
m -
U







0
H


H M




0( M




H
0 I






U










H
M Q
H Z


1i Ln CN c













N ON N v r n
N0001 0












o o-1 o














o 0 o rN o
O i N O
OH0O D N








NQ CO04 O
om(mn (










00o0N i
o*o* *


H 0H
0 (N 00
o* *o










Question Four

What significant correlations exist among the following variables:

average hearing loss, speech reception threshold, speech discrimination,

receptive vocabulary, auditory language, syntactic comprehension, read-

ing ability, speech articulation proficiency, written language, per-

formance intelligence, age, speechreading skill, and grade point average?


The data presented in answer to this question concern correla-

tions among variables. A complete intercorrelation matrix of all num-

eric variables in the study may be found in Appendix D. All tabulations

illustrating the correlations between variables include the coefficient

of correlation and the significance level of each coefficient of corre-

lation. In each table the coefficient of correlation appears first

and the significance level appears directly under it.

In order to facilitate discussion and illustration, a certain

number of variables have been arranged in clusters. (A cluster is an

arrangement of variables that have some common bond.) These variables

have been identified as clusters:

Severity of the hearing loss:

Average loss of hearing
Speech reception threshold
Speech discrimination score

Receptive language:

Receptive vocabulary
Auditory language
Syntactic comprehension
Reading ability

Expressive language:

Speech articulation proficiency
Written language










Because they were not easily grouped, the variables performance

intelligence, age, speechreading skill, and grade point average were

illustrated individually.


Severity of the hearing loss

Variables which measured the severity of the hearing loss were

average loss of hearing, speech reception threshold, and speech discri-

mination score. All three of these variables were significantly related

to speechreading skill. The subjects with more severe hearing impair-

ments were apparently better speechreaders than were the subjects with

less severe hearing impairments. Average loss of hearing and SRT were

positively correlated with speechreading while speech discrimination

score and speechreading skill were negatively correlated. These data

are illustrated in Table XXV.


TABLE XXV
CORRELATIONS BETWEEN THE VARIABLES ASSOCIATED WITH THE SEVERITY
OF THE HEARING LOSS AND SPEECHREADING SKILL



SPEECHREADING SKILL*


SEVERITY OF THE
HEARING LOSS:

Average Loss of Hearing 0.551
0.007

Speech Reception Threshold 0.393
0.044

Speech Discrimination -0.417
0.032




*The upper number in each pair represents the coefficient of corre-
lation between the two variables and the lower number represents the
corresponding significance level.







88


There was no significant relationship between the severity of

the hearing loss and any type of linguistic skill measured in this study

even though the 30 dB to 88 dB range of hearing impairments in the study

is relatively large. This finding supports the position that the degree

of actual hearing deficit is less important than other factors may be

in determining the language facility of the individual.

A significant negative correlation existed between speech dis-

crimination score and age. This unexpected relationship may be attri-

buted to the fact that one 18-year-old subject was unable to respond

to the discrimination test at the limits of the audiometer. This

subject was assigned an arbitrary discrimination score of 20 percent,

which was considerably lower than any other score on the test. This

extreme score may have distorted the relationship between the variables

of age and discrimination score.


Receptive language

The variables representing receptive language were: receptive

vocabulary, syntactic comprehension, auditory language, and reading.

Receptive vocabulary, represented by the PPVT score, measures one-word

receptive vocabulary, and the other three variables assess the ability

to comprehend sentences. It has been previously demonstrated that the

hard of hearing subjects in this study performed well below the established

means for normal hearing students on these receptive language variables,

that socioeconomic status was significantly related to two of the four

receptive language variables, and that the severity of the hearing loss

was not significantly related to any of the receptive language variables.

Significant positive correlations existed between most of the








language variables in the study. Tables XXVI, XXVII, and XXVIII illus-

trate the correlations between the receptive language variables and

other variables in the study. Significant positive relationships

existed between the subjects' understanding of the meaning of single

words (receptive vocabulary) and their ability to comprehend syntactic

clues to sentence meaning, their ability to follow oral directions (audi-

tory language), and their ability to read. Table XXVI demonstrates that

the significance level of the relationship between receptive vocabulary

and other receptive language variables ranged from 0.001 to 0.009.


TABLE XXVI
INTERCORRELATIONS AMONG RECEPTIVE LANGUAGE VARIABLES



RECEPTIVE* SYNTACTIC AUDITORY READING
VOCABULARY COMPREHENSION LANGUAGE ABILITY


RECEPTIVE 0.620 0.497 0.568
VOCABULARY 0.001 0.009 0.002

SYNTACTIC
COMPREHENSION 0.620 0.326 0.584
0.001 0.099 0.002

AUDITORY
LANGUAGE 0.497 0.326 0.417
0.009 0.099 0.031

READING
ABILITY 0.568 0.584 0.417
0.002 0.002 0.031



The upper number in each pair represents the coefficient of
correlation between the two variables and the lower number represents
the corresponding significance level.











The only intercorrelation between the receptive language variables

that failed to reach significance at the 0.05 level was that between

auditory language and syntactic comprehension. All other variables

representing receptive language abilities were significantly correlated.

Table XXVII illustrates the coefficients of correlation between

receptive and expressive language variables. Expressive language was

assessed by speech articulation proficiency and written language. Speech

articulation proficiency refers to the total articulation score on the

Screening Deep Test of Articulation, and written language is represented

by the syntax quotient on the Picture Story Language Test.


TABLE XXVII
CORRELATIONS BETWEEN THE RECEPTIVE LANGUAGE VARIABLES
AND THE EXPRESSIVE LANGUAGE VARIABLES




SPEECH ARTICULATION* WRITTEN
PROFICIENCY LANGUAGE


RECEPTIVE VOCABULARY 0.447 0.752
0.013 0.0001

SYNTACTIC COMPREHENSION 0.220 0.382
0.278 0.051

AUDITORY LANGUAGE 0.392 0.625
0.044 0.0009

READING ABILITY 0.219 0.385
0.280 0.049




*The upper number in each pair represents the coefficient of
correlation between the two variables and the lower number represents
the corresponding significance level.







91

A positive correlation, significant at the 0.0001 level, existed

between the subjects' receptive vocabulary and the grammatical accuracy

of their written language. The ability to follow oral directions

(auditory language) and the grammatical accuracy of the subjects' written

work were significantly correlated at the 0.0009 level. Receptive voca-

bulary was significantly related to speech articulation proficiency at

the 0.02 level, and auditory language and speech articulation proficiency

were correlated at the 0.05 level.

Table XXVIII demonstrates the correlations between the receptive

language variables and age, performance IQ, and grade point average.

No coefficient of correlation between the receptive language variables

and age reached significance; however, in each case the correlation was

in a negative direction. Receptive vocabulary and performance IQ were

significantly correlated at the 0.039 level. Receptive vocabulary was

also related significantly to grade point average at the 0.02 level.











TABLE XXVIII
CORRELATIONS BETWEEN THE RECEPTIVE LANGUAGE VARIABLES
AND AGE, PERFORMANCE INTELLIGENCE,
AND GRADE POINT AVERAGE


PERFORMANCE GRADE POINT
AGE* INTELLIGENCE AVERAGE


RECEPTIVE -0.245 0.401 0.444
VOCABULARY 0.224 0.039 0.021

SYNTACTIC -0.289 0.326 0.299
COMPREHENSION 0.148 0.099 0.134

AUDITORY
LANGUAGE -0.057 0.172 0.292
0.775 0.597 0.144

READING -0.134 0.152 0.013
ABILITY 0.517 0.537 0.944


*The upper number in each pair
correlation between the two variables
the corresponding significance level.


represents the coefficient of
and the lower number represents




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs