_,

V

,_

,

__: ? E _,

"‘

fun

van's

't‘i

 

0.“.

I .0
.. a.
.0. 6

x...

I...

‘40

Nth.»

fix...

THESIS

LIBRARY

Michlgan State

University

 

ROOM use om

Lid am 7’33"“; Pun-w -
U 4:”: ‘4

 

ABSTRACT

AN INVESTIGATION OF THE USE OF QUANTIFICATION
IN ARTICULATION TESTING OF
FUNCTIONAL ARTICULATORY DISORDERS

by Norma J. Smith

Testing for articulatory disorders constitutes an
important part of the work of the public school speech
correctionist. Effective instruments in carrying out
such testing are imperative as children with articulatory
disorders can compose up to 85 per cent of the case load
in a public school. .

The main purposes of this study were to construct a
short descriptive articulation test which would also be
more comprehensive and to devise a method of obtaining the
percentage of an individual's improvement.

An articulation test which yielded three indicies
was constructed. These were an.Articulation Score, a
Stimulability Score and an Improvement Index. The test
was administered to a sample of 25 school children. .A
SO-second sample of their conversational speech was recorded.
This tape was Judged by 12 graduate students in Speech and
Hearing Science. The Judges used two types of rating scales;
an ungrouped rating scale and a grouped rating scale. The
hypotheses tested concerned the relationship between the
number of consonant sounds misarticulated and the Judgement

of the severity of that articulatory disorder along an

Norma J. Smith

ungrouped rating scale and along a grouped rating scale.

From the data obtained it was concluded that a
positive significant relationship at the .05 level of
confidence did exist between the number of test sounds mis-
articulated and the Judgement of severity of the articula-
tory disorder using both rating scales.

Further research regarding the indices obtained from
the test, particularly the Improvement Index, may yield more
information as to the possibility of their serving as an

estimate of the prognosis for speech therapy.

AN INVESTIGATION OF THE USE OF QUANTIFICATION
IN ARTICULATION TESTING OF
FUNCTIONAL ARTICULATORY DISORDERS

By

Norma J. Smith

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF ARTS

Department of Speech

1965

TABLE OF CONTENTS

Chapter Page
I . STATEMENT OF THE PROBLm. O O O O O O O O O O O 1
Introduction. . . . . . . . . . . . . . . . 1

Purpose of Study . . . . . . . . . . . . . . 2
HYPOthCSCS o o o e o o o o o o o e o e 3
Importance of Study. . . . . . . . . . . . . A
Definition of Terms. . . . . . . . . . . . . 5
Organization of the Thesis . . . . . . . . . 8

II. REVIEW OF THE LITERATURE. . . . . . . . . . . . 9
Introductiono . . . . . . . . . . . 9
Incidence of.Articulatory Disorders. . . lO
Articulation Testing . . . . . . . . . . ll

Maturation and Testing . . . . . . 13
Stimulability and.Articulation Testing . . . 1h
Measurement and Articulatory Defects . . . . 16
IntFOdUCtion o o e o o o o o o o o o o e 16

Later Research . . . . . . . . . . 18

Scaling and Articulatory Defects . . . . 22

Listener Rating and.Articulatory Defects . . 2h

III. SUBJECTS, MATERIALS, EQUIPMENT AND PROCEDURES . 28

SUbJCCtS o o o e o o o o o o o e o o o o e o 28
Materials and Equipment. . . . . . . . . . . 28
Procedures 0 C O C O C C O O O O O O O O 9 O 29

sumary. O O O O O O O O O O C O O O O O 32

o
o
o
o
C»)
(A

IV. RESULTS AND DISCUSSION . .
Results. 0 O O O O O O O O O O O O O O O O O 33
Discussion . . . . . . . . . . 36

V. SUMMARY AND CONCLUSIONS . . . . . . . . . . . . ul

Smarye o e o e o o o e o e o . . . . . . . 1-4-1
ConCluSIOns. C C O C O O C O O O O O O C [4-2
Implications for Further Research. . . . MS

BIBLIOGRAPI-IY. O O O O C O O O O O O O O O O O O O I O LLLI’
APPENDICES. O O O O O O O O O O O O O O O O O 0 O O 0 LL?

ii

CHAPTER I

STATEMENT OF THE PROBLEM

Introduction

 

Speech is a very important asset to mankind. Its use
among family, friends and at national and international
conference tables is almost priceless. Successes or failures
in our communication through the medium of speech determines
to a great extent the way the world turns.

The effectiveness of speech depends not alone upon mere
skill in using words but also upon the mechanics involved
in the utterance of these words. The manner in which a
verbal communication is articulated may hold a relative de-
gree of importance to what is said. Hence speech may be more
effective if it is correctly articulated.

The authors of the book Speech Handicapped School
Children maintain that disorders of articulation account
for the maJority of speech defects found among public
school children. The revised edition of the book cites
70 to 85 per cent of the children seen by the public school

correctionist as having defects of the articulatory type.1

 

#—

lWendell Johnson, Spencer J. Brown, James Curtis,
Clarence U. Edney and Jacqueline Keaster, Speech Handi-
capped School Children, 2nd ed. rev.; (New YorK} Harper
and raw, 1956), p. 9; .

1

The maJority of these children have functional articula-
tory disorders. Anderson gives 75 per cent as the inci-
dence of articulatory disorders found among the speech de-
fectives in the public school population.1 A.questionnaire-
type survey of the clinical practices and remedial proce-
dures of a nationwide representative sample of 7&9 speech
therapists revealed that the caseload of the average
therapist was 130. This survey also revealed that the
average percentage of the articulation cases for the
therapists who reported was 81 per cent.2

With such a high percentage of the Speech therapist's
caseload being composed of children with articulatory dis-
orders testing for such disorders becomes important. An
effective and comprehensive articulation test particularly
for use in describing the articulatory behavior of children

would be an aid to the public school speech correctionist.

Purpose of Study

 

The purpose of this study is two-fold: (l) to devise
a short, descriptive and comprehensive articulation test
which may be used in describing the articulatory disorders

of individuals and (2) to suggest a method whereby an index

 

1Virgil A. Anderson, Improving the Child's Speech,
(New York: Oxford university Press, 1953), p. I26.

gMyfanwy E. Chapman, Ester Herbert, Charlotte Avery,
John Selmar, "Clinical Practice: Remedial Procedures,"
Journal of Speech and Hearing Disorders, Monograph Supplem

ment 8 (I961T, pt 59.

of the individual's improvement may be obtained.

Such a test would help to assure the maxium benefit
of the speech therapy pragram to the maJority of those
who receive its services. The test may be of assistance
to the public school therapist in the following ways:
(1) selecting a caseload, (2) giving an indication of the
child's ability to correct his misarticulated sounds,
(3) aiding the therapist in planning therapy, (h) describ-
ing the severity of the articulatory problem. I

Hypotheses

 

The clear, intelligible articulation of speech sounds
is one aspect of effective communication. When an ex-
cessive number of misarticulated sounds appear in one's
speech they may be noticed by the listener. If this is
true it would suggest that some type of relationship may
exist between the number of misarticulated sounds in one's
speech and the awareness of these errors by a listener.

The present study investigates the following two hypotheses.

Hypotheses.--There is a relationship between the

 

number of consonant sounds misarticulated and the Judge-
ment of the severity of that articulatory disorder along

an ungrouped rating scale.

There is a relationship between the number of consonant
sounds misarticulated and the Judgement of the severity of

that articulatory disorder along a grouped rating scale.

u

The terms grouped and ungrouped, as pertaining to

rating scales, will be defined subsequently.

Importance of Study
Articulation defects have been recognized as the most
prevalent of speech defects among public school children.
Studies concerning many aspects of articulatory disorders ?2
have been done. In comparison relatively few of these
studies have been directly concerned with the quantifi-

cation of articulatory defects.

 

‘flh‘uh'im ——— — _ _

The preposed study differs from similar ones in two
ways. First, the Descriptive.Articulation Test used in
this study contains the most frequently misarticulated

1

sounds according to Roe and Milisen. Six sounds from the

Wood Articulation Index with a weighted value of more than
.90 in two or more positions were also included in the test.2
Secondly, this study seeks to provide information which

‘will be helpful in devising a method whereby some indin
cation of an individual's ability to improve misarticulated
sounds may be obtained.—

It is hoped that from this study information may be

gained as to the usefulness and practicableness of a

 

1Vivian Roe and Robert Milisen, "The Effect of Matura-
tion Upon Defective Articulation In Elementary Grades,"
Journal of Speech and Hearing_Disorders, 7 (19h2), 9. us.

2Charles Van Riper and John Irwin, ‘Voice and.Articu-
lation, (Egglewood Cliffs, N.J.: Prentice-Hall, Inc.,
, p. l .

S

descriptive articulation test which yields certain numeri-
cal indices. It is further heped that from such a test as
the proposed one more comprehensive information may be
gained which will aid the therapist in planning and ex-
ecuting more effective therapy. The desire of the writer
is that any information which may be gained from this study
be of such a nature as to contribute to the development of

and greater understanding of articulation testing.

Definition of Terms

For the purposes of this study certain terms employed
have been defined in the following manner:

Defective Articulation.--The incorrect production
of a speech sound involving either substitutions, omissions,
distortions or additions. Substitution is the replacing
of one phoneme with another. OmisSion is leaving out a
phoneme where it should occur. Distortion is the modi-
fying of the production of a speech sound so that the
acoustic result is inaccurate. .Addition is interpolating
or adding sounds which are not part of the word.1

Using Powers's definition a functional articulatory
disorder is "an inability to produce correctly all of the

standard speech sounds of the language, an inability for

which there is no appreciable structural, physiological

ne-
L.

 

 

F7.

IMargaret Hall Powers, “Functional Disorders of
.Articulation: Symptomatology and Etiology,“ Handbook 0f

Speech Patholo , ed. Lee E. Travis, (New York:.AppIeton-
Century Crofts, Inc., 1957). PP. 713, 71h.

 

or neurOIOQical basis in the speech mechanism or its

supporting structures."1

The Articulation Score yielded by the test is defined

 

as the percentage of test sounds misarticulated by the
'testee following picture stimulation. This score is
obtained by using the following formula in which twenty-
one is the total number of sounds contained in the test

and P represents the number of sounds which the subject

 

misarticulates by picture stimulation.

.Articulation Score = 100 X 2:"?

The Stimulability Score indicates the percentage of

 

test sounds which the testee continues to misarticulate
after the examiner has correctly produced the test sound
three times. This score is obtained by using a formula in
which the letter V represents the number of incorrectly
produced sounds which were misarticulated after voice
stimulation.

.. .. m _, ,. ‘7l-V
Stimulability Score -. 100 X 5-27-

The Improvement Index indicates the percentage of
correction of misarticulated sounds. This index is found
by employing the following formula in which A represents
the Articulation Score and 5 represents the Stimulability
Score:

. . a - u SqA
ImprOVement Index — 100 X —§—-

libid” p. 708.

The picture test which ‘was constructed for this in—

vestigation is called the Descriptive Articulation Test.

 

This test is composed of twentynone frequently misarticu-
lated consonant sounds. .All test sounds are presented in
the medial position only. An Articulation Score (A),
Stimulability Score ‘8) and an Improvement Index (I) may
be obtained with the test. The test form is presented

in Appendix c. '

Judged Severity is defined as listener reaction to

 

the speech samples of functional articulatory disorders.
These reactions are indicated along a nine-point scale
in which one represents least severe articulation and
nine represents most severe articulation.

The ungrouped rating scale used in the study is a

 

nineapoint scale along which the Judges ratings of the

speech samples are placed. The grouped rating scale is

 

a nineupoint scale which has been divided into three groups
of 3 points each and each group has been given a verbal
description. The above rating scales are presented in
Appendices D and E.

The trained speech therapists involved in the study
are graduate students in the area of Speech and Hearing
Science in the Department of Speech at Michigan State

University.

 

 

Organization of the Thesis

 

Chapter I contains the statement of the problem which
led to the present study. Sub-headings include an intrOm
duction to the study and the purpose for the study. The
hypotheses considered in this study along with the impor-
tance of the study are also discussed in Chapter 1. This
chapter concludes_with the definitions of certain terms

used in the study. The review of the literature pertinent

 

to this study composes Chapter 11. Chapter Iii discusses

 

the materials, subjects, equipment and procedures em«
ployed in the collection of the data for the study. The
results of the study are presented in Chapter IV along
with a discussion of these results. Chapter V consists
of a summary of the study and conclusions drawn from the

study. Implications for further research are also dis-

cussed in this final chapter.

CHAPTER II

REVIEW OF THE LITERATURE

Introduction

 

Many speech therapists may hold their own views in
regard to the efficiency of and indeed, the possibility

of effectively quantifying articulation defects. _Some

 

research studies have been done in the area of quantifica-

 

tion and articulation testing. Among them are studies

1

by Perrin, Morrison,2 Parobeck,3 Jordan,LL and Carter

5

and Buck. These and other studies have investigated

 

1Elinor Horwitz Perrin, "The Rating of"Defective
Speech by Trained and untrained Observers," ‘Journal of
Speech and Hearing Disorders, 19 (l95h),- p. 48.

,?Shelia Morrison, "MeaSuring the Severity of Articu-
lation Defectiveness," Journal of Speech and Hearing Dis~
orders, 20 (1955), p. ShT. .

3Donna J. Parobeck, "An Investigation of the Utility
of Wood's Articulation Index as an Independent Measure of
Articulation Proficiency,“ (unpublished M.A. thesis,
Bowling Green State University, 1956).

 

 

“Evan P. Jordan, "Articulation Test Measures and'“
Listener Ratings of Articulation Defectiveness," Journal
of Speech and Hearing Research, 3 (1960), p. 303._

SEunice T. Carter and McKenzie Buck, "Pregnostic
Testing for Functional Articulation Disorders Among Children
in the First Grade," Journal of Speech and Hearing Dis-
orders, 23 (1958), p. l2h.

10

directly or indirectly the applicability of quantification
to diagnostic articulatory testing.

The studies discussed in the remainder of this chapter
are organized chronologically as far as possible under
major headings and sub-headings which are relevant to the

present study.

Incidence of Articulatory Disorders.--Investigations
into the incidence of speech defects among public school
children have revealed that a substantial incidence of
such defects do exist. Of these children with speech de-
fects those with functional articulatory defects comprise
the majority.

In 1916, Wailin conducted an investigation by means
of a questionnaire among the public school pupils in
St Louis. Of the 89,057 pupils included in the study
2.8 per cent or 2,536 were judged by their teachers as
having speech defects.1

Louttit made a survey of speech defects among the
public schools in lndiana in 1936.2 Questionnaire blanks
were sent to the principals of 3,717 public schools in the
state. Of the 1,223 blanks which were returned the gross

incident of speech defects among a represented enrollment

 

1J.E. Wallace Wallin, ”A Census of Speech Defectives
Among 89,057 Public School Pu ils: A.Preliminary Report,”
School and Society, 3 (1916 , p. 213.

. 2c. M. Louttit and E. c. Hall, "1936 Survey br
Speech DGfects Among Public School Children of Indiana,"
Journal of Speech Disorders, 1 (1936), P. 73.

 

ll

of 199,839 children was 3.7 per cent. And of those children

said to have speech defects 79 per cent had articulatory
disorders.

The report of a speech survey during the school year
l9hO-l9hl made by Mills and Streit in Holyoke, Massachuetts
among the first three grades revealed 73.2 per cent as
the total incidence of misarticulated sounds.1

Seventy per cent of the speech defects in children
are of the articulatory type. This was the finding of
Wood in his study involving parental maladjustment and
functional articulatory defects among children.2

These and other similar studies have led to an agree~
ment that speech defects of an articulatory nature are
the most prevelant in school children. In order to detect
more accurately such articulatory problems special
articulation tests have been developed. In addition to
the standard tests many speech therapists prefer to dem
vise their own individual tests.

Articulation Testing.-~According to Templin the

 

articulation of speech sounds is assessed for two purposes:

u...

IA. C. Mills and H. Strait, "l9h2 Report of a Speech
Survey - Holyoke, Massachuetts," Journal of Speech Diem
orders, 7 (l9h2), p. 161.

2Kenneth S. Wood, "Parental Maladjustment and
Functional Articulatory Defects in Children," Journal
of Speech and Hearing Disorders, 11 (19h6), P. 256.

 

 

12

(l) “to determine correctness or incorrectness of specific
sounds" and (2) "to determine the general adequacy of arti-
culation,"1 However, Templin regards the chief purpose
of testing as screening individuals with acceptable arti-
culation from those with unacceptable articulation. She
developed a nonmdiagnostic articulation test for ages three
through eight which measured the child's spontaneous
utterance in response to a picture, his utterance repeated
after the examiner and his repeated utterance without the
picture. The test consisted of fifty items. Norms were
obtained for this test on hBO subjects.2
The diagnostic type of test is more comprehensive
than the nonodiagnostic articulation test. This type of
test was utilized in the Carter and Buck study concerning
a method for the prognostic testing of functional articu-
lation disorders among first grade children.3 The children
were administered three tests. Test one was designed to
evoke a spontaneous response. Test two was an imitation
type test in which the child was asked to watch the examiner,
listen carefully and to repeat the word in the same manner

it was presented. The third test utilized the same sounrs

 

lMildred C. Templin, "A.NonuDiagnostic Articulation
Test," Journal of Speech and Hearing Disorders, 12 (19h?),
P. 392.

2Mildred C. Templin, ”Norms On a Screening Test of
Articulation for Ages 3-8," Journal of Speech and Hearing
Disorders, 18 (1953), p. 323.

 

3Carter and Buck, Op. cit.

13

in tests one and two but in nonsense syllables. Their
basic reason for employing test three was to evaluate

the child's ability to change a defective articulation

pattern instantaneously.1

Prognostic Testing of articulatory disorders has
been used with younger children. Farquhar investigated
the prognostic value of imitative and auditory discrimina-
tion tests with kindergarten children.2

Maturation and Testing,--Studies seem to show that

 

physical maturation does appear to have some effect on
functionally misarticulated sounds.

In l9h2 Roe and Milisen administered a modification
of the Detroit Articulation Picture Test to 1,989 children
in grades one through six in the public schools of nine
Indiana cities. One of their conclusions was that the

"mean number of errors decreased as the grade level increas~

ed."3

,A continuation of the above study was done by Sayler

with grades seven through twelve.LL The data showed a

1Carter and Buck, 0E° 91339 p. 127.

2Mary Stuart Farquhar, ."Prognostic Value of Imitative
and Auditory Discrimination Tests, Journal of Speech and
Hearing Disorders, 26 (1961), p. 3h2.

 

 

3Vivian Roe and Robert Milisen, "The Effect of
Maturation Upon Defective Articulation in Elementary Grades,"
Journal of_Speech and Hearing Disorders, 7 (19h2), p. 50.

uHelen K. Sayler, "The Effect of Maturation Upon
Defective Articulation in Grades Seven through Twelve,"
Journal of Speech and Hearing Disorders, in (19h9), p. 207.

1h

minimal amount of improvement between grades seven and ten
and between grades ten and eleven. There was practically
no change between grades eleven and twelve.

The Steer and Drexler study was a longitudinal study
of the articulation of 93 kindergarten children over a
five year period of time. They concluded that "if the
tests (articulation tests) were administered atithe be-

ginning and'end of the kindergarten year, allowing more

time for maturational effects, the predictive value of the

improvement scores would be strengthened."1

These studies suggest that during the child's younger

years maturation is taking place relatively rapidly. Signi-

ficant changes can be detected in the child's ability to
correctly articulate any functionally misarticulated

sounds which he might have. As maturation slows down,

comparatively little improvement can be noted in the correc-

tion of functionally misarticulated sounds.

Stimulability and Articulation Testing

 

Comprehensive diagnostic articulation testing includes

a stimulability test. In such a test the child is stimu»
lated with his misarticulated sound several times; then he

is asked to produce the sound himself. This test was

 

l

Articulation Ability From Kindergarten Tests," Journal
of Speech and Hearing Disorders, 25 (1960), p. 397.

M. D. Steer and Hazel G. Drexler, "Predicting Later

 

15

designed to indicate the childis ability to improve his
misarticulated sounds. According to Milisen often "children
are able to improve as many as 85 per cent of their mis-
articulated sounds after receiving only a few integral

"1 Milisen stresses the importance of vivid:

stimulations.
ness, forcefulness, and completeness in stimulation as
the reproduction of the sound by the child is dependent upon
the effectiveness of the stimulation. He regards the
Stimulability Test as one which records the changes occuru
ing in response to stimulation for each misarticulated
sound in isolation, nonsense syllables, and words.2

Templinls NonuDiagnostic Articulation Test also con“
tains a stimulability test. One of the measures obtained
in her test was the childls utterance of his misarticulated
sound repeated after the examiner.3

Though called by a different name the Carter and Buck
Prognostic Test for Functional.Articulation Disorders in-
cludes a stimulability test. Test two of the Carter and

Buck test is called an Imitation Test. The child being

tested is asked to watch the examiner, to listen carefully

 

1Robert Milisen, "Methods of Evaluation and Diagnosis
of Speech Disorders," handbook of Speech Patholog*, ed. Lee
Edward Travis, (New York: AppletonnCentury Crofts, Inc.,
1957)., p0 2919 .

2Robert Milisen, "A Rationale for Articulation Dis-
orders,” Journal of Speech and Hearing Disorders, Mono»

graph Supplement h (19SH), pp. 10, Th.

3Templin, op. cit., p. 393.

 

 

 

 

16

and to repeat the word in the same manner as it was presented
to him.1

in their folloWuup study to the Carter and Buck study
Steer and Drexler employed the same procedure. They be=
lieved that "one way of measuring the child‘s ability to
learn better speech patterns is to stimulate the child with
the correct sound and to observe how many of his errors he
can modify or eliminate.2

The Prognostic value of imitative and auditory dis~
crimination tests was investigated by Farquhar. Among her
conclusions she found that children with severe articula~
tory defects could receive profit from strong stimulation
of the sound in isolation before the sound is presented to

them in words.3

Measurement and Articulatory Defects

 

Introduction.umEariy research in the area of articu-

 

latory defects did not emphasize exact measurement as such.
The evaluation of articulatory defects was largely sub-
jective. In an effort to establish satisfactory accuracy

in articulation testing. Henderson suggests that four
statistical measures be determined. They are "consistency

of the examinerls judgements, objectivity in the-examiner's

 

1Carter and Buck, op. cit.

2Steer and Drexler, op. cit.

3F3,rqu:h.ar5 0P0 Cit-Loy pa 3&6.

l7

Judgements, consistency of subject responses and the co»

efficient of reliability."1

Recognizing the need for more

objectivity in articulation testing Henderson recommends

that extensive and well controlled research be initiated

by way of "determining the best technique for establish»

ing objectivity in judgement and setting up standards of IF
consistency and objectivity in judgement."2 ;

In a study of maturation and articulation defects in

 

the elementary grades Roe and Milisen simply noted the 4“
number of defective sounds. In addition to concluding that E
the mean number of errors decreased as the grade level -
increased they also noted that the percentage of children
making various kinds and degrees of errors decreased from
the first grade to the sixth grade on 80 per cent of the
sounds tested.3

An attempt at more objectivity in the measurement
of articulatory defects was made in 19h3 by Curry, Kennedy,
Wagner and Wilke when they constructed a phonographic

n. ,W
scale to measure defective articulation.‘ The scale was

 

 

1Florence M. Henderson, "Accuracy In Testing The
Articulation of Speech Sounds,” JOurnal of Educational
Research, 31 (1938).

21b11.

 

3Roe and Milisen. og._gii.

LLRobert Curry, Lou Kennedy, Loretta Wagner and Walter
Wilke. "A Phonographic Scale for the Measurement of Defective
Articulation," Journal of Speech and Hearing_Disorders, fl
(19u3), p. 123.

18

constructed with steps along a scale of defectiveness

from normal to unintelligible speech. Using the method of
paired comparisons a group of twenty-five Judges selected

the better of two recorded speech samples. The resulting
phonographic scale was then made available commerically.

This scale was designed so that the quality of articulation
of a speech defective could be rated against the scale
samples.

Later Research.uwin l9h6 a weighted index of sounds
I

 

was devised by Wood. His objective was to obtain a nu»
merical evaluation of articulatory ability. An Articulau
tion Index bases on the relative frequency of occurrence of
a given sound in the speech of children was devised. The
study was conducted with 50 children who were considered
functional articulatory cases. The Articulation Index
obtained was the sum of relative weights of the sounds
that the child could articulate correctly. This was done
equally for sounds appearing in initial, medial and final
positions.

Two years later Henrikson presented an analysis of
Woodie Articulation Index. His study questioned the
occurrence of sounds in the speech of children in equal

frequency for each position. Two of his conclusions

were:

 

1Kenneth S. Wood, "Parental Maladjustment and‘
Functional Articulatory Defects in Children," Journal of
Speech and Hearing Disorders, 11 (l9h6), pp. ESS~275.

 

I9

. . .Prorating consonant sounds on the
assumption that they occur equally or approximately
equally in all positions in a word is not justi-
fied and that using such a prorating as the basis
for constructing an index of progress is not Jus-
tified and the value of conclusions drawn from 1
using such an index is correspondingly questionable.
In defense of his Articulation Index Wood gives two

factors which he regards as necessitating equal prorating.

Initial and final consonants in words tend
to become medial when words are combined in the
phrase units which characterize speech. . .It
has been observed that a person who learns an
initial [t] has in part learned the medial and
final [t] even though he has not at a given point
developed the ability 0E habit of producing it in
the last two positions.

However, he does say that in a more exact numerical ex«
pression which will more nearly represent the social ade-
quacy of an individual": speech one cannot overlook the
fact that some consonant sounds occur more frequently in
the language than do others.

Just as phonemes are practically meaningless by them-
selves Wood maintains that merely counting the number of
defective phonemes has limited meaning. An auditory con-
cept and mastery of a sound is not indicated when the
sound is produced in isolation, syllables or words.

However, his study did indicate that an individual's abigity

 

1Ernest H. Henrikson, ”An Analysis of Wood's Articulation
Index," Journal of Speech and Hearing Disorders,, 13 (l9h8),
p. 235. .

2Kenneth S. Wood, "Measurement of Progress in the Correc~
tion of Articulatory Speech Defects," Journal of Speech and
Hearing Disorders, 1h (19h9), pp. 173, 17h.

 

 

 

 

20

to pronounce a sound correctly in an isolated word in re:
sponse to pictorial or object stimuli often adequately
represents his ability to produce the sound accurately in
connected speech. He concludes that the final test in
measuring articulatory progress is "whether or not the
sounds are consistently produced correctly in the stream
of spontaneous speech."l

Parobeck investigated the utility of the Wood.Articu-
lation Index in measuring articulation proficiency. .A
fifteen-point scale was divided into five groups of three
points each. In addition each group was given a verbal
description as follows: Group I a Normal speech, Group II a
Mildly defective speech, Group 111 a Moderately severe '
articulatory defect, Group IV a Severe articulatory de-
fect and Group V a Very severe articulatory defect.
Using picture stimuli Parobeck tested the articulation of
52 children. A three minute segment of each childls speech
was taped and judged. Parobeck also rated the speech
samples. The average correlation between Parobeck’s
rating of the speech samples and that of her three judges
was r = .92.2

Barker lists five criteria concerning the usefulness

of an measure of articulation. "(1) It should include a

 

Ibid., p. 171.

 

2Parobeck. OE. cit.

21

consideration of all Speech sounds. (2) It should represent
speech adequacy in a quantitative manner. (3) It should

be numerically accurate and allow for statistical manipum
lation. (h) It should be simple and convenient to use.

(5) It should be easily interpreted."1

Barker's study
investigated the correlation between an.Articulation Score
based on the relative frequency of occurrence of all speech
sounds with the subjective evaluations of an individual's
articulatory proficiency. The correlation between the A
judges' ratings and the.Articulation Score based on the
three consonant positions was r= .9h. The correlation

was r= .91; when only initial and final syllable positions
‘were used. She assumes that if thiscorrelation is high
enough the Articulation Score is a valid measure of articula-
tory proficiency.

In 1962 Barker and England followed up her earlier
study with some further developments. This study added
consonant blends and consonant and [r] sounds. .A
simplified form for faster computation was employed. A
comparison of the simplified form and the Judges ratings

indicated a Pearson product moment correlation coefficient

of .92.2

1Janet O'Neill Barker, "A Numerical Measure 0f Arti-
culation," Journal of Speech and Hearinngisorders, 25
(1960). p. 79.

2Janet Barker and Gene England, “A.Numerica1 Measure
of.Articulation: Further Developments,“ Journal of Speech
and Hearing Disorders, 27 (1962), pp. 23, 2h.

 

 

 

 

22

Scaling and Articulatory Defects.--Studies involving
scaling methods and procedures have been concerned with.
one or more of the following: equal appearing intervals,
successive intervals, randomization-of—segments and/or
direct magnitude estimation.

.A study by Sherman and Moodie placed particular em-
phasis on the method of equal appearing intervals. .A
nineupoint scale extending from one for least severe to
nine for most severe was employed. High fidelity tape re-
cordings five seconds in length of the continuous speech
of children between five and ten years of age were made
using the test items. The speech segments represented a
range of articulation from normal to severely defective.
On the basis of the results they concluded that the method
of equal appearing intervals is most useful for scaling
articulation defectiveness. The scale values obtained by
this method were said to be “reliable, relatively easy to
compute, and in close agreement with the internally con-
sistent scale values obtained by the method of successive
intervals.1

In 1960 several procedures for scaling articulation
were investigated by Sherman and Cullinan. The main

purposes of this study were:

 

 

1Dorothy Sherman and C. Moodie, "Four Psychological
Scaling Methods Applied to Articulation Defectiveness,"
Journal of Speech and Hearinngisorders, 22 (1957),

 

23

(a)To evaluate the reliability of mean scale
values of articulation defectiveness based upon
single observer consecutive ratings at ten second
intervals during one minute speech samples, (b)To
evaluate reliability of individual observer ratings
of articulation defectiveness based upon single
ratings of oneaminute speech samples and (c)To
evaluate how well the scale values obtained by
the above two procedures agree with each other
and with the scale values obtained by the rando-
mizationmofnsegments method.

They found that the intrawclass correlation coefficient for
the group who rated at consecutive intervals was .89 and
the intrauclass correlation coefficient for the group who
rated each sample as a whole was also .89. They also found
that for both procedure a and b measures based upon indivi-
dual observer responses were satisfactorily reliable.

Also in 1960 Prather investigated the scaling of de-
fective articulation by direct magnitude estimation. The
purpose was to study the psychological scaling method of
direct magnitude estimation for obtaining measures of
defectiveness of articulation along a ratio scale. The
test items consisted of twentymseven tape recorded five»
second segments taken from the speeches of children.

These speech segments ranged from normal articulation to
severely defective articulation. Scale values were ob-
tained from listener responses. These scale values were

then compared with one another and with corresponding sets

which had previously been obtained by the methods of equal

 

1Dorothy Sherman and w. Cullinan, "Several Procedures
for Scaling Articulation," Journal of Speech and Hearinngau
search, 3 (1960), pp. l9lwl98.

 

21+

appearing intervals, pair comparisons and constant sums.
The results indicated that there was close correspondence

between sets of scale values for all comparisons.1

Listener Rating and Articulatory Defects

 

Perrin sought to investigate whether there was any

a n
. , _'

difference in the ratings of severity given to functional
2 ‘ 1

articulation defects by trained and untrained observers.

 

Disc recordings were made of fifteen children reading a 1‘
prepared typewritten passage. These recordings were then F
paired (according to the procedure for paired comparia

sons) and presented to the groups of raters. The untrained

group consisted of twentyasix students enrolled in an inn

troductory psychology course. The trained group consisted

of thirteen graduate students enrolled in a course in

clinical methods. The rank order correlation obtained

between the trained and untrained ratings was .82. Both

trained and untrained observers agreed significantly with»

in their rankings, but untrained observers were slightly

more in agreement with themselves than trained observers.

The correlation coefficients between rank order obtained

from the actual count of articulation errors and the judges

 

IElizabeth Moodie Prather. "Scaling Defectiveness of
Articulation by Direct MagnitudewEstimaticn,” Journal of
§peech and Hearing Research. 3 (1960), pp..380=392.

 

 

O
“Perrin, op. cit.

25

ranking were .98 for clinicians and .82 for nonuclinicians.

Morrison recorded the conversational speech of 66
children between the ages of five and ten years. Using
both naive and sophisticated observers the tape recorded
speech samples were presented to them for rating along a
nineapoint scale of equal appearing intervals. Median
scale values and measures of strength (Kendall's Q) re-
lationship were obtained. Reliability coefficients were
.97 for five second segments and .98 for ten second seg-
ments. Ratings of expert listeners correlated well with
the ratings of naive listeners. The two major purposes
of Morrison's study were:

To investigate the reliability of measures of
articulation defectiveness obtained by the method
of equalwappearing intervals from responses of
groups of observers to short segments of continious
speech and to construct a severity scale of artia

culation defectiveness with recorded short segments
of continuous speech.

In 1955 Sherman and Morrison studied the reliability
of the Morrison Scale with the following conclusions:

(1) A trained individual observer can
reliably rate defective speech on a one to nine»
point scale by listening to a one minute sample
of the speech.

(2) Precise mean scale values of severity of
defective articulation can be obtained frmn the
responses of a trained individual observer.

(3) Absolute values of ratings are not
necessarily comparable from one observer to
another.

(u) Five second segments and ten second

1Morrison, op. cit.

 

 

26

segments of defective speech can be rated equally
reliably.

An analysis of the relationships existing between
some factors associated with defective articulation and
listener reaction was the concern of an investigation by
Jordan. The measures obtained from the test responses
were (1)number of defective speech sounds, (2) frequency
of these sounds in the language, (3)phonetic consistency
of the speech sound errors, (h)type of sound error, (5)posi~

tion and function of the misarticulated consonant sounds,

 

(6)phonetic category of misarticulated sounds and (7)sounds
misarticulated in blends. These measures were obtained from
150 children with articulatory deviations ranging from

mild to severe. After an evaluation by means of a multiple
regression analysis of the articulation test responses

and the measures of articulation defectiveness from listener
ratings of connected speech the results indicated that:

(1)Articulation rest responses, under the
conditions of this experiment, provide valid
information on articulatory behavior in connected
speech.

(2)Reactions of listeners to articulation
defectiveness are primarily dependent upon fre-
quency with which articulation deviations occur
and degree of articulation deviations.

(3)Articulation test measures of the
number of defective sounds and items are both

 

1Dorothy Sherman and Shelia Morrison, "Reliability
of Individual Ratings of Severity of Defective.Articulation,"
Journal of Speech and Hearing Disorders, 20 (1955), p. 356.

 

27

highly related to measures of defectiveness of

articulation derived from listener responses to
connected speech.

In this chapter an attempt has been made to pres

sent briefly several aspects of articulation testing

which

were thought to be relevant to the present study.

 

 

1

Jordan, op. cit., p. 319.

 

CHAPTER III

SUBJECTS, MATERIALS, EQUIPMENT AND PROCEDURES

Subjects

The subjects for this study consisted of twentynfive
children with functional articulatory defects as defined
in Chapter I. The sample was composed of 10 females and
15 males whose age range was from six to nine years. All
of the subjects were enrolled in public elementary schools
in Haslett, Michigan. Twentymone of the subjects were
taken from the first four grades of Central Elementary
School. In addition to these, four subjects from the

second grade at Wilkshire Elementary School were used.

The entire sample was composed of children who were currently

receiving speech therapy in their respective schools.

Materials and Equipment
The materials and equipment employed in this study

are listed below.

i. The Descriptive Articulation Test defined and
described in Chapter 1 was employed to obtain the raw

data.

2. A.model T1500 Wollensak tape recorder was utilized

in recording connected speech from the subjects.

28

. .F‘J

 

 

29

3. Twenty-one consonant picture cards containing
pictures with the test sounds in the medial position were
used in the Descriptive Articulation Test to elicit spon-
taneous responses from the subjects. ( See Appendix.A )

h. The samples of continuous speech of the subjects
was recorded on Burgess magnetic recording tape.

C
5. One dozen Saturday Evening Post Pictures were

. Fr
49“
4

available if needed to be used in eliciting a sample of

 

free speech from the subjects. wh
6. The responses of each subject were recorded on I
an Articulation Test Score Sheet. Twentyafive such sheets
were used. ( See Appendix C )
7. The judges indicated their ratings of the speech
samples on twentyufour rating sheets. ( See Appendices
D and E l
8. An Ampex tape recorder, model PRlO, was employed
in playing the speech samples through earphones to the

judges.

Procedures

 

Each child was given the Descriptive Articulation
Test which was composed of twenty—one consonant sounds.
The test was administered to each child individually.
The consonant sounds contained on the test were evoked by
means of pictures of objects which contained the sound

being tested in the medial position. This resulted in

 

30

an Articulation Score for each child. This score consisted
of the per cent of test sounds which were misarticulated
by the child.

The Stimulability Score was obtained in the followu
ing manner: For each sound misarticulated by picture
stimulation the child was asked to listen carefully as the
examiner repeated that particular sound three times in iso»
lation. The child watched the examiner and was then asked
to repeat the sound once. Each child‘s raw score consisted
of the number of misarticulated sounda by picture stimula=
tion which he also incorrectly articulated in isolation
after vocal Stimulation by the examiner.

‘After the administration of the Descriptive Articulaw
tion Test a sample of the childls connected speech was
elicited through conversation with the child and/or with
the use of the Saturday Evening Poet pictures. A.recording
of this speech sample was made on magnetic audio tape.

A magnetic audio tape containing a SOwsecond speech sample
of each subject was edited from the tape of each subject's
connected speech with a fivemsecond silent interval hem“
tween each recorded speech sample. The purpose of this
silent interval wag to allow the judges time to record their
rating of each speech sample.

The tape of continuous speech samples was then played
to a group of twelve judges who rated them according to the

severity of articulation. Theae judges were graduate atuu

 

dents in the Speech and Hearing Science area of the De-
partment of Speech at Michigan State university. They
rated the twentyafive SO-second samples of speech on two
occasions using a different type of rating sheet on each
occasion. At the first rating a nineapoint scale with the

number 1 representing least severe articulation and the

.j- ‘5}?!

number 9 representing most severe articulation was used.

The speech samples were rated on this continuum from least ' }

 

1"

severe to most severe. Each sheet contained the following y
directions which were read to the Judges:

You are being asked to rate twenty-five
SOusecond speech samples in regard to severity
of articulation only! You are to place your ra-
tinge along a nineupoint scale which you will find
below. One (1) represents least severe and nine (9)
represents most severe. Please indicate your rating
by refering to the scale and placing the number re-
presenting your rating beside the number of the
apprOpriate speech sample. Give only one number
for each speech sample. If you change your mind
completely erase or strike out all previous marks.

On the second occasion of rating the speech samples the
judge's rating sheet also employed a nine-point scale.

This scale was constructed in a different manner. On

this sheet the nineupoint scale was divided into three ra-
ting groups and each group was given a verbal description.
This was similar to the scale which was utilized in Parobeck's
study. The groups and descriptions are given below.

Rating Group Description Hating

-u‘auu- won—nun“.

 

 

1 Least severe

(44 R?- an:

39

 

 

Rating_Group Deecription Rating
II Moderately severe 5
6
7
III Most severe 8
9

The Judges were asked to place their rating of each speech
Sample beside the appropriate subject number. Only one
rating was given for each speech sample. The rating
sheets ueed in this experiment may be seen in Appendices

D and E.

Summarz.wawentyufive children with functional arti-
culation disorders were given the Descriptive Articulation
Test which was composed of 21 consonant sounds. In addi-
tion, a soasecond speech sample of each child was obtained
and recorded. An Articulation Score, Stimuiability Score
and Improvement Index was computed for each child.

The SOusecond sampiee of each_child?s connected
speech were rated on two separate Occaeions by twelve graduate
students in Speech and Hearing Science. The first rating
was done on a ninewpoint scale with the number one represent»
ing least severe and the number nine representing meet severe.
At the second rating the ninewpoint scale was divided into
three groups and each 'group wae given the descriptions
least severe. moderately severe'nnd most severe respectively.
The Articulation Score, Stimulability Score, Improvement Index
and mean rating for connected speech were recorded for each

of the twentymfive children.

CHAPTER IV

RESULTS AND DISCUSSION

Reeults

 

 

As stated in Chapter I the present study had two
purposes. They were, to devise a short, descriptive and
compreheneive articulation test which could be used in

describing the articulatory disorders of individuals and

 

to suggest a method which might give a meaningful indi-
cation of the childis improvement.

Judgee Ratingé.—~The twenty-five speech samples of

 

children with articulatory dieorders were rated by twelve
judges. The results of the ratings are given below in
Table I.

TABLE I

Mean. Range. Standard Deviation of Judge's Ratings

 

 

Rating Range Mean Standard Deviation
I lug 30'39 1090
II 1‘39 307? 2.10

At the first rating a nine-point Scale was used with
one representing least severe and nine representing most

severe. The same nineupoint ecale was used at the eecond

 

3h

rating; however, the nine points were organized into groups
of three. Each group was given a verbal description as
followss Group One a least severe; Group Two a moderately
severe; and Group Three 9 most severe.

Descriptive Articulation Teet Indices.=-The test em-

 

ployed in this study was designed to yield an Articulation
Score, a Stimulability Score and an Improvement index. The
formulae for calculating theee indices are presented in
Chapter I. The results of these indicea for each of the
twentymfive subjects are given below in Table 2.

TAbLE 2

DESCRIPTIVE ARTICULATICN TEST SCORE INDICES FOR SAMPLE

_._ ‘ _.-. 4.. ‘ __‘_:._H w I p.-

Subjeet .Artieulation Stimulability Improvement

 

 

 

Score Score Index
1 100 100 O
2 100 100 O
3 100 100 O
u 100 100 O
5 95 100 5
6 95 9% O
7 90 200 10
8 90 100 19
9 90 100 10
10 86 ‘ 100 EU
11 86 95 12
12 86 9o ' u
13 76 95 20
1h 76 95 2x
15 76 95 20
16 76 90 15
17 71 130 2,
18 71 95 24
19 67 95 ' 2?
20 62 100 3d
21 62 8 6
22 37 95 in
23 52 100 hb

35

Subject .Articulation Stimulability Improvement
Score Score Index
211 52 100 he
25 52 90 AZ

 

Four of the subjects tested receivethrticulation
Scores of 100 per cent. .Although receiving speech therapy
these individuals were working primarily on "carrymover."
The mean.Articulation Score obtained from the sample was
79.72 per cent while the standard deviation of the Articula=
tion Scores wan 15.90 per cent. The twentymfive subjects
yielded a mean Stimulability Score of 97.25 per cent with
a standard deviation of n.2u per cent. sThe mean Improve-
ment Index was 2u.08 per cent with a etandard deviation of
11.31 per cent.

Hypothesea.a=The following two hypotheses were tested

 

in this present study:

(1) There is a relationship between the number of
consonant aounde misarticulated and the judgement of the
severity of that articulatory disorder along an ungrouped
rating scale.

(2) There is a relationehip between the number of
consonant sounds misarticulated and the judgement of the
severity of that articulatory disorder along a grouped
rating scale.

To determine if a relationship exists between the
number of consonant sounds misarticulated on the test and

the judgement of the severity of the articulatory dieorder

36

a Pearson productamoment correlation was obtained for each
of the rating scales used. For the ungrouped nine-point
rating scale the correlation coefficient (r) was .88. _ A
The correlation coefficient (r) for the grouped nine—point
rating scale was .86.

.A test for the significance of these correlations was
made by testing the null hypotheses that each of the two
observed correlation coefficients did not differ from zero.
The respective F values computed for the correlations being
tested were 3.39 for the judgements of articulatory severity
by means of the ungrouped rating scale and 2.83 for the
judgements of articulatory severity by means of the grouped
rating scale. Both null hypotheses were therefore rejectm
ed at the .05 level of confidence. The rejection of the
null hypotheses indicates that some type of relationship
does exist between the number of misarticulated sounds and
the judgement of the severity of the articulatory disorder
by means of both the ungrouped rating scale and the grouped
rating scale. Figures 1 and 2 illustrate the relationship
between the Articulation scores obtained and the mean

severity rating for each subject for each of the two ratings.

Discussion

 

From the analysis of the data obtained in this study
it is shown that there is a significant relationship be-

tween Articulation Score and mean ratingnscale values at

 

 

E

UR

N SC

{IO

CULA'
“fl
(.3

‘.

ARTI

< J
O

F?

O\
C)

 

 

 

 

 

 

i‘ 5b
r= .88
o
0
O o
0
00 O
0 o
00
3 C)
O
O
O
o
I o O o
s
1 .2 3 u 5 6 7 8 9
Mean Severity Rating
Fig. 1.maRelationship between mean severity

ratings and the Articulation Score.

ARTICULATION SCORE

100

 

bu
CI)

 

Q) a:
O
r? .86
G
C O
'o
O O o
" C
C ()0
O
O
0<)
O
C o
._._,_ k,.. a l ‘ a “14 ‘-,_u.‘..__ 1 _l 1
1 2 3 24 5 6 7 8

Mean Severity Rating

Fig. Zen-Relationship between mean severity
ratings and the Articuiation Score.

39

the .05 level of conficence.
SqA

The formula 100 X —§—-was used to obtain the Improve-
ment Index for two reasons: (1) It would not give scores
below zero or above 100 and (2) it expresses a person's
improvement by auditory stimulation as a percentage of the
best articulation score he can get (namely that obtained
by auditory stimulation). It is thought that the Improve-
ment Index may yield an estimate of the prognosis for
speech therapy, but this would have to be demonstrated
through subsequent longitudinal research.

As stated in Chapter I the test sounds were administered
in the medial position only. This was done because in
connected speech all sounds become medial with the exception

of initial and final sounds. Wood maintains this point of

view in his article on Measurement of Progress in the Correc~

 

tion of Articulatory Speech Defects.

 

In accordance with Milisenls view that children are
able to improve their misarticulated speech sounds after
receiving stimulation; all subjects tested were able to
make improvement on one or more of their misarticulated
speech sounds after receiving voice stimulation. Nine
of the subjects tested achieved a Stimulability Score of

100 per cent.

 

1Wood, "Measurement of Progress," op. cit.

to

Figures 1 and 2 indicate the mean rating given by
the judges for each of the twenty-five subjects on the
ungrouped and grouped rating scales respectively. It
may be noted that with the grouped rating scale the judges
tended to rate the articulatory defect as being more severe
than when rating the same articulatory defect on an un-
grouped rating scale. It was not determined in this study
whether these slight differences in rating constituted a

real difference or were simply chance differences.

 

 

CHAPTER V

SUMMARY AND CONCLUSIONS

Summary
Speech defects of an Articulatory nature are very *3
Prevalent among public school p0pulations. Morethan fa,

half of the case load of the public school therapist con-

 

sists of children with articulatony defects. The majority -r{
of these children have functional articulatory defects.
Testing for articulatory defects therefore becomes impor-
tant to the therapist. .A review of the literature reveals
that various aspects of the problem of testing for articu~
latory defects have been studied including the using of
some system of numerical notation to measure the severity
of articulatory disorders.

In this study an attempt was made to devise a short
but descriptive and comprehensive articulation test for
use in articulation teating. A second purpose was to suggest
a method whereby a meaningful index of the child's improve»
ment might be obtained. The accomplishment of these pur-
poses would aid the therapist in selecting a case load,
noting the child's progress and in planning therapy for the
child.

The hypotheses tested in this study concerned the
hi

 

h2

relationship between the number of consonant sounds mis-
articulated and the judgement of the severity of that
articulatory disorder along an ungrouped rating scale and
along a grouped rating scale.

Twentyufive subjects from two public elementary
schools in Haslett, Michigan were utilized as subjects for
the study. These children possessed functional articula-
tory disorders and were all currently receiving speech
therapy.

The Descriptive.Articulation Test which was construc-
ted for this study was administered to each of the sub-
jects. The test consisted of twentyaone pictures with the
test sounds in the medial position. .A recording of each
'childis free speech was obtained and recorded on audio tape.
This tape was then edited and played to twelve judges who
rated the speech samples along an ungrouped ninempoint rating
[scale and along a grouped ninenpoint rating scale. Each of
the three groups in the grouped rating scale was given a
verbal description. The data gathered for each child con-
sisted of an Articulation Score, a Stimuliability Score

and an Improvement Index.

Conclusions

 

From the experimental design set forth in this study
and the data obtained the following conclusions have been

drawn:

us

1. There is a positive significant relationship at the
.05 level of confidence between the number of consonant
sounds misarticulated and the Judgement of severity of the
articulatory disorder using an ungrouped rating scale.

2. There is a positive significant relationship at
the .05 level of confidence between the number of consonant .
sounds misarticulated and the Judgement of severity of the Tim
articulatory disorder on a grouped rating scale.

A:

Implications for Further Research.--In this study 'aa

 

twenty-one of the most frequently misarticulated sounds
were tested. The present research design was not concerned
with the maturational appearance of speech sounds. A.study
utilizing a test in which the maturational appearance of
sounds is considered might possibly yield different and

more comprehensive results. Such a study might yield in-
formation which would be useful in testing the articulation
of children below the age of eight.

The taped speech samples used in this study consisted
of the subJect's free speech. A.study in which prepared
passages containing the test sounds would allow control
over the frequency of the test sounds appearing in the in-
dividual's speech sample and consequently might yield
different ratings by Judges.

As stated in Chapter IV further research in the form
of a longitudinal study might reveal that the Improveu

ment Index may also serve as an estimate of prognosis in therapy.

 

BIBLIOGRAPHY

Books

 

Anderson,‘Virgil.A. Improvingthe Child's Speech. New
York: Oxford university—Press, 1953.

 

Johnson, Wendell, Spencer J. Brown, James Curtis, Clarence
Edney and Jacqueline Keaster. :Speech Handicapped
School Children. 2nd rev. ed.7_New YErE: Harper
and Row, I935.

 

van Riper, Charles and John Irwin. Vbice and Articulation.
Englewood Cliffs, N. J. : PrenticeeHalI, Inc., 1956.

 

Articles and Periodicals

 

Barker, Janet O'Neill. 'A.Numericai Measure of Articulation,"
Journal of Speech and Hearing Disorders, 25 (1960),
pp. 79-88. a .

Barker, Janet O'Neill and Gene England. "A.Numerical
Measure of.Articulation: Further DeveIOpments," '
Journal of Speech and Hearing Disorders, 27 (1962),

pp. 23;27.

Carter, Eunice T. and McKenzie Buck. "Prognostic Testing
for Functional Articulation Disorders Among Children
in the First Grade,“ Journal of Speech and Hearing
Disorders, 23 (1958),.pp. léh-TEK. .

 

Chapman, Myfanwy E. et al. "Clinical Practice: Remedial
Procedures," Journal of Speech and Hearing Disorders,

Monograph Supplement 8 (1961), pp. 38-77.

Curry, Robert, Lou Kennedy, Loretta Wagner and waiter Wilke,
“A Phonographic Scale for the Measurement of Defective
Articulation," Journal of Speech and Hearing Disorders,
8 (19h3). pp. 123il26.

Farquhar, Mary Stuart. "Prognostic Value of Imitative and
.Auditory Discrimination Tests," Journal of Speech
and Hearing Disorders, 26 (1961), pp. SMEJKHV.

 

uh

MS

Henderson, Florence M. “Accuracy In Testing The Articula-
tion of Speech Sounds," Journal of Educational
Research, 31 (1939),

 

Henrikson, Ernest H. "An Analysis of Wood's.Articulation
Index," Journal of Speech and Hearing Disorders,

13 (19h87) PPo 233E235.

Jordan, Evan P. ”Articulation Test Measures and Listener
Ratings of Articulation Defectiveness," Journal of
Speech and Hearing Research, 3 (1960), PP. SOS-319.

 

Louttit, C.M. and E.C. Hall. "1936 Survey of Speech De-
fects Among Public School Children of Indiana,".
Journal of Speech Disorders, 1 (1936), pp. 73-80.

 

Milisen, Robert. "Methods of Evaluation and Diagnosis of
Speech Disorders," Handbook of Speech Pathology,
ed. Lee Edward Travis, New Vork: Appleton-Century
Crogts Inc., 1957, Pp. 267-309.

. "A Rationale for Articulation Disorders,"
JOurnal of Speech and Hearing Disorders, Monograph
Supplement *h (l95h), pp. SéiB. ‘

 

 

Mills, A. C. and Ho Streit. "19h2 Report of a Speech Survey-
Holyoke, Massachuetts," Journal of Speech Disorders,

7 (19h2). pp. 161m167.

Morrison, Shelia. "Measuring the Severity of Articulation
Defectiveness," Journal of Speech and Hearing Diem
orders, 20 (19557, pp. 3h7u351.

Perrin, Elinor Horwitz. "The Rating of Defective Speech by
Trained and Untrained Observers," .Journal of Speech
and Hearing Disorders, 19 (l95h), ppiwh8-31.

 

 

Powers, Margaret Hall. "Functional Disorders of Articula—
tion: Symptomatology and Etiology," Handbonk 0f
Speech Pathology, ed. Lee Edward Travis, New york:
AppletonuCentury Crofts Inc., 1957, pp. 707-768.

 

 

. Prather, Elizabeth Moodie. "Scaling Defectiveness cf Arti-
culation by Direct Magnitude-Estimation," Journal
of Speech and Hearing Research, 3 (1960), pp. 330-392.

 

Roe, Vivian and Robert Milisen. "The Effect of Maturation
Upon Defective Articulation In Elementary Grades,"
Journal of Speech and HearingDisorders, 7 (l9h2),

Pp. 37~50-

M6

Sayler, Helen K. "The Effect of Maturation Upon De-
fective.Articulation in Grades Seven through TWelve,"

Journal of Speech and Hearing_Disorders, 1h (19h9),
pf 202“207 o

Sherman, Dorothy and Catherine Moodie. "Four Psycholow
gical Scaling’Methods Applied to Articulation De-
fectiveness,’ Journal of Speech and Hearing Dis=
orders, 22 (1957), PP. 6964706.

and Walter Cullinan. "Several Procedures for
Scaling Articulation," Journal of Speech and Hearing
Research, 3 (1960), pp. 191-198.

and Shelia Morrison. "Reliability of Indivi-
dual Ratings of Severity of-Defective Articulation,"
Journal of Speech and Hearing Disorders, 20 (1955),

PPa 4332‘3580

Steer, M. D. and Hazel G. Drexler. "Predicting Later Articu-
lation.Ability From Kindergarten Tests," Journal
of Speech and Hearing Disorders, 25 (1960), pp. 391-397.

 

 

 

Templin, Mildred C. "Norms on a Screening Test of Articu-
lation for Ages 3:8," JOurnal of Speech and Hearipg

Disorders, 18 (1953), pp. 323-331.

. “A Non-Diagnostic Articulation Test," Journal
of Speech and Hearing Disorders, 12 (19h?), pp."3131396.

 

 

Wallin, J. E. Wallace. “A Census of Speech Defectives Among
89,057 Public School Pupils: A Preliminary Report,"
School and Society, 3 (1916),

Wood, Kenneth S. "Parental Maladjustment and Functional
’Articulatory Defects in Children," Journal of Speech
'and Hearing Disorders, 11 (l9h6 , pp. 255G275.

"Measurement of Progress in the Correction of
Articulatory Speech Defects," JOurnal of Speech and
HearingADisorders, 1h (l9h9), pp. l7l=f7h.

 

Unpublished Material

Parobeck, Donna J. "An Investigation of the Utility of
Wood's Articulation Index as an Independent Measure
of Articulation Proficiency," Unpublished M-A.
Thesis, Bowling Green State University, 1956.

APPENDI CE S

1+?

ha

APPENDIX.A

SOUNDS AND STIMULUS PICTURES USED IN THE
DESCRIPTIVE ARTICULAIION TEST

Sound Picture
1. [3] pencil
2. [z] scissors
3. [5] dishes
h. [ti] matches
5. [3] television
6. [d2] orange box
7. U%] birthday cake
8. [9] feather
9. [r] carrot
10. [1] balloons
11. [f] telephone
12. [v] seven
13. [k] turkey
1h. [9] wagon
15. [J] onions
16. [p] zipper
179 [b] baby
18. [m] hammer
19. [n] banana
20. [n] rings
21. [w] sandwich

#9

APPENDIX B

NUMBER OF SOUNDS MISARTICULATED BY PICTURE

STIMULATION AND JUDGES RATINGS

 

 

Number Misarticulated

Mean

Mean

 

Sounds First rating Second rating
10 8.00 8.h2
10 7.33 7.08
lO 6.33 7.h2

9 3.83 3.08
8 5.67 6.08
8 5.08 5.83
7 5.91 6.92
6 3.25 3.50
6 gal—L2 2075
S 3.82 3.75
S 3.08 3.08
5 3.66 h.08
S 3.92 h.83
3 2.h2 3.25
3 3.08 3.00
3 2.58 h.17
2 1.50 2.00
2 2.h2 2.33
2 2.75 3.17
1 1.1? 1.17
l 1.83 2.33
O 3.00 1.00
O 1.50 2.00
O 1.00 1.00
0 1.50 1.92

 

50
APPENDIX C

ARTICULATION TEST SCORE SHEET

Name
Sex Grade Age

 

 

 

Articulation Score Stimulability Score
Improvement Index

Sounds Tested

 

(m omission, sound substituted, 0 distortion, + addition)

 

 

 

 

 

 

picture voice picture voice
stimulation stimulation stimulation stimulation
1. [s] ____ 12. [v]
20 [z] ___‘___ 13. [k] __
3. [I] __ 111. [g] __
h. [3] 15. [31 __
S. ”5]“ 1.6. [p] ________
6. [d3]________ 17. [b] __ -..__._._..-
7. [9] ____ 18. [m] ____
a, [2,} __ 19. [n] __ _______,,,,,
9. [r] ______ 20. [n] __
10. [1] __ 2.1. [W] _._._._. .............
11. [r]
1. Number misarticulated by picture stimulation (P) =
2. Number misarticulated by voice stimulation (V) =
3. Articulation Score A 3 100 X §%%5
11. Stimulability Score. s = 100 K 2:5
5. Improvement Index I = 100 X iifl

5.)

51

.APPENDIX D

LISTENERUS RATING SHEET

Name Degree in progress

 

DIRECTIONS:

You are being asked to rnte twenty-five‘SO-second
speech samples in regard to severity of articulation
onlyl You are to place your ratings along a nine-point
scale which you will find below. One (1) represents
least severe and nine (9) represents most severe. 'Plense
indicate your rating by refering to the scale and placing
the number representing your rating beside the number of
the appropriate speech sample. Give only one number for
each speech sample. If you change your mind completely
erase or strike out all previous marks.

least most
severe severe
1 2 3 9

 

SPEECH SAMPLE
NUMBER

SPEECH SAMPLE
NUMBER

1. i3.
2. lu.
3. 15.
u. 16.
5. l7.
6. 18.
7. 19.
8. 20.
9. 2 .
10. 22.
ll. 23.
12. 2h.

52
.APPENDIX E

LISTENERUS RATING SHEET NO. 2

 

 

 

Name

Rating Group Description Rating

1

I Least severe 2

' 3

u

11 Moderately severe S

6

7

III Moet severe 8

9

Place the above rating values below in the apprOpriate epace
according to your rating of the epeech sample. Only one
rating for each sample.

SUBJECT NUMBER
1. 11. 21.

2. 12. 22.
3. 13. 23.
It. I, It. 214..
5. 15. 25.
6. . 16.
7. 17.
8. 18.
9.~ 19.

10. 20.

 

:ququ

7